JP2000011188A - Method for processing image and device therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing method and device for improving image quality by exactly extracting an edge part by increasing the extracting precision of the edge part. SOLUTION: Gradient indicating the direction and intensity of the edge of the picture element under consideration of an image constituted of color image signals to be image-processed and the surrounding picture elements is calculated from the pixel values of the picture element under consideration and the surrounding pixels (a step 1). Then, the connectivity of the picture element under consideration and the surrounding picture elements is calculated from the direction of the stored gradient (a step 2). The gradient of the picture element under consideration is calculated for each R, G, and B channel so that the directivity of each R, G, and B of the picture element under consideration can be calculated. Thus, whether or not the picture element under consideration is an edge part can be judged from the calculated connectivity and directivity for each R, G, and B, and the edge part can be extracted (a step 5).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing method and apparatus, and more particularly, to sharpness enhancement and grain suppression for a color image or a color image signal obtained by photoelectrically reading an object. The present invention relates to an image processing method and apparatus for performing edge extraction for performing image processing such as.

[0002]

2. Description of the Related Art A color image recorded on a negative film, a reversal film, a color print, or the like is converted into a CC image.
An image signal is read photoelectrically by a photoelectric conversion element such as D to obtain image signals for each of the three primary colors red (R), green (G), and blue (B), and converted into digital signals. 2. Description of the Related Art A digital photo printer has been put into practical use as a digital color image reproduction system that performs various image processings and reproduces it on a recording material such as color paper or a display means such as a CRT.

According to this digital photo printer, even if a color image is photographed under inappropriate photographing conditions such as underexposure or overexposure and recorded on a negative film, a reversal film or a color print,
By performing image processing on the obtained original image signal, it is possible to reproduce a color image having a desired color and gradation. In addition, a color image recorded on a negative film, a reversal film, a color print, or the like can be reproduced as a color image having different colors and gradations, if desired.

As such image processing, for example, a method of performing image processing on an image signal representing a given image to enhance the sharpness of the image (sharpness enhancement),
There is a graininess suppression method for suppressing graininess of a film to reduce image roughness. Prior to performing the sharpness enhancement and the graininess suppression, it is necessary to extract an edge portion or a grainy portion. Conventionally, various methods are known for extracting an edge portion and the like.

As an edge extraction method, for example,
A method based on rate matching (matching) is known.
This is a function of the pixel values of the surrounding pixels of the target pixel to be determined.
For distribution, some shape, for example, 3 × 3 matrix
Prepare a template consisting of
It is to determine whether it is closest to the template. now,
Let x be the pixel value of the pixel of interest and its surrounding pixels_j(J = 1, ...
・, N), and k templates are prepared
And the value of the i-th template is w_j ⁱ(J =
1,..., N). At this time, i = 1,..., K
For the sum of products p_iIs calculated. p_i= Σ_jw_j ⁱ・ X_j (Σ is j = 1, ..., n
The largest of these p_iA template that matches what gives
And

For example, as shown in FIG. 11A, for an image composed of a target pixel and eight peripheral pixels around it, eight templates a1 to a8 as shown in FIG.
Is given. It is assumed that the directions of edges indicated by arrows b1 to b8 correspond to the templates a1 to a8. The operator (operator) represented by such a template is called a Sobel operator. At this time, the product sums p1 to p8 are calculated. First, the calculation for the template a1 is as follows. p1 = 1 × 1 + 2 × 2 + 1 × 3 + 0 × 9 + 0 × 10 + 0
× 4 + (-1) × 8 + (-2) × 7 + (− 1) × 5 = −
19

Similarly, the product sums p2 to p8 are calculated for the other templates a2 to a8. p2 = -8, p3 = 11, p4 = 20, p5 = 19, p6 = 8, p7 = -11, p8 = -20.

