JPH0837604A - Image processor - Google Patents

Abstract

PURPOSE:To improve the differences of image colors and definition between both images of an endoscope image recorder A and an observing monitor B by performing the color conversion through an image processing part in order to secure coincidence between the regenerated image of the recoder A and the displayed image of the monitor B and then correcting the feeling of sharpness. CONSTITUTION:An electronic endoscope 6 of an electronic endoscope device produces an image in an organism through an objective lens 17 by the illumination light of a light source part 7A led by a light guide 15. This image of the organism is converted into the image signals and outputted by a CCD 18, and these signals are processed by a signal processing part 7B and displayed in an image on an observing monitor 8. Furthermore the displayed image is processed by an image processor 3 ana then reproduced and recorded by an endoscope image recorder 4. Meanwhile the image processing part of the processor 3 performs the color conversion to secure coincidence between the image displayed on the monitor 8 and the image reproduced and recorded by the recorder 4 and corrects the image sharpness. Thus the differences of colors and sharpness are improved between both images of the monitor 8 and the recorder 4 respectively.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image for improving the image quality of an image reproduced by an image recording / reproducing apparatus when an image signal obtained by capturing an image of a subject by an image capturing means is recorded in the image recording / reproducing apparatus. Regarding a processing device.

[0002]

2. Description of the Related Art Conventionally, an electronic endoscope has been widely used which has a slender insertion portion to be inserted into a lumen, images a subject at the tip of the insertion portion, displays the subject image on a monitor, and observes and treats the subject. Has been.

There is an image recording device for recording an electronic endoscopic image picked up by this electronic endoscope, for example, on a video printer for recording the endoscopic image or on a CRT built in the image recording device. A film recorder or the like has been used that captures a reproduced endoscopic image and records it as a film image.

In the conventional example, while various image recording devices such as a video printer or a film recorder are used, an operator usually reproduces the image on a monitor connected to the electronic endoscope. Observing the endoscopic image.

[0005]

However, there is a difference between the color or sharpness of an endoscopic image normally observed by an operator and the color or sharpness reproduced by various image recording devices. Due to the difference in image reproduction process in various image recording devices, a difference in color (color difference) perceived by humans or a difference in sharpness occurs. Due to this color difference or difference in sharpness, an operator may feel a sense of discomfort in the image reproduced by the image recording apparatus.

Further, the above color difference is colorimetrically improved,
Even if the deterioration of the frequency transfer characteristic (hereinafter referred to as MTF) by the image recording device is improved, the color and sharpness desired by the operator can be measured by the memory of the operator for the endoscopic image reproduced on the monitor. In some cases, the color difference may be improved and the deterioration of the MTF may be improved. In such a case,
The image quality is adjusted to obtain the desired color or sharpness by the operator, but the plurality of video signals input to the image recording device are color spaces dependent on the electronic endoscope and represent human perceived colors. Since it is not a perceptual color space and the image reproduction process of the image recording apparatus is complicated, it is not easy for an operator to reproduce a desired color and sharpness by adjusting a plurality of video signals.

Furthermore, particularly in an image such as an endoscopic image, which is characterized by a color distribution and in which a subtle change in color tone needs to be reproduced, the color distribution of the image to be corrected is intensively focused. High-precision correction and subtle adjustment are required. In recent years, a general-purpose device that corrects color differences has been proposed, but it is a device that improves the image quality on a wide range of colors, such as landscapes and human images, on average, and has the high accuracy required for endoscopic images. It is hard to say that the correction and the fine adjustment are sufficient.

The present invention has been made in view of the above circumstances, and determines the difference between the color of an image reproduced on a monitor which inputs a video signal from an image pickup device and the color of an image reproduced by an image recording device. An object of the present invention is to provide an image processing apparatus which improves the difference in sharpness of an image reproduced colorimetrically or reproduced on a monitor and the sharpness of an image reproduced by an image recording apparatus in terms of MTF difference. .

Further, the image quality targeted by the operator varies from the result of colorimetrically improving the color difference or correcting the sharpness depending on the memory, the observation conditions, etc. of the operator with respect to the image reproduced on the monitor. Even in such a case, it is an object of the present invention to provide an image processing apparatus that enables an operator to reproduce a desired color or a desired sharpness.

[0010]

The image processing apparatus of the present invention comprises an image pickup means for picking up an image of a subject and outputting an image signal, and an image recording means for recording the image signal output by the image pickup means on a recording medium. And an image display unit for displaying an image of the subject by inputting an image signal output from the image pickup unit, wherein the image display unit displays the image signal based on the image signal output from the image pickup unit. Image signal processing means for processing the image signal output from the image pickup means based on image processing data that equalizes the image quality of the displayed image and the image obtained from the image recording means is provided. Image processing means for equalizing the image quality of the image displayed on the image display means based on the image signal output from the imaging means and the image obtained from the image recording means. By processing the image signal output from the image pickup means based on the data, the color of the image reproduced on the image display means for inputting the video signal from the image pickup means and the color of the image reproduced by the image recording means. Is improved colorimetrically, or the difference between the sharpness of the image reproduced on the display means and the sharpness of the image reproduced by the image recording means is MT.
It is possible to improve the difference in F, and the image quality targeted by the operator is improved colorimetrically by the operator's memory of the image reproduced on the image display means, or Even when the sharpness varies from the corrected result, it is possible to allow the operator to reproduce the target preferable color or the desired sharpness.

[0011]

Embodiments of the present invention will be described below with reference to the drawings.

1 to 11 relate to a first embodiment of the present invention, FIG. 1 is a block diagram showing the configuration of an endoscope system including an image processing apparatus of the first embodiment, and FIG. 2 is a diagram of FIG. FIG. 3 is a block diagram showing a detailed configuration of the endoscope system, FIG. 3 is a configuration diagram showing a configuration of the image processing apparatus of FIG. 1, and FIG. 4 is a configuration of software of an image processing unit of the central processing unit of FIG. FIG. 5 is a software configuration diagram, FIG. 5 is a software configuration diagram showing the software configuration of the color conversion unit of FIG. 4, FIG. 6 is a flowchart explaining the processing flow of the color conversion unit of FIG. 4, and FIG. FIG. 8 is a conceptual diagram illustrating the color distribution of a general endoscopic image in the CIE1976 (L ^* u ^* v ^* ) color space used in the color conversion unit. FIG. 8 is software of the sharpness correction unit in FIG. FIG. 9 is a software configuration diagram showing the configuration of FIG. Figure Legend for explaining the concept of FIG. 10 is a flow chart, FIG. 11 is a flowchart for explaining a processing procedure sharpness correcting unit of FIG. 4 is an explanatory view illustrating a sharpness correction by sharpness correction unit of FIG.

As shown in FIG. 1, the endoscope system 1 includes
An electronic endoscope apparatus 2 (imaging unit) including an imaging unit that images a subject, and an image processing apparatus 3 (image signal that performs color conversion processing and / or sharpness correction processing on a captured endoscopic image. Processing means) and an endoscopic image recording device 4 (image recording / reproducing means) for recording an endoscopic image.

The electronic endoscope apparatus 2 is a signal for performing signal processing on the electronic endoscope 6 which is an image pickup means for picking up an image of a subject, a light source section 7A for supplying illumination light to the electronic endoscope 6 and the image pickup means. The observation device 7 has a built-in processing unit 7B and an observation monitor 8 (image display means) for displaying an image signal output from the observation device 7.

The electronic endoscope 6 has an elongated insertion part 11 to be inserted into a living body 9, an operation part 12 formed at a rear end of the insertion part 11, and a universal part extended from the operation part 12. The observation device 7 includes a connector 14 that is composed of a cable 13 and is provided at the end of the universal cable 13.
Can be connected to.

A light guide 15 is inserted into the insertion portion 11 (see an enlarged view of the tip of the electronic endoscope 6 in a circle in FIG. 1), and a connector 14 is connected to the observation device 7 to thereby form a light source portion 7A. Illuminating light is supplied to the incident end surface. Then, the illumination light is transmitted by the light guide 15 and is emitted forward from the end face of the electronic endoscope 6 on the side of the distal end portion 16 to illuminate the target site in the living body 9. The illuminated target portion is imaged by the objective lens 17 provided on the tip portion 16 on the CCD 18 arranged at the image forming position, and photoelectrically converted. The objective lens 17 and the CCD 18 form an image pickup section 19 as an image pickup means.

The image signal photoelectrically converted by the CCD 18 is subjected to signal processing by the signal processing section 7B in the observation device 7 to generate an image signal, which is output to the observation monitor 8 and the image. The image output to the processing device 3 and image-processed by the image processing device 3 is recorded in the endoscope image recording device 4.

Next, the structures of the light source section 7A and the signal processing section 7B in the observation device 7 will be described.

As shown in FIG. 2, the light source section 7A is provided with a lamp 21 that emits light in a wide band from ultraviolet light to infrared light. A general xenon lamp, a strobe lamp, or the like can be used as the lamp 21, and the xenon lamp, the strobe lamp, or the like emits not only visible light but also a large amount of ultraviolet light and infrared light.

Power is supplied to the lamp 21 from a power source 22 so that the lamp 21 emits light. A rotary filter 24, which is rotationally driven by a motor 23, is arranged in front of the lamp 21. Filters that transmit light in the respective wavelength regions of red (R), green (G), and blue (B) for normal observation are arranged on the rotary filter 24 along the circumferential direction.

The rotation of the motor 23 is controlled by a motor driver 25 and is driven. The light that has passed through the rotary filter 24 and is time-sequentially separated into lights in the R, G, and B wavelength regions is further incident on the incident end of the light guide 15, and the light is guided through the light guide 15 to the inside of the electron. The endoscope 6 is guided to the emission end face on the side of the distal end portion 16 and emitted forward from this emission end face to illuminate an observation site or the like.

The return light from the subject (subject) such as the observation site due to the illumination light is transmitted by the objective lens 17 to the CC
An image is formed on D18 and photoelectrically converted. A drive pulse from a driver 31 in the signal processing unit 7B is applied to the CCD 18 via a signal line 26,
An electric signal (video signal) corresponding to the image of the subject photoelectrically converted by the drive pulse is read out.

The electric signal read from the CCD 18 is input to the preamplifier 32 provided in the signal processing section 7B via the signal line 27. still,
The preamplifier 32 may be provided in the electronic endoscope 6 and configured.

In the signal processing unit 7B, the video signal amplified by the preamplifier 32 is input to the process circuit 33 to be γ.
Signal processing such as correction and white balance is performed, and A /
It is adapted to be converted into a digital signal by the D converter 34.

This digital video signal is, for example, red (R), green (G), blue (B) by the selection circuit 35.
Are selectively stored in the three memories (1) 36a, memory (2) 36b, and memory (3) 36c corresponding to the respective colors. R stored in the memory (1) 36a, the memory (2) 36b, and the memory (3) 36c,
The G and B color signals are simultaneously read out, converted into analog signals by the D / A converter 37, and output to the observation monitor 8 as R, G and B color signals via the input / output interface 38, and this observation is performed. The observation part is displayed in color by the monitor 8. In the following, the expression “interface” is referred to as “I / F”.

A timing generator 39 is provided in the signal processing unit 7B to generate the timing of the entire system, and the timing generator 39 synchronizes the above-mentioned circuits such as the motor driver 25, the driver 31, and the selection circuit 35. Has been taken.

In this embodiment, the output terminals of the memory (1) 36a, the memory (2) 36b, the memory (3) 36c and the sync signal output terminal of the timing generator 39 are connected to the image processing apparatus 3.

Next, the structure of the image processing apparatus 3 will be described.
As shown in FIG. 3, the image processing apparatus 3 includes a ROM 45 that stores programs and the like, and a central processing unit 40 that executes processing in accordance with the programs stored in the ROM 45.
And an information input device 41 for inputting data and the like to the central processing unit 40, and a main storage device 42 composed of a RAM for storing data and the like by the processing of the central processing unit 40.
An image input I / F 43 for inputting R, G, B image data from the signal processing unit 7B, an external storage device 44 for storing data image-processed by the image processing device 3, and an endoscope image recorded. And an image recording device I / F 46 which is connected to the endoscope image recording device 4 for inputting / outputting data, and these devices are connected by a bus 48.

The information input device 41 is composed of a keyboard and the like, and can input data such as the type of the electronic endoscope 6 and the like. The image input I / F 43 includes a memory (1) 36a, a memory (2) 36b, and a memory (2) 36b in the image processing unit 7B.
It is connected to the memory (3) 36c, is synchronized by the timing generator 39, and receives these image data.

The central processing unit 40 has an image processing unit 47 which operates according to a program stored in the ROM 45,
As shown in FIG. 4, the image processing unit 47 displays the color of the image reproduced by the endoscopic image recording device 4 on the observation monitor 8.
The color conversion unit 50 that performs color conversion so as to match the color of the image displayed on the display, and / or the sharpness correction unit 51 that performs the sharpness correction process of the image reproduced by the endoscopic image recording apparatus 4.
Etc.

In FIG. 4, the color conversion unit 50 is located in front of the sharpness correction unit 51, but this position can be reversed.

By the way, the external storage device 44 is characterized in that the storage capacity of the external storage device 44 is larger than that of the main storage device 42, but the input / output speed of the data is slow. Therefore, the external storage device 44 stores a large amount of parameter data or the like necessary for color conversion processing, sharpness correction processing, etc., and is read out only when necessary, or color conversion processing or sharpness correction is performed. It is used to store a large amount of data generated by correction processing and the like. The external storage device 44 is composed of, for example, a hard disk, a magneto-optical disk, a flash memory, etc., and the main storage device 42 is composed of a semiconductor memory such as RAM.

Next, the image processing section 47 of the central processing unit 40.
The color conversion unit 50 in FIG.

In FIG. 3, in the image processing device 3, the image data input via the image input I / F 43 is temporarily stored in the main storage device 42. In addition, after reading the RGB data forming the image data from the main storage device 42,
Further, the color conversion result is transferred to the central processing unit 40, the color conversion unit 50 is executed by the central processing unit, the color conversion result is temporarily stored in the main storage device 42, and then is transferred to the image recording device 4 via the image recording device I / F 46. Is output.

When the color conversion processing is executed by the central processing unit 40 of FIG. 3, the software configuration of the color conversion unit 50 corresponding to the program stored in the ROM 45 is RGB as shown in FIG. Data R, G, B are converted to CIE19
76 (L ^* u ^* v ^* ) color space coordinate conversion unit 60 for converting and outputting to L ^* , u ^* , v ^* color space coordinates, and color space coordinate conversion unit 60
The CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* output from the CIE 1976 (3 * 3) matrix operation are executed to C.
IE1976 (L ^* u ^* v ^* ) color space coordinates L ^* ', u ^* ',
Matrix operation unit 61 that outputs v ^* ', and CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* ', u ^* ', v ^* ' standardized RGB data output from the matrix operation unit 61 R ',
The color space coordinate reverse conversion unit 62 for converting into G ′ and B ′.

The RGB data will be described below as being standardized to 8 bits, but the standardization number is not limited to 8 bits, and even if the standardization number is more than 8 bits,
Or it may be small.

Then, the color space coordinate conversion unit 60 of the color conversion unit 50 expresses the inputted RGB data R, G, B as a color perceived on the observation monitor 8 by CIE.
Convert to 1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* .

More specifically, in the color space coordinate conversion section 60, as shown in FIG. 6, the RGB data R, G,
Conversion from B to CIE (XYZ) tristimulus values X, Y, Z is performed, and subsequently, in step S2, CIE (XYZ) tristimulus values X, Y, Z to CIE1976 (L ^* u ^* v ^* ) color space coordinates. Convert to L ^* , u ^* , v ^* . Step S at this time
The conversion formula in 1 is represented by formula (1) (Reference 1: The University of Tokyo Press, Color Science Handbook, page 936) and step S2.
The conversion formula in (1) is shown in Formula (2) (Reference 1: The University of Tokyo Press, Color Science Handbook, pages 139 to 140).

[0039]

[Equation 1]

[Equation 2] The matrix elements in the equation (1) are coefficients when the xy chromaticity coordinates of the image-receiving three primary colors defined by NTSC are used, and if the image-receiving three primary colors change, the matrix elements are determined again. Then, considering the γ characteristic of the monitor, RGB
It is also possible to correct the data and apply the corrected RGB data to the equation (1).

Next, the processing shifts from the color space coordinate conversion unit 60 to the matrix calculation unit 61, and as shown in step S3 of FIG. 6, the matrix calculation by the 3 × 3 matrix M in the CIE1976 uniform color space. CIE1976 (L ^* u
^* v ^* ) color space coordinates L ^* , u ^* , v ^* are converted to CIE1976 (L ^*
u ^* v ^* ) Convert to color space coordinates L ^* ', u ^* ', v ^* '. The conversion formula at this time is shown in Formula (3).

[0041]

(Equation 3) Then, the CIE1976 output from the matrix calculation unit 61.
The (L ^* u ^* v ^* ) color space coordinates L ^* ', u ^* ', v ^* 'are converted into RGB data R'through the color space coordinate inverse conversion unit 62 through steps S4 and S5 of FIG. , G ', B'. That is, in step S4, the inverse transformation of step S2 is performed, and the CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* ', u ^* ', v ^* 'are converted to CIE (XYZ) tristimulus values X',
Conversion into Y'and Z'is performed. Then, in step S5, the inverse conversion of step S1 is performed, and CIE (XYZ)
RGB data R ', G', B'from tristimulus values X ', Y', Z '.
Is converted to.

Then, in step S6, it is determined whether the RGB data R ', G', B'is within a predetermined range (for example, 0 to 255 when the standardization number is 8 bits).
If the RGB data R ′, G ′, B ′ are within a predetermined range,
The RGB data R ′, G ′, B ′ is output as it is, and the processing is ended. Further, when the RGB data R ′, G ′, B ′ is out of the predetermined range, the process proceeds to step S7 and is rounded off in step S7, or the standardization process such as rounding down or rounding up at the lower and upper limits of the standardized range is performed. Is done, and finally,
The normalized RGB data R ′, G ′, B ′ is output from the color space coordinate reverse conversion unit 62, and the process is terminated.

Next, a method of determining the matrix element (mij) of the matrix M in the equation (3) will be described with reference to FIG.

FIG. 7 shows the RGB data R ′, G ′, B ′, the color of the image displayed on the observation monitor 8 and the same RGB data R ′, G ′, B ′ in the image recording device 4. The relationship between the colors of the reproduced image that is input and reproduced is CIE1976.
(L ^* u ^* v ^* ) color space coordinates.

In FIG. 7, point P (L ^* ', u ^* ', v ^* ')
Is a CIE1976 (L ^* u ^* v ^* ) color space in which RGB data R ′, G ′, B ′ is expressed as a color perceived when the displayed color is observed by inputting it to the observation monitor 8. Coordinates. Further, the point Q (L ^* , u ^* , v ^* ) is a color that is perceived when the reproduced reproduction image is observed by inputting the RGB data R ′, G ′, B ′ to the image recording device 4. , CIE1
976 (L ^* u ^* v ^* ) color space coordinates. Also, point Q
In order to display (L ^* , u ^* , v ^* ) on the observation monitor 8, RGB data that must be input to the observation monitor 8 are R, G, and B.

Here, the meaning represented by the Euclidean distance between two points in the CIE1976 (L ^* u ^* v ^* ) color space will be briefly described.

The CIE1976 (L ^* u ^* v ^* ) color space is also called a uniform perceptual color space, and the CIE1976 (L ^*
u ^* v ^* ) Euclidean distance between two points in the color space is
It corresponds to the color difference perceived by humans. That is, the same RGB data R ′, G ′, and B ′ are output to the observation monitor 8 and the image recording device 4.
Even if it is input to, a color difference corresponding to the Euclidean distance between the points P and Q occurs.

Therefore, in other words, the relationship between the points P and Q is paraphrased. The CIE1976 (L ^*) which represents the colors displayed by inputting the RGB data R, G, B to the observation monitor 8 ^.
u ^* v ^* ) In order to reproduce the color space coordinate Q on the reproduced image, RGB data R ′, G ′, B ′ for displaying the color represented by the point P on the observation monitor 8 are used. Image recording device 4
You just have to type in. Therefore, CIE19
In order to reproduce the color displayed on the observation monitor 8 which is represented by the 76 (L ^* u ^* v ^* ) color space coordinate Q on the reproduced image, CIE1976 (L ^* u ^* v ^{*). *} ) RGB for predicting the color space coordinates P and displaying the CIE1976 (L ^* u ^* v ^* ) color space coordinates P on the observation monitor 8.
This RGB data R ', converted into data R', G ', B'
It suffices to input G ′ and B ′ to the image recording device 4.

In order to predict the point P from the point Q, the CIE1976 (L ^* u ^*) represented by the relationship between the point Q and the point P is used ^.
v ^* ) A large number of color space coordinate data over the entire color space are prepared in advance as data 1 and data 2, and the prediction coefficient from data 1 to data 2 is obtained and determined by multiple regression analysis, which is a known technique. The matrix operation may be performed using the created matrix. At this time, the data 1 and the data 2 respectively correspond to the explanatory variable and the objective variable in the multiple regression analysis.

The mapping relationship between the data 1 and the data 2 depends on the difference between the color reproduction process of the image recording apparatus 4 including the film development process and the color reproduction process of the observation monitor 8 and the like.
In the 76 (L ^* u ^* v ^* ) color space, it is generally represented by a nonlinear mapping model. Therefore, for the multiple regression analysis, a non-linear multiple regression model including terms of the second or higher order is generally used.

However, a large number of general endoscopic images (eg, endoscopic images not coated with a stain) are prepared as samples, and CIE1976 (L ^*
When the frequency distribution in the u ^* v ^* ) color space is examined, it is found that they are concentrated within a certain range, as shown in FIG. 7 as a color distribution of a general endoscopic image. CIE1 like this
Within a limited range within the 976 (L ^* u ^* v ^* ) color space, it is possible to approximate the mapping relationship for data 1 and data 2 as a linear one. Therefore, as the multiple regression model in the multiple regression analysis, a linear multiple regression model using only the primary term can be used.

Next, a method of creating data 1 and data 2 corresponding to the explanatory variable and the objective variable in the multiple regression analysis will be described.

A color chip of a large number of reproduced images including a color distribution of a general endoscopic image is prepared, and C is measured by a colorimeter or the like.
The IE1976 (L ^* u ^* v ^* ) color space coordinates are measured, and data 1 is created based on this data. On the other hand, from the RGB data input to the image recording device 4 when creating the color chart, CIE1976 (L ^* u ^* v) on the observation monitor 8 is used.
^* ) The color space coordinates are calculated according to equations (1) and (2) and used as data 2.

Using the created data 1 and data 2,
By the multiple regression analysis based on the linear multiple regression model, the matrix M
And the matrix calculation unit 61 performs matrix calculation using the calculated matrix M. Then, a matrix calculation unit 61 that performs a matrix calculation using the matrix M determined by the multiple regression analysis.
Will work in the sense of minimum color difference.

Next, the image processing section 47 of the central processing unit 40.
The sharpness correction unit 51 in FIG.

The sharpness correction processing is performed by the central processing unit 40 of FIG.
In the case of executing the above, the software configuration of the sharpness correction unit 51 corresponding to the program stored in the ROM 45 is the same as that of the input original image data of R, G, B as shown in FIG. A filtering execution unit 65 that applies filtering to the data conversion unit 66 and a data conversion unit 66 that adjusts the value of the processing result.

The sharpness correction processing of the apparatus for displaying an endoscopic image on the CRT screen and taking a picture or a slide will be described below. Here, the value of each pixel in the image is R (x, y), G (x, y) and B (x, y).

In the image photographing apparatus, the frequency transfer characteristic (Modulation Transfer Function: hereinafter referred to as MTF) is desired to be 1.0 in all frequency bands as shown in FIG. 9 (a). (This is hereinafter referred to as an ideal MTF). On the other hand, in a device (hereinafter, referred to as a photo-taking device) for taking a photograph / slide of an original image displayed on a CRT screen, its MTF is generally as shown in FIG.
As shown in (b). Such deterioration of the MTF causes a lack of sharpness in image quality.

Therefore, the sharpness correction unit 51 corrects the MTF of the original image by using filtering to perform a process for obtaining an image more suitable for observation.

First, a method for correcting the MTF shown in FIG. 9B to the ideal MTF shown in FIG. 9A will be described. Here, a known technique, inverse filtering, is applied in the spatial frequency domain.

The original image is f (x, y), and M of the photographic device is
Letting h (x, y) be a point spread function (hereinafter referred to as PSF) that gives TF, the obtained captured image g (x, y) is g (x, y) = f (x, y) * H (x, y) ... (4). * Represents a convolution operation. It is also assumed here that the PSF is position invariant. When this is expressed as a calculation in the spatial frequency domain, it becomes G (u, v) = F (u, v) · H (u, v) (5). Where F (u, v), H (u, v) and G
(U, v) denote the Fourier transform of f (x, y), h (x, y) and g (x, y), respectively. In addition, H (u,
v) is equivalent to the MTF shown in FIG. Multiplying both sides of Expression (5) by 1 / H (u, v), G ′ (u, v) = G (u, v) / H (u, v) = (F (u, v) · H (U, v)) / H (u, v) = F (u, v) (6) is obtained, which coincides with the photographing result by the photographing apparatus having the MTF of FIG. 9A. That is, a filter having a spatial frequency characteristic opposite to that of the PSF in the imaging system of the photographic device (hereinafter referred to as a correction filter) M (u, v) = 1 / H (u, v) (7) is previously set. F, which is the Fourier transform of the original image f (x, y)
(U, v) is multiplied, and the processing result F ′ (u, v) = F (u, v) · M (u, v) (8) is inverse-Fourier-transformed. By photographing y), it becomes possible to realize a photographed image by the ideal MTF.

The PSF of the photographing device may be derived in advance from the result of photographing a point image or a line image.

Therefore, the filtering execution unit 65 shown in FIG. 8 executes the filtering in the spatial frequency domain for each RGB original image data.

The process flow will be described with reference to FIG. As shown in FIG. 10, in the filtering execution unit 65, the respective RGB original image data are input,
The two-dimensional Fourier transform in step S11 is applied. This corresponds to the conversion from f (x, y) in the equation (4) to F (u, v) in the equation (5). continue,
Filtering is performed in step S12.

The coefficient values of the filter used in step S12 are stored in advance in the ROM 45 in FIG. 3 or the external storage device 41. Here, the ideal MTF is corrected by correcting the MTF in FIG. 9B. In order to realize the above, a correction filter M (u, v) having a frequency transfer characteristic shown in FIG. 11A (corresponding to a so-called inverse filter) is used. The coefficient value of the filter is determined so as to satisfy the equation (7) in each frequency band.

H (u, v) = 0 (9) That is, when there is a frequency band that is completely lost in the captured image, for example, M (u, v) = 1.0 (10) ) Exception processing such as The processing in step S12 corresponds to the above-mentioned expression (8) and generates the filtering processing result F ′ (u, v).

Subsequently, an inverse Fourier transform is applied to F '(u, v) in step S13 to obtain a photographing original image f' (x, y).

The processing in steps S11 to S13 described above is performed by the original image r (x, y), g of each RGB data.
(X, y) and b (x, y), and PSFs hr (u, v), hg (u, v), and hb (u, v) that are PSFs for the respective data, and are used as originals for photographing. Image R ',
Obtain G'and B '.