As a result, p4 is maximum, and FIG.
As shown in (b), the direction from the upper right to the lower left indicated by the arrow b4 of the template a4 is the direction of the edge (density gradient, that is, the direction of the gradient) of the pixel of interest 10, and the value 20 of the product sum p4 is the gradient intensity. Becomes
Thus, the gradient (vector amount) of the target pixel is calculated. Next, the direction of this gradient and 90 °
The same calculation is performed on the 3 × 3 image centered on each pixel with respect to the adjacent pixels 1 and 5 of the pixel of interest 10 in the direction of Calculate the direction of the gradient. If at least one of the two adjacent pixels 1 and 5 is within 45 ° of the direction of the target pixel 10, the target pixel 10 is determined to be an edge, that is, to have connectivity. .

[0009]

However, although the conventional edge part extraction method can extract the edge part to some extent, in some cases, the granular part is detected as the edge part, and the quality of the finished image deteriorates. There was a problem. For example, as shown in FIG. 9A, in addition to a white line which is an original edge portion, a granular portion is determined to be an edge portion, and many white particles may remain.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned conventional problems, and an image processing method and an image processing method capable of improving the extraction accuracy of an edge portion, extracting an edge portion more accurately, and improving the image quality than before. It is an object to provide a device.

[0011]

According to an aspect of the present invention, there is provided an image processing method for performing a predetermined image processing on a color image signal, the method comprising the steps of: From the pixel value of the pixel of interest and its surrounding pixels of the composed image, a gradient representing the direction and intensity of the edge of the pixel of interest and its surrounding pixels is calculated and stored, and the gradient is stored from the direction of the stored gradient. By calculating the connectivity between the target pixel and its surrounding pixels, and calculating the gradient of the target pixel for each of the R, G, and B channels, each RG of the target pixel is calculated.
B. calculating an orientation for each B, and determining whether or not the pixel of interest is an edge from the connectivity and the orientation for each RGB, thereby extracting an edge portion. It provides a processing method.

In calculating the connectivity between the target pixel and its peripheral pixels, the gradient direction of the target pixel is compared with the gradient direction of the neighboring pixels of the target pixel. It is preferable that the connection is established when the directions of the predetermined number of pixels match the predetermined condition. In addition, the case where the direction of a predetermined number of pixels among the neighboring pixels matches the predetermined condition means that 5 directions of the direction of the neighboring pixels are satisfied.
It is preferable that 0% or more is within 45 ° of the direction of the target pixel. Alternatively, the case where the directions of a predetermined number of pixels among the neighboring pixels match predetermined conditions,
60% or more, 70% or more, 80% of the direction of the neighboring pixels
Or 90% or more is 45 ° from the direction of the pixel of interest.
Or within 30 °. Further, the neighboring pixels are centered on the pixel of interest,
1 × 3 pixels, 1 × 5 pixels, 3 × 5 pixels, which are more present in the direction orthogonal to the direction of the gradient of the pixel of interest,
The pixel is preferably a pixel obtained by removing a central pixel from one of 3 × 7 pixels and 5 × 9 pixels. The method of taking the neighboring pixels and the condition for determining the connectivity of pixels including the predetermined number of pixels whose direction matches the predetermined condition and the predetermined condition where the directions of the predetermined number of pixels match are determined by the document type. , The document size, the print magnification, the camera type, and the exposure condition.
Further, in calculating the connectivity between the pixel of interest and its surrounding pixels, the direction of the gradient of the pixel of interest and 9
It is preferable that there is connectivity when the angle of the gradient direction of the adjacent pixels on both sides forming the direction of 0 ° with the direction of the gradient of the pixel of interest is within 45 °.

According to another aspect of the present invention, there is provided an image processing apparatus for performing a predetermined image processing on a color image signal, the image processing apparatus comprising: Means for calculating a gradient representing the direction and intensity of the edge of the pixel of interest and its surrounding pixels from the pixel values of the pixel of interest and its surrounding pixels, means for storing the calculated gradient, and Means for calculating the connectivity of the pixel of interest and its surrounding pixels from the direction of the gradient, and calculating the gradient of the pixel of interest for each of the R, G, and B channels, thereby providing Means for calculating directionality, means for determining whether or not the pixel of interest is an edge portion from the connectivity and the directionality for each of RGB,
The image processing apparatus according to the present invention is characterized in that edge portions are extracted by providing the image processing apparatus.