Then, the data converter 66 is supplied with RGB signals.
Original images R ', G', and B'for shooting in each data are input. The data conversion unit 66 executes post-processing for converting data having a value exceeding the display range by filtering. That is, the data below the decimal point is rounded off, or the value less than 0 or 256 or more when the displayable range is, for example, 0 to 255 is rounded down. The respective photographed images R ', G'and B'after applying this processing are output as processing results.

As described above, according to the electronic endoscope system of the present embodiment, the RGB data R, G, B input in the color space coordinate conversion section 60 are displayed on the observation monitor 8. CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* representing a reproduced color are converted into CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u by the matrix calculation unit 61. ^* , V ^*
CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* ′,
CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* ', u ^* ', which are converted into u ^* ', v ^* ' and output from the matrix calculation means 61.
v ^* 'is converted into RGB data R', G ', B'by the color space coordinate inverse conversion unit 62, and this RGB data R', G ', B'is replaced with RGB data R, G, B'. To
By inputting into the image recording device 4, the observation monitor 8
The color difference between the color reproduced in and the color of the reproduced image is CIE197.
It is improved in the sense of minimum color difference in the 6 (L ^* u ^* v ^* ) color space.

In addition, CIE1976 (L ^*
u ^* v ^* ) By utilizing the characteristic that the color distribution in the color space is localized, the matrix calculation in the matrix calculation unit 61 is performed by 3
It becomes possible to execute by small-scale calculation of × 3, and it becomes possible to reduce the calculation cost.

In this embodiment, CIE1976
(L ^* u ^* v ^* ) color space is used, but CIE1976
Instead of the (L ^* a ^* b ^* ) color space or the color space recommended by CIE, for example, reference 2 ((publishing) ADDISION W
ESLEY, INTERRACTVE COMPUTERGRAPHICS, BURGER, GILLIES
HS composed of three attributes of color, that is, the three axes of hue, saturation, and brightness, as described on pages 333 to 338).
Even if a perceptual color space such as V or HLS color space is used,
The color difference between the color reproduced on the observation monitor 8 and the reproduced image is improved.

In addition, in the sharpness correction unit 51, as shown in FIG.
The processing in steps S11 to S13 shown in FIG.
Original images r (x, y), g (x, y) and b (x, y) of each GB data, and hr (x, y), hg (x, y) and hb which are PSFs for the respective data. (X, y)
Since the original images for photographing R ′, G ′ and B ′ have been obtained by applying the same to the above, the sharpness of the endoscopic image reproduced on the observation monitor 8 and the endoscopic image storage device 4 reproduce it. It is possible to improve the difference between the sharpness of the endoscopic image and the MTF difference.

When a noise component is mixed in the original image f (x, y), the noise component may be emphasized when the ideal inverse filter is applied. In this case, since the noise component is generally included in the high frequency band, the spatial frequency characteristic of the applied filter is
For example, the change is made as shown in FIG. By controlling the degree of correction in the high frequency band, it becomes possible to correct the MTF in the frequency band practically required while preventing unnecessary noise components from being emphasized.

In the sharpness correction section 51, the MTF of the photographic device is an ideal MTF or an MT close to it.
Although the method of correcting the sharpness of the captured image by setting it to F has been described, here, by changing the filter coefficient,
A method for obtaining an image suitable for observation by controlling the MTF more positively will be described.

In general, the low frequency component constitutes the brightness information of the entire image, and the frequency component affecting the sharpness of the image exists in a high frequency band to some extent. Further, in the higher frequency band, the noise component is dominant over the image information component.
Therefore, the spatial frequency characteristic of the filter used in the filtering execution unit 65 (denoted as M ′ (u, v)) is
In the low frequency band, the ideal MTF is dominant, and in the high frequency band that affects the sharpness, the noise component is dominant such that H (u, v) · M ′ (u, v) ≧ 1.0 (11) In the high frequency band, H (u, v) · M ′ (u, v) ≦ 1.0 (12) is set. FIG. 11C is an example of the spatial frequency characteristic having such a property. As a result of applying the filter having the spatial frequency characteristic of FIG. 11C to the MTF shown in FIG. 9B, the MTF obtained is as shown in FIG.
As shown in (d), the sensed image is more sharpened.

M '(u, v) is the PSF due to the difference in the camera and film used in the above-mentioned photographic device.
In addition to the change, the increase / decrease of the frequency band and the spatial frequency characteristic is adjusted.

Further, if the original image is, for example, an image picked up by an endoscope, M '(u, v) is controlled by information such as the number of pixels of the solid-state image pickup element based on the model of the endoscope used. Good. Since the frequency bands to which the equations (11) and (12) are applied are changed based on the difference in resolution, a more preferable captured image can be obtained.

As a modification of the filtering execution unit 65, a FIR filter which is a digital filter described below can be used. That is, the filtering execution unit 65 executes filtering by the FIR filter on each of the RGB original image data.

First, the outline of the principle of filtering will be described. In the filtering by the FIR filter, the arithmetic processing corresponding to the equation (6) or the equation (8) in the above filtering in the spatial frequency domain is performed in the spatial domain.

From the image convolution theorem, the equation (8) can be expressed as f '(x, y) = f (x, y) * m (x, y) (13). * Represents a convolution operation. That is, the original image f ′ (x, y) for photographing is the original image f (x, y) and the inverse Fourier transform m (x, y) of the correction filter M (u, v).
It is realized by the convolution operation with and. m (x, y) corresponds to the coefficient value of the FIR filter in this embodiment. The actual filtering by the FIR filter is a mask operation of mask size n1 × n2 (n1 and n2 are positive numbers that are not 0). This is from m (x, y)
This corresponds to a cut-out window of a size capable of reproducing the spatial frequency characteristic of M (u, v) with practically sufficient accuracy. The filtering execution unit 65 receives the RGB original image data, and performs filtering corresponding to Expression (13).

Then, the data converter 66 is supplied with RGB signals.
Original images R ', G', and B'for shooting in each data are input. The data conversion unit 66 executes post-processing for converting data having a value exceeding the display range by filtering. That is, the data below the decimal point is rounded off, or the value less than 0 or 256 or more when the displayable range is, for example, 0 to 255 is rounded down. The respective photographed images R ', G'and B'after applying this processing are output as processing results.

When setting the coefficient of the FIR filter applied in the modification of the filtering execution unit 65, for example, when correcting the MTF of the photographic device to the ideal MTF, the correction filter Mr (u, v), Mg (u,
v) and Mb (u, v) are the respective inverse Fourier transforms mr (x, y), mg (x, y) and mb (x,
y) is sent to the filtering execution unit 65 as the coefficient of the FIR filter.

When a noise component is mixed in the original image f (x, y), the noise component may be emphasized when the ideal inverse filter is applied. In this case, since the noise component is generally included in the high frequency band, the spatial frequency characteristic of the applied FIR filter is changed, for example, as shown in FIG. By controlling the degree of correction in the high frequency band, it becomes possible to correct the MTF in the frequency band practically required while preventing unnecessary noise components from being emphasized.

Further, when the statistical property of noise is known, a Wiener filter or a least squares filter may be applied instead of the inverse filter.

By changing the coefficient of the FIR filter and controlling the MTF more positively, an image suitable for observation can be obtained.

In general, the low frequency component constitutes the light and dark information of the entire image, and the frequency component affecting the sharpness of the image exists in a high frequency band to some extent. Further, in the higher frequency band, the noise component is dominant over the image information component.
Therefore, the F used in the filtering execution unit 65
The spatial frequency characteristic of the IR filter (M ′ (u, v) is expressed by the above equation (11) in the low frequency band and in the high frequency band which affects the sharpness,
In the high frequency band where the noise component is dominant, it is set so as to become the expression (12). That is, the FIR filter having the spatial frequency characteristic shown in FIG. 11C may be used.
As a result of applying the FIR filter having the spatial frequency characteristic of FIG. 11C to the MTF shown in FIG. 9B,
The obtained MTF is as shown in FIG. 11 (d) in the same manner as described above, and the photographed image has a sharpened feeling.

M '(u, v) is the PSF due to the difference in the camera and film used in the above-mentioned photographic device.
In addition to the change, the increase / decrease of the frequency band and the spatial frequency characteristic is adjusted.

Further, if the original image is, for example, an image picked up by an endoscope, M '(u, v) is controlled based on information such as the number of pixels of the solid-state image pickup device based on the model of the endoscope used. Good. Since the frequency bands to which the equations (11) and (12) are applied are changed based on the difference in resolution, a more preferable captured image can be obtained.

Therefore, similar effects can be obtained also in this modification.

Next, the second embodiment will be described. 12
1 and FIG. 13 relate to the second embodiment, and FIG. 12 is an explanatory diagram for explaining the sharpness correction by the sharpness correction unit of the second embodiment.
FIG. 3 is a software configuration diagram showing the software configuration of the sharpness correction unit that performs the sharpness correction of FIG. Since the second embodiment is different from the first embodiment only in the processing configuration of the sharpness correction unit 51, only the different points will be described.

The sharpness correction unit 51 of the second embodiment suppresses the deterioration of the SN ratio of the picked-up image caused by the correction processing of the sharpness, and the sharpness correction of the second embodiment will be described below. The section 51 will be described.

It is assumed that the mixing of the noise component into the image f (x, y) is uncorrelated with f (x, y) and is additive. That is, the original image without noise is f0
If (x, y) and noise are ν (x, y), then f (x, y) = f0 (x, y) + ν (x, y) (14) The noise component is mainly distributed in the high frequency band, and is often mixed in the frequency band that gives the sharpness of the image. In this case, the enhancement of sharpness causes the enhancement of noise. If the SN ratio in the original image is high enough, this problem can be ignored,
Due to the property of additive noise, the S / N ratio differs depending on the bright part (bright part) and the dark part (dark part) in the image. Therefore, even if the amount of noise that can be ignored in the bright portion of the image is increased, it may be noticeable in the dark portion, which may rather deteriorate the image quality of the captured image.

On the other hand, the region where sharpness is desired in the image is generally a region having a certain degree of brightness.

Therefore, in the sharpness correction unit 51 of this embodiment,
The bright portion of the image is sufficiently corrected for the sharpness, and the dark portion is processed so as to obtain a photographed image with less noise by suppressing the degree of the correction.

First, the concept of processing in the sharpness correction unit 51 will be described.

In this embodiment, a sharpness correction component image, which will be described in detail later, is weighted and added to the original image, and the weighting coefficient value is controlled based on the information regarding the brightness of the image.

Let f (x, y) be the original image, fs (x, y) be the sharpness correction component image, and w be the weighting coefficient value. The sharpness correction component is the MTF of the photographic device with respect to the ideal MTF.
12 is a frequency band component lost due to the MTF of FIG. 9B described in the first embodiment.
The image information corresponding to the shaded area in (a) is shown. or,
As for the MTF shown in FIG. 11C, FIG.
The image information corresponding to the shaded area in (d) is shown.

The sharpness correction component image can be easily obtained by filtering. In the first embodiment, the correction filter M
Although the creation of (u, v) has been described, in the present embodiment, bandpass filtering based on M (u, v) is applied to the original image. That is, when M (u, v) shown in FIG. 11A is taken as an example, the spatial frequency characteristic of the bandpass filter Mp (u, v) used is shown in FIG. 12B.
It will be as shown in. These are obtained by setting Mp (u, v) = M (u, v) -1.0 (15).

The sharpness correction component image f which is the processing result of the bandpass filtering using the created Mp (u, v).
If f '(x, y) is a processing result image obtained by weighting and adding s (x, y) to the original image f (x, y), f'
(X, y) is expressed as f ′ (x, y) = f (x, y) + w · fs (x, y) (16). In the equation (16), if w = 1.0, then f ′ (x, y) is an image for obtaining the ideal MTF described in the sharpness correction unit 65 of the first embodiment and its modification. It matches the original image f '(x, y). Also, w <1.
A value of 0 results in suppressing the sharpness correction effect.
Therefore, by changing w for each pixel based on the information on the brightness of the image, it becomes possible to control the sharpness correction effect in the bright portion and the dark portion in the image.

FIG. 12C shows an example of setting the weighting coefficient value w. Here, it is assumed that the luminance of each pixel represented by the following equation is used as the information component regarding the brightness.

Y = 0.3R + 0.59G + 0.11B (17) In FIG. 12C, the horizontal axis represents luminance Y (brightness information),
The vertical axis represents the weighting coefficient w, the threshold value for Y is defined as T, w <1.0 is set for Y <T, and w = 1.
By setting it to 0, it is possible to suppress an increase in the noise feeling in the dark part. Such a method of giving w is, for example, w = (1/2) × {1-cos (Y · π / T)} (0 ≦ Y <T) w = 1.0, where w is a function of Y. (T ≦ Y) ... (18). Formulas (17) and (18)
Is applied for each pixel. Further, as the information about the brightness, not the luminance Y but image data of any of RGB may be used.

Next, the sharpness correction unit 51 in this embodiment.
The software configuration of will be described.

As shown in FIG. 13, the sharpness correction unit 51 of the second embodiment synthesizes the filtering unit 71 for filtering input RGB data and the original image and the filtering result image. Performing synthesis unit 72
And a data conversion unit 73 for adjusting the processing result output value.

In the filtering unit 71, the sharpness correction component images Rs (x, y), Gs (x, y) and B are applied to the RGB data in the input original image.
Apply filtering to generate s (x, y). Specific means for implementing the filtering are filtering in the spatial frequency domain described in the first embodiment and filtering by the FIR filter which is the digital filter described in the modification of the sharpness correction unit of the first embodiment. Either of these may be used.

The weight coefficient value w is set on the basis of the information on the brightness of the image, and here, the value of the G pixel is used as the information on the brightness. Regarding the setting of w, for example, the function w = (1/2) × {1-cos (G (x, y) · π / T)} (0 ≦ G (x, y) <T) w = 1.0 (T ≦ G (x, y)) (19) may be used.

In the synthesizing unit 72, R in the original image
A calculation R '(x, using the respective GB data, the sharpness correction component images Rs (x, y), Gs (x, y) and Bs (x, y), and the weighting factor value w output from the filtering unit 71. y) = R (x, y) + w · Rs (x, y) (20) G ′ (x, y) = G (x, y) + w · Gs (x, y) (21) B ′ ( x, y) = B (x, y) + w · Bs (x, y) (22) is performed.

In the data conversion section 73, data in which R '(x, y), G' (x, y) and B '(x, y) input for each exceeds the display range. Perform post-processing to convert the. That is, rounding off the data below the decimal point, and the displayable range is, for example, 0 to 255.
If the value is less than 0 or a value of 256 or more is truncated.

Other configurations and operations are the same as those in the first embodiment.

As described above, according to the second embodiment, in addition to the effect of the first embodiment, by performing the arithmetic processing using the weighting coefficient value w in the synthesizing unit 72, the sharpness correction processing is accompanied. It is possible to prevent deterioration of the SN ratio of the captured image.

The filtering unit 7 in the present embodiment.
The filter coefficient applied in 1 is, for example, band-pass filters Mpr (u, v), Mpg (u, v) and Mp if filtering in the spatial frequency domain is used.
If digital film filtering is used for each value in b (u, v), filter coefficients mpr (x, y), mpg giving characteristics of each correction filter
(X, y) and mpb (x, y).

If the original image is, for example, an image picked up by an endoscope, Mp '(u, v) or m depending on information such as the number of pixels of the solid-state image pickup element based on the model of the endoscope used.
You may control p '(x, y). Since the frequency bands to which the equations (11) and (12) are applied are changed based on the difference in resolution, a more preferable captured image can be obtained.

In the present embodiment, the sharpness correction component image is generated by bandpass filtering. However, the correction filter M (u, It is also possible to achieve the same effect by using the processing result image according to v).

That is, the sharpness correction component image fs which is the processing result of the original image f (x, y) and the correction filter.
In the weighting calculation with the weighting coefficient value w with (x, y), the above equation (16) is changed to f ′ (x, y) = (1.0−w) · f (x, y) + w Fs (x, y) ... (23), and the weighting coefficient value w may be controlled by the information component indicating the brightness of the image. The method of controlling and setting the weighting factor value w may be realized by the method described above.

In the present embodiment, the weighting coefficient value w is controlled for each pixel of the image. However, for example, the image is divided into a plurality of blocks and each block of the information component representing the brightness is divided into blocks. You may control by the average value in. In this case, the same weighting factor value w is applied to each pixel in one block.

Next, a third embodiment will be described. 14
15 and FIG. 15 relate to the third embodiment, FIG. 14 is a software configuration diagram showing the software configuration of the color conversion unit, and FIG. 15 is a flowchart explaining the processing flow of the color conversion unit of FIG. Since the third embodiment is almost the same as the first embodiment, only different configurations will be described, the same configurations will be denoted by the same reference numerals, and description thereof will be omitted.

The main difference between this embodiment and the first embodiment is that
The color space coordinate inverse conversion unit 62 of the color conversion unit 50 of the first embodiment.
This is because an abnormal value processing unit is provided in the subsequent stage (see FIG. 5).

As described in the first embodiment, the matrix M is determined so as to be the target of color conversion for the color of the color distribution of the general endoscopic image shown in FIG. The endoscopic image may include points different from the color distribution of a general endoscopic image. As a result of performing color conversion processing on such a color, RGB data (abnormal value) exceeding the standardized range of RGB data may be output, and an extreme color change may occur. In particular, when the halation or the like is converted into a red color as a result of an extreme color change, the reproduced image reproduced will give a considerable discomfort. It is not desirable that the impression of the reproduced image is impaired due to color change such as halation.

Therefore, in the present embodiment, with a simple configuration, a color that is not the target of color conversion is determined, and for colors that are not the target of color conversion, the original RGB data R, G, B are output. , Prevent extreme color changes.

As shown in FIG. 14, the color conversion unit 50 of this embodiment has the RGB output from the color space coordinate inverse conversion unit 62.
It is determined whether or not the data R ′, G ′, B ′ is within the standardized range, and if it is within the standardized range, the RGB data R ′,
G ', B'is output, and if it is out of the standardized range, RGB
An abnormal value processing unit 81 for outputting the data R, G, B is provided. The other structure is the same as that of the first embodiment.

Next, the operation of this embodiment thus constructed will be described focusing on the abnormal value processing section 81.

As shown in FIG. 15, the processing in the color conversion section 50 is performed in steps S1 to S1 described in FIG. 6 of the first embodiment.
The description up to S5 is omitted because it is the same. Then, the R output from the color space coordinate inverse conversion unit 62 in step S5.
It is determined whether or not the GB data R ′, G ′, B ′ is within the predetermined standardized range (for example, 0 to 255 when standardized to 8 bits) in step S21. If the determination is made and if it is out of the range, in step S22, the RGB data R ′, G ′, B ′ are RGB.
After being replaced with the data R, G, B, the abnormal value processing unit 81 eventually outputs the original RGB data R, G, B.

On the other hand, when the RGB data R ', G', B'output from the color space coordinate inverse conversion unit 62 is within the standard range, the RGB data R ', G', B'is output as it is. .

The other operations are the same as in the first embodiment.

As described above, according to this embodiment, in addition to the effect of the first embodiment, it is possible to improve the color difference for the color to be color-converted by the color conversion processing, and as shown in FIG. By providing the value processing unit 81 in the subsequent stage of the color space coordinate inverse conversion unit 62 and inputting the RGB data R, G, B to the abnormal value processing unit 81, the RGB data R ′ for the color that is not the target of color conversion. , G ′, B ′ are discriminated by determining whether they are within the standardized range, and the original RGB data R, G, B are identified.
By outputting, it is possible to prevent an extreme color change with a simple configuration.

Next, a fourth embodiment will be described. FIG.
11 is a flowchart illustrating a processing flow of a color conversion unit according to the fourth embodiment. The fourth embodiment is almost the same as the first embodiment, and the difference between this embodiment and the first embodiment is the first embodiment.
In the matrix calculation unit 61 of the color conversion unit 50 of the embodiment, 3 × 3
Instead of executing the matrix operation of, the matrix operation of, for example, 3 × 10 including the second-order and higher terms is executed.
Since other configurations and operations are the same as those in the first embodiment,
Only the operation of the matrix calculation unit 61 will be described, and the same configurations will be denoted by the same reference numerals and description thereof will be omitted.

As described in the first embodiment, CIE1976 (L ^* u ^* v ^* ) for general endoscopic images.
The color distribution in the color space tends to exist within a limited range as shown in the color distribution of the general endoscopic image in FIG. 7. However, depending on the situation, an image in which a stain such as methylene blue is sprayed may be input,
It is clear that such an image shows a tendency different from the color distribution described above.

Therefore, in order to determine the matrix elements in the matrix calculating means 101 that can be adapted to a specific situation such as a general endoscopic image and an image in which a stain is scattered, 3
The prediction error may increase in the linear multiple regression model of × 3.

Therefore, the color conversion unit 50 of the present embodiment also handles the case where an image having a color distribution other than the color space occupied by a general endoscopic image is input, such as the case where a stain is scattered. Yes, improve the color difference.

That is, referring to FIG. 5 and using FIG. 16, the operation of the matrix calculation unit 61 of the present embodiment will be described. As shown in FIG. 16, the output is from the color space coordinate conversion unit 60. The CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* are the quadratic term in step S31,
Alternatively, values of the third order or higher are created. Subsequently, in step S32, the matrix calculation shown in the following equation (24) is executed, and the CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* ', u ^* ', v ^* 'are output. It

[0131]

[Equation 4] Note that the equation (24) shows an example up to the quadratic term including the constant term.

The matrix M in the matrix calculating means 101 is determined using a non-linear multiple regression model in which multiple regression models include higher-order terms of second order or higher. Other functions are the same as those in the first embodiment.

As described above, according to the present embodiment, in addition to the effect of the first embodiment, a non-linear multiple regression model based on higher-order terms of the second or higher order is used for matrix determination, so It is possible to obtain a matrix M corresponding to a specific situation such as an endoscopic image and an image in which a stain is scattered, and a matrix calculation unit 61 performs a high-order matrix calculation using this matrix M. The color difference between the color reproduced on the observation monitor 8 and the color of the reproduced image is CIE1976 (L ^* u
^* v ^* ) In the meaning of the minimum color difference in the color space, it is possible to improve the color difference corresponding to a specific situation such as a general endoscopic image and an image coated with a stain.

Next, a fifth embodiment will be described. FIG. 17
21 to 21 relate to the fifth embodiment, FIG. 17 is a software configuration diagram showing the software configuration of the color conversion unit, and FIG.
Conceptual diagram illustrating the relationships of the observation monitor gamut and the image recording apparatus gamut applied to the color conversion unit of FIG. 17 u ^* v ^* plane,
FIG. 19 is a flowchart for explaining the processing flow of the color conversion unit of FIG. 17, and FIG. 20 is a configuration of the gamut data of the endoscope image storage device applied to the color conversion unit of FIG. 17 and the hue (H) saturation ( C) Lightness (V) means CIE1976 (L ^* u
^* v ^* ) A conceptual diagram described in the color space, and FIG. 21 is a software configuration diagram showing a configuration of a modified example of the color gamut compression unit in FIG.

The fifth embodiment is almost the same as the first embodiment, and the difference from the first embodiment is that the color space in the color conversion section 50 of the present embodiment is as shown in FIG. Converter 6
The color gamut compression unit 86 is added between 0 and the matrix calculation unit 61. In the present embodiment, the configuration and operation other than the color gamut compression unit 86 are the same as in the first embodiment, so the operation of the color gamut compression unit 86 will be mainly described.

The range of colors that can be reproduced on the observation monitor 8 (hereinafter, observation color gamut) and the range of colors that can be reproduced on the image reproduced by the image recording device 4 (hereinafter, image recording device color gamut) are CIE1976. (L ^* u ^* v ^*) of color space L ^*
FIG. 18 schematically shows on a cutting plane (hereinafter, u ^* v ^* plane) in FIG.

As shown in FIG. 18, when the image recording device gamut is narrower than the observation monitor gamut, among the colors displayed on the observation monitor 8, the colors located outside the image recording device gamut are In the image recording device 4, the CIE1976 (L ^*
u ^* v ^* ) cannot be expressed in color space coordinates.

Then, for such data, the first
When the color conversion processing described in the embodiment is executed, the matrix calculation unit 6
Since the data outside the color gamut of the image recording apparatus is not used as the explanatory variable for determining the matrix M in 1, a remarkable color reproduction error may occur.

Therefore, in this embodiment, the observation monitor color gamut is compressed so that the observation monitor color gamut is included in the image recording apparatus color gamut, and an appropriate color reproduction is performed in the color gamut of the image recording apparatus 4. The color difference is improved for the entire input endoscopic image.

The operation of the fifth embodiment will be described below centering on the color gamut compression section 86.

The color gamut compression method between devices having different color gamuts such as the observation monitor 8 and the image recording device 4 described in the present invention is roughly divided into the color gamut outside the image recording device. There is a case where the coordinate conversion is performed only for the color existing in, and a case where the coordinate conversion is performed for all the colors. Further, in each case, there is a case where the coordinate conversion is performed in the saturation direction without changing the hue and the lightness, a case where the coordinate conversion is performed in the lightness and the saturation direction while keeping only the hue, and the coordinate The conversion method also includes linear conversion or non-linear conversion, and various methods based on combinations thereof have been proposed.

Hereinafter, in the description of the operation of the color gamut compression unit 86,
As an example, a method of performing coordinate conversion in the saturation direction only with respect to a color outside the color gamut of the image recording apparatus without changing its hue and lightness will be described.

As shown in FIG. 19, in the color gamut compression unit 86, the CIE 19 output from the color space coordinate conversion means 60 is output.
The 76 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* are converted into H, C, and V by polar coordinate conversion in step S41. The conversion formula is shown in formula (25).

[0144]

(Equation 5) Polar H, C, as shown in FIG. 20, respectively, u ^* v ^* rotation angle from at u ^* axis in the plane, and u ^* v ^* corresponding to the Euclidean distance from the at the origin in the plane. The polar coordinate H is a metric hugh angle (Reference 1: The University of Tokyo Press,
Color Science Handbook, pages 142-143), which is equivalent to the hue among the three attributes of color, and has polar coordinates C.
Is called Metric Chroma (Reference 1: The University of Tokyo Press, Color Science Handbook, pages 142-143) and corresponds to the saturation among the three attributes of color. V is called metric lightness, and V = L ^* , which corresponds to the lightness of the three color attributes.

The color gamut data of the image recording device input from step S45 in FIG. 14 represents the color gamut data of the image recording device 4, and the main memory device 42,
Alternatively, it is read from the external storage device 44.

[0146] The image recording apparatus gamut data, as shown in FIG. 20, L ^* (V) for each quantized u ^* v ^* plane in the direction of quantized u ^* v at θ intervals in the H direction ^* It is composed of data of Euclidean distance Cg (H, V) from the origin on the plane. This Cg (H, V) becomes data representing the color gamut of the image recording device 4 on each u ^* v ^* plane.

H, C, V output from step S41
And the image recording device color gamut data from step S45, in step S42, the color space coordinate conversion unit 6
It is determined whether or not the CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* output from 1 are included in the color gamut of the image recording apparatus 4 or not included.

Judgment is made based on H output from step S41,
The color gamut data of the image recording device 4 in step S45 is referred to by V.

Then, the referred color gamut data Cg (H,
V) and C output from step S41 are compared to obtain C
If g (H, V) <C, the CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* output from the color space coordinate conversion unit 60 are the colors of the image recording device 4. It will exist outside the area, and the saturation will be converted.