Here, the means for calculating the connectivity between the pixel of interest and its surrounding pixels compares the direction of the gradient of the pixel of interest with the direction of the gradient of the neighboring pixels of the pixel of interest, and When the directions of the predetermined number of pixels satisfy the predetermined condition, it is preferable to determine that there is connectivity. In addition, the case where the direction of a predetermined number of pixels among the neighboring pixels matches the predetermined condition means that 5 directions of the direction of the neighboring pixels are satisfied.
Preferably, 0% or more is within 45 ° of the direction of the target pixel. Alternatively, the case where the directions of a predetermined number of pixels among the neighboring pixels match predetermined conditions,
60% or more, 70% or more, 80% of the direction of the neighboring pixels
Or 90% or more is 45 ° from the direction of the pixel of interest.
Or within 30 °. Further, the neighboring pixels are centered on the pixel of interest,
1 × 3 pixels, 1 × 5 pixels, 3 × 5 pixels, 3 ×, extending in a direction orthogonal to the gradient direction of the target pixel
The pixel is preferably a pixel obtained by removing a central pixel from any one of 7 pixels and 5 × 9 pixels. The method of taking the neighboring pixels and the condition for determining the connectivity of pixels including the predetermined number of pixels whose direction matches the predetermined condition and the predetermined condition where the directions of the predetermined number of pixels match are determined by the document type. ,
Preferably, it is determined according to at least one of the document size, print magnification, camera type, and exposure conditions. Further, the means for calculating the connectivity between the pixel of interest and its surrounding pixels includes: a direction of a gradient of the pixel of interest and a direction of a gradient of adjacent pixels on both sides forming a 90 ° direction;
It is preferable that the connection is made when the angle between the gradient direction of the target pixel and the gradient direction is within 45 °.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an image processing method and apparatus according to the present invention will be described in detail based on preferred embodiments shown in the accompanying drawings.

FIG. 1 is a block diagram showing an embodiment of a digital photo printer to which an image processing apparatus for implementing the image processing method of the present invention is applied. A digital photo printer (hereinafter referred to as a photo printer) 10 shown in FIG.
A scanner (image reading device) 12 for photoelectrically reading an image captured on the film F, image processing such as edge extraction of image data (image information) read by the scanner 12, operation of the entire photo printer 10, and An image processing device 14 for performing control and the like, and a photosensitive material (photographic paper) is image-exposed with a light beam modulated according to the image data output from the image processing device 14, developed and processed (finished).
And an image recording device 16 for outputting an image as a print. The image processing apparatus 14 includes a keyboard 18a and a mouse 18 for inputting various conditions such as input, setting, selection and instruction of processing, and instructions such as color / density correction.
and an operation system 18 having an image b, a monitor 20 for displaying an image read by the scanner 12, various operation instructions, a setting / registration screen for various conditions, and the like.

The scanner 12 is a device for photoelectrically reading an image photographed on a film F or the like one frame at a time.
A variable aperture 24, a diffusion box 26 for making the reading light incident on the film F uniform in the surface direction of the film F, a carrier 28 of the film F, and an imaging lens unit 30
, An image sensor 32 having a three-line CCD sensor corresponding to reading of image densities of R (red), G (green), and B (blue), an amplifier (amplifier) 33, and an A / D (analog / digital). ) A converter 34.

In the photo printer 10, a dedicated carrier 28 which can be freely attached to the main body of the scanner 12 is provided with a film F such as a new photographic system (Advanced Photo System) or a 135 size negative (or reversal) film.
Are prepared in accordance with the type and size of the film, the form of the film such as strips and slides, and the like, and can be adapted to various films and processes by replacing the carrier 28. An image (frame) photographed on a film and provided for printing is conveyed by the carrier 28 to a predetermined reading position. Also, as is well known, a magnetic recording medium is formed on the film of the new photographic system, and a cartridge ID, a film picker, and the like are recorded. And various data such as the model of the developing machine can be recorded. The carrier 28 corresponding to the film (cartridge) of the new photo system includes:
The magnetic information reading means is arranged, reads the magnetic information when the film is transported to the reading position, and sends these various kinds of information to the image processing device 14.