If Cg (H, V) ≧ C, CIE1976 (L ^* u ^*) output from the color space coordinate conversion unit 60 ^.
v ^* ) color space coordinates L ^* , u ^* , v ^* are present in the color gamut of the image recording device 4, and CIE1976 (L ^* u ^* v
^* ) Color space coordinates L ^* , u ^* , v ^* are directly L ^* , u ^* , v
It is output from the color gamut compression unit 86 as ^* .

In step S 42, the CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* output from the color space coordinate conversion unit 60 are outside the color gamut of the image recording device 4. If determined, in step S43, polar coordinates C
Is converted to C '.

The conversion method is the gamut data S1 of the image recording apparatus.
Let Cg be the Cg (H, V) referenced from 5. That is, all the colors existing outside the color gamut of the image recording device 4 move to the outermost part of the color gamut of the image recording device in the saturation (C) direction while preserving the hue (H) and the lightness (V). Equivalent to that.

In step S44, the CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u with respect to the polar coordinates H, C, V.
Inverse conversion into ^* , v ^* is performed, and CIE 19
76 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* are output.

The other operations are the same as those in the first embodiment.

As described above, according to this embodiment, in addition to the effects of the first embodiment, in the color gamut compression unit 86, the CIE197 is used.
6 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* are determined to be included in the image recording device color gamut, and for data not included in the image recording device color gamut, the hue, By changing the saturation so that it moves to the outside of the color gamut of the image recording device while keeping the lightness, it is possible to prevent a significant color reproduction error as a result of executing color conversion for the data outside the color gamut of the image recording device. However, it is possible to obtain appropriate color reproduction as a whole in the color gamut of the image recording apparatus.

It should be noted that another color gamut compression method is applied to the color gamut compression unit 8.
Even when used in No. 6, it is possible to prevent the occurrence of a significant color reproduction error as a result of performing color conversion on the data outside the color gamut of the image recording device, and obtain an appropriate color reproduction as a whole within the color gamut of the image recording device. it can.

Further, since each color gamut compression method has a peculiar property, a color gamut compression section 86a as a modification may be constructed as shown in FIG. That is, as shown in FIG. 21, the user selects the color gamut compression method by the information input device 41 such as a keyboard according to the situation,
The selected color gamut compression method is input to the switching unit 92 via the parameter input unit 91, and the switching unit 92 uses, for example, three first to third compression units 93, 94 having different compression methods depending on the input parameters. It is also possible to configure so as to select the compression unit that executes the compression method selected by the operator from 95.

As a result, it is possible to prevent a remarkable color reproduction error from occurring by performing the color conversion on the data outside the color gamut of the image recording device, and to compress the color gamut within the color gamut of the image recording device as a whole according to the purpose of the operator. Appropriate color reproduction can be obtained by switching the method.

Note that FIG. 21 shows a case where three color gamut compression methods are switched as an example, but the number of color gamut compression methods is not limited to three.

Next, a sixth embodiment will be described. FIG.
23 and FIG. 23 relate to the sixth embodiment, FIG. 22 is a software configuration diagram showing the configuration of the matrix operation unit of the color conversion unit, and FIG. 23 is the CIE1976 (L ^* u ^*) applied to the matrix operation unit of FIG.
v ^* ) is a conceptual diagram illustrating a plurality of divided areas in a color space.

The sixth embodiment is almost the same as the first embodiment, and the difference of the present embodiment from the first embodiment is the configuration of the matrix operation 61 of the color conversion section 50, as shown in FIG. The matrix calculation unit 61 of the present embodiment includes a plurality of first to third matrix calculation units 101, 102, and 103 that perform matrix calculation in parallel, and a plurality of first to third matrix calculation units provided in parallel. A plurality of results from 101, 102, and 103 are added to a combining unit 105 that combines and outputs one result by a weighted linear sum operation using W with the weighting coefficient set by the weighting coefficient setting unit 104. . Therefore, in this embodiment, the first to third matrix operation units 101, 102, 1
03, the weighting factor setting unit 104, and the synthesizing unit 105 will be described focusing on the configuration and action.

Matrix M used in the matrix calculation unit 61
The use of multiple regression analysis, which is a well-known technique to determine the, is described in the first example. In this multiple regression analysis,
If there is a strong non-linear characteristic in the relationship between the data corresponding to the explanatory variable and the data corresponding to the objective variable, it is possible that a considerable amount of prediction error occurs, and this prediction error is caused by the color difference in the color conversion device. It is one of the main factors that determine the accuracy of improvement. Therefore, in multiple regression analysis, it is conceivable to use a nonlinear multiple regression model using higher-order terms in order to make more accurate predictions, but as the number of elements of matrix M increases The calculation cost increases.

Therefore, in this embodiment, CIE1976
By dividing the (L ^* u ^* v ^* ) color space into a plurality of regions and increasing the linearity of the data pair corresponding to the explanatory variable and the objective variable in each region, accuracy is maintained, The calculation cost of the color conversion unit 50 is suppressed.

As shown in FIG. 22, the matrix calculation unit 61 of the fifth embodiment outputs the C output from the color space coordinate conversion means 60.
IE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , and v ^* are input, and first to third matrix operation units 10 perform matrix operation.
1, 102, 103, a weighting factor setting unit 104 for setting the weighting factor W, and first to third matrix computing units 101, 10
2 and 103 output a plurality of CIE1976 (L ^*
u ^* v ^* ) The color space coordinates are input, and the weighting factor setting unit 104
The CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* ′, u ^* ′, v ^* ′ are output by performing a weighted linear sum operation using the weighting coefficient W output from And is configured. The other structure is the same as that of the first embodiment.

In the description of this embodiment, three matrix calculation units are set, but the number is not limited to three, and three or more or three or less may be set.

Hereinafter, the operation of this embodiment will be described focusing on the first to third matrix computing units 101, 102, 103, the weighting coefficient setting unit 104, and the combining unit 105.

CI output from the color space coordinate conversion unit 60
E1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* are
It is input to the first to third matrix calculation units 101, 102 and 103, and matrix calculation is executed. First to third matrix calculation unit 10
1, 102, 103 from CIE1976 (L ^* u
^* v ^* ) color space coordinates L ^* 1, u ^* 1, v ^* 1 and L ^* 2, u ^* 2
, V ^* 2 and L ^* 3, u ^* 3, v ^* 3 are output.

Next, the weighting factor setting unit 104 sets and outputs the weighting factor W. CIE1976 (L ^* u ^*
v ^* ) color space coordinates L ^* 1, u ^* 1, v ^* 1 and L ^* 2, u ^* 2,
The v ^* 2, L ^* 3, u ^* 3, v ^* 3 and the weighting factor W are input to the synthesizing unit 105, calculation based on the equation (26) is performed, and output to the color space coordinate inverse transforming unit 102. It

[0169]

(Equation 6) Subsequently, in the weighting factor setting unit 104, the weighting factor W =
A method of setting [w1, w2, w3] will be described with reference to FIG.

FIG. 23 shows the CIE1976 (L ^* u ^* v ^* ).
In the u ^* v ^* plane formed by cutting with L ^* which is a color space, it represents a region divided into a plurality of regions. Actually, C
Since the regions are divided in the IE1976 (L ^* u ^* v ^* ) color space, each region becomes a solid. Here, for the sake of simplicity, the description will be made on the u ^* v ^* plane. In addition, in FIG. 22, three first to third matrix operation units 101, 102, and 10 are provided.
Since 3 is set, FIG. 23 is also divided into three regions, a region 1, a region 2, a region 3, and a boundary region where each region is in contact with the boundary region.

The weighting factor setting unit 104 outputs the CIE1976 (L ^* u ^* v ^* ) output from the color space coordinate conversion unit 60.
The color space coordinates L ^* , u ^* , v ^* serve as addresses, and the weighting coefficient W set for each area is referred to. The weighting factor w set uniquely in each area is set to w1 = 1.0 and w2 = w3 = 0.0 in the area 1, and w2 = 1.0 and w1 in the area 2. = W3 = 0.0
, And in region 3, w3 = 1.0,
It is set to w1 = w2 = 0.0.

First to third matrix operation units 101, 102, 1
The matrix in 03 is determined by using the data existing in each of the divided areas. Therefore, the CIE1976 (L ^* u ^*) output from the color space coordinate conversion unit 60 is determined ^.
v ^* ) When the color space coordinates L ^* , u ^* , v ^* are present in the area 1, the weight coefficient setting unit 104 determines that wL1 = wu1 = wv1 =
A weighting coefficient that is 1.0 and 0.0 otherwise is output,
Eventually, the output of the first matrix operation unit 101, which performs the operation using the matrix determined using the data existing in the region 1,
It is directly input to the color space coordinate inverse conversion unit 62. Similarly, the CI output from the color space coordinate conversion unit 60
Even when the E1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , and v ^* are present in the area 2 or the area 3, the outputs of the second matrix operation unit 102 and the third matrix operation unit 103, respectively. Is input to the color space coordinate reverse conversion unit 62 as it is.

As described above, when the weighting coefficient is set in each area, if the color distribution of the endoscopic image is concentrated in one of the areas, no problem occurs. When the color distribution of the endoscopic image extends over a plurality of areas, pseudo contours may occur in the colors near the boundaries of the areas. This is because the mapping direction corresponding to the color conversion in each area is different, so that the area does not change smoothly.

Therefore, a boundary area in which the respective areas are in contact is provided, and in the boundary area, the weighting factor is, for example, w1 =
By determining in advance such as setting w2 = w3 = 1/3 so that region 1, region 2, and region 3 have the same effect, the discontinuity in the mapping direction between regions can be determined. It can be smoothed.

As described above, according to this embodiment, in addition to the effects of the first embodiment, first to third matrix operations corresponding to a plurality of areas set in the CIE1976 (L ^* u ^* v ^* ) color space are performed. A color conversion unit 5 that performs highly accurate color difference improvement by performing a linear sum operation on a plurality of outputs from the units 101, 102, and 103 using a weighting coefficient set corresponding to each region.
The cost of 0 can be reduced.

Further, a boundary area is provided between the respective areas, and in the boundary area, for example, by using a weighting coefficient determined as w1 = w2 = w3 = 1/3, each area is You can smoothly connect the color conversion results
Generation of pseudo contours can be suppressed.

Next, the seventh embodiment will be described. FIG.
FIG. 19 is a software configuration diagram showing a software configuration of a color conversion unit according to the seventh embodiment.

The seventh embodiment is almost the same as the first embodiment, and the main difference from the first embodiment of the present embodiment is the configuration of the color conversion unit 50. The color conversion unit 50 of the present embodiment is The color space coordinate conversion unit 60, the matrix calculation unit 61, and the color space coordinate inverse conversion unit 62 in the color conversion unit 50 of the first embodiment are configured by only the RGB matrix calculation unit.

In FIG. 5, the RGB data R, G, B input to the color space coordinate conversion unit 60 and the color space coordinate inverse conversion unit 6 are shown.
There is a one-to-one relationship with the RGB data R ′, G ′, B ′ output from No. 2. Therefore, RGB data R, G, B
If a matrix for predicting the RGB data R ′, G ′, B ′ is created from, the color space coordinate conversion unit 60 and the color space coordinate inverse conversion unit 62 can be omitted, and the overall configuration can be simplified.

Therefore, in this embodiment, the RGB data R,
A matrix for converting G, B into RGB data R ′, G ′, B ′ is created, and the matrix operation using this matrix improves the color difference with a simple configuration as a whole.

As shown in FIG. 24, the color conversion section 50 of the seventh embodiment comprises an RGB matrix calculation section 111 which receives RGB data as input and outputs RGB data R ′, G ′, B ′. The other structure is the same as that of the first embodiment.

Next, regarding the operation in this embodiment, R
The GB matrix calculation unit 111 will be mainly described.

The RGB data input from the main storage device 42 is subjected to 3 × 3 matrix calculation in the RGB matrix calculation unit 111 based on the equation (27), and RGB data R ′, G ′, B'is output.

[0184]

(Equation 7) Next, a method of determining matrix elements in the RGB matrix calculation unit 111 will be described.

As described in the first embodiment, CIE197
Data 1 and Data 2 used to create the prediction matrix in the 6 (L ^* u ^* v ^* ) color space are the observation monitor 8
Regarding the CIE1976 (L ^* u ^* v ^* ) color space coordinates reproduced above, data 1 corresponds to RGB data R, G, B, and data 2 corresponds to RGB data R ', G', B '. It corresponds. Therefore, in the first embodiment, the data 1
And the data 2 were used to predict the matrix in the CIE1976 (L ^* u ^* v ^* ) color space, and the RGB data R, G,
Predict using B and RGB data R ', G', B ',
The matrix used in the RGB matrix calculation unit 111 is determined. Other functions are the same as those in the first embodiment.

As described above, according to the present embodiment, in addition to the effects of the first embodiment, the color conversion unit 50 is provided with the color space CIE1976 (L ^* u ^* v ^* ) color space coordinate conversion unit 6 in the first embodiment.
0, the matrix calculation unit 61, and the color space coordinate inverse conversion unit 62
By configuring with the GB matrix calculating means 111, the color difference can be improved with a simplified configuration.

Next, the eighth embodiment will be described. Figure 25
FIG. 26 relates to the eighth embodiment, FIG. 25 is a software configuration diagram showing the software configuration of the color conversion unit, and FIG. 26 is a software configuration diagram showing the configuration of the matrix calculation unit of FIG.

The eighth embodiment is almost the same as the first embodiment, and the main difference between the first embodiment and the first embodiment is the configuration of the color conversion section 50, which includes the parameter input section and the matrix element reading section. And the addition of the matrix element changing unit. The configuration of this embodiment is the same as that of the first embodiment except for the parameter input unit, the matrix element changing unit, and the matrix element reading unit. Therefore, the parameter input unit, the matrix element changing unit, and the matrix element reading unit are The configuration and operation will be mainly described.

As described in the first embodiment, the matrix elements of the matrix calculation unit 61 are used to reproduce colors and reproduced images on the observation monitor 8 in the CIE1976 (L ^* u ^* v ^* ) color space. It was decided to minimize the color difference of the reproduced colors,
The CIE1976 (L ^* u ^* v ^* ) color space coordinates of the data 1 and the data 2 used when determining the matrix are the observation conditions, the emission characteristics of the observation monitor 8 and the photosensitive material used for the image recording device 4. Different types have different CIs
Since it has E1976 (L ^* u ^* v ^* ) color space coordinates, the matrix obtained using data 1 and data 2 will also be different. Therefore, it is necessary to prepare a different matrix for each combination of the observation conditions, the emission characteristics of the observation monitor 8, the type of recording material used in the image recording device 4, and the like.

Therefore, in this embodiment, the observation conditions, the light emission characteristics of the observation monitor 8, the type of recording material used in the image recording apparatus 4, etc. are input as parameters, and matrix operation is performed based on these parameters. By rewriting the matrix M into an appropriate matrix, the color difference can be improved even when the observation conditions, the light emission characteristics of the observation monitor 8, the type of the photosensitive material used in the image recording device 4, and the like change.

As shown in FIG. 25, the color conversion section 50 of this embodiment uses the observation conditions input by the operator through the information input device 41 such as a keyboard, the emission characteristics of the observation monitor 8 and the image recording device 4. The observation condition stored in the external storage device 44 or the main storage device 42 in response to a parameter input unit 121 for inputting data such as the type of recording material to be input as a parameter and a parameter output from the parameter input unit 121. A matrix element reading unit 122 for reading an appropriate matrix from a plurality of matrices determined for each combination of parameters such as the light emission characteristics of the observation monitor 8 and the type of photosensitive material used in the image recording device 4, and matrix calculation Part 61
And a matrix element changing unit 123 that changes the matrix used in the above to the matrix read by the matrix element reading unit 122. The other structure is the same as that of the first embodiment.

Next, the operation of this embodiment will be described.
Data such as the observation conditions input from the information input device 41 such as a keyboard, the emission characteristics of the observation monitor 8 and the type of photosensitive material used in the image recording device 4 are stored in the parameter input unit 1.
At 21, the color conversion unit 50 inputs the parameter. Then, based on the parameters input via the parameter input unit 121, the matrix element reading unit 122
External storage device 44 such as a magnetic disk or an optical disk,
Alternatively, an appropriate matrix element is read from the main storage device 42, and the matrix element changing unit 123 changes the matrix used in the matrix operation unit 61 to the appropriate matrix read based on the parameter. As a result, the matrix in the matrix calculation unit 61 is in a state adapted to the combination of the observation conditions, the emission characteristics of the observation monitor 8, the type of recording material used in the image recording apparatus 4, and the like. Other functions are the same as those in the first embodiment.

As described above, according to this embodiment, in addition to the effects of the first embodiment, the observation conditions, the light emission characteristics of the observation monitor 8,
Even if the combination such as the type of recording material used in the endoscopic image recording apparatus changes, the matrix element reading unit 122 causes the parameter storage unit 121 to input these conditions as parameters and the external storage An appropriate matrix element is read from the device 44 or the main storage device 42 based on the parameter data, and the matrix element changing unit 123 changes the matrix used by the matrix operation unit 61 to the matrix read by the matrix element reading unit 122. As a result, the color difference between the observation monitor 8 and the image recording device 4 can be improved.

Further, instead of rewriting the matrix of the matrix calculation unit 61, the matrix calculation unit 61 may be configured as shown in FIG. 26, and the observation conditions, the emission characteristics of the observation monitor 8,
A plurality of matrix operation units 126 corresponding to combinations of types of recording materials used in the image recording apparatus 4 are provided, and the matrix operation units 126 are selected from the plurality of matrix operation units 126 by the parameter input unit 121 and the switching unit 127. By selecting an appropriate matrix calculation unit according to the parameter input from the parameter input unit 121, the matrix calculation unit 61 can be operated faster than reading and changing the matrix element, and the observation condition, It is possible to improve the color difference applied to the combination of the emission characteristics of the observation monitor 8 and the type of recording material used in the endoscopic image recording apparatus.

Although FIG. 26 shows an example in which three matrix operation units are provided in parallel, the present invention is not limited to this.

Next, a ninth embodiment will be described. FIG.
30 to 30 are related to the ninth embodiment, FIG. 27 is a software configuration diagram showing the software configuration of the color conversion unit, and FIG.
27 is a flowchart illustrating the flow of processing of the color conversion unit in FIG. 27, FIG. 29 is a flowchart illustrating the flow of a modified example of processing of the color conversion unit in FIG. 27, and FIG. 30 is a preferred conversion of lightness according to the flowchart of FIG. It is a conceptual diagram explaining.

The ninth embodiment is almost the same as the first embodiment, and the main difference from the first embodiment is the configuration of the color conversion section 50. In the color conversion section 50 of the present embodiment, That is, a hue (H), saturation (C), lightness (V) change unit and a parameter input unit are added between the color space coordinate conversion unit 60 and the matrix calculation unit 61. In this embodiment, the configuration and operation other than the hue (H), saturation (C), lightness (V) changing unit and parameter input unit are the same as those in the first embodiment, so the hue (H) ), Saturation (C), brightness (V) changing unit and parameter input unit, and the configuration and operation thereof will be described.

In the first embodiment, the color conversion is performed so that the color of the reproduced image and the color of the observation monitor 8 have the minimum color difference in the CIE1976 (L ^* u ^* v ^* ) color space. Such color conversion may not always be optimal. Depending on the purpose of observing the endoscopic image, the memory of the endoscopic image possessed by the observer, the number of years of experience of the observer in observing the endoscopic image, the target color reproduction is CIE1976 (L ^* u ^* v ^* ) color. It is known that the minimum color difference in space varies from the target color reproduction, and it is generally called a preferable color reproduction.

In view of this, in the present embodiment, the color conversion section 50 is provided with means for color adjustment in the preceding stage of the matrix calculation means 61, so that preferable color reproduction is realized in the image storage device 4.

As shown in FIG. 27, the color conversion unit 50 of this embodiment outputs the CIE output from the color space coordinate conversion unit 60.
1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , and v ^* are input, and hue (H), saturation (C), and lightness in the CIE1976 (L ^* u ^* v ^* ) color space are input. (V) is changed and CIE
1976 (L ^* u ^* v ^* ) color space coordinates L ^* d, u ^* d, v ^* d
, The hue (H), saturation (C), and lightness (V) changing unit 131, and the hue (H), saturation (C), and lightness (V) change amounts are input, and the hue (H) , A saturation (C), and a brightness (V) changing unit 131, and a parameter input unit 132 for outputting to the changing unit 131, and other configurations are the same as those in the first embodiment.

Next, the operation of the hue (H), saturation (C), lightness (V) changing unit 131 and the parameter input unit 132 in this embodiment will be described with reference to FIG.

In FIG. 28, step S50 is an operation in the parameter input section 132, and other than that, it is an operation in the hue (H), saturation (C), and lightness (V) changing section 131. is there.

In step S51, the color space coordinate conversion unit 60
CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* output from H representing the polar coordinates on the u ^* v ^* plane.
And C. The conversion formula is the above-mentioned formula (25).

As described in the fourth embodiment, the polar coordinate H corresponds to the hue among the three attributes perceived by humans, and the polar coordinate C corresponds to the saturation. The L ^* Of Ironosanzokusei, because it is converted to correspond to the brightness, treat the L ^* a lightness V.

As described above, CIE1976 (L ^* u
^* v ^* ) Converting the color space coordinates L ^* , u ^* , v ^* into three attributes of color, hue (H), saturation (C), and lightness (V) is the most intuitive for human beings. This is because it is an easy color adjustment method.

[0206] In step S50, the parameter input unit 132 determines the amount of change in hue (H), saturation (C), and brightness (V) input by the operator using the information input device 41 such as a keyboard.
Input as ΔH, ΔC, and ΔV via
H, ΔC and ΔV are output to step S52.

In step S52, the hue (H), saturation (C), and lightness (V) are changed, and the changed hue (H '), saturation (C'), and lightness (V ') are changed. Output. In step S53, conversion from hue (H ′), saturation (C ′), and lightness (V ′) to CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* d, u ^* d, v ^* d is performed. Performed, matrix operation unit 6
It is output to 1.

In the matrix calculation unit 61, the CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* d, after the hue, saturation and lightness are changed,
CIE1 so that u ^* d and v ^* d are reproduced in the reproduced image.
Color conversion is performed in the 976 (L ^* u ^* v ^* ) color space in the sense of minimum color difference.

[0209] With the change in step S52, C
IE1976 (L ^* u ^* v ^* ) color space coordinates L ^* d, u ^* d,
It is determined in step S54 whether v ^* d is a color outside the color gamut of the observation monitor 8 and CIE1976 (L ^* u ^*
v ^* ) color space coordinates L ^* d, u ^* d, v ^* d are the observation monitor 8
With respect to the color that goes out of the color gamut of the above, in step S55, as described in the fourth embodiment, the hue and the lightness are not changed and the color is moved to the outermost part of the observation monitor color gamut in the saturation direction. The color gamut compression processing is performed, and movement into the observation monitor color gamut is performed. Then, the CIE1976 (L
^* u ^* v ^* ) color space coordinates are changed to CIE1976 (L ^* u)
^* v ^* ) Color space coordinates L ^* d, u ^* d, v ^* d are output to the matrix calculation unit 61.

On the other hand, the CI output from step S53
E1976 (L ^* u ^* v ^* ) color space coordinates L ^* d, u ^* d, v ^*
If d is within the color gamut of the observation monitor 8, the CIE is used as it is.
1976 (L ^* u ^* v ^* ) color space coordinates L ^* d, u ^* d, v ^* d
Is output to the matrix calculation unit 61.

Incidentally, step S50 and step S52
Then, the change of the hue, the saturation, and the lightness was merely an additive change, but this part is shown in step S in FIG.
By replacing with 61 and S62, it is possible to perform with a function including a non-linear change.

With respect to the reproduced image in which the endoscopic image is recorded, what kind of brightness change is preferred is obtained by sensory evaluation, and as a result, as shown in FIG.
When divided into equal parts, from dark to dark, middle, and light,
When the dark part is used, the dark part has a relatively high contrast (hard contrast) and tends to be brighter, the middle part has a slightly weaker contrast (soft tone), and the bright part has a tendency to be hard and dark. I understood. Based on the result of such a sensory evaluation experiment, step S6 of FIG.
1 is a basic function form (change function) for changing hue, saturation, and brightness
, And parameters PH, PC, P for changing the shape of the functional form such as the positions of the inflection points of fH, fC, fV corresponding to the changing functions of hue, saturation, and lightness in the parameter input unit 132.
Enter V. Then, in step S62, the hue, saturation and lightness are changed based on the changing functions fH, fC and fV. Other functions are the same as those in the first embodiment.

As described above, according to the present embodiment, in addition to the effects of the first embodiment, the parameter input unit 132 and the hue (H), saturation (C), lightness (V) changing unit 131 are arranged in the matrix computing unit. By providing in the preceding stage of 61, it becomes possible to adjust the color in a direction more preferable for a viewer of the reproduced image on which the endoscope image is recorded, and it is possible to realize preferable color reproduction.

Hue (H), saturation (C), lightness (V)
By performing the change by adjusting the parameter of the change function based on the sensory evaluation experiment, preferable color reproduction can be realized.

Next, the tenth embodiment will be described. FIG.
FIG. 1 is a software configuration diagram showing the configuration of the matrix calculation unit of the color conversion unit according to the tenth embodiment.

The tenth embodiment is almost the same as the first embodiment, and the main difference between this embodiment and the first embodiment is the configuration of the matrix operation unit 61. Therefore, only the different configuration will be described.
The same components are designated by the same reference numerals and description thereof is omitted.

The matrix M of the 3 × 3 matrix calculation executed by the matrix calculation means 61 has a color distribution of an endoscopic image of CIE197.
As described in the first example, the property of being localized in the 6 (L ^* u ^* v ^* ) color space is used to determine by multiple regression analysis using a linear multiple regression model.

Therefore, in the present embodiment, assuming that the color distribution of the endoscopic image changes, which cannot be handled by one matrix M, the matrix M adapted to the color distribution of the input endoscopic image.
, And perform color conversion using this matrix to improve the color difference.

The matrix calculation unit 61 of this embodiment, as shown in FIG. 31, outputs the CIE output from the color space coordinate conversion unit 60.
A histogram creation unit 141 that creates a histogram in the CIE1976 (L ^* u ^* v ^* ) color space using the 1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , and v ^* , and a histogram creation A histogram data updating unit 142 for updating the histogram data stored in the main storage device 42 or the external storage device 44 using the histogram created by the unit 141, and stored in the main storage device 42 or the external storage device 44. A histogram data reading unit 143 for reading the updated histogram, a color chart data selecting unit 144 for selecting color chart data for determining the matrix M using the histogram data read by the histogram data reading unit 143, Color chart data selection section 1
A matrix determination unit 145 that determines a matrix M by multiple regression analysis using a set of explanatory variables and objective variables output from 44;
The matrix determining unit 145 includes a matrix element changing unit 146 that changes the matrix element determined by the matrix determining unit 145, and a matrix operation executing unit 147 that actually executes an operation using the matrix element changed by the matrix element changing unit 146. To be done. The other structure is the same as that of the first embodiment.

Next, the operation of this embodiment will be described.
CIE1976 output from the color space coordinate conversion means 60
The histogram creating unit 141 creates histogram data using the (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , and v ^* . Then, the histogram data updating unit 142 updates the histogram data in the CIE1976 (L ^* u ^* v ^* ) color space of the endoscopic image. The updated histogram data is read by the histogram data reading unit 143, and the color chart data selection unit 144 uses the read histogram data to select appropriate data for determining the matrix.