When an image captured on the film F is read by the scanner 12, uniform reading light emitted from the light source 22 and adjusted in light amount by the variable aperture 24 and the diffusion box 26 is transmitted by the carrier 28. Incident on the film F located at the predetermined reading position,
By transmitting the light, projection light carrying an image photographed on the film F is obtained. The input of the color image signal is
The invention is not limited to a device that reads light transmitted through a film, but may be a reflection document or a device using an image captured by a digital camera.

In the illustrated example, the carrier 28 is a 24
It corresponds to a long film F (strips) such as a 35-size film or a cartridge of a new photo system. As schematically shown in FIG. 2A, the film F is located at a predetermined reading position. While the image sensor 32 is conveyed in the sub-scanning direction orthogonal to the extending direction (main scanning direction) of, for example, an RGB three-line CCD sensor, the longitudinal direction of the film F being conveyed. Conveying roller pairs 28a and 28 interposed in the sub-scanning direction
b, and a mask 28d having a slit 28c that extends in the main scanning direction and that corresponds to the reading position and regulates the projection light of the film F into a predetermined slit shape. The film F is positioned at the reading position by the carrier 28, and is conveyed in the sub-scanning direction, and receives the reading light.
As a result, the film F is two-dimensionally slit-scanned by the slit 28c extending in the main scanning direction, and the image of each frame captured on the film F is read.

The projection light of the film F is imaged on the light receiving surface of the image sensor 32 by the imaging lens unit 30. As shown in FIG. 2B, the image sensor 32
Is a line CCD sensor 32 for reading an R image
Line CCD sensor 32G that reads R and G images
CCD sensor 32 for reading image and B image
This is a so-called three-line color CCD sensor having B, and each line CCD sensor extends in the main scanning direction as described above. The projection light of the film F is separated into three primary colors of R, G and B by the image sensor 32 and read photoelectrically. The R, G, and B output signals output from the image sensor 32 are amplified by an amplifier 33 and sent to an A / D converter 34, where the A / D converter 34 has, for example, 12 bits of RGB, respectively.
After being converted into digital image data, the image processing device 14
Is output to

In the scanner 12, when reading an image photographed on the film F, a pre-scan (first image reading) for reading at a low resolution and a fine scan (first image reading) for obtaining image data of an output image are performed. (2nd image reading). Here, the pre-scan is performed under preset pre-scan reading conditions so that the image sensor 32 can read all images of the film F to be scanned by the scanner 12 without saturation. On the other hand, the fine scan is performed under the fine scan reading conditions set for each frame from the pre-scan data so that the image sensor 32 is saturated at a density slightly lower than the minimum density of the image (frame). The prescan and fine scan output image signals are basically the same image data except that the resolution and the output image signal level are different.

Note that the scanner 12 used in the photo printer 10 is not limited to the one that performs the slit scanning reading, and may be the one that performs the surface reading that reads the entire surface of the film image of one frame at a time. Good. In this case, for example, an area sensor such as an area CCD sensor is used, and means for inserting R, G, and B color filters are provided between the light source 22 and the film F, and inserted into the optical path of the light emitted from the light source 22. The reading light transmitted through the color filters is irradiated onto the entire surface of the film F, and the transmitted light is imaged on the area CCD sensor to read the entire image of the film by sequentially switching the R, G, and B color filters. Accordingly, the image photographed on the film F is separated into three primary colors and read.

As described above, the digital image data signal output from the scanner 12 is output to the image processing device 14 that performs the image processing method of the present invention. FIG. 3 shows a block diagram of the image processing device (hereinafter, referred to as a processing device) 14. Here, the processing device 14 includes a scanner correction unit 36, a LOG converter 38, a pre-scan (frame)
Memory 40, fine scan (frame) memory 4
2. It has a pre-scan data processing section 44, a fine scan data processing section 46 for performing edge portion extraction and various image processing features of the present invention, and a condition setting section 48.
FIG. 3 mainly shows a portion related to image processing. The processing device 14 includes a CP that controls and manages the entire photo printer 10 including the processing device 14.
U, a memory for recording information necessary for operation of the photo printer 10 and the like are provided.
Are connected to each part via the CPU or the like (CPU bus).