As the structure of the histogram data, the number of pixels of the endoscopic image, which is quantized with appropriate accuracy and is distributed in the unit space of the CIE1976 (L ^* u ^* v ^* ) color space, is recorded as data. .

In the color chart data selection unit 144, the center coordinate of the unit space having a frequency equal to or higher than the threshold becomes the address for referring to the color chart data by the threshold processing on the frequency.

The data structure of the color chart data is such that the reference address determined by the color chart data selection unit 144 corresponds to the explanatory variable and the objective variable described in the first embodiment, that is, data 1 and data 2, respectively. Are recorded in pairs.

Using the data output from the color chart data selection unit 144, the matrix determination unit 143 determines the matrix by multiple regression analysis. By outputting the determined matrix to the matrix calculation executing unit 147 via the matrix element changing unit 146, the matrix calculating unit 61 is in a state adapted to the color distribution of the endoscopic image currently input. Other functions are the same as those in the first embodiment.

As described above, according to the present embodiment, in addition to the effects of the first embodiment, the histogram creating unit 141, the histogram data updating unit 142, and the histogram data reading unit 1 are provided.
43, a color chart data selection unit 144, a matrix determination unit 145, and a matrix element change unit 146 are provided to select the color chart data according to the color distribution of the input endoscopic image, and the matrix determination unit 14
5, the matrix used by the matrix operation executing unit 147 is determined, and the matrix element changing unit 146 changes the matrix, thereby realizing the color conversion unit 50 that adaptively changes the input endoscopic image. It is possible to improve the color difference.

Although the first to tenth embodiments have been described above, the abnormal value processing section 82 described in the third embodiment has been described.
Is added to the fourth, fifth, sixth, and seventh embodiments.
The present invention can also be applied to the embodiments, the eighth embodiment, the ninth embodiment, and the tenth embodiment, and has an effect of improving the color difference of the color to be color-converted while preventing an extreme color change.

Also, the matrix operation unit 6 described in the fourth embodiment.
The matrix operation in 1 is changed from a 3 × 3 linear operation to a non-linear matrix operation using a higher-order term of the second or higher order, that is, the third embodiment, the fifth embodiment, the sixth embodiment, and the seventh embodiment. Examples, 8th embodiment, 9th embodiment, and 10th embodiment are also applicable,
The color difference between the color reproduced on the observing monitor 8 and the color of the reproduced image is the minimum color difference in the CIE1976 (L ^* u ^* v ^* ) color space. It is possible to achieve highly accurate color difference improvement that corresponds to a specific situation such as an image in which an agent is sprayed.

Further, the color gamut compression section 8 described in the fifth embodiment.
The addition of 6 is also applicable to the third, fourth, sixth, eighth, ninth, and tenth embodiments, and is applied to data outside the color gamut of the image recording apparatus. As a result of executing the color conversion, it is possible to prevent a significant color reproduction error from occurring, and it is possible to obtain an appropriate color reproduction as a whole in the color gamut of the image recording apparatus.

Further, as described in the sixth embodiment, the plurality of matrix operation units 101 to 103 are provided, and the addition of the weighting coefficient setting unit 104 and the composition unit 105 is performed in the third, fourth and fifth embodiments. The present invention can be applied to the examples, the seventh embodiment, the eighth embodiment, the ninth embodiment, and the tenth embodiment, and the calculation cost of the color conversion unit 50 that improves the color difference with high accuracy can be reduced. .

Further, the color conversion unit 50 described in the seventh embodiment is used.
Replacing the color space coordinate conversion unit 0, the matrix calculation unit 61, and the color space coordinate inverse conversion unit unit 62 in the above with the RGB matrix calculation unit 111 is equivalent to the third embodiment, the fourth embodiment, the sixth embodiment,
It is also applicable to the eighth and tenth embodiments, and the color difference can be improved with a simplified configuration.

The addition of the parameter input unit 121, the matrix element changing unit 123, and the matrix element reading unit 122 described in the eighth embodiment is the same as the third embodiment, the fourth embodiment, the fifth embodiment, and the sixth embodiment. The present invention can also be applied to the embodiments, the seventh embodiment, and the ninth embodiment, in which the combination of the observation conditions, the emission characteristics of the observation monitor 8 and the kind of the storage material used in the endoscopic image storage device changes. Can improve the color difference.

The addition of the hue (H), saturation (C), lightness (V) changing unit 131 and the parameter input unit 132 described in the ninth embodiment is the same as in the third and fourth embodiments. ,
The present invention is applicable to the fifth, sixth, eighth, and tenth embodiments as well, and can realize preferable color reproduction.

Further, the histogram creating section 141, the histogram data updating section 142, the histogram data reading section 143, and the color chart data selecting section 1 described in the tenth embodiment.
44, the matrix determining unit 145, and the matrix element changing unit 146 can be applied to the third embodiment, the fourth embodiment, the fifth embodiment, and the ninth embodiment as well. The color conversion unit 50 that adaptively changes the endoscopic image can be realized, and the color difference can be improved.

In the first to tenth embodiments described above, the embodiments in which the processing in the image processing apparatus 3 is realized by software by the central processing unit 40 have been described. However, in the following, the processing in the image processing apparatus 3 is performed by hardware. An example in the case of the configuration will be described.

First, the eleventh embodiment will be described. 32 to 36 relate to the eleventh embodiment, FIG. 32 is a block diagram showing a configuration of an image processing apparatus, FIG. 33 is a block diagram showing a configuration of a modified example of the image processing apparatus of FIG. 32, and FIG. 3 is a block diagram showing the configuration of the color conversion circuit of FIG. 35, FIG. 35 is a block diagram showing the configuration of the sharpness correction circuit of FIG.
6 is a block diagram showing the configuration of the filtering unit in FIG.

The eleventh embodiment is almost the same as the first embodiment, and the main difference between this embodiment and the first embodiment is the configuration of the image processing apparatus 3. Therefore, only different configurations will be described and the same configurations will be described. Are denoted by the same reference numerals and description thereof is omitted.

In the image processing apparatus 3 of this embodiment, the images recorded in the memory (1) 36a, the memory (2) 36b, and the memory (3) 36c (see FIG. 2) in the signal processing unit 7A of the observation apparatus 7 are recorded. The data R, G, B are as shown in FIG.
It is temporarily stored in the memory 161 via the input I / F 160. Using the stored image data, the selector 16
2 and 164, the color conversion circuit 163 and / or the sharpness correction circuit 165 performs color conversion processing and / or sharpness correction processing of the image. Then, the image data subjected to the color conversion processing and / or the sharpness processing is stored in the memory 166.

The image data stored in the memory 166 is read out, converted into an analog signal by the D / A converter 167, and output to the image recording device 4 via the output I / F 168a.

Alternatively, the image data stored in the memory 166 is read out and the digital image data is output to the image recording apparatus 4 via the output I / F 168b.

As described above, the memory 161 and the memory 16
By performing the signal processing serially with the circuit 6, the circuit configuration can be integrated into one system, and the cost can be reduced.

Further, the image processing apparatus 3 includes the color conversion circuit 1
63, the arithmetic processing controller 169 for controlling the arithmetic operations of the sharpness correction circuit 165, and the memories 161, 166.
A memory controller 170 for controlling the reading and writing of the data is provided. The selector 162 uses the control signal output from the parameter input device 172 via the parameter input device I / F 171 to output the color conversion circuit 1
63, one of the selector 164 selectively selects the memory 161.
It is designed to be connected to.

The selector 164 is an I for parameter input device.
One of the sharpness correction circuit 165 and the memory 166 is selectively connected to the selector 164 by a control signal output from the parameter input device 172 via the / F171.

The control signal output from the parameter input device 172 via the parameter input device I / F 171 is input to the arithmetic processing controller 169, and the color conversion circuit 163 is converted by the parameter input from the parameter input device 172. And / or it is possible to control the operation of the sharpness correction circuit 165.

Here, it is possible to input data to the color conversion circuit 163 and / or the sharpness correction circuit 165, or to output data from the color conversion circuit 163 and / or the sharpness correction circuit 165. In order to do so, the configuration of a modified example of the image processing device 3 of the 11th embodiment is shown in FIG.

That is, the image processing apparatus 3 of the eleventh embodiment.
In the modified example, when changing the parameters and various coefficient data used in the processing executed by the color conversion circuit 163 and / or the sharpness correction circuit 165, a storage means for storing a plurality of types of parameters and various coefficient data is provided. The color conversion circuit 163 and / or the sharpness correction circuit can be read by reading parameters and various coefficient data as necessary to change the processing method of the color conversion circuit 163 and / or the sharpness correction circuit 165. This makes it possible to store the data output from 165.

In the modified example of the image processing apparatus 3 of the eleventh embodiment, as shown in FIG. 33, the memory 175 including a RAM and the external storage device 173 including a magnetic disk and a magneto-optical disk are Color conversion circuit 163 and /
Alternatively, a plurality of types of parameters and various coefficient data used in the processing executed by the sharpness correction circuit 165 are stored. The memory 175 and the external storage device 173 are controlled by the controller 176 by signals input to the controller 176 via the information input device 172 such as a keyboard and the information input device I / F 171. Then, the memory 175 and the external storage device 173 output data to the color conversion circuit 163 and / or the sharpness correction circuit 165 or input data from the color conversion circuit 163 and / or the sharpness correction circuit 165 by the control signal. To do.

In this embodiment, the endoscopic image obtained by the electronic endoscope 6 is processed by the image processing device 3 and the processed result is output to the image recording device 4. Further, although the color conversion circuit 163 is shown to be positioned before the sharpness correction circuit 165,
This position can be reversed.

The image recording device 4 is composed of, for example, a video printer or a film recorder.

Next, the color conversion circuit 163 will be described. RGB data and CIE1976 (L ^* u ^* v
^* ) Each color space coordinate signal is standardized to 8 bits, but the standardization number is not limited to 8 bits, and may be 8 bits or less or 8 bits or more.

As shown in FIG. 34, the color conversion circuit 163.
Inputs the RGB data from the selector 162, and
^{E1976 (L * u * v *} ) color space coordinates L ^*, u ^*, v ^* and outputs a signal L ^* a LUT181a, u ^* for LUT181b
And a v ^* LUT 181c for color space coordinate conversion, and the CIE1976 (L ^* u ^* v ^* ) color space coordinate L ^* , u ^* , v ^* signals output from the LUT 181 for color space coordinate conversion. (L ^* u ^* v ^* ) color space coordinates L ^* ', u ^* ', v ^* 'signals for conversion into L ^* ' LUT
182a, u ^* 'LUT 182b and v ^* ' LUT 18
2c for color conversion LUT 182 and color conversion L
CIE1976 (L ^* u output from the UT182
^* v ^* ) Color space coordinates L ^* ', u ^* ', v ^* 'R'output to selector 164 which converts signals into RGB data R', G ', B'.
LUT 183a, G'LUT 183b and B'LU
LUT 18 for inverse color space coordinate conversion configured by T183c
And 3.

Each LUT 181 to 183 is shown in FIG.
Memory 17 in a modification of the image processing apparatus 3 shown in FIG.
5 or LUT data can be input from the external storage device 173.

Next, the sharpness correction circuit 165 will be described.

As shown in FIG. 35, the sharpness correction circuit 16
Reference numeral 5 denotes a filtering unit 201 that filters input RGB image data, a coefficient value setting unit 202 that sets a filtering coefficient value, a data input unit 203 that inputs data for setting a coefficient value, and each block. The control unit 204 for controlling

The sharpness correction circuit 165 is the one in which the sharpness correction unit 51 of the first embodiment is realized by hardware, and the principle of sharpness correction is the same as that of the first embodiment (FIGS. 9 and 10). Since it has been described in 11), the description is omitted.

As shown in FIG. 36, the filtering unit 201 stores the filtering execution unit 210 including R, G, and B filtering execution units 210r, 210g, and 210b for each RGB data and the value of the processing result. R, G, B data conversion means 211r, 211 for adjustment
The data conversion means 211 is composed of g and 211b.

In the filtering execution unit 210,
Filtering in the spatial frequency domain is performed on each RGB original image data. .

The filtering process by the filtering execution unit 210 is the process in steps S11 to S13 of FIG. 10 shown in the first embodiment.
Original images r (x, y), g (x, y) and b (x, y) of each GB data, and hr (x, y), hg (x, y) and hb which are PSFs for the respective data. (X, y)
To obtain original images R ', G'and B'for shooting.

Next, the data conversion unit 211 inputs the original image data R ', G'and B'for each RGB data, and the data conversion unit 211 filters the values exceeding the display range by filtering. Perform post-processing to convert data with. That is, the data below the decimal point is rounded off, or the value less than 0 or 256 or more when the displayable range is, for example, 0 to 255 is rounded down. The respective photographed images R ', G'and B'after applying this processing are output as processing results.

The details of this processing have been described in the filtering execution unit 65 of the first embodiment, and therefore will be omitted.

Next, the operation of the coefficient setting section 202 will be described. In the coefficient setting means 202, the coefficients of the filters applied in the filtering section 201, that is, the correction filters Mr (u, v), Mg (u, v) and Mb.
Each value in (u, v) is sent to the filtering unit 201.

Further, the data input section 203 sends information on coefficient setting to the coefficient setting means. In a photographic device, the PSF differs depending on the camera, film, etc. used. Therefore, each data is used as information used for coefficient setting, and an appropriate correction filter is installed. The other operations of the sharpness correction circuit 165 are the same as those of the sharpness correction unit 51 of the first embodiment, and therefore the description thereof will be omitted.

Subsequently, the color conversion circuit 16 according to the present embodiment.
The operation of No. 3 will be described. The principle of color conversion has been described in the first embodiment, and will not be repeated.

First, the RGB data R, G, B are input to the LUT 125 for color space coordinate conversion, and CIE1976 (L ^*
u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* signals are converted.

The color space coordinate conversion LUT 181 calculates the RG based on the equations (1) and (2) (see the first embodiment).
B data and CIE1976 (L ^* u
^* v ^* ) Create data paired with the color space coordinate signal. Then, the RGB to the color space coordinate conversion LUT 181.
The CIE1976 so that the data becomes the reference address.
(L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* data is created by writing in advance.

Subsequently, in the matrix calculation LUT 182, the calculation in the equation (3) is performed by the CIE1976 (L ^* u ^*).
v ^* ) color space coordinates L ^* , u ^* , v ^* serve as reference addresses, and CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* ',
It is executed by outputting u ^* 'and v ^* '.

The matrix calculation LUT 126 is the CIE197.
The data in the 6 (L ^* u ^* v ^* ) color space is calculated according to the equation (3) or the equation (4), and the CIE1 is calculated.
976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* and CIE
1976 (L ^* u ^* v ^* ) color space coordinates L ^* ', u ^* ', v ^* 'are paired to create data, and CIE1976 (L ^* u ^*) is created ^.
v ^* ) color space coordinates L ^* , u ^* , v ^* serve as reference addresses, and CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* ',
Created to output u ^* ', v ^* '.

In the color space coordinate reverse conversion LUT 183, C
IE1976 (L ^* u ^* v ^* ) color space coordinates L ^* ', u ^* ',
Convert v ^* 'into standardized RGB data R', G ', B'.

As described above, according to this embodiment, in addition to the effects of the first embodiment, the color conversion circuit 163 is provided for color space coordinate conversion L.
By configuring the UT 181, the matrix calculation LUT 182, and the color space coordinate inverse conversion LUT 183 by hardware, it is possible to operate the color conversion device faster than the software configuration of the first embodiment, and the observation monitor 8 The color conversion circuit 163 that improves the color difference between the reproduced image and the reproduced image can be operated at high speed.

The third and fifth to ninth embodiments can be realized by hardware by configuring as follows.

That is, the abnormal value processing unit 8 in FIG.
If 1 is composed of an LUT, the third embodiment can be realized by hardware as in the case of this embodiment.

If the color gamut compressing section 86 in FIG. 17 is constructed by an LUT, the fifth embodiment can be realized by hardware as in the case of this embodiment.

A plurality of matrix operation units 101 in FIG.
102 and 103 are configured by LUTs, the synthesizing unit 105 is configured by a multiplier and an adder, and the weighting coefficient setting unit 104 is output from the color space coordinate conversion LUT 181 as CIE1976.
(L ^* u ^* v ^* ) The weighting coefficient is read from the memory 175 or the external storage device 173 in the modified example of the image processing apparatus 3 shown in FIG. 33 in which the color space coordinates L ^* , u ^* , and v ^* data are used as control signals. By using the controller, the sixth embodiment can be realized by hardware.

The seventh embodiment can be realized by hardware by configuring the RGB matrix calculation unit 111 in FIG. 24 with an LUT.

The matrix element changing unit 123 and the matrix element reading unit 122 in FIG. 25 are constituted by a controller which reads LUT data from the memory 175 or the external storage device 173 in the modification of the image processing apparatus 3 shown in FIG. The parameter input unit 121 to the parameter input device I
The eighth embodiment can be realized by hardware by configuring with / F171.

Hue (H), saturation (C), and
The brightness (V) changing unit 131 is set to the color space coordinate conversion LUT 18.
A CIE 1976 (L ^* u ^* v ^* ) color space coordinate L ^* , u ^* , v ^* data and a parameter data output from the memory 1 175 in the modified example of the image processing apparatus 3 shown in FIG. Alternatively, the external storage device 173
The parameter input unit 132 serves as a controller for reading LUT data from the I / F 171 for the parameter input device.
With the above configuration, the ninth embodiment can be realized by hardware.

Next, the twelfth embodiment will be described. FIG.
7 is a block diagram showing the configuration of the color conversion circuit according to the twelfth embodiment.

The twelfth embodiment is almost the same as the eleventh embodiment, and the difference between this embodiment and the eleventh embodiment is the configuration of the color conversion circuit. Therefore, only the different configurations will be described and the same configurations will not be described. The same reference numerals are given and the description is omitted.

For example, when the input signal is standardized to 8 bits, the matrix operation LUT 182 shown in FIG. 34 of the 11th embodiment requires a memory capacity of 24 bits, and the entire color conversion circuit 163 is required. The cost will be high. Therefore, by reducing the number of bits of the signal input to the LUT, the cost of the color conversion circuit 163 as a whole is reduced while maintaining the accuracy of color difference improvement of the color conversion circuit 163 within an allowable range.

The color conversion circuit 163 of the twelfth embodiment is shown in FIG.
4 corresponds to an example in which the color conversion unit 50 of the seventh embodiment shown in FIG. 4 is configured by an LUT, and as shown in FIG.
L'for R'which inputs G and B and outputs RGB data R '
UT221 and LU for G'which outputs RGB data G '
T222 and B'LUT for outputting RGB data B '
223 and 223.

The operation of each LUT 221, 22, 223 is controlled by the arithmetic controller 169. Further, the LUTs 221, 22 and 223 can input LUT data from the memory 175 in the modification of the image processing apparatus 3 of the eleventh embodiment shown in FIG. 33 or the external storage device 173.

Since the operation concerning each LUT 221, 222, 223 is the same as that of the RGB matrix operation unit 111 in the seventh embodiment, here, each LUT 221, 222, 2 will be described.
A method of setting the number of input bits 23 will be mainly described.

As described in the seventh embodiment, in order to determine the matrix for obtaining the RGB data R ', G', B'from the RGB data R, G, B, the known technique of multiple regression analysis is used. . For example, in multiple regression analysis using a linear multiple regression model, the predictor variable R ′ is predicted by using the explanatory variable R
The weighted linear sum of G and B is used, and the weighting factors of R, G, and B obtained as a result of multiple regression analysis are
It is a well-known fact that it corresponds to the degree of contribution of G and B to the objective variable R '.

Therefore, even if the input bit is reduced for the explanatory variable having a low degree of contribution to the objective variable,
Since the effect on the entire color conversion processing is low, the R'LUT does not input R, G, and B 8 bits each, but rather the input bits according to the degree of contribution to R ', for example, 8 bits, 7 bits, Change it to 7 bits.
The same applies to the G'and B'LUTs.

The example in which the RGB matrix operation unit 111 in the seventh embodiment is configured by the LUT has been described above. However, other than the seventh embodiment, when the constituent elements are replaced by the LUT, all of them can be applied. Is.

As described above, according to this embodiment, in addition to the effect of the eleventh embodiment, the input bits of each LUT are changed according to the degree of contribution of the input signal to the output signal, so that the total number of input bits is smaller. By using the LUT configured in (3), the overall memory capacity can be reduced, and the cost of the color conversion circuit 163 can be reduced.

Next, a thirteenth embodiment will be described. FIG.
FIG. 8 is a block diagram showing the configuration of the color conversion circuit according to the thirteenth embodiment.

The thirteenth embodiment is almost the same as the twelfth embodiment, and the difference from the twelfth embodiment of the present embodiment is the configuration of the color conversion circuit. Therefore, only the different configurations will be described and the same configurations will not be described. The same reference numerals are given and the description is omitted.

The color conversion circuit 163 of the thirteenth embodiment is shown in FIG.
5 corresponds to an example in which the color conversion unit 50 of the eighth embodiment shown in FIG. 5 is configured by an LUT, and color conversion and color three attributes adapted to various conditions described in the eighth embodiment are changed to obtain a preferable color reproduction. In order to realize it with the color conversion circuit 163 configured by the LUT,
It is necessary to rewrite the LUT data of the matrix calculation unit 61 every time the content of the calculation is changed, or to provide a plurality of matrix calculation LUTs corresponding to various changes in parallel and switch to the parameter input. Depending on capacity,
The rewriting time and the cost of the color conversion circuit 163 may not be ignored.

On the other hand, depending on various conditions and the degree of change of the three attributes of color, it is possible to replace the original color conversion by a relatively simple low-order matrix operation.

Therefore, in this embodiment, when the change from the original color conversion, such as the change of various conditions and the three attributes of color, can be replaced by a relatively simple low-order matrix operation, various conditions and colors are obtained. Change the three attributes in the previous stage of the color conversion LUT,
Alternatively, a matrix operation circuit (hereinafter referred to as a space converter) provided in a subsequent stage can be used, and the color conversion circuit 163 can be used for a variety of conditions and changes in three color attributes with a simple configuration.

The color conversion circuit 163 of the thirteenth embodiment is similar to that of FIG.
As shown in FIG. 8, the R ′ LUT 221 that performs color conversion respectively.
And a space converter 231 that executes a matrix operation on the RGB data output from the LUT 222 for G'and the LUT 223 for B ', and if the output data from the space converter 231 exceeds the standard range, standardization processing is performed. And a limiter 232 for performing mapping within a predetermined standardized range.
This is the same as the embodiment.

In the color conversion circuit 163 of the thirteenth embodiment, the color conversion LUTs 221, 222 and 223 are the memory 16
The RGB data input from 1 is used as a reference address, and R
It outputs GB data R ', G', B '. LU for color conversion
The RGB data R ′, G ′, B ′ output from T221, 222, and 223 is subjected to matrix calculation by the space converter 231, and RGB data R ″, G ″,
B ″ and the space converter 231.
If the RGB data R ″, G ″, B ″ output from the data exceeds the predetermined standardization range, the limiter 232 prevents the color change due to the inversion of the data, so that the rounding process is performed within the standardization range. A mapping corresponding to the standardization processing of is performed.

On the other hand, the parameter input device I / F 171
It is possible to read and rewrite the coefficient data of the space converter 231 from the external storage device 173 or the memory 175 in the modification of the image processing device 3 of the eleventh embodiment shown in FIG. 33 based on the parameters input via the. It is possible.

In FIG. 38, the space converter 231 and the limiter 232 are shown as color conversion LUTs 221, 222 and 2.
Although the example provided in the latter stage of 23 is shown, the space converter 231 and the limiter 232 are provided in the color conversion LUT 2
Even if it is provided in the preceding stage of 21, 222 and 223, the same operation is performed.

As described above, according to this embodiment, in addition to the effects of the twelfth embodiment, various conditions, changes in the three color attributes, and the like are dealt with by the space converter 231 provided before or after the color conversion LUT. With a simple configuration, the color conversion circuit 163 can be adapted to various conditions and changes in the three attributes of color.

Next, a fourteenth embodiment will be described. FIG.
9 is a block diagram showing the configuration of the color conversion circuit according to the fourteenth embodiment.

The fourteenth embodiment is almost the same as the eleventh embodiment, and the difference from the eleventh embodiment of this embodiment is the configuration of the color conversion circuit. Therefore, only the different configuration will be described and the same configuration will be described. The same reference numerals are given and the description is omitted.

In the configuration of FIG. 37 for explaining the twelfth embodiment, the contents of the LUT can be rewritten using the data line, but depending on the memory capacity of the LUT, rewriting time is required. Therefore, in the present embodiment, the contents of the LUT are changed at high speed by switching by the control signal input to the LUT.

The color conversion circuit 163 of the fourteenth embodiment has a seventh
This corresponds to an example in which the color conversion unit 50 of the embodiment is configured by an LUT,
As shown in FIG. 39, RGB data R, G, B and a control signal are input and RGB data R ′ is output R ′.
LUT 241, G'LUT 242 for outputting RGB data G ', and B'L for outputting RGB data B'.
It is composed of the UT 243.

LUT 241 for R'and LUT 242 for G '
The operations of the LUT 243 for B'and B'are almost the same as those of the LUT 221 for G'221, the LUT 222 for G'and the LUT 223 for B'of the twelfth embodiment. Therefore, here, the LUT 241 for R'and the LUT 242 for G'are provided. And and B '
The operation of the control signal input to the for-use LUT 243 will be described.

LUT 241 for R'and LUT 242 for G '
Data of a plurality of systems are stored in advance in the LUTs 243 for B'and. Control signal is R'LUT
The number of bits enough to switch the data of a plurality of systems stored in the 133, the G ′ LUT 134, and the B ′ LUT 135 is secured.

The control signal is output from the arithmetic processing controller 169 shown in FIG. 32 and is a signal input from the parameter input device 172 such as a keyboard.

The operator inputs a changeover to a color conversion method by changing various conditions or three attributes of colors with the parameter input device 172, and the arithmetic processing controller 169 changes over the LUT data of a plurality of coefficients based on the parameters. The control signal is output to the color conversion circuit 163. This control signal corresponds to the R'LUT 241 and the G'LUT 24.
2 and B'are input to the LUT 243 for B'and a plurality of lines of L
The operation of the LUT is controlled so as to refer to appropriate LUT data from the UT data.

Although the operation of the fourteenth embodiment has been described above, the same structure and operation as the present embodiment can be applied to the embodiments which can be constituted by the LUT, other than the structure shown in the present embodiment. Is.

As described above, according to this embodiment, in addition to the effects of the twelfth embodiment, a signal line for inputting a control signal is provided to each LUT in which LUT data of a plurality of systems are stored in advance, and the LUT is provided. It is possible to switch the color conversion method at high speed by controlling the reference operation of (1) by the control signal.

Next, a fifteenth embodiment will be described. FIG.
0 is a block diagram showing a configuration of a filtering unit of the sharpness correction circuit according to the fifteenth embodiment.

The fifteenth embodiment is almost the same as the eleventh embodiment, and the difference from the eleventh embodiment of this embodiment is the configuration of the sharpness correction circuit 165. Therefore, only the different configuration will be described and the same configuration will be described. Are denoted by the same reference numerals and description thereof will be omitted.

The sharpness correction circuit 165 of the fifteenth embodiment is
This corresponds to an example in which a modification of the sharpness correction unit 51 of the first embodiment shown in FIG. 8 is configured by hardware, and the principle of this sharpness correction has been described in the first embodiment (FIG. 11). , Description is omitted.