The R, G, and B image signals input from the scanner 12 to the processing device 14, for example, 12-bit digital image data, are input to the scanner correction unit 36.
The scanner correction unit 36 performs DC offset correction to correct sensitivity variation and dark current of each pixel of RGB digital image data due to the three-line CCD sensors 32R, 32G, and 32B of the image sensor 32 of the scanner 12,
It performs data correction of the read image data such as darkness correction, defective pixel correction, and shading correction. The digital image signal that has been subjected to the correction processing of the sensitivity variation and dark current for each pixel by the scanner correction unit 36 is output to the LOG conversion processor. The LOG converter 38 converts the digital image data into gradation data by performing logarithmic conversion processing, and converts the digital image data into digital image density data. For example, the LOG converter 38 corrects the digital image data using the lookup table (LUT) by the scanner correction unit 36. For example, the converted 12-bit digital image data is converted into 10-bit (0 to 1023) digital image density data.

The digital image density data converted by the LOG converter 38 is stored (stored) in the pre-scan memory 40 if it is pre-scan image data and in the fine scan memory 42 if it is fine scan image data. . The pre-scan memory 40 is obtained by pre-scanning the film F by the scanner 12,
This is a frame memory for storing or storing the low-resolution image density data of the entire frame of the film F subjected to various data correction and logarithmic conversion processing for each of the RGB colors. The prescan memory 40 needs at least a capacity capable of storing image density data of one frame of the film F in three colors of RGB, but may have a capacity capable of storing image density data of a plurality of frames, It may have a large number of memories each having a capacity of one frame. The pre-scan image data stored in the pre-scan memory 40 is read by the pre-scan data processing unit 44.

On the other hand, the fine scan memory 42 stores the high-resolution image density data of the entire frame of the film F obtained by the fine scan of the film F by the scanner 12 and subjected to various data correction and logarithmic conversion processing. This is a frame memory for storing or storing for each color. The fine scan memory 42 has a capacity to store at least image density data of three colors of RGB of two images of the film F, and while writing image density data of one frame, image density of another one frame is written. It is preferable that the data is read out and the fine scan data processing section 46 simultaneously performs various processes for extracting the edge portion, which is a feature of the present invention. However, the present invention is not limited to this, and the present invention is not limited to this. May have a capacity to store the image density data of each frame and process the image data frame by frame. Further, a memory having a large number of memories for one frame may be used, for example, a toggle memory. The fine scan image data stored in the fine scan memory 42 is read out by the fine scan data processing unit 46.

A prescan data processing unit 44 that performs various image processes required for display on the monitor 20 on the prescan image data stored in the prescan memory 40
Has an image processing unit 50 and an image data conversion unit 52. Here, the image processing unit 50 includes a condition setting unit 4 described below.
The image data read by the scanner 12 and stored in the pre-scan memory 40 in accordance with the image processing conditions set by the scanner 8 in the desired image quality,
In order to reproduce a color image on the RT display screen, predetermined processing such as gradation correction, color conversion, density conversion, and the like is performed by a lookup table (hereinafter, represented by LUT) or matrix (hereinafter, represented by MTX) calculation. This is for performing image processing. The image data conversion unit 52 includes the image processing unit 5
0 is thinned out as necessary to match the resolution of the monitor 20, and
Using a D (three-dimensional) LUT or the like, the data is converted into image data corresponding to the display on the monitor 20 and displayed on the monitor 20. The processing conditions in the image processing unit 50 are set by a condition setting unit 48 described later.

On the other hand, the fine scan image data stored in the fine scan memory 42 is added to the image recording device 1.
The fine scan data processing unit 46 for performing various image processing necessary for outputting as a color print from the image processing unit 6 and extracting an edge portion of the present invention includes an image processing unit 54 and an image data conversion unit 56. Here, the image processing unit 54
The image data read by the scanner 12 and stored in the fine scan memory 42 according to the image processing conditions set by the condition setting unit 48 to be described later are added to the color paper with desired density, gradation and color tone as color prints. A LUT, an MTX calculator, a low-pass filter, an adder / subtractor, and the like are used to reproduce a color image or a high-quality image in which sharpness enhancement, graininess suppression, and the like have been performed after edge portion extraction, which is the object of the present invention. It is for performing various image processing such as color balance adjustment, gradation adjustment, color adjustment, density adjustment, saturation adjustment, electronic scaling, and sharpness enhancement (edge enhancement; sharpening). The details will be described later. I do.

The image data conversion unit 56 includes an image processing unit 54
Is converted into image data corresponding to image recording by the image recording device 16 using, for example, a 3DLUT and supplied to the image recording device 16. The image recording device 16 includes a fine scan data processing unit 46.
Is output as a finished print in which a color image is reproduced based on the image data output from the printer.

The processing conditions in the image processing section 54 are set in the condition setting section 48. The condition setting unit 48
Various processing conditions in the fine scan data processing unit 46 are set. The condition setting unit 48 includes a setup unit 58, a key correction unit 60, and a parameter integration unit 62. The setup unit 58 sets the fine scan reading conditions using the prescan image data and supplies the conditions to the scanner 12, and creates the image processing conditions for the prescan data processing unit 44 and the fine scan data processing unit 46. (Calculation) and supplies the result to the parameter integration unit 62.

More specifically, the setup section 58 reads out pre-scan image data from the pre-scan memory 40, creates a density histogram, average density, LATD (large area transmission density), and highlights from the pre-scan image data. Calculation of image feature amounts such as (lowest density) and shadow (highest density) is performed. Based on the calculated image feature amount, the image sensor 32 has a density slightly lower than the lowest density of the image.
In order to saturate the (line CCD sensors 32R, 32G, 32B), the reading conditions of the fine scan, for example, the light amount of the light source 22, the aperture value of the variable aperture 24, and the (each line CCD sensor 32R, 32G, 32B)
) Set the accumulation time. The fine scan reading condition may change all the elements corresponding to the output level of the image sensor with respect to the prescan reading condition, or change only one element such as the aperture value. Alternatively, only a plurality of factors such as the aperture value and the accumulation time may be changed. Furthermore, the setup unit 5
Reference numeral 8 sets image processing conditions such as the above-described color balance adjustment and gradation adjustment according to a density histogram, an image feature amount, and an instruction from an operator performed as necessary.

The key correction unit 60 includes an adjustment amount such as density (brightness), color, contrast, sharpness, and saturation set by a key (not shown) provided on the keyboard 18a and the operation system 18, and a mouse 18b. The amount of adjustment of image processing conditions (for example, LU
A correction amount of T) is calculated, parameters are set, and the parameters are supplied to the parameter integration unit 62. The parameter integration unit 62 receives the image processing conditions set by the setup unit 58, and transmits the supplied image processing conditions to the image processing unit 50 of the pre-scan data processing unit 44 and the image processing unit 54 of the fine scan data processing unit 46. After setting, the image processing condition set for each part is corrected (adjusted) or the image processing condition is reset according to the adjustment amount calculated by the key correction unit 60.

Next, the image processing unit 54 of the fine scan data processing unit 46 for extracting an edge portion, which is a feature of the present invention, will be described in detail. FIG. 4 shows the image processing unit 5
FIG. 4 is a block diagram schematically showing an example of the fourth embodiment. As shown in FIG. 4, the image processing unit 54 includes a color density gradation conversion unit 64 that converts the density, color, and gradation of the image data, a saturation conversion unit 66 that converts the saturation of the image data, Digital magnification conversion (electronic scaling) means 6 for converting the number of pixels
8. An edge part extracting means 70 for extracting an edge part, and an image processing block 72 for performing various image processing such as sharpness enhancement, graininess suppression, and conversion of image data density dynamic range as necessary.