In this embodiment, the filtering execution unit is composed of a digital filter, and the filtering by the digital filter is advantageous in real-time processing. Here, an FIR filter is used as the digital filter.

As shown in FIG. 40, the filtering section 201 has R, G, and B Fs for each RGB data.
IR filtering execution units 250r, 250g, 250
FIR filtering execution unit 250 composed of b, and R, G, B data conversion means 25 for adjusting the value of the processing result.
Data conversion means 25 including 1r, 251g, and 251b
1 and 1. The other structure is the same as that of the eleventh embodiment.

In the filtering execution unit 250,
Filtering in the spatial frequency domain is performed on each RGB original image data. The flow of the processing has been described in the modified example of the first embodiment, and will be omitted.

Next, the coefficient setting unit 202 in this embodiment
The operation of will be described. In the coefficient setting unit 202, the coefficients of the FIR filtering execution units 250r, 250g, 250b applied in the filtering unit 201 are set. For example, if the MTF of a photography device is ideal M
When correcting to TF, the above-mentioned correction filter Mr is used.
Mr (x, y), mg which are the inverse Fourier transforms of (u, v), Mg (u, v) and Mb (u, v), respectively.
(X, y) and mb (x, y) are sent to the filtering unit 201 as FIR filter coefficients.

Further, the data input section 203 sends information on coefficient setting to the coefficient setting section 202. In the photography equipment, the PSF depends on the camera and filter used.
Therefore, each data is used as information used for coefficient setting, and an FIR filter that gives an appropriate spatial frequency characteristic of the correction filter is set. The other actions are the same as in the eleventh embodiment.

Therefore, according to the fifteenth embodiment, in addition to the effects of the eleventh embodiment, it is advantageous in realizing the processing in real time.

In order to realize the sharpness correction with the configuration shown in FIG. 3, the series of processes or part of the processes described above may be operated in the central processing unit 40 in FIG.

Next, a sixteenth embodiment will be described. FIG.
1 to 44 relate to the 16th embodiment, FIG. 41 is a block diagram showing a configuration of a sharpness correction circuit, FIG. 42 is a block diagram showing a configuration of a combining unit of FIG. 41, and FIG. 43 is a data conversion unit of FIG. FIG. 44 is a block diagram showing the configuration of FIG. 41, and FIG. 44 is a block diagram showing the configuration of a modification of the combining unit of FIG.

The sixteenth embodiment is almost the same as the eleventh embodiment, and the difference between this embodiment and the eleventh embodiment is the configuration of the sharpness correction circuit 165. Therefore, only the different configuration will be described and the same configuration will be described. Are denoted by the same reference numerals and description thereof will be omitted.

In the sixteenth embodiment, the sharpness correction circuit 165 of the fifteenth embodiment is intended to suppress the deterioration of the SN ratio of the picked-up image which is caused by the sharpness correction processing.
Corresponds to an example in which the sharpness correction unit 51 of the second embodiment shown in FIG. 13 is configured by hardware, and the principle of this sharpness correction has been described in the second embodiment, so description will be omitted. .

Next, the sharpness correction circuit 1 in the present embodiment.
The configuration and operation of 65 will be described.

The sharpness correction circuit 165 of the sixteenth embodiment is
As shown in FIG. 41, a filtering unit 261 that filters each input RGB data, a weighting coefficient value w setting unit 262 that sets a weighting coefficient value w, an original image and a filtering processing result by the weighting coefficient value w. It is configured to include a synthesizing unit 263 for synthesizing an image and a data converting unit 264 for adjusting a processing result output value. The other structure is the same as that of the eleventh embodiment.

In the filtering section 261, the sharpness correction component images Rs (x, y), Gs (x, y) and Bs are set for each RGB data in the input original image.
Apply filtering to generate (x, y) (see the second embodiment). As a concrete means for realizing the filtering, either the filtering in the spatial frequency domain described in the eleventh embodiment or the filtering by the FIR filter which is the digital filter described in the fifteenth embodiment may be used. .

The weighting factor value w setting unit 262 sets the weighting factor value w described in the second embodiment based on the information on the brightness of the image.

In the synthesizing unit 263, each RGB data in the original image, and the sharpness correction component images Rs (x, y) and Gs (x, y) output from the filtering unit 261.
, Bs (x, y), and the weighting coefficient value w are used for the calculation (see equations (20) to (22)).

Therefore, as shown in FIG.
3, in the R, G, B multiplication units 281r and 281b, the sharpness correction component images Rs (x, y), Gs (x,
y) and Bs (x, y) are multiplied by the weighting coefficient value w. Then, in the adders 282r and 28b for R, G, and B, the multipliers 281r and b for R, G, and B are used.
Output from w · Rs (x, y), w · Gs (x,
y) and w · Bs (x, y) and R (x,
y), G (x, y) and B (x, y) are added. These calculation results R '(x, y), G' (x, y)
And B ′ (x, y) are output to the subsequent data conversion unit 264.

As shown in FIG. 43, the data conversion unit 264 is composed of R, G, B data conversion units 291r and 291b. In the data conversion units 291r-b, R '(x, y), G' (x,
y) and B '(x, y) perform post-processing for converting the data beyond the display range. In other words, the data that can be displayed after the decimal point is rounded off or the displayable range is 0, for example.
When the value is up to 255, the value less than 0 or a value of 256 or more is truncated. Each photographing image R ″ after applying this processing,
G ″ and B ″ are output as the processing result. The data conversion unit 264 may be configured by, for example, an LUT (look-up table) including a ROM (read-only memory).

Next, the coefficient setting unit 202 in this embodiment
The operation of will be described. In the coefficient setting unit 202, the coefficient of the filter applied in the filtering unit 261, that is, the band pass filter Mpr (u, v), if using the filtering in the spatial frequency domain,
If the values in Mpg (u, v) and Mpb (u, v) are to be filtered by digital film, filter coefficients mpr (x, y), mpg (x, y) giving the characteristics of each correction filter. ) And mpb (x,
y) is sent to the filtering unit 261.

Further, the data input 203 transmits information on coefficient setting to the coefficient setting unit 102. In a photographic device, the PSF differs depending on the camera, film, etc. used, so each data is used as information used for coefficient setting, and an appropriate correction filter is set. Further, if the original image is, for example, an image picked up by an endoscope, Mp ′ (u, v) or mp ′ (x, y) is determined according to information such as the number of pixels of the solid-state image pickup device based on the model of the endoscope used. May be controlled. Since the frequency bands to which the expressions d8 and d9 are applied are changed based on the difference in resolution, a more preferable captured image can be obtained. The other actions are the same as in the eleventh embodiment.

Therefore, in the present embodiment, in addition to the effect of the eleventh embodiment, it is possible to suppress the deterioration of the SN ratio of the picked-up image, which is caused by the sharpness correction processing, as in the second embodiment.

Although the sharpness correction component image is generated by band-pass filtering in this embodiment, as in the second embodiment, the correction filter M (u) described in the eleventh and fifteenth embodiments is used. , V), it is possible to realize the same effect by using the processed image.

That is, the sharpness correction component image fs which is the processing result of the original image f (x, y) and the correction filter.
The second weighting operation using the weighting coefficient value w with (x, y)
The weighting factor value w may be controlled by the information component representing the brightness of the image instead of the equation (23) described in the embodiment.
It may be realized by the control method of the weighting factor value w and the above-mentioned determined method.

Along with this, the synthesizing unit 2 in FIG.
63 is changed to the one shown in FIG. That is, FIG.
As shown in, the weighting factor in the first multiplication unit 301 is applied to the sharpness correction component images Rs (x, y), Gs (x, y) and Bs (x, y) output from the filtering unit 261. The multiplication with the numerical value w is performed. Also, for the input original images R (x, y), G (x, y) and B (x, y),
The second multiplication unit 302 performs multiplication with (1.0-w). Then, in the addition unit 303, the first multiplication 3
01 and the second calculation means 302 add the respective calculation results, and then R '(x, y), G' (x, y) and B '(x,
y) and sends it to the data conversion means 264.

The synthesizing unit 26 described in this embodiment is used.
In order to implement 3, the multiplication unit and the addition unit, which are the constituent elements, are L using a RAM (read only memory), for example.
A UT (Look Up Table) may be used.

In this embodiment, the control of the weighting coefficient value w is performed for each pixel of the image. However, for example, the image is divided into a plurality of blocks and each block of the information component indicating the brightness is divided into blocks. You may control by the average value in. In this case, the same weighting factor value w is applied to each pixel in one block.

In order to realize the sharpness correction with the configuration shown in FIG. 3, the series of processes described above or a part of the processes may be operated in the central processing unit 40 shown in FIG.

Next, a seventeenth embodiment will be described. FIG.
5 and 46 relate to the seventeenth embodiment, FIG. 45 is an explanatory view for explaining the mask size of the FIR filter, and FIG. 46 is FIG.
6 is a block diagram showing a configuration of a filtering unit including the FIR filter of FIG.

The seventeenth embodiment is almost the same as the eleventh embodiment, and the difference from the eleventh embodiment of this embodiment is the configuration of the sharpness correction circuit 165. Therefore, only the different configuration will be described and the same configuration will be described. Are denoted by the same reference numerals and description thereof will be omitted.

The seventeenth embodiment relates to the filtering units 250 and 261 in the fifteenth and sixteenth embodiments,
This is particularly relevant in configurations using FIR filters.

The two-dimensional FIR filter has a mask size of n1.times.n2 (n1 and n2 are non-zero positive numbers). Here, for simplification, n1 = n2 = n is set, and an FIR filter having a mask size of n × n will be described as an example.

Generally, in the hardware implementation of the FIR filter, the mask size has a great influence on the hardware cost. For example, an FIR filter with a mask size of 9 × 9 requires four filters in a device capable of performing filtering with an FIR filter with a mask size of 5 × 5 (this size is common for hardware devices). That is, a total of 12 pieces are required for RGB 3 system data.
Also, as the number of devices increases, many peripheral hardware such as delays are required. Therefore, increasing the mask size in the FIR filter causes a problem of increasing hardware cost.

Therefore, in the filtering unit using the FIR filter, even if the mask size is increased, a structure for minimizing the increase in hardware cost due to the increase in mask size is realized.

In the present embodiment, the spatial frequency characteristic of one FIR filter is realized by the cascade connection of a plurality of FIR filters having smaller mask sizes (hereinafter, divided filters), thereby achieving the above object. To achieve.

Mask size (4k + 1) × (4k + 1)
(However, k = 1, 2, 3, ...) The FIR filter coefficient is FI of mask size (2k + 1) × (2k + 1).
It can be expressed by a convolution operation of two R filters.

For example, in FIG. 45, when k = 1, that is, the FIR filter m0 (x, y) with a mask size of 5 × 5.
The coefficient of (FIG. 45 (a)) is converted into
R filter m1 (x, y) (FIG. 45 (b)) and m2
This is an example configured by convolution calculation with (x, y) (FIG. 45 (c)).

The processing result image f ′ (x, y) obtained by applying these FIR filters of mask size 3 × 3 to the original image f (x, y) in cascade connection is f ′ (x, y). = (F (x, y) * m1 (x, y)) * m2 (x, y) (28) However, * represents a convolution operation. On the other hand, from m0 (x, y) = m1 (x, y) * m2 (x, y) (29), the equation (28) becomes f ′ (x, y) = f (x, y) * (m1 It can be expressed as (x, y) * m2 (x, y) = f (x, y) * m0 (x, y) (30), and it can be seen that it matches the processing result by the filter m0 (x, y). .

Further, in terms of spatial frequency characteristics, equation (28)
Can be expressed as F ′ (u, v) = (F (u, v) · M1 (u, v)) · M2 (u, v) (31), and M0 (u, v) = M1 (u, v) · M2 (u, v) (32) From F ′ (u, v) = F (u, v) · (M1 (u, v) · M2 ((u, v)) = F (u, v) · M0 (u, v) (33) is shown.

Here, F (u, v), F '(u, v),
M0 (u, v), M1 (u, v) and M2 (u, v) are f (x, y), f '(x, y) and m0 (x,
y), m1 (x, y) and m2 (x, y) are Fourier transforms.

As in the above description, the mask size is 9 × 9.
FIR filter (k = 2) has a mask size of 5
It can be configured by convolution operation of two × 5 FIR filters. Further, from the application of those FIR filters in a cascade connection, a processing result equivalent to one-time filtering processing with the FIR filter of mask size 9 × 9 can be obtained.

Therefore, the required mask size 5 ×
The number of devices that perform the filtering by the FIR filter of No. 5 becomes two, and the hardware cost is significantly reduced.

Here, an example applied to the filtering unit 250 in the sharpness correction device 201 of the fifteenth embodiment will be described. The mask size of the applied FIR filter is 9
× 9, and two FIR filters each having a mask size of 5 × 5 (first FIR filter and second FIR filter, respectively)
It is realized by a cascade connection of the R filter).

Filtering unit 201 in the present embodiment
46, as shown in FIG. 46, a first filtering execution unit 310 including first filtering execution units 310r and b for R, G and B and a second filtering execution unit 311r and b for R, G and B. The second filtering execution unit 310 and the data conversion unit 211 including the R, G, and B data conversion units 211r and 211b for adjusting the value of the processing result.

In the first filtering execution section 310, the first F
IR filters mr1 (x, y) to mb1 (x, y)
Perform filtering by.

In the second filtering execution section 311, for each R′G′B ′ image data resulting from the filtering processing result by the first filtering execution section 310,
Second FIR filters mr2 (x, y) to mb2
Perform filtering by (x, y).

Mr1 (x, y) and mr2 (x, y),
mg1 (x, y) and mg2 (x, y), mb1 (x,
y) and mb2 (x, y) are the frequency characteristics Mr0 of the correction filter when they are applied in cascade connection, respectively.
(U, v), Mg0 (u, v) and Mb0 (u, v) are set.

In the data conversion unit 211, each R ″ G ″ B ″ image data which is the filtering processing result by the second filtering execution unit 311 is input, where it has a value exceeding the display range due to filtering. Post-processing for converting the data is performed, that is, rounding off the data below the decimal point, and truncating values less than 0 or 256 or more when the displayable range is, for example, 0 to 255. The respective photographed images R "', G"' and B "'after application are output as processing results. The data conversion unit 211 may be configured by, for example, a LUT (look-up table) formed by a ROM (read-only memory).

Next, the coefficient setting unit 202 in the present embodiment.
The operation of will be described. In the coefficient setting unit 202, the coefficient of the FIR filter applied in the first filtering execution unit 310 and the second filtering execution unit 311 is set. For example, when the MTF of the photographing device is corrected to the ideal MTF, the correction filter Mr described above is used.
Mr1 (x, y) and mr2 (x, which are combinations of FIR filters for realizing the respective spatial frequency characteristics of (u, v), Mg (u, v) and Mb (u, v).
y), mg1 (x, y) and mg2 (x, y), mb1
(X, y) and mb2 (x, y) are sent to the first filtering execution unit 310 and the second filtering execution unit 311.

The data input section 203 also sends information on coefficient setting to the coefficient setting section 202. In the photography equipment, depending on the camera and film used, PSF
Therefore, each data is used as information used for coefficient setting, and an FIR filter that gives an appropriate spatial frequency characteristic of the correction filter is set.

When the original image f (x, y) contains a noise component, the noise component may be emphasized when the ideal inverse filter is applied. In such a case, as described in the fifteenth embodiment, the correction filter M0 (u, v) to be applied is applied.
It suffices to change the spatial frequency characteristic of and control the degree of correction in the high frequency band. The first and second FIR filters m1 (x, y) and m2 (x, y) may be set so as to give a spatial frequency characteristic of M0 (u, v) when applied in cascade connection. As a result, it is possible to correct the MTF in the frequency band practically required while preventing unnecessary noise components from being emphasized.

Further, when the statistical property with respect to noise is known, the first and second FIR filters which give a spatial frequency characteristic by creating a Wiener filter or a least squares filter instead of applying an inverse filter. May be set.

Further, similarly to the one described in the eleventh embodiment, the coefficient of the FIR filter set in the coefficient setting section 202 is changed and the MTF is controlled more positively, which is suitable for observation. It is possible to obtain a clear image.

That is, by setting the first and second FIR filters so as to give the spatial frequency characteristic shown in FIG. 11C, the picked-up image becomes more sharp.

M0 '(u, v) is the data input unit 20
According to 3, the PSF is changed due to the difference between the camera and the film used in the above-mentioned photographing device, and the increase / decrease of the frequency band and the spatial frequency characteristic is adjusted.

Further, as described in the eleventh to sixteenth embodiments, if the original image is, for example, an image picked up by an endoscope, the number of pixels of the solid-state image pickup element based on the model of the endoscope used, etc. Information may be used to control M '(u, v).

Also, in the configuration of the sharpness correction circuit 263 described in the sixteenth embodiment, the filtering unit 2 is also included.
It is obvious that the configuration of 61 can be realized by the cascade connection of the filtering execution units by the plurality of FIR filters.

Although the number of the division filters is two in the present embodiment, it is also possible to employ a constitution with three or more division filters. For example, FI with a mask size of 13 × 13
The spatial frequency characteristic of the R filter is realized by applying three FIR filters having a mask size of 5 × 5 in a cascade connection.

To realize the sharpness correction with the configuration shown in FIG. 3, the series of processes or part of the processes described above may be operated in the central processing unit 40 in FIG.

Next, the eighteenth embodiment will be described. FIG.
FIG. 7 is a block diagram showing the configuration of the filtering unit according to the 18th embodiment.

The eighteenth embodiment is almost the same as the seventeenth embodiment, and the difference between the seventeenth embodiment and the seventeenth embodiment is the configuration of the filtering unit 201 of the sharpness correction circuit 165. Therefore, only the different configuration will be described. However, the same components are denoted by the same reference numerals and the description thereof is omitted.

There may occur a problem depending on the word length of the input data of the hardware device used for executing the filtering by the FIR filter. For example, the 17th
In FIG. 46 of the embodiment, it is assumed that the word length of the input data of the hardware device is 8 bits, that is, 0 to 255. Further, the input RGB original image data also takes values of 0 to 255. The coefficient values of the first FIR filtering execution unit 310 are all 0 or more and 1
If the sum is equal to or less than 1 and the sum of coefficient values is 1, the value of each R′G′B ′ image data input to the second filtering execution unit 311 also takes a range of 0 to 255. However, depending on the coefficient of the first FIR filtering execution unit 310, if the word length of the output data of the hardware device is sufficiently long with respect to the data of the filtering processing result, the first FIR filtering execution units 310r to 310b.
The processing result output from 0 to 0 is permitted as an input to the second FIR filtering execution unit 311r or 311b.
It is considered that the range of 255 is exceeded.

Therefore, in the present embodiment, the sharpness correction circuit 1
The configuration of the filtering unit 201 of FIG.
It is configured as shown in. That is, the first filtering execution unit 330r and the second filtering execution unit 332r and b that perform filtering by the first FIR filter and the first filtering execution unit 330r and 330b that execute filtering by the second FIR filter. A data conversion unit 331 composed of the data conversion units 331r and 331b is provided between the second filtering execution unit 332.

The data conversion units 331r-b round, truncate or compress the word length of the output data from the first filtering execution units 330r-b so as to match the word length of the input data to the second filtering units 332r-b. Etc. are processed. For realizing the data conversion units 331r or 331b, for example, an LUT (look-up table) using a RAM (read-only memory) may be used. The processing results of the second filtering units 332r and 332b are input to the data converting units 211r and b, respectively, and post-processing is performed to convert data having a value exceeding the display range by filtering. That is, the data after the decimal point is rounded off, or less than 0 or 2 when the displayable range is 0 to 255, for example.
Truncate values of 56 or more. Data conversion unit 211
Each of r and b may be configured by a LUT (look-up table) formed by a ROM (read-only memory), for example.

As described above, in the eighteenth embodiment, the hardware cost reduction is realized by the cascade connection configuration of the FIR filters while solving the problem concerning the limitation of the data distribution range depending on the input word length of the hardware device. .

In order to realize the sharpness correction with the configuration in FIG. 3, the series of processes or part of the processes described above may be operated in the central processing unit 40 in FIG.

Next, a nineteenth embodiment will be described. FIG.
8 is a block diagram showing a configuration of a filtering unit according to the nineteenth embodiment.

The nineteenth embodiment is almost the same as the seventeenth embodiment. Since the difference between the seventeenth embodiment and the seventeenth embodiment is the configuration of the filtering unit 201 of the sharpness correction circuit 165, only the different configuration will be described. The same components are designated by the same reference numerals and the description thereof will be omitted.

There may be a problem depending on the word length of the input data of the hardware device used for executing the filtering by the FIR filter. For example, the 17th
In FIG. 46 of the embodiment, it is assumed that the word length of the input data of the hardware device is 8 bits, that is, 0 to 255. Further, the input RGB original image data also takes values of 0 to 255. If the coefficient values of the first FIR filter are all 0 or more and 1 or less, and the sum of the coefficient values is 1, the R'G'B 'image data of R'G'B' to be input to the second filtering means is input. The value is also 0
It takes a range of 255.

However, depending on the coefficient of the first FIR filter, if the word length of the output data of the hardware device is sufficiently long with respect to the data of the filtering processing result, the first filtering execution units 310r to 310b.
The processing result output from 0 to 255 that is allowed as an input to the second filtering execution unit 311r or 311b.
It is considered that the range is exceeded.

Therefore, in this embodiment, the filtering section 201 is constructed as shown in FIG.

That is, in FIG. 48, the second FI
The filtering execution means by the R filter is composed of the second filtering execution units 335r1 and r2 to b1 and b.
It consists of 2. That is, two hardware devices are provided as second filtering execution means for each RGB data. The second filtering execution units 335r1 and r2 through b1 and b2 include synthesis units 336r through b for inputting the processing results.

For each of the second filtering execution units 335r1 and r2 or b1 and b2, the lower 8-bit data R of the filtering processing result R'G'B 'image data by the first filtering execution unit 334r or b. ′ L, G′l, B′l and the remaining high-order bit data R′h, G′h, B′h are input, respectively. The division into the low-order and high-order bit data is realized by the distribution units 341r and b provided in the subsequent stages of the first filtering execution units 334r and b. Then, the filtering by the second FIR filter is executed for each, and the processing results of the lower and upper bits of the R ″ G ″ B ″ image data are input to the combining units 336r and 336b in the subsequent stage.

In the synthesizing sections 336r and 336b, the processing results for the lower and upper bits are synthesized to generate the second FI.
This is the processing result by the R filter. The processing results by the combining units 336r and 336b are the data conversion units 337r and 337b.
Post-processing is performed to convert the data that is input to each and is converted into the data having the value exceeding the display range by filtering. That is, the data below the decimal point is rounded off, or the value less than 0 or 256 or more when the displayable range is, for example, 0 to 255 is rounded down. The data conversion means 336r or 336b may be configured by, for example, an LUT (look-up table) using a ROM (read-only memory).

Also, the processes in the combining units 336r and 336b and the data converting units 337r and 337b are collectively LUT.
It may be configured to be executed by (not shown).

As described above, the hardware cost can be reduced by the cascade connection configuration of the FIR filters while preventing the accuracy reduction of the filtering result data by the first FIR filter due to the limitation of the data word length.

In order to realize the sharpness correction with the configuration in FIG. 3, the series of processes or part of the processes described above may be operated in the central processing unit 40 in FIG.

Next, a twentieth embodiment will be described. FIG.
9 is a block diagram showing a configuration of a filtering unit according to the twentieth embodiment.

The twentieth embodiment is almost the same as the seventeenth embodiment, and the difference from the eleventh embodiment of the present embodiment is the configuration of the filtering unit 201 of the sharpness correction circuit 165. Therefore, only the different configuration will be described. The same components are designated by the same reference numerals and the description thereof will be omitted.

In the eleventh to nineteenth embodiments, the filtering section 201 has a three-system parallel configuration for each RGB image data.

In the present embodiment, the filtering means 3
01 is realized by a processing configuration of one system by applying a series of processing in time series for each RGB image data, and the hardware cost is reduced.

As shown in FIG. 49, the filtering unit 201 of the sharpness correction circuit 165 of this embodiment executes filtering with the selection unit 338 which selects any one of the input RGB original image data. The filtering execution unit 339 and the data conversion unit 340 that adjusts the value of the processing result by the filtering unit 339.

Next, the operation of the filtering section 201 in this embodiment will be described. The selection unit 338 is connected to the control unit 304 and selects any one of the input RGB original image data.

In the filtering execution unit 339,
The original image data of any one of RGB selected by the selection unit 338 is input and filtering is applied.

The processing result of the filtering execution unit 339 is input to the data conversion unit 340, and post-processing for converting data having a value exceeding the display range by filtering is executed. That is, the data that can be displayed after the decimal point is rounded off,
When the value is 255, the value less than 0 or 256 or more is truncated. Each image R'or G'or B'after the application of this processing is output as the processing result. The data conversion unit 340 is an L (read only memory) ROM, for example.
You may comprise by UT (look-up table).

When the series of processes for any one of the selected RGB image data is completed, the selecting section 338 finishes.
Select the next image data.

In addition, the coefficient setting section 202 transmits the coefficient of the correction filter corresponding to any one of the selected RGB image data to the filtering execution section 339.

In the above, the filtering unit 201 in the eleventh embodiment has been described as an example. Also for the twelfth to nineteenth filtering units 201,
Similarly, by applying a series of processes in chronological order for each of the RGB image data, it is possible to realize with a single-system processing configuration.

Next, a twenty-first embodiment will be described.

In the following embodiments, the image processing section 47 of the central processing unit 40 provided in the image processing apparatus 3 shown in FIG.
In, an example in which the functional configuration is changed will be described.

The color of the image (reproduced image) reproduced by the image recording device 4 and the color of the image reproduced on the observation monitor 8 may differ in color specification values defined by CIE. By reducing this color difference (color difference), color reproduction can be improved. Therefore, the twenty-first embodiment shows an example of the configuration of the color conversion means capable of improving the color tone by performing color conversion processing on the image signal input to the image recording device 4.

FIG. 50 is a block diagram showing the conceptual configuration of software of the image processing unit in the central processing unit, FIG. 51 is a block diagram showing the functional configuration of the color conversion means according to the 21st embodiment, and FIG. 52 is the 21st embodiment. 53 is a flowchart for explaining the operation of the color conversion means in FIG.
FIG. 54 is a flow chart showing an example of a gamut compression method, FIG. 54 is a gamut compression method, and FIG. 54 is a flow chart showing an example of a gamut compression method in the (L ^* u ^* v ^* ) color space. FIG. 56 is an explanatory diagram for explaining the concept in the CIE1976 (L ^* u ^* v ^* ) color space, and FIG. 56 is a conceptual diagram for explaining the configuration of a device-independent color conversion system.

In the image processing section 47 of the central processing unit 40, as shown in FIG. 50, the color of the image reproduced by the endoscope image recording device 4 matches the color of the image displayed on the observation monitor 8. Color conversion means 350 for performing color conversion so as to obtain a color desired by the observer, and / or sharpness correction means 351 for performing sharpness correction processing of an image reproduced by the endoscopic image recording device 4. And / or device characteristic correction means 352 for absorbing the variation in the color reproduction characteristic of the endoscopic image recording device 4 and the like.

In FIG. 50, the color conversion means 350 is
Although it is located upstream of the sharpness correction unit 351, this position can be reversed. Further, either one of the color conversion means 350 and the sharpness correction means 351 can be used. In this case, the device characteristic correction unit 352 is located immediately after the color conversion unit 350 or the sharpness correction unit 351.

In the image processing apparatus 3 of FIG. 3, the image data input through the image input I / F 43 is temporarily stored in the main storage device 42. Further, after the RGB data forming the image data is read from the main storage device 42,
Further transferred to the central processing unit 40, the central processing unit 40
The color conversion processing is executed by the color conversion means 350 provided in the image processing unit 47. The color conversion result is temporarily stored in the main storage device 42 and then output to the image recording device 4 via the image recording device I / F 46.

Next, a schematic configuration of the color conversion means 350 will be described as a specific example of the image processing section 47. FIG. 51 is a diagram showing a software configuration of the color conversion means 350 corresponding to the program stored in the ROM 45 when the color conversion processing is executed by the central processing unit 40 of FIG.

The color conversion means 350 uses the RGB data R,
CIE1 which is a numerical value representing a color displayed on the observation monitor 8 when G and B are input to the observation monitor 8.
976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* are converted and output, and CIE1976 (L ^* u ^*) output from the color space coordinate conversion means 400 ^.
v ^* ) mapping operation for compressing the color gamut of the observation monitor 8 to the color space coordinates L ^* , u ^* , v ^* until the color gamut of the observation monitor 8 is included in the color gamut of the image recording device 4. Do CIE1
976 (L ^* u ^* v ^* ) color space coordinates L ^* ′, u ^* ′, v ^* ′, and CIE1976 (L ^* u ^* v ^*) output from the color gamut compression unit 401 ^. ) With respect to the color space coordinates L ^* ′, u ^* ′, v ^* ′, the color perceived when the output image of the image recording device 4 is observed is CIE1976 (L ^*).
u ^* v ^* ) inverse color space coordinate transformation for outputting input signal RGB data R ′, G ′, B ′ to the image recording device 4 to obtain L ^* ′, u ^* ′, v ^* ′ in color space coordinates. Means 40
2 and.

The RGB data is described below as being standardized to 8 bits, but the standardization number is not limited to 8 bits, and may be more or less than 8 bits. .

The operation of the twenty-first embodiment will be described below centering on the color conversion means 350.

As shown in FIG. 51, in the color space coordinate conversion means 400, the input RGB data R, G, B is represented by C, which represents the color perceived on the observation monitor 8.
Convert to the IE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* . FIG. 52 shows a flowchart for explaining the operation of the color conversion means 350.

An outline of the CIE1976 (L ^* u ^* v ^* ) color space will be described below.

It is necessary to quantify the color sense in order to accurately transmit the color sense, which is a pure perceptual amount, to the outside. Attempts to quantify color perception have been made in the past, and several methods have been proposed. Among them, the CIE color system proposed by the International Commission on Illumination (CIE) is one of the most commonly used quantification methods. Here, CI
CIE1976 (L ^* u defined in E color system
^* v ^* ) uniform color space (hereinafter, CIE1976 (L ^* u
^* v ^* ) A method for handling a color space) will be described.

The CIE1976 (L ^* u ^* v ^* ) color space is
Pixel measurement lightness L ^* , which represents the psychological brightness of the object surface, and three axes that are orthogonal to u ^* and v ^* that form the color plane. The Euclidean distance between any two points in the CIE1976 (L ^* u ^* v ^* ) color space has a strong correlation with the color difference. Therefore, the CIE1976 (L ^* u ^* v ^* ) color space can be regarded as a color space with good uniformity in terms of color difference (uniform color space).

The RGB data R, G, B to CIE 1976 (L ^*) which is an internal operation of the color space coordinate conversion means 400 ^.
u ^* v ^* ) A coordinate conversion method into color space coordinates L ^* , u ^* , v ^* will be described below.

In the color space coordinate conversion means 400, as shown in FIG. 52, first, in step S101, the RGB data R, G, B are converted to CIE (XYZ) tristimulus values X, Y ,.
Convert to Z. Then, in step S102, the CI
From E (XYZ) tristimulus values X, Y, Z to CIE1976
(L ^* u ^* v ^* ) conversion to color space coordinates L ^* , u ^* , v ^* .

The conversion from the RGB data R, G, B to the CIE (XYZ) tristimulus values X, Y, Z in step S101 is performed by the matrix operation of a cubic matrix of the third order as shown in equation (34). .

[0413]

(Equation 8) Here, gmR (R), gmG (G), gmB (B) are
RGB data R, G, B are observed by the γ characteristic monitor 8
Is a value converted into the emission brightness of each of the R, G, and B phosphors. The matrix elements in the equation (34) are determined from the xy chromaticity coordinates of the RGB primary colors on the observation monitor 8.

Further, the conversion in step S101 is
In addition to the method based on the calculation formula defined by formula (34), RG
There is also a method using a reference table (lookup table, hereinafter abbreviated as LUT) from B to XYZ. Below R
The CIE (X
YZ) A method of creating an LUT that outputs tristimulus values X, Y, Z will be described.

Observation monitor 8 for RGB data R, G, B
To present a uniform color patch with an area. Using the color light of this color patch as a target, CIE (XYZ) tristimulus value measurement using a color luminance meter, etc.,
Record X, Y, Z. By repeating the above procedure in the entire color space composed of RGB data (hereinafter, referred to as an observation monitor signal color space), the CIE (XYZ) tristimulus value X with the RGB data R, G, B as reference addresses. , Y, Z
It is possible to create an LUT that outputs

In general, preparing a combination of RGB data and CIE (XYZ) tristimulus values of the number which completely fills the observation monitor signal color space is to measure the CIE (XYZ) tristimulus value measurement time and LUT. It is not realistic from the viewpoint of the amount of data. Therefore, the observation monitor signal color space is sampled at appropriate intervals to create an LUT that outputs CIE (XYZ) tristimulus values only for the RGB data located at the grid points. When RGB data corresponding to an address other than the grid point is input to the LUT, it is obtained by interpolation from the output values of the neighboring sampling points.

At step S101, the RGB data R, G,
After being converted from B to CIE (XYZ) tristimulus values X, Y, Z, the CIE (XYZ) tristimulus values X, Y, Z are CIE1976 (L ^* u ^* v ^* ) color space coordinates in step S102. Converted to L ^* , u ^* , v ^* . This step S102
The conversion in can be performed by the conversion formula shown in Formula (2). In the equation (2), Yn, u'n, v'n are the luminance of the reference white and CIE1976UCS chromaticity coordinates u ', v', respectively.

As described above, the LUT for outputting the CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* is created by using the RGB data R, G, B as reference addresses in advance. The operation of the color space coordinate conversion means 400 in steps S101 and S102 can be executed by one LUT.

Next, in the color gamut compression means 401, as shown in FIG.
CIE1976, as shown in step S103 of 2.
From (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* to CIE19
Conversion to 76 (L ^* u ^* v ^* ) color space coordinates L ^* ′, u ^* ′, v ^* ′ is performed. The internal operation of the color gamut compression means 401 will be described with reference to FIGS.

FIG. 53 shows a range of colors reproducible on the observation monitor 8 (hereinafter referred to as an observation monitor gamut) and a range of colors reproducible on the reproduced image by the image recording device 4 (hereinafter referred to as an image. (Referred to as a recording device color gamut) is referred to as CIE197.
6 (L ^* u ^* v ^* ) is a schematic representation on a cutting plane in the color space, where (a) shows the L ^* -u ^* plane, and (b) shows the u ^* -v ^* plane. ing.

As shown in FIG. 53, when the color gamut of the image recording device is narrower than the color gamut of the monitor for observation, among the colors displayed on the monitor 8 for observation, the color located outside the color gamut of the image recording device is In the image recording device 4, CIE1976 (L ^* u ^* v ^* )
It cannot be expressed in color space coordinates. In addition, the color space coordinate reverse conversion means 402 located after the color gamut compression means 401
In the inverse conversion in, the conversion accuracy is guaranteed within the color gamut of the image recording device, but the CIE1976 (L ^* u ^* v ^* ) color space coordinates input to the color space coordinate inverse conversion means 402 are outside the image recording device color gamut. In the case of a color, there is a possibility that the color will be converted into a color extremely different from the surrounding color.

Therefore, for a color outside the color gamut of the image recording device, it is necessary to move to a region in the CIE1976 (L ^* u ^* v ^* ) color space where the color space coordinate inverse conversion means 402 guarantees conversion accuracy. In addition, CIE1976 (L ^* u ^* v
^* ) When moving the position in the color space, it is desirable to maintain the gradation and color tone as much as possible. Therefore,
It is necessary to perform a mapping operation (hereinafter referred to as color gamut compression) so that the monitor color gamut for observation is included in the color gamut of the image recording apparatus without changing the gradation and color tone as much as possible.

An example of such a color gamut compression method will be described with reference to FIGS. 54 and 55. 54 is a flowchart of a data conversion method by gamut compression, and FIG. 55 is a conceptual diagram of gamut compression.

First, as shown in step S106 in FIG. 54, a reproduction range in the psychological measurement lightness L ^* direction (hereinafter,
L ^* dynamic range) compression. The conversion formula at this time is shown in Formula (35).

[0425]

[Equation 9] In the equation (35), L ^* MAX and L ^* MIN are L ^* maximum and minimum values of the observation monitor 8, and this width is the width of the observation monitor 8.
L ^* dynamic range. On the other hand, L ^* max, L ^*
min is the L ^* maximum value and the minimum value of the image recording device 4, and this width is the L ^* dynamic range of the image recording device 4.
Here, L ^* represents the psychological measurement lightness before conversion, and L ^* s represents the psychological measurement lightness after conversion.

[0426] As can be seen from FIGS. 53 and 55, the narrower general the observation monitor 8 L ^* of dynamic range image recording apparatus than 4 L ^* dynamic range. In such a case, the above L ^* dynamic range compression is performed,
It is necessary to adjust the brightness gradation.

As shown in FIG. 55, as the L ^* dynamic range is compressed, the observation monitor color gamut becomes the color gamut after the L ^* dynamic range compression. As the color gamut moves, C
The IE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* are the CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* s, u ^* s,
Go to v ^* s.

Next, in step S107, conversion to VHC by cylindrical coordinate conversion, that is, lightness, hue, and saturation is performed. The coordinate conversion in this step S107 is performed by the equation
This can be performed by the conversion formula shown in 5).

Then, in step S108, saturation is compressed. It is generally known that the color balance is less disrupted when the color gamut is compressed by moving the saturation than when the color gamut is compressed by moving the hue. Therefore, here, the color gamut is compressed by maintaining the hue and compressing the saturation. The conversion formula at this time is shown in Formula (36).

[0430]

[Equation 10] In the equation (36), Cmax is the maximum saturation at H and V in the color gamut of the image recording device. C represents the saturation before the saturation compression, and C ′ represents the saturation after the saturation compression.

Note that the saturation conversion formula shown in formula (36) is a linear conversion, but in order to minimize the movement of colors included in the color gamut of the image recording apparatus, an appropriate nonlinear function is used. It is also possible to convert by.

Finally, in step S109, L ^* ', u ^* ', v ^* 'are obtained from VHC' by the inverse transformation of equation (25). As shown in FIG. 55, the color gamut after L ^* dynamic range compression finally matches the image recording device color gamut by saturation compression, and the CIE1976 (L ^* u ^* v ^* ) color space coordinate is L ^*. ′, U ^* ′, v ^* ′.

Next, the internal operation of the color space coordinate reverse conversion means 402 will be described. The CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* ′, u ^* ′, v ^* ′ output from the color gamut compression means 401 are color space coordinate inverse transformation means 402.
In, the data is converted into RGB data R ′, G ′, B ′ which is input to the image recording device 4.

In the color space coordinate reverse conversion means 402, as shown in FIG.
In S104 in the first, first, CIE1976 (L ^* u
^* v ^* ) color space coordinates L ^* ′, u ^* ′, v ^* ′ to CIE (X
YZ) conversion to tristimulus values X ', Y', Z ', and subsequently, in S105, CIE (XYZ) tristimulus values X', Y ',
Conversion from Z'to RGB data R ', G', B'is performed. The conversion in step S104 may be the inverse conversion of the conversion equation shown in equation (2).

[0435] In step S105, CIE (XY
Z) RGB data R'from tristimulus values X ', Y', Z ',
The conversion into G ′ and B ′ may be performed by using a method using an LUT or a method using a color prediction formula based on a color reproduction model of the image recording device 4. The conversion method using the LUT will be described below.

In order to convert the CIE (XYZ) tristimulus values into RGB data, R input to the image recording device 4 is used.
An LUT that outputs tristimulus values XYZ using GB data as an input address is created in advance. First, CIE (XY
Z) tristimulus values X ′, Y ′, Z ′ and LUT output value X ″,
The minimum value search is performed under the condition of the minimum Euclidean distance of Y ″ and Z ″. As a result of the search, the input address R ″,
In the area having an appropriate width centering on G "and B", the first
The detailed search with a width smaller than the search interval is performed under the minimum condition. The input address obtained as a result is R ′, G ′,
B '. In this way, L for estimating the CIE (XYZ) tristimulus value of the output image of the image recording device 4 from the RGB data
By using the UT to perform a minimum search, the CIE
Conversion from (XYZ) tristimulus values to RGB data is performed.

The LUT for the inverse conversion changes the RGB data input to the image recording device 4 to output an image, and the CIE (XYZ) of the output image (hereinafter referred to as a color chart) by a color luminance meter. ) Measuring tristimulus values for example about 5
By performing 000 sets of data, it can be created by filling a set of RGB data and CIE (XYZ) tristimulus values at an appropriate interval in the RGB data color space.

At this time, the color space coordinate conversion means 400
Similar to the LUT creation method used in, the data other than the RGB data used to create the LUT is used as the LU address.
When input to T, the CIE (XYZ) tristimulus value can be obtained by interpolation calculation using neighboring sampling points.

Also, in the color space coordinate reverse conversion means 402, RGB data is used as an input address and CIE1976.
An LUT for outputting (L ^* u ^* v ^* ) color space coordinates L ^* ′, u ^* ′, v ^* ′ is created in advance, and steps S104 and S1 are performed.
It is also possible to execute 05 with one LUT.

As described above, in the color space coordinate conversion means 400, the input RGB data R, G, B is input.
Is C, which represents the reproduced color on the observation monitor 8.
IE1976 (L ^* u ^* v ^* ) color space coordinates are converted into L ^* , u ^* , v ^* , and color gamut compression is performed by the color gamut compression means 401 to obtain C.
The IE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* are converted to CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* ′,
CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* ′, u which are converted into u ^* ′, v ^* ′ and output from the color gamut compression unit 401.
^* ′, V ^* ′ is RG by the color space coordinate inverse transformation means 402.
B data R ', G', B'converted, and this RG
By inputting the B data R ', G', B'instead of the RGB data R, G, B to the image recording device 4, the image reproduced on the observation monitor 8 and the image recording device 4 are reproduced. The color difference from the image is improved.

As shown in this embodiment, by using the CIE color system, the observation monitor 8 and the image recording device 4 are provided.
It is possible to handle colors uniformly among different devices such as. FIG. 56 shows the concept of such a device-independent color conversion system.

The device A drive signal color space corresponds to the observation monitor color space input to the observation monitor 8 in this embodiment. The device B drive signal color space corresponds to the image recording device color space input to the image recording device 4 in this embodiment. These color spaces are color expressions unique to the device, and even if the color space coordinates are the same, they do not always represent the same color. This is because the device A and the device B have different device characteristics.

Therefore, a color space (hereinafter referred to as a standard color space) unrelated to the device characteristics and the device-specific color reproduction method.
It is possible to handle colors in the standard color space by preparing a color space conversion profile describing the conversion method of and for each device and operating this color space conversion profile.

The color space profile conversion means 400 and the color space coordinate inverse conversion means 402 correspond to the color conversion profile in the above-described embodiment. By configuring the color converting means 350 as shown in FIG. 56, even if the device characteristics of the observing monitor 8 and the image recording device 4 change, a color space conversion profile corresponding to the device characteristics is prepared. It is possible to respond flexibly without making a large change to the whole.

In this embodiment, the CIE1976 (L ^* u ^* v ^* ) color space is used as the standard color space, but the CIE1976 (L ^* a ^* b ^* ) color space and the CIE Even when, for example, a perceptual color space such as the HSV or HLS color space composed of the three attributes of the above-mentioned color, that is, the three axes of hue, saturation, and lightness is used instead of the color space to be estimated The color difference between the image reproduced on the observation monitor 8 and the output image of the image recording device 4 can be improved.

Next, a twenty-second embodiment will be described. Figure 5
7 is a block diagram showing the functional configuration of the color converting means according to the 22nd embodiment, FIG. 58 is a block diagram showing the internal configuration of the color correction effect adjusting means provided in the 22nd embodiment, and FIG.
FIG. 7 is a conceptual diagram illustrating the operation of the color correction effect adjusting means.

Color converting means 350 in the twenty-second embodiment.
51 is substantially the same as that of the twenty-first embodiment shown in FIG. 51, and the main difference from the twenty-first embodiment is that the color correction effect adjusting means 403 is provided after the gamut compression means 401. . In the present embodiment, the configuration and operation other than the color correction effect adjusting means are the same as those of the 21st embodiment, so the description of the similar parts will be omitted and only the different parts will be described.

As described in the twenty-first embodiment, although the color tone is improved by the color conversion means 350, the color sensation is extremely subjective and the color conversion designed for a standard observer. It is conceivable that there is an observer who cannot tolerate the color correction result. It is also conceivable that there may be an observer who shows a relatively good evaluation of the output result without performing the color conversion, rather than the color conversion result performed by the configuration shown in FIG. Therefore, in the 22nd embodiment, the 21st
In addition to the effect of the embodiment, a configuration example of the color conversion means capable of adjusting the color correction effect with a simple configuration and capable of outputting the color conversion result intended by the observer is shown.

FIG. 57 shows the schematic arrangement of the color conversion means in the 22nd embodiment. The color conversion unit 350 is an L output from the color gamut compression unit 401 after the color gamut compression unit 401.
CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* c by adjusting the correction effect for ^* ′, u ^* ′, v ^* ′ according to the correction effect parameter Ce input from the information input device 41. , U ^* c, v ^* c are output from the color correction effect adjusting means 403.

The operation of the 22nd embodiment will be described below with reference to FIG.
8 and FIG. 59, the color correction effect adjusting means 403 will be mainly described.

FIG. 58 shows the internal structure of the color correction effect adjusting means 403. The RGB data R, G, B are RG'd to the image recording device 4 in the image recording device output color estimating means 405.
CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* b, u where the B data R, G, B are input and the output color is represented.
It is converted into ^* b and v ^* b. And CIE1976 (L ^*
The u ^* v ^* ) color space coordinates L ^* b, u ^* b, v ^* b are input to the interior dividing point calculation means 404. The internal dividing point calculation means 404 outputs the CIE1976 output from the color gamut compression means 401.
(L ^* u ^* v ^* ) color space coordinates L ^* ′, u ^* ′, v ^* ′ and the CIE 19 output from the image recording device output color estimation means 405.
Using the 76 (L ^* u ^* v ^* ) color space coordinates L ^* b, u ^* b, v ^* b and the correction effect adjustment parameter Ce output from the information input device 41, (L ^* ′, u ^* ′) , V ^* ′) and (L ^* b,
The internal division point (L ^* c, u ^* c, v ^* c) of u ^* b, v ^* b) is calculated based on the internal division ratio Ce.

FIG. 59 shows that (L ^* ', u ^* ',
v ^* '), (L ^* b, u ^* b, v ^* b), (L ^* c, u ^* c,
The positional relationship of the three points of v ^* c) is CIE1976 (L ^* u
^* v ^* ) shown in the color space.

When the correction effect parameter Ce is 1.0, the internally dividing point (L ^* c, u ^* c, v ^* c) is (L ^* ',
u ^* ′, v ^* ′), and the correction effect is maximized. on the other hand,
When Ce is 0.0, the internal division points (L ^* c, u ^* c, v ^* c
) Matches (L ^* b, u ^* b, v ^* b), and the result is the same as when no color conversion is performed.

The image recording device output color estimation means 405.
For the internal operation in CIE, from the RGB data in the color space coordinate reverse conversion means 402 of the twenty-first embodiment, to CIE.
(XYZ) tristimulus values, and CIE1976 (L ^* u
^* v ^* ) Similar to the method of estimating color space coordinates, CIE1976 from RGB data by a LUT or the like created in advance.
Conversion to (L ^* u ^* v ^* ) color space coordinates is performed.

As described above, the color correction effect adjusting means 403
Is provided between the color gamut compression means 401 and the color space coordinate inverse transformation means 402, in addition to the effects of the twenty-first embodiment, the single parameter Ce input from the information input device 41 allows simple and uncorrected correction. It is possible to adjust the color correction effect between the state and the color conversion result obtained by the configuration of the twenty-first embodiment, and it is possible to deal with an observer who cannot accept the color conversion result by the standard setting.

Next, a twenty-third embodiment will be described. Figure 6
0 is a block diagram showing the functional configuration of the color converting means according to the 23rd embodiment, and FIG. 61 is a conceptual diagram for explaining the operation of the light source effect adjusting means provided in the 23rd embodiment.

Color conversion means 350 in the twenty-third embodiment.
Is substantially the same as that of the twenty-first embodiment shown in FIG. 51. The main difference from the twenty-first embodiment is that the light source effect parameter Le is input from the information input device 41 and the color space coordinate inverse conversion means 402 is input. The light source effect adjusting means 406 for rewriting the LUT used in the above is added. In the present embodiment, the configuration and operation other than the light source effect adjusting means are the 21st
Since it is the same as the embodiment, the description of the similar parts is omitted,
The operation of the light source effect adjusting means will be mainly described.

Color space coordinate inverse transformation means 40 shown in FIG.
From the RGB data used in 2 CIE (XYZ)
The LUT for estimating the tristimulus value is a light source when observing the output image of the image recording device 4, and is created by measuring the color chart with a color luminance meter.

However, it is conceivable that the light source for actually observing the image output from the image recording device 4 is different from the light source used for colorimetry of the color chart for creating the LUT. In such a case, measure the color chart under a new light source,
Although it is necessary to recreate the LUT, it is not realistic to recreate the LUT each time when trying to support a large number of light sources. Further, even if LUTs corresponding to many light sources are created in advance, if the number of set light sources increases, the LUT
Also increases the memory area for storing.

Therefore, in the twenty-third embodiment, in addition to the effects of the twenty-first embodiment, a simple light source effect adjusting method is used.
An example of the configuration of the color conversion means capable of coping with the change of the light source when observing the image output from the image recording device 4 is shown.

FIG. 60 shows a schematic structure of the color conversion means in the twenty-third embodiment. The color conversion means 350 receives the light source effect adjustment parameter Le from the information input device 41, creates an appropriate LUT according to this light source effect adjustment parameter Le, and rewrites the LUT used by the color space coordinate inverse conversion means 402. And a means 406.

The light source effect adjusting means 4 will be described below with reference to FIG.
The operation in 06 will be described. FIG. 61 shows the CIE (XYZ) tristimulus values of the color chart measured with a plurality of types of light sources in the XYZ space.

In FIG. 61, P1 (X1, Y1, Z1
) Is the CIE (XYZ) of the color chart measured with the first light source
The tristimulus value, P2 (X2, Y2, Z2), is the CIE (XYZ) tristimulus value of the color chart measured by the second light source, P3 (X3
, Y3, Z3) is the CI of the color chart measured with the third light source.
E (XYZ) tristimulus values. Here, as many sets of P1, P2, and P3 as the number of color chips used when creating the LUT are prepared.

Assuming that the light source color changes from the first light source to the third light source through the second light source, the locus of the tristimulus values is the movement distance from P1 → P2 → P3. L
It is expressed by the relational expression shown in Expression (37) using e (light source effect adjustment parameter) as a parameter.

Xi = fiX (Le) Yi = fiY (Le) (37) Zi = fiZ (Le) Here, Xi, Yi, and Zi are designated by the light source effect adjustment parameter Le of the i-th color chart. 3 represents an approximation of tristimulus values measured under a new light source. fi X (Le), fi Y (L
e) and fi Z (Le) are loci using the light source effect adjustment parameter Le in the i-th color chart as a variable.

By previously storing the locus representing such a light source change as a function with the data of the form shown in the equation (37), the light source effect adjustment parameter Le can be designated to change the first light source to the third light source. It is possible to obtain color chart data that approximates the measurement result when the color chart is measured with the virtual light source between the light sources. Once the color chart data is prepared, as described in the twenty-first embodiment, the LUT for estimating the CIE (XYZ) tristimulus values of the output image of the image recording device 4 from the RGB data.
Can be created.

Depending on the approximation accuracy, the function f
As the simplest form of i X (Le), fi Y (Le), and fi Z (Le), it is conceivable to calculate the internal division point of tristimulus values measured under two light sources. In this case, the internal division ratio becomes the light source effect adjustment parameter Le.

The light source effect adjusting means 406 inputs the light source effect adjusting parameter Le output from the information input device 41, and based on the light source effect adjusting parameter Le, the formula (3)
The color chart data is newly created by 7). Then, the LUT for estimating the tristimulus value of the color chart under the virtual light source set by the light source effect adjustment parameter Le from the RGB data
Is created and replaced with the LUT used in the color space coordinate inverse transformation means 402.

As described above, by adding the light source effect adjusting means 406, in addition to the effect in the twenty-first embodiment, the color converting means 350 can be set by the light source effect adjusting parameter Le input from the information input device 41. It is possible to deal with the change of the light source when observing the output of the image recording device 8 without re-measurement of the color chart and only by providing a memory for the function data shown in Expression (37). Becomes

Next, a twenty-fourth embodiment will be described. Figure 6
2 is a block diagram showing the functional configuration of the color conversion means according to the twenty-fourth embodiment, and FIG. 63 is a conceptual diagram explaining the operation of the color adaptation conversion means provided in the twenty-fourth embodiment.

Color converting means 350 in the twenty-fourth embodiment.
Is substantially the same as that of the twenty-first embodiment shown in FIG. 51, and the main difference from the twenty-first embodiment is that the chromatic adaptation conversion means 407 is added. In the present embodiment, the configuration and operation other than the chromatic adaptation conversion means are the same as those of the twenty-first embodiment, so description of similar parts will be omitted, and the operation of the chromatic adaptation conversion means will be mainly described.

The white perceived by the observer on the observation monitor 8 and the white perceived when the output image of the image recording device 4 is observed under a halogen light source are the same white under each observation environment. feel. Such a visual mechanism is called chromatic adaptation. In other words, there is a corresponding point in the color space where the same color is perceived in each observation environment due to the visual mechanism called chromatic adaptation. If these corresponding points are known, it is possible to perform color correction by positively utilizing the difference in the observation environment and the visual mechanism.

Therefore, in the twenty-fourth embodiment, in addition to the effects of the twenty-first embodiment, a mechanism for color adaptation is incorporated in the color conversion means so that the time when the observation monitor is observed and the image output from the image recording apparatus is A configuration example of a color conversion unit capable of performing color correction in consideration of a difference in observation environment when observing is shown.

FIG. 62 shows a schematic structure of the color conversion means in the twenty-fourth embodiment. The color conversion unit 350 receives the CIE1976 (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , v ^* output from the color space coordinate conversion unit 400 at a stage subsequent to the color space coordinate conversion unit 400, the L ^*, u ^*, v is the color adaptation corresponding points ^{* CIE1976 (L * u * v} *) color space coordinates ^{L * ', u *',} v * ' color adaptation for outputting the color gamut compression means 401 The conversion means 407 is included.