In the image processing section 54, the color density gradation conversion means 64 converts the image data into density data, color data and gradation data according to an LUT or the like. Further, the saturation conversion means 66 includes a color density gradation conversion means 6.
4 is to convert the saturation data of the image data obtained in step 4 in accordance with MTX operation or the like. The electronic scaling unit 68 interpolates the image data subjected to the saturation conversion by the saturation conversion unit 66 according to the size of the color image output to the color paper in the image recording device 16 and according to the output pixel density. In this case, the number of pixel data of the image data is increased or decreased by thinning or thinning. The edge part extracting means 70 determines whether the pixel of interest is an edge part or a granular part, and extracts an edge part. The image data subjected to the edge part extraction is input to the image processing block 72. Is done. In the image processing block 72, various other image processes are performed, such as performing sharpness enhancement on the edge portion extracted above, and performing grain suppression on the granular portion.

FIG. 5 is a block diagram schematically showing an embodiment of the edge portion extracting means 70. As shown in FIG. 5, the edge part extracting means 70 includes a color tone converting means 74, a luminance signal generating means 76, a template 78 used for calculating a gradient, and a means 8 for calculating a gradient.
0, means 81 for storing the calculated gradient, means 82 for calculating the connectivity of the pixel of interest, etc., means 84 for calculating the directionality of each RGB, and means 86 for determining and extracting an edge portion.

The tone conversion means 74 converts the tone signal level of the image data into an original image signal. The luminance signal generation unit 76 converts the R, G, and B color signals of the original image signal whose color tone has been converted by the color tone conversion unit 74 into a luminance signal.
The R, G, and B color signals are weighted and converted to a luminance signal Y. For example, the following expression is exemplified as the conversion expression. Y = 0.3R + 0.59G + 0.11B

The template 78 is for obtaining a gradient (gradient, corresponding to a so-called differential) representing the direction and intensity of the edge of the target pixel and its peripheral pixels. For example, in the case of a template composed of a 3 × 3 matrix, , And a Sobel operator as shown in FIG. 10 and are not particularly limited. Other than this, for example, the Prewitt operator, Kirsch
sch) operator and the like for calculating the magnitude and direction of the density gradient at the target pixel. It should be noted that the matrix constituting the template used in the present invention, which is these operators, is not limited to 3 × 3 as shown in FIG. 10, and its size and shape are not particularly limited. It may be. As a matrix of the template of the present invention, for example, even a square matrix such as 5 × 5, 7 × 7, 9 × 9, etc.
Non-square matrices such as x7 and 5x9 can also be used.

The gradient calculating means 80 calculates the gradient of each pixel of interest and its surrounding pixels using the template 78 for the image data converted into the luminance signal by the luminance signal generating means 76 as described above. For example, in the case of image data of 3 × 3 pixels centering on the pixel of interest, a product sum is calculated for these image data using eight types of 3 × 3 templates respectively indicating eight directions shown in FIG. The gradient of the pixel of interest is calculated with the obtained product sum and the direction indicated by the template in which the product sum has the maximum value as the intensity and direction, respectively. Similarly, a gradient is calculated for the peripheral pixels. Next, the gradient storage unit 81 stores the value of the gradient of each pixel of interest and its surrounding pixels calculated by the gradient calculation unit 80. The connectivity calculating unit 82 calculates the connectivity of the target pixel from the stored gradients of the target pixel and the peripheral pixels. That is, the connectivity calculating means 82 compares the direction of the gradient of the pixel of interest with the direction of the gradient of the neighboring pixels of the pixel of interest, determines the presence or absence of the connectivity according to the determination condition of the connectivity, and determines the connectivity. To be calculated, specifically, when the direction of the gradient of a predetermined number of pixels among these neighboring pixels matches the predetermined condition, it is determined that there is connectivity. The condition for determining connectivity will be described later. Also,
The directivity calculation means 84 for each RGB uses the template 78 for each of R, G, and B for each pixel of interest, and
The gradient is calculated for each direction and the directionality is calculated.

The edge part judging / extracting means 86 determines the pixel of interest from the edge part based on the connectivity of the pixel of interest calculated by the connectivity calculating means 82 and the direction calculated by the direction calculating means 84 for each of RGB. , Or not , that is, whether it is a granular part or not , an edge part is extracted. The image data for which the edge portion has been determined and extracted is sent to the image processing block 72, where image processing (frequency processing) such as sharpness enhancement or graininess suppression processing is performed.