The operation of the twenty-fourth embodiment will be described below with reference to FIG. 63, centering on the color adaptation conversion means 407.

In FIG. 63, the corresponding points of chromatic adaptation are CIE197.
6 (L ^* u ^* v ^* ) color space. In FIG. 63, (L ^* b, u ^* b, v ^* b) is the CIE1976 (L ^* u ^* v ^* ) color space of an image obtained by inputting RGB data R, G, B into the image recording device 4. Coordinates.
(L ^* , u ^* , v ^* ) is CIE1976 (L ^* u) when RGB data R, G, B are input to the observation monitor 8.
^* v ^* ) Color space coordinates.

The operation of the color conversion means 350 of this embodiment is shown in FIG.
3 in the CIE1976 (L ^* u ^* v ^* ) color space. When color conversion is not performed, the color of the output image of the image recording device 4 is (L ^* b, u ^* b, v ^* b), and the color displayed on the observation monitor 8 is (L ^* b, u). A color difference corresponding to the Euclidean distance between ^* b, v ^* b) and (L ^* , u ^* , v ^* ) occurs. Therefore, the color correction is (L ^* b, u ^* b, v
^This is done with the goal of bringing ^* b) closer to (L ^* u ^* v ^* ).

[0478] In this case, when considering the color adaptation, color reproduction target is ^{(L *, u *, v} *) from ^{(L * ', u *'} , v * ') replaces the. The ^{(L * ', u *'} , v * ') , when the visual environment to observe the image output from the image recording apparatus 4 are sufficiently conformable, on observation monitor 8 (L ^*, u ^* ,
v ^* ) is a point that is perceived as the same color, and this point is called a corresponding color.

The color adaptation converter 407 outputs the CIE1976 (L ^* u) output from the color space coordinate converter 400.
^* v ^* ) Color space coordinates L ^* , u ^* , v ^* are ^assigned to the corresponding color L ^* ′,
It is converted into u ^* ′ and v ^* ′ and output to the gamut compression means 401.

A method for obtaining the corresponding color is described in, for example, Document 3 (Japan Society of Applied Physics, Optical Communication, Color Properties and Technology, pages 100 to 106), von-K.
It is possible to obtain an approximate corresponding color by a model such as a ries formula or a CIE prediction formula. Alternatively, the color space may be appropriately sampled, the corresponding color at each sampling point may be obtained by a subjective evaluation experiment, and the corresponding color other than the sampling point may be obtained by interpolation from surrounding sampling points.

As described above, the color space coordinate conversion means 400
By adding the color adaptation conversion means 407 between the color gamut compression means 401 and the color gamut compression means 401, in addition to the effect of the twenty-first embodiment, the observation conditions for observing the output images of the observation monitor 8 and the image recording device 4 respectively. It is possible to carry out color correction in consideration of the difference.

Next, a twenty-fifth embodiment will be described. Figure 6
4 is a block diagram showing the functional configuration of the color conversion means according to the twenty-fifth embodiment, and FIG. 65 is a conceptual diagram explaining the operation of the gamut compression center changing means provided in the twenty-fifth embodiment.

Color converting means 350 in the twenty-fifth embodiment.
Is substantially the same as that of the twenty-first embodiment shown in FIG. 51, and the main difference from the twenty-first embodiment is that the gamut compression center changing means 208 is added. In the present embodiment, the configuration and operation other than the gamut compression center changing means are the second one.
Since it is the same as that of the first embodiment, description of similar portions will be omitted, and the operation of the gamut compression center changing means will be mainly described.

In the above-described embodiment, the white reproduced on the observation monitor 8 and the white (the color of the normal light source) under the observation condition of the output image of the image recording apparatus 4 have a corresponding color relationship, and the color space The same coordinates are assumed above. However, depending on the observer, it is possible that the adaptation state is not perfect and the image on the observation monitor remains strong. In such a case, it is necessary to intentionally adjust the white point according to the observer's preference, rather than simply adjusting the white point. In addition, it is necessary to change the gamut compression center accordingly.

Therefore, in the twenty-fifth embodiment, in addition to the effects of the twenty-first embodiment, the white point is adjusted according to the preference of the observer, and the color gamut compression center is changed accordingly, so that a desired value can be obtained. An example of the configuration of a color conversion unit capable of performing color correction will be shown.

FIG. 64 shows a schematic structure of the color conversion means in the twenty-fifth embodiment. The color conversion unit 350 receives the compression center changing parameter Ctr input from the information input device 41 at the subsequent stage of the color space coordinate conversion unit 400, and outputs the CIE1976 (L ^* u ^* v) output from the color space coordinate conversion unit 400.
^* ) Perform a mapping operation on the color space coordinates L ^* , u ^* , and v ^{* according} to the change of the observation monitor color gamut after changing the compression center,
1976 (L ^* u ^* v ^* ) color space coordinates L ^* ′, u ^* ′, v ^* ′
The color gamut compression center changing unit 408 for outputting the color gamut compression unit 401 to the color gamut compression unit 401.

Hereinafter, the operation of the twenty-fifth embodiment will be described with reference to FIG.
The color gamut compression center changing unit 408 will be mainly described with reference to FIGS.

The compression center of color gamut compression is initially set to the achromatic color axis connecting Pmax and Pmin in FIG.
However, when adjusting the white point according to the observer's incomplete adaptation to the viewing environment or the observer's preference,
Along with that, the compression center connecting Pmax and Pmin must be changed.

A method of changing the gamut compression center will be described below. Here, the xy chromaticity coordinates of the reference white and black which are the adaptation fields of view of the observation monitor 8 are respectively defined as xmax p and ymax p.
, Xmin p, ymin p. In addition, the image recording device 4
The xy chromaticity coordinates of the reference white and black when observing the output image are defined as xmax q, ymax q, xmin q and ymin q, respectively. First, by using the compression center changing parameter Ctr received by the color gamut compression center changing unit 408 from the information input device 41, the reference white and black of the reference image when observing the output image of the image recording device 4 is expressed by the following equation (38). The xy chromaticity coordinates are converted into xmax n, ymax n, xmin n, and ymin n, respectively.

Xmax n = xmax q + Ctr · (xmax p −xmax q) ymax n = ymax q + Ctr · (ymax p −ymax q) xmin n = xmin q + Ctr · (xmin p −xmin q) (38) ymin n = yminq + Ctr. (yminp-yminq) This relational expression (38) is nothing but to find the internal division point designated by the internal division ratio Ctr.

Next, chromaticity coordinates xmax n, ymax n, xmi
From the luminances of nn, ymin n, Pmax and Pmin, QMmax and QMmin are obtained as shown in FIG. QMmax and QM
Letting QNmax and QNmin be the points at which the line segment connecting min intersects the color gamut of the image recording device, the line segment connecting QNmax and QNmin becomes the new gamut compression center.

In the color gamut compression center changing means 408, the mapping operation for converting the color gamut compression centers Pmax to Pmin into QNmax to QNmin is input from the color space coordinate converting means 400 to CIE1976 (L ^* u ^* v ^* ). Color space coordinates L ^* , u
CIE1976 after conversion by acting on ^* , v ^*
The (L ^* u ^* v ^* ) color space coordinates L ^* ′, u ^* ′, v ^* ′ are output to the color gamut compression means 401.

As described above, the gamut compression center changing means 40
8 is provided between the color space coordinate conversion unit 400 and the color gamut compression unit 401, and the color gamut compression center changing unit 4 is set by the compression center changing parameter Ctr input from the information input device 41.
CIE1976, which is output from the color space coordinate conversion unit 400, for the mapping conversion accompanying the change of the gamut compression center in 08.
By performing the operation on the (L ^* u ^* v ^* ) color space coordinates L ^* , u ^* , and v ^* , in addition to the effect of the twenty-first embodiment, the observer uses the color conversion means 350 to control the image recording device 4. If the output image observation environment is not completely adapted, or if it is desired to adjust the white point according to the preference of the observer, it is possible to easily cope with it.

Although the twenty-first to twenty-fifth embodiments have been described above, the addition of the color correction effect adjusting means 403 described in the twenty-second embodiment is not limited to the twenty-third and twenty-fourth embodiments.
It can be applied to the twenty-fifth embodiment, and the correction effect can be easily adjusted between the uncorrected state and the color conversion result obtained with the configuration shown in the twenty-first embodiment, and the color setting can be performed with the standard setting. It is possible to deal with an observer who cannot accept the conversion result.

Further, the addition of the light source effect adjusting means 406 described in the 23rd embodiment can be applied to the 22nd, 24th and 25th embodiments, and the color converting means 3
50 by using the light source effect adjustment parameter Le obtained from the information input device 41, without re-measurement of the color chart, and by only providing the memory for the function data shown in Expression (37). It becomes possible to deal with the change of the light source when observing the output of.

The addition of the color adaptation conversion means 407 described in the twenty-fourth embodiment is the same as that in the twenty-second and twenty-third embodiments.
The present invention can be applied to the twenty-fifth embodiment as well, and color correction can be performed in consideration of the difference in observation conditions when observing the output images of the observation monitor 8 and the image recording device 4.

The addition of the color gamut compression center changing means 408 described in the twenty-fifth embodiment is also applicable to the twenty-second, twenty-third and twenty-fourth embodiments, and the color conversion means 350. When the observer is not completely adapted to the output image observing environment of the image recording device 4, or when he or she wants to adjust the white point according to the observer's preference, it is possible to easily cope with the above.

Next, the twenty-sixth embodiment will be described. Figure 6
6 is a block diagram showing the functional configuration of the device characteristic correction means according to the 26th embodiment, and FIG. 67 is a flowchart for explaining the operation of the device characteristic correction means according to the 26th embodiment.

The twenty-sixth embodiment shows a first configuration example in which the device characteristic correction means 352 is added to the color conversion means 350 or the sharpness correction means 351 at the subsequent stage in the configuration of the twenty-first embodiment. is there. In the present embodiment, the configuration and operation other than the device characteristic correction means are the same as those of the 21st embodiment, so description of similar parts will be omitted and the device characteristic correction means will be mainly described.

The image processing apparatus 3 defines the standard output characteristics of the image recording apparatus 4 (hereinafter referred to as standard output characteristics),
Color conversion means 350 designed on the premise of standard output characteristics
Alternatively, the image quality correction video signal generated by the sharpness correction unit 351 is output to the image recording device 4.

However, the output characteristics of the image recording apparatus 4 are different in each machine even if they are of the same model. Due to the fact that the image reproduction characteristics of such equipment (hereinafter referred to as equipment characteristics) vary from machine to machine, stable image quality correction results cannot be obtained and the difference in output results from machine to machine is subjective. There may be cases where it is not acceptable.

Therefore, in the twenty-sixth embodiment, in addition to the effects of the twenty-first embodiment, the video signal to be input to the image recording apparatus is processed without changing the output characteristics of the image recording apparatus, and the unprocessed video is processed. An example of the configuration of a device characteristic correction means capable of correcting the output characteristic of the image recording apparatus with respect to the signal so as to approximate the standard output characteristic will be shown.

The color conversion means 350 in the image processing section 47.
FIG. 66 shows a schematic configuration of the device characteristic correction means 352 provided in the latter stage of the above. Device characteristic correction means 3 of the 26th embodiment
A reference numeral 52 designates a standard device characteristic reproducing means 411 for outputting RGB data as an input and outputting luminance data YR, YG, YB, and luminance data YR, YG, YB from the standard device characteristic reproducing means 411 for calculation. The gain adjusting means 412 for outputting the luminance data YR ', YG', YB 'and the luminance data YR', YG ', YB' from the gain adjusting means 412 are input to perform an operation to obtain RGB. The data R ', G', and B'are recorded on the image recording device 4
And a device characteristic reverse reproducing means 413 for outputting to.

The operation of the twenty-sixth embodiment will be described below with reference to FIG.
A description will be given along the flow of the processing shown in.

The device characteristic correction means 352 firstly, in step S110, RG in the standard device characteristic reproduction means 411.
The brightness data YR, YG, and YB are calculated by calculating the B data R, G, and B, respectively. The arithmetic expression in step S110 is shown in Expression (39).

YR = FR (R) YG = FG (G) (39) YB = FB (B) Here, FR (R), FG (G), FB in the equation (39).
(B) is a function (hereinafter referred to as a standard output characteristic function) that represents the standard output characteristic of the image recording apparatus 4 that is arbitrarily determined when the image processing apparatus 3 is designed. In this embodiment, FR (0) = FG (0) = FB (0) = 0 is designed.

Luminance data YR obtained by the above calculation
, YG, YB are output to the gain adjusting means 412.

Next, in step S111, the gain adjusting means 412 performs the following calculation for each of the luminance data YR, YG, and YB to linearly transform the luminance to obtain the luminance data YR ', YG', and YB '. To calculate. Step S11 performed by the gain adjusting means 412
The arithmetic expression of the processing in 1 is shown in Expression (40).

YR '= YR / h YG' = YG / h (40) YB '= YB / h Here, a method of calculating the parameter h used in the equation (40) will be described. First, the characteristics of the output brightness YR, YG, YB for the RGB data R, G, B of the image recording device 4 connected to the image processing device 3 must be obtained by brightness measurement and regression analysis which is a known technique. . The output characteristic of the image recording device 4 thus obtained is expressed by the following equation (4
It is assumed that the function (1) is satisfied (hereinafter, referred to as an output characteristic function).

YR = GR (R) YG = GG (G) (41) YB = GB (B) In obtaining the output characteristic function, the image recording device 4
Output characteristics of GR (0) = GG (0) = GB (0) = 0
However, there is no problem in practical use.

In this embodiment, the RGB data is represented by 8
If bits are used, the range of values that RGB data can take is 0 to 255. That is, when the RGB data is 255, the highest point of output brightness in the image recording device 4 is set. Therefore, h is defined as shown in Expression (42) below.

H = max [FR (255) / GR (255), FG (255) / GG (255), FB (255) / GB (255)] (42) Here, the equation (42) is FR (255) / GR (25
5), FG (255) / GG (255), FB (25
5) / GB (255), the largest one is h.

Thus, h is calculated and the luminance data Y
YR ', YG', YB 'are obtained by dividing R, YG, YB by h according to the equation (40). This calculation means that the luminance within the range that the standard output characteristic can take is linearly converted and compressed or expanded to the luminance within the range that the output characteristic of Expression (41) can take.

Luminance data YR obtained by the above calculation
′, YG ′ and YB ′ are output to the device characteristic reverse reproduction means 413.

[0515] Then, in step S112, the device characteristic reverse reproduction means 413 produces luminance data YR ', YG', YB.
'By performing the following calculations for each, RGB
Data R ', G', B'are calculated. This step S1
The arithmetic expression of the processing in 12 is shown in Expression (43).

R ′ = GR ⁻¹ (YR ′) G ′ = GG ⁻¹ (YG ′) (43) B ′ = GB ⁻¹ (YB ′) The function GR ⁻¹ used in the equation (43) , GG ^-1 , GB
^-1 is an inverse function of GR, GG and GB used in the equation (41). That is, the brightness data YR ', YG', and YB 'are calculated by the calculation processing of the equation (43) so that the brightness data YR', YG ', and YB' can be output to a certain image recording apparatus 4 by using RGB values.
It means that it is converted into data R ', G', B '.

The RGB data R ′ thus obtained,
G ′ and B ′ are output to the image recording device 4. With the above processing, when performing color conversion for improving the color difference between the image reproduced on the observation monitor 8 and the image reproduced on the image recording device 4, the output characteristic that varies between the machines is the standard output characteristic. Is corrected to approximate

The functions FR, FG and FB used by the standard device characteristic reproducing means 411, the constant h used by the gain adjusting means 412, and the functions GR ^-1 , GG ^-1 , GB used by the device characteristic inverse reproducing means 413. ^The setting of ^-1 can be changed by inputting it from the information input device 41 as shown in FIG.

As described above, in the device characteristic correction means 352, the RGB data R, G, B are temporarily transferred to the image recording device 4.
Luminance data YR, YG, Y based on the standard output characteristic function of
B, and the brightness data YR, YG, YB are linearly converted so as to be within the range of the output characteristics of the image recording device 4,
Further, by converting the RGB data R ′, G ′, B ′ using the inverse function of the output characteristic function of the image recording device 4 and outputting the RGB data to the image recording device 4, the image recording device 4 having different output characteristics is converted. The output can be approximated to the output according to the standard output characteristic.

The operations performed by the standard equipment characteristic reproducing means 411, the gain adjusting means 412, and the equipment characteristic reverse reproducing means 413 are combined into one, and R
The GB data R, G, B are converted to RGB data R ', G', B '.
It is also possible to prepare a function for directly converting to and to reduce the number of man-hours of the operation to one, and the same effect can be obtained.

Next, a twenty-seventh embodiment will be described. Figure 6
8 is a block diagram showing the functional configuration of the device characteristic correction means according to the 27th embodiment, FIG. 69 is a flowchart for explaining the operation of the device characteristic correction means in the 27th embodiment, and FIG. 70 is the output characteristic function of the image recording apparatus. It is a conceptual diagram explaining a determination method.

The twenty-seventh embodiment has a device characteristic correcting means 352.
Is a second configuration example in which a timer 416 and a switching means 41 are included in the device characteristic correction means 352 of the twenty-sixth embodiment.
5, switching means 415, gain adjusting means 412,
The output value of the timer 416 is input to the device characteristic reverse reproduction means 413 to change the operation characteristic according to the passage of time. In the present embodiment, the configuration and operation other than the device characteristic correction means are the same as those of the 21st embodiment, so description of similar parts will be omitted and the device characteristic correction means will be mainly described.

The output characteristic of the image recording apparatus 4 becomes stable after the preparatory operation time (hereinafter referred to as warming time) T has elapsed since the power was turned on. The state in which the output characteristics of the image recording device 4 are stable at this time is hereinafter referred to as a stable state. The output during the preparatory run has a deviation compared to the output during the stable state, and this deviation becomes smaller with the passage of time and gradually approaches the output during the stable state, and stabilizes after the warming time T. It becomes a state. Hereinafter, the state of the image recording device 4 during the preliminary operation will be referred to as an unstable state.

The image processing device 3 is configured to perform the image quality correction process on the assumption that the image recording device 4 is in a stable state. However, in actual use, the image recording device 4
After turning on the power, the user wants to use it without waiting for the warming time, that is, wants to use it even in an unstable state. When the image recording device 4 is in an unstable state, the output result of the image recording device 4 based on the color data output by the image processing device 3 is different from that in the stable state of the image recording device 4. In this way, in the unstable state, the device characteristics fluctuate over time, so a stable image quality correction result cannot be obtained, and there is a possibility that a deviation from the output result in the stable state may be subjectively unacceptable. is there.

Therefore, in the twenty-seventh embodiment, in addition to the effects of the twenty-first embodiment, the output of the image recording apparatus in the unstable state immediately after power-on becomes equal to the output of the image recording apparatus in the stable state. As described above, a configuration example of the device characteristic correction unit capable of converting the color data and correcting the color tone deviation will be described.

FIG. 68 shows a schematic structure of the device characteristic correcting means in the 27th embodiment. The device characteristic correction means 352
In addition to the configuration of the twenty-sixth embodiment, a timer 416 that outputs the elapsed time τ from the time when the image recording apparatus 4 is powered on, and R
Switching means 415 for switching the output of the two systems by using the GB data R, G, B and the elapsed time τ output from the timer 416 as inputs.

Of the two outputs of the switching means 415, one is connected to the standard equipment characteristic reproducing means 411 and the other is directly connected to the output portion of the equipment characteristic correcting means 352. The RGB data R, G, and B are output to. The elapsed time τ output from the timer 416 is supplied not only to the switching unit 415 but also to the gain adjusting unit 412 and the device characteristic reverse reproducing unit 413.

R of the standard device characteristic reproducing means 411 and later
The procedure of the arithmetic processing of the GB data R, G, B is executed according to the procedure shown in the twenty-sixth embodiment, and finally the RGB data R ', G', B'is output to the image recording device 4. It is like this. The other configuration of the device characteristic correction means 352 is the same as that of the twenty-sixth embodiment, and the description thereof is omitted here.

The operation of the 27th embodiment will be described below with reference to FIG.
And FIG. 70.

The timer 416 holds the elapsed time τ since the power of the image recording apparatus 4 was turned on. The elapsed time τ is
It is set to 0 at the same time when the power of the image recording device 4 is turned on, and updated with the lapse of time. This elapsed time τ is output to the switching unit 415, the gain adjusting unit 412, and the device characteristic reverse reproducing unit 413.

FIG. 69 shows the flow of processing in the device characteristic correction means 352 of the 27th embodiment. First, step S11
3, the switching unit 415 determines the output destination of the input RGB data R, G, B by the standard device characteristic reproduction unit 411 or the image depending on whether the elapsed time τ input from the timer 416 is longer than the warming time T. Switch to the recording device 4. Here, when τ is smaller than T, the output destination of the RGB data R, G, B is the standard device characteristic reproducing means 411.
If τ is T or more, the output destination of the RGB data R, G, B is the image recording device 4.

The warming time T is constant data of the actual measurement time from the power-on of the image recording device 4 to the stable state, and the information input device 41 causes the switching means 415.
Has been entered and retained.

The RGB data R, G, B output to the image recording device 4 by the output switching performed by the switching means 415 is completely subjected to the arithmetic processing by the device characteristic correcting means 352 of the present embodiment shown below. It is not applied.

Now, referring to FIG. 70, step S114
A method of calculating the output characteristic function shown in the following Expression (44), which is necessary for the following description, will be described.

[0535]

[Equation 11] In equation (44), KR (t), KG (t), KB
(T), gR (t), gG (t), gB (t) are time t
Is a function that takes as input, and can be obtained by measuring the output brightness of the image recording device 4 with respect to RGB data together with the elapsed time after the power is turned on and using the following method. . This equation (44) is called an output characteristic function of the image recording device 4.

FIG. 70 shows the passage of time t and the RGB data R,
It shows the relationship with the brightness Y corresponding to G and B.
Note that only R data is shown here.

The curved surface extending in the RYt coordinate system shown in FIG. 70 measures the brightness Y of the output of the image recording device 4 corresponding to the change of R at G = 0 and B = 0 at each time t, and measures it. It is a plot. Note that t = 0 is the time when the power is turned on, and the warming time T is used to indicate that the image recording device 4 is in a stable state at t = T. In addition, for G and B, GYt
Similar curved surfaces can be drawn in the coordinate system and the BYt coordinate system.

In this RYt coordinate system, the relationship of the change in Y corresponding to the change in R at a certain time t is expressed by the following equation (4)
As shown in 5), estimation is performed using a known technique, that is, a non-linear regression method.

[0539]

(Equation 12) In the equation (45), at = tc and bt = tc are the time t = tc.
Is an estimated parameter in. This estimate should be
For the sequence of numbers at and bt.

Then, a linear regression method or a linear regression method is adopted so that the relationship between changes in a and b corresponding to changes in t becomes a linear or non-linear function of t such that a = KR (t) and b = gR (t), respectively. Estimated by nonlinear regression method, KR (t), gR (t)
Ask for. Thus, KR used in equation (44)
(T) and gR (t) can be obtained.

Similarly, a curved surface on the GYt coordinate system is applied to KG (t) and gG (t), and a curved surface on the BYt coordinate system is applied to KB (t) and gB (t). By doing, KG (t), gG (t), KB (t), gB (t)
Can be requested.

Next, the operation after step S114 in FIG. 69 will be described using the output characteristic function of the equation (44) described above.

At step S114, the standard equipment characteristic reproducing means 411 inputs the RGB data R, G,
The brightness data YR, YG is calculated for each B
, YB is calculated. The calculation formula in step S114 is shown in formula (46).

[0544]

(Equation 13) The equation (46) is equal to the equation (44) in which the warming time T is substituted for t. Therefore, the RGB data R,
G and B are converted into luminances YR, YG, and YB that would be output when the image recording device 4 is in a stable state, respectively.

[0545] The standard equipment characteristic reproducing means 411 has KR
(T), KG (t), KB (t), gR (t), gG
(T) and gB (t) are input from the information input device 41,
This is held as constant data. Also, YR
(R, t), YG (G, t), YB (B, t) are the second
Formula (39) used for the calculation in the standard device characteristic reproducing unit 411 in the device characteristic correcting unit 352 described in the sixth embodiment.
FR (R), FG (G), and FB (B), respectively. Luminance data YR, YG obtained by this calculation
, YB are output to the gain adjusting means 412.

Next, in step S115, the gain adjusting means 412 performs the same operation as the operation performed by the gain adjusting means 412 in the device characteristic correcting means 352 described in the twenty-sixth embodiment. However, the gain adjusting means 4 of the present embodiment
In 12, the elapsed time τ output by the timer 416 is used.

If 8 bits are used to represent RGB data in this embodiment, FR (255), FG (25) used for calculating h in the 26th embodiment.
5) and FB (255), using equation (44) and T, respectively, YR (255, T), YG (255,
T), YB (255, T) are used instead. Also,
For GR (255), GG (255), GB (255), using equation (44) and τ, YR (2
55, τ), YG (255, τ), YB (255, τ)
Is used instead.

These constants YR (255, T), YG (25
5, T), YB (255, T) and the function YR (255,
τ), YG (255, τ), YB (255, τ) are input from the information input device 41.

Based on h thus obtained, luminance data Y
R ', YG', and YB 'are calculated in the same manner as in the twenty-sixth embodiment and output to the device characteristic reverse reproducing means 413.

Then, in step S116, the device characteristic reverse reproducing means 413 uses the luminance data YR ', YG', Y.
By performing the following calculation for each B ′, RGB
Data R ', G', B'are calculated. This step S1
The arithmetic expression of the processing in 16 is shown in Expression (47).

[0551]

[Equation 14] Expression (47) is calculated as an inverse function when t of the output characteristic function of Expression (44) is regarded as a constant.
′, YG ′, YB ′ and τ are input. The calculation processing of this expression (47) is performed by the luminance data YR ', YG', Y
It means that B ′ is regarded as the output luminance of the image recording device 4 at the time τ after the power is turned on, and the RGB data R ′, G ′, B ′ that outputs such luminance is obtained. Here, the functions R (YR ', τ), G (YG',
τ) and B (YB ′, τ) are input from the information input device 41.

The RGB data R ′ thus obtained,
G ′ and B ′ are output to the image recording device 4. Through the above processing, the output characteristic that changes with the passage of time in the unstable state immediately after the power is turned on is corrected to be equal to the output characteristic in the stable state.

As described above, the device characteristic correction means 352 determines whether the image recording apparatus 4 is in a stable state or an unstable state according to the time elapsed after the power of the image recording apparatus 4 is turned on. In the stable state, the RGB data is converted and output so as to obtain an output equivalent to the output of the image recording apparatus 4 in the stable state, so that the output of the image recording apparatus is in the stable state even in the unstable state. The result is similar to the output when, and the deviation of the color tone in the unstable state can be eliminated.

In this embodiment, the stable state of the image recording apparatus 4 is judged by the elapsed time after the power is turned on, but other judgment parameters such as the internal temperature of the image recording apparatus 4 may be used.

[0555] Also, a power-on detecting means for detecting the power-on of the image recording device 4 which is applied after the power of the endoscope system is turned on, and a signal line for transmitting a signal indicating that the power-on is detected. Are provided in the image processing apparatus 3, and a detecting means for detecting the signal is provided in the timer 416.
The timer 416 may be set to 0 at the same time when the signal indicating the addition of the image recording device 4 to the endoscope system is sensed.

Also, equation (44), equation (46), and equation (4)
7) is not limited to the above-mentioned one, but may be one that can approximate the output characteristics of the image recording device 4 to a higher level.

[Supplementary Note] (1) The image processing apparatus according to claim 1, wherein the image recording means has a function of reproducing an image recorded in the recording medium.

(2) Image pickup means (electronic endoscope 6 in FIG. 1) for picking up an image of a subject and outputting the image signal, and image display means for displaying the image of the subject based on the image signal output by the image pickup means. An image signal recording apparatus including (observation monitor 8 in FIG. 1) and image recording means (endoscopic image recording apparatus 4 in FIG. 1) for recording the image signal output by the imaging means on an image signal recording medium. In the image processing, the image displayed on the image display unit based on the image signal output from the image pickup unit and the image displayed based on the image signal recorded on the image signal recording medium have the same image quality. An image signal recording device, comprising: an image signal converting device (image processing device of FIG. 1) for processing an image signal output from the image capturing device based on data and inputting the image signal to the image recording device.

(3) The image signal recording apparatus according to attachment 2, wherein the image signal converting means processes the color of the image signal recorded on the image signal recording medium.

(4) In an image signal recording apparatus for recording a subject image picked up by the image pickup means on a recording medium, a color signal generation means (signal processing of FIG. 1) for generating a plurality of color signals based on the output of the image pickup means. Section 7B), display means for displaying the subject image based on the plurality of color signals (observation monitor 8 in FIG. 1), and display means for displaying the color signals generated by the color signal generating means. Based on the image processing data, a recording unit (an endoscope image recording device 4 in FIG. 1) that records image processing data that equalizes the image quality of a subject image reproduced from the recording medium, An image signal recording device comprising a recording control unit (image processing device 3 in FIG. 1) for processing the plurality of color signals and recording the image signal on the recording medium.

(5) Color signal generating means (signal processing section 7B in FIG. 1) for generating a plurality of color signals based on the output of the image capturing means for capturing an image of the subject, and the color signal generated by the color signal generating means. A recording unit (an endoscope image recording apparatus 4 in FIG. 1) for recording image data based on the image data based on the color signal generated by the color signal generating unit. An image signal processing device (image processing device 3 in FIG. 1) for generating the image data having equivalent image quality by performing signal processing on the color signal generated by the color signal generating device.

(6) Input means for inputting a plurality of video signals picked up by the image pickup means (image input I / F 4 in FIG. 3)
3 or the input I / F 160 of FIG. 32) and a correction unit (image processing unit 47 of FIG. 3 or the color conversion circuit 163 of FIG. 32 and / or of the image processing unit 47 of FIG. Sharpness correction circuit 165) and output means (image storage device I / F 46 in FIG. 3 or output I / F 46 in FIG. 32) for outputting the plurality of video signals whose image quality has been corrected by the correction means to an image recording device. F168
a)) and an image processing apparatus comprising:

(7) The image processing apparatus according to attachment 6, wherein the correction means performs at least one of color conversion and sharpness correction on a plurality of video signals.

(8) Note 6 provided with control means (central processing unit 40 in FIG. 3 or arithmetic processing controller 169 in FIG. 32) for controlling the correction means on the basis of a plurality of video signals input by the input means. Or the image processing device according to 7.

(9) Parameter input means for inputting a parameter instructing the correction content of the correction means (information input device 41 in FIG. 3 or parameter input device I in FIG. 32).
/ F171), and the control means controls the correction means based on the plurality of video signals and the parameters input by the input means.

(10) The correction means is provided with one or more series of mapping conversion means for converting an image of a plurality of video signals inputted by the input means.
The image processing device according to item 1.

(11) An image output from any one of an image formed by a plurality of video signals input from the input unit or an image converted by the mapping conversion unit, or input by the parameter input unit. 7. The image processing apparatus according to appendix 6, further comprising mapping conversion control means for controlling the operation of the mapping conversion means by using the parameter as a control signal.

(12) Of the plurality of mapping conversion means,
At least one is provided in parallel, and there is provided a combining means for combining a plurality of outputs from the one or more mapping conversion means provided in parallel into one output, and the mapping conversion control means is provided at least in parallel. The signal input to one mapping conversion unit is a signal space formed by a plurality of video signals input from the input unit, or one or more of the mapping conversion units provided in series are provided in parallel. Area determination means for performing area determination of which area of the signal space formed by the signal output from the mapping conversion means positioned in front of the mapping conversion means, and the result determined by the area determination means. According to Supplementary Note 10, the operation determining means for determining the operation in the synthesizing means and the operation changing means for applying the action determined in the operation determining means to the synthesizing means are added. Mounting image processing apparatus.

(13) Of the plurality of mapping conversion means,
At least one is provided in parallel and at least one is provided, and the switching means is provided in the preceding stage of the plurality of mapping conversion means provided in parallel at least one, and the control means controls the switching means to perform parallel operation. 11. The image processing apparatus according to appendix 10, which selects an operating mapping conversion unit from among the mapping conversion units provided.

(14) Of the mapping conversion means provided one or more in series, the mapping conversion means at the last stage has the output of the mapping conversion means at the front stage of the mapping conversion means at the last stage dynamic of the image recording device. It is determined whether it is within the range, and if it is outside the dynamic range, it is determined that the output of the mapping conversion unit in the preceding stage of the mapping conversion unit in the last stage is an abnormal value, and the mapping conversion into the dynamic range is performed. Image processing device.

(15) Among one or more mapping conversion means provided in series, the mapping conversion means provided closest to the input means is a color space coordinate system for converting an input signal into a perceptual color space coordinate system. Note that one of the mapping conversion means provided after the color space coordinate conversion means is a color space coordinate inverse conversion means for converting the perceptual color space coordinates into a signal that can be input by the recording device. The image processing device according to item 10.

(16) The mapping conversion means provided in the subsequent stage of the color space coordinate conversion means is a gamut compression means for performing mapping conversion accompanying the gamut compression processing, and the mapping conversion means provided after the gamut compression means. 16. The image processing device according to attachment 15, wherein the conversion unit is the color space coordinate reverse conversion unit.

(17) At least one of the one or more mapping conversion means provided in series is a mapping conversion means capable of executing mapping conversion of a plurality of systems, and the control means is provided with the plurality of mapping conversion means. 11. The image processing device according to appendix 10, which controls switching of mapping conversion methods in mapping conversion means capable of executing systematic mapping conversion.

(18) At least one of the mapping conversion means provided one or more in series is:
11. The image processing device according to appendix 10, which changes the quantization accuracy of a plurality of input signals according to the degree of contribution of the input signals to the output signal.

(19) The control means calculates the frequency distribution of signals in a signal space formed by a plurality of input signals, and the frequency distribution of signals calculated by the counting means. On the basis of one or more of the mapping conversion means provided in series, the operation determination means for determining the operation of at least one of the mapping conversion means and the operation determined by the operation determination means are connected in series. 11. The image processing apparatus according to appendix 10, further comprising at least one of the mapping conversion means provided to at least one of the mapping conversion means and an operation application means.

(20) The correcting means comprises a filter setting means for setting a filter for the input video signal, a filtering means for applying filtering, and an output means for outputting a filtering result. The image processing device according to attachment 6.

(21) The image processing apparatus according to attachment 20, wherein the filtering means comprises filtering execution means in the spatial frequency domain.

(22) The image processing apparatus according to attachment 20, wherein the filtering means is digital filtering execution means.

(23) The image processing apparatus according to attachment 20, wherein the filter setting means comprises an inverse filter setting means for setting an inverse filter for the frequency characteristic of the image recording apparatus.

(24) The correcting means combines the input video signal with a filter setting means for setting a filter, a filtering means for applying filtering, and the input video signal and a filtering processing result. 7. The image processing apparatus according to appendix 6, which is configured to include a synthesizing unit and an output unit that outputs the signal synthesized by the synthesizing unit.

(25) The image processing apparatus according to appendix 24, wherein the filtering means comprises filtering execution means in the spatial frequency domain.

(26) The image processing apparatus according to attachment 24, wherein the filtering means is digital filtering execution means.

(27) The image processing apparatus according to appendix 24, wherein the correction means includes control means for controlling the synthesizing means based on the input video signal.

(28) The correcting means combines the input video signal with a filter setting means for setting a filter, a filtering means for applying filtering, and the input video signal and a filtering processing result. The image processing device according to attachment 6, which is configured to include a mapping conversion unit.

(29) The image processing apparatus according to attachment 24, wherein the correction means is provided with a mapping conversion means for mapping the output of the synthesizing means.

(30) The filtering means comprises a cascade connection of a plurality of digital filters.
Or the image processing device described in 26.

(31) The filtering means performs mapping conversion of the output signal from the preceding digital filter,
31. The image processing device according to attachment 30, which is configured to include a mapping conversion unit that is an input signal to the digital filter in the subsequent stage.

(32) The image according to appendix 31, wherein the mapping conversion means is configured to convert the input video signal so as to be distributed in a quantization region which can be input to the digital filter in the subsequent stage. Processing equipment.

(33) The image processing apparatus according to attachment 30, wherein the filtering means has a plurality of digital filters in the subsequent stage connected in parallel to the digital filter in the previous stage.

(34) The filtering means is composed of a dividing means for dividing and inputting the output signal from the digital filter in the preceding stage to the plurality of digital filters connected in parallel in the latter stage, and the plurality of digital filters. Supplementary Note 33 comprising a combining means for combining the processing results
The image processing device according to item 1.

(35) The image processing according to any one of claims 20 to 34, further comprising: mapping conversion means for performing mapping conversion of an image based on a plurality of video signals input from the output means. apparatus.

(36) An image processing method including a correction step (color conversion unit 50 or sharpness correction unit 51 in FIG. 4) for performing image quality correction on a plurality of video signals picked up by the image pickup means.

(37) The correction step is a mapping conversion step of performing mapping conversion on the plurality of video signals input in the input step (color space coordinate conversion unit 60 in FIG. 5).
A color conversion step of performing color conversion on the video signal subjected to the mapping conversion in the mapping conversion step (matrix operation unit in FIG. 5); and an inverse operation for the video signal subjected to the color conversion in the color conversion step. Supplementary Note 36 configured by including an inverse mapping conversion step for performing mapping conversion (color space inverse coordinate conversion unit 60 in FIG. 5)
The image processing method described in.

(38) The image according to appendix 36, wherein the correction step includes a filtering step (filtering execution unit 65 in FIG. 8) for filtering the plurality of video signals input in the input step. Processing method.

(39) In the image processing apparatus according to claim 1, the image signal processing means is an input color which is a three-dimensional space formed by a base vector representing an image signal created in advance and output from the imaging means. A color space coordinate conversion unit that has a conversion reference table that associates each point of the space with each point of the perceptual color space, maps the image signal output from the imaging unit to the perceptual color space, and the perceptual color space In the color space conversion means for performing the above-mentioned processing (processing by the image signal processing means of claim 1) as an image signal to be processed, and a basis vector expressing an image signal which is created in advance and can be input to the image recording apparatus. A reverse conversion lookup table that associates each point in the output color space, which is a three-dimensional space, with each point in the perceptual color space is opposite to the mapping direction from the output color space to the perceptual color space (hereinafter, forward mapping direction). Map of An image processing device, comprising: a color space coordinate inverse transforming unit that associates a perceptual color space with an output color space by using the direction (hereinafter, reverse mapping direction).

(40) The color space inverse conversion means is an input point scanning means for scanning points in the input color space, the scanned points converted by the inverse conversion reference table, and the imaging means. A distance detection means for detecting the Euclidean distance of a point obtained by processing the image signal output from the Euclidean distance in the perceptual color space, and a minimum value calculating means for calculating the minimum value of the Euclidean distance. 40. The image processing apparatus according to appendix 39, wherein the perceptual color space and the output color space are associated with each other by making the scanning point that gives the minimum value a point processed in the perceptual color space.

(41) The input point scanning means uses a plurality of scanning modes in which a scanning area can be set and the scanning pitch in the input color space is gradually reduced, and the input color based on the result of the minimum value calculation means. The image processing according to appendix 40, further comprising: a scanning area setting unit that sets a scanning area in a space, wherein a scanning point that sequentially reduces the scanning area and that provides the minimum value of the Euclidean distance is searched. apparatus.

The above Supplementary Notes 39 to 41 are the disclosure contents of the twenty-first embodiment.

(42) In the image processing device described in appendix 39, further, in response to the reference color image input means, the correction amount input means, and the reference color image signal input from the reference color image input means, Processing amount correction means for correcting the processed image signal processed by the color space coordinate conversion means according to the correction amount input by the correction amount input.

(43) As the reference color image signal, the original image signal output from the image pickup means and before being processed by the in-color-space converting means is used as a reference table for inverse conversion of the color-space coordinate inverse-converting means. Appendix 42, which is an image signal obtained by using
The image processing device according to item 1.

(44) The correction amount input means is means for inputting a weighting factor for weighted addition of the reference color image signal and the processed image signal processed by the color space conversion means, and the processing amount The correcting means determines a point corresponding to the weighting factor on a curve connecting a reference point representing the reference color image signal and a processed point representing the processed image signal in a color space for three-dimensionally displaying colors. The image processing apparatus according to appendix 43, wherein the processed image signal is corrected by outputting it as a corrected processed image signal.

(45) In the processing amount correction means,
The image processing device according to attachment 44, wherein a curve connecting the reference point and the processed point is a line segment, and an internally dividing point obtained by internally dividing the line segment by the weight ratio is output as the corrected processed image signal. .

The above supplementary notes 42 to 45 are the disclosure contents of the twenty-second embodiment.

(46) The image processing apparatus described in appendix 39 is further characterized by further comprising reference table rewriting means for rewriting the inverse conversion reference table based on the input correction amount.

(47) The image processing described in appendix 46, wherein the correction amount input to the reference table rewriting means is based on the type of light source for observing the image output from the image recording device. apparatus.

(48) The image processing device described in appendix 47, wherein the conversion table rewriting means stores a plurality of conversion tables based on the color and luminance of a plurality of light sources.

The above Supplementary Notes 46 to 48 are the contents disclosed in the twenty-third embodiment.

(49) In the image processing device described in appendix 39, the color space conversion means has a chromatic adaptation conversion means for copying the input image signal to a chromatic adaptation corresponding point.

Appendix 49 is the disclosure content of the twenty-fourth embodiment.

(50) The image processing apparatus described in appendix 39 is characterized by further comprising processing operation correcting means for correcting the processing operation of the in-color-space converting means based on the input correction amount. .

(51) The correction amount input to the processing operation correction means is data designating a fixed point that is not converted by the processing in the perceptual color space performed by the in-color space conversion means. The image processing apparatus according to attachment 50.

The above Supplementary Notes 50 and 51 are the disclosure contents of the twenty-fifth embodiment.

(52) In the image processing device described in appendix 39, further, there is provided a luminance data conversion means for receiving the output signal of the color space coordinate inverse conversion means and converting the output signal into a plurality of luminance data. Gain adjustment means for independently adjusting the gain of each of the plurality of luminance data, and luminance inverse conversion means for converting an output signal of the gain adjustment means into output color space coordinates of the color space coordinate inverse conversion means. It is characterized in that it has a device characteristic correcting means for performing.

(53) The image processing apparatus described in appendix 52 is characterized by further comprising gain setting means for setting the gain of the gain adjusting means based on the input / output characteristics of the image recording means.

(54) In the image processing apparatus described in appendix 53, further, there is provided monitor means for monitoring the operating state of the image recording apparatus, and the gain setting means measures the set value of gain by the monitor means. It is characterized in that it is changed according to the operating state.

(55) The monitor means is a temperature measuring means for measuring the operating temperature of the image recording apparatus, and the gain setting means changes the set value of the gain according to the operating temperature measured by the temperature measuring means. The image processing device according to appendix 54.

(56) The monitor means is a timer means for measuring the elapsed time from the power-on of the image recording means, and the gain setting means changes the set value of the gain according to the elapsed time measured by the timer means. 55. The image processing device according to appendix 54, wherein

(57) The device characteristic correcting means has a switching means for switching the output signal of the color space coordinate reverse converting means in front of the luminance data converting means, and the switching means controls the timer means. When the elapsed time is less than or equal to a predetermined time, the output signal of the color space coordinate inverse conversion means is sent to the luminance data conversion means, and when the elapsed time exceeds the predetermined time, it is directly sent to the image recording means. Note 56 characterized by performing a switching operation so as to send
The image processing device according to item 1.

The above Supplementary Notes 52 to 57 are the 26th and 27th.
It is the disclosure content of the embodiment.

[0620]

As described above, according to the image processing apparatus of the present invention, the image signal processing means causes the image displayed on the image display means based on the image signal output from the image pickup means, and the image recording means. Since the image signal output from the image pickup means is processed based on the image processing data that makes the image displayed on the image display means equivalent to the image signal reproduced from the reproducing means, the image signal from the image pickup means is processed. The difference between the color of the image reproduced by the image display means and the color of the image reproduced by the image recording / reproducing means is improved colorimetrically, or the sharpness and image recording of the image reproduced by the display means are improved. The difference in the sharpness of the image reproduced by the reproducing means can be improved in the difference in MTF, and the operator can set a target by the memory of the operator for the image reproduced on the image display means. An effect that enables an operator to reproduce a desired preferable color or a desired sharpness even when the image quality is changed from the result of colorimetrically improved color difference or the corrected sharpness. There is.

[Brief description of drawings]

FIG. 1 is a configuration diagram showing a configuration of an endoscope system including an image processing apparatus according to a first embodiment.

2 is a block configuration diagram showing a detailed configuration of the endoscope system of FIG.

3 is a configuration diagram showing a configuration of the image processing apparatus of FIG.

FIG. 4 is a software configuration diagram showing a software configuration of an image processing unit of the central processing unit of FIG.

5 is a software configuration diagram showing the software configuration of the color conversion unit in FIG.

FIG. 6 is a flowchart illustrating a processing flow of a color conversion unit in FIG.

7 is a CIE1976 used in the color conversion unit of FIG. 4;
A conceptual diagram for explaining the color distribution of a general endoscopic image in the (L ^* u ^* v ^* ) color space.

FIG. 8 is a software configuration diagram showing the software configuration of the sharpness correction unit in FIG.

FIG. 9 is an explanatory diagram illustrating the concept of sharpness correction by the sharpness correction unit in FIG.

FIG. 10 is a flowchart illustrating a processing flow of the sharpness correction unit in FIG.

11 is an explanatory diagram for explaining sharpness correction by the sharpness correction unit in FIG. 4. FIG.

FIG. 12 is an explanatory diagram for explaining sharpness correction by the sharpness correction unit according to the second embodiment.

FIG. 13 is a software configuration diagram showing the software configuration of a sharpness correction unit that performs the sharpness correction of FIG.

FIG. 14 is a software configuration diagram showing a software configuration of a color conversion unit according to the third embodiment.

FIG. 15 is a flowchart illustrating a processing flow of a color conversion unit in FIG.

FIG. 16 is a flowchart illustrating a processing flow of a color conversion unit according to the fourth embodiment.

FIG. 17 is a software configuration diagram showing a software configuration of a color conversion unit according to the fifth embodiment.

18 is a conceptual diagram illustrating a relationship between an observation monitor color gamut and an image recording device color gamut applied to the color conversion unit in FIG. 17 on a u ^* v ^* plane.

FIG. 19 is a flowchart illustrating a processing flow of the color conversion unit in FIG.

20 shows the structure of the gamut data of the endoscopic image storage device applied to the color conversion unit of FIG. 17 and the meaning of hue (H) saturation (C) lightness (V) in CIE1976 (L ^* u ^* v ^*). ) Conceptual diagram explained in color space

FIG. 21 is a software configuration diagram showing a configuration of a modified example of the color gamut compression unit in FIG.

FIG. 22 is a software configuration diagram showing the configuration of a matrix calculation unit of the color conversion unit according to the sixth embodiment.

FIG. 23 is a CIE19 applied to the matrix calculation unit of FIG. 22;
76 (L ^* u ^* v ^* ) color space conceptual diagram for explaining a plurality of divided regions

FIG. 24 is a software configuration diagram showing a software configuration of a color conversion unit according to the seventh embodiment.

FIG. 25 is a software configuration diagram showing the software configuration of the color conversion unit according to the eighth embodiment.

FIG. 26 is a software configuration diagram showing a configuration of the matrix calculation unit of FIG. 25.

FIG. 27 is a software configuration diagram showing the software configuration of the color conversion unit according to the ninth embodiment.

FIG. 28 is a flowchart for explaining the processing flow of the color conversion section in FIG.

FIG. 29 is a flowchart illustrating the flow of a modified example of the processing of the color conversion unit in FIG. 27.

FIG. 30 is a conceptual diagram illustrating a preferable conversion of brightness according to the flowchart of FIG. 29.

FIG. 31 is a software configuration diagram showing the configuration of the matrix calculation unit of the color conversion unit according to the tenth embodiment.

FIG. 32 is a block diagram showing the arrangement of an image processing apparatus according to the eleventh embodiment.

33 is a block diagram showing the configuration of a modified example of the image processing apparatus shown in FIG. 32.

34 is a block diagram showing the configuration of the color conversion circuit of FIG. 32.

35 is a block diagram showing the configuration of the sharpness correction circuit of FIG. 32.

36 is a block diagram showing the configuration of the filtering unit in FIG. 35.

FIG. 37 is a block diagram showing the configuration of a color conversion circuit according to a twelfth embodiment.

FIG. 38 is a block diagram showing the configuration of a color conversion circuit according to a thirteenth embodiment.

FIG. 39 is a block diagram showing the configuration of the color conversion circuit according to the fourteenth embodiment.

FIG. 40 is a block diagram showing the configuration of a filtering unit of the sharpness correction circuit according to the fifteenth embodiment.

FIG. 41 is a block diagram showing the configuration of a sharpness correction circuit according to the sixteenth embodiment.

42 is a block diagram showing the configuration of the combining unit in FIG. 41.

43 is a block diagram showing the configuration of the data conversion unit in FIG. 41.

FIG. 44 is a block diagram showing a configuration of a modified example of the combining unit in FIG. 41.

FIG. 45 is an explanatory diagram illustrating a mask size of the FIR filter according to the seventeenth embodiment.

46 is a block diagram showing the configuration of a filtering unit including the FIR filter shown in FIG. 45.

FIG. 47 is a block diagram showing the configuration of a filtering unit according to the eighteenth embodiment.

FIG. 48 is a block diagram showing the configuration of a filtering unit according to the nineteenth embodiment.

FIG. 49 is a block diagram showing the configuration of a filtering unit according to the twentieth embodiment.

50 is a block diagram showing a conceptual configuration of software of an image processing unit in the central processing unit of FIG.

FIG. 51 is a block diagram showing the functional arrangement of a color conversion means according to the twenty-first embodiment.

FIG. 52 is a flowchart for explaining the operation of the color conversion means in the twenty-first embodiment.

FIG. 53 is a conceptual diagram illustrating the relationship between the color gamut of the observation monitor and the image recording device in the CIE1976 (L ^* u ^* v ^* ) color space.

FIG. 54 is a flowchart showing an example of a color gamut compression method.

FIG. 55 shows the concept of color gamut compression based on CIE1976 (L ^* u ^* v).
^* ) Illustration explained in color space

FIG. 56 is a conceptual diagram illustrating the configuration of a device-independent color conversion system.

FIG. 57 is a block diagram showing the functional arrangement of a color conversion means according to the 22nd embodiment.

FIG. 58 is a block diagram showing the internal structure of a color correction effect adjusting means provided in the 22nd embodiment.

FIG. 59 is a conceptual diagram illustrating the operation of the color correction effect adjusting means in FIG. 58.

FIG. 60 is a block diagram showing the functional arrangement of a color conversion means according to the 23rd embodiment.

FIG. 61 is a conceptual diagram for explaining the operation of the light source effect adjusting means provided in the 23rd embodiment.

FIG. 62 is a block diagram showing the functional arrangement of a color conversion means according to the twenty-fourth embodiment.

FIG. 63 is a conceptual diagram explaining the operation of the chromatic adaptation conversion means provided in the twenty-fourth embodiment.

FIG. 64 is a block diagram showing the functional arrangement of color conversion means according to the twenty-fifth embodiment.

FIG. 65 is a conceptual diagram explaining the operation of the gamut compression center changing means provided in the twenty-fifth embodiment.

FIG. 66 is a block diagram showing a functional configuration of a device characteristic correction means in a twenty-sixth embodiment.

FIG. 67 is a flowchart for explaining the operation of the device characteristic correction means in the twenty sixth embodiment.

FIG. 68 is a block diagram showing the functional configuration of a device characteristic correction means in the twenty-seventh embodiment.

FIG. 69 is a flowchart for explaining the operation of the device characteristic correction means in the twenty-seventh embodiment.

FIG. 70 is a conceptual diagram illustrating a method of determining an output characteristic function of an image recording device in a device characteristic correction unit of the 27th embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Electronic endoscope system 2 ... Electronic endoscope apparatus 3 ... Image processing apparatus 4 ... Endoscopic image recording apparatus 6 ... Electronic endoscope 7 ... Observing apparatus 7A ... Light source section 7B ... Signal processing section 8 ... Observation Monitor 40 ... Central processing unit 41 ... Information input device 42 ... Main storage device 43 ... Image input I / F 44 ... External storage device 45 ... ROM 46 ... Image storage device I / F 47 ... Image processing unit 50 ... Color conversion unit 51 ... Sharpness correction unit 60 ... Color space coordinate conversion unit 61 ... Matrix calculation unit 62 ... Color space inverse coordinate conversion unit 65, 71 ... Filtering unit 66, 73 ... Data conversion unit 72 ... Compositing unit 81 ... Abnormality processing unit 86, 86a Color gamut compression unit 160 Input I / F 163 Color conversion circuit 166 Sharpness correction circuit 168a, 168b Output I / F 169 Computation controller 171 Parameter input device I / F 350 Color conversion means 351 ... Sharpness correction means 352 ... Equipment characteristic correction means 400 ... Color space coordinate conversion means 401 ... Color gamut compression means 402 ... Color space coordinate inverse conversion means 411 ... Standard equipment characteristic reproduction means 412 ... Gain adjustment means 413 ... Device characteristic reverse reproduction means

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification number Office reference number FI Technical indication location H04N 1/409 1/46 7/18 U H04N 1/46 Z (72) Inventor Hirokazu Nishimura Tokyo 2-43-2 Hatagaya, Shibuya-ku, Tokyo Olympus Optical Co., Ltd. (72) Inventor Issei Nakamura 2-43-2 Hatagaya, Shibuya-ku, Tokyo Olympus Optical Co., Ltd. (72) Inventor Yoshitaka Miyoshi 2-43-2 Hatagaya, Shibuya-ku, Tokyo Inside Olympus Optical Co., Ltd.

Claims (1)

[Claims] 1. An image pickup means for picking up an image of a subject and outputting an image signal, an image recording means for recording an image signal output by the image pickup means on a recording medium, and an image signal output from the image pickup means are input. And an image displayed on the image display means for displaying the image of the subject, the image displayed on the image display means based on the image signal output from the image pickup means, and the image obtained from the image recording means. An image processing apparatus comprising an image signal processing means for processing an image signal output from the image pickup means on the basis of image processing data having an image quality equivalent to that of an image.