The image processing apparatus according to the present invention and the digital photo printer using the same are basically configured as described above. Hereinafter, the operation of the image processing apparatus and the digital photo printer of the present invention and the image processing method of the present invention will be described.

The operator operates the film F (frame to be read).
Is loaded into the scanner 12, the film F is set at a predetermined position of the carrier 28, necessary information such as finishing information and a print size to be created is input, and then a print creation start is instructed. Thereby, the aperture value of the variable aperture 24 of the scanner 12 and the image sensor 3
2 (the line CCD sensors 32R, 32G, and 32B) are set according to the prescan reading conditions.
Thereafter, the carrier 28 conveys the film F in the sub-scanning direction at a speed corresponding to the pre-scan, and the pre-scan is started, the film F is slit-scanned, the projection light forms an image on the image sensor 32, and the film F Is separated into R, G, and B and photoelectrically read at a low resolution. Here, the pre-scan and the fine scan may be performed one frame at a time, or the pre-scan and the fine scan may be performed continuously for all frames or for a predetermined plurality of frames. In the following example, one frame of image reading will be described as an example to simplify the description.

The output signal of the image sensor 32 by the prescan is amplified by the amplifier 33 and
4 and converted into digital image data, which is then output to the image processing device 14 of the present invention. The digital image data input to the image processing device 14 is
After a predetermined correction such as a dark current of the image sensor 32 is performed in step (b), the correction value is sent to the LOG converter 38 and the density range corresponding to the pre-scan, for example, the density range of 4 with the density D is set to 10b.
The data is converted using the LUT so as to be assigned to the data of the it, and is converted into prescan image data. The prescan image data is stored in the prescan memory 40.

When the pre-scan image data is stored in the pre-scan memory 40, the setup unit 58 of the condition setting unit 48 reads it out, creates a density histogram,
Calculate image features such as highlights and shadows, etc.
Setting the scanning conditions for fine scan, the scanner 12
And various image processing conditions such as gradation adjustment and gray balance adjustment are set and supplied to the parameter integration unit 62. Parameter integration unit 6 receiving the image processing conditions
2 sets this in predetermined portions (hardware and software) of the pre-scan data processing unit 44 and the fine scan data processing unit 46.

When performing a test, prescan image data is read from the prescan memory 40 by the prescan data processing unit 44, image-processed under the image processing conditions set in the image processing unit 50, and The data is converted by the data converter 52 and displayed on the monitor 20 as a simulation image. The operator looks at the display on the monitor 20, confirms (verifies) the image, that is, the processing result, and adjusts the color, density, gradation, and the like using the adjustment keys and the like set on the keyboard 18a as necessary. The input of this adjustment is sent to the key correction unit 60, which calculates the correction amount of the image processing condition according to the adjustment input, and sends it to the parameter integration unit 62. The parameter integration unit 62 determines whether the image processing unit 50
And 54 LUT and MTX are corrected. Therefore, the image displayed on the monitor 20 also changes according to the correction, that is, the adjustment input by the operator.

When the operator determines that the image of the frame is proper (OK), the operator instructs the start of printing using the keyboard 18a or the like. Thus, the image processing conditions are determined, the aperture value of the variable aperture 24 is set in the scanner 12 in accordance with the reading conditions of the fine scan, and the carrier 28 moves the film F at a speed corresponding to the fine scan. It is conveyed and fine scan is started. When the test is not performed, the fine scan data processing unit 4 by the parameter integration unit 62
When the setting of the image processing condition in the image processing unit 54 of Step 6 is completed, the image processing condition is determined, and the fine scan is started. The fine scan is performed in the same manner as the pre-scan except that the reading conditions such as the aperture value of the variable aperture 24 are different. The output signal from the image sensor 32 is amplified by the amplifier 33, and the A / D converter of the processing device 14 The digital density data is converted into digital density data at 34, subjected to predetermined processing by a scanner correction unit 36, and sent to a LOG converter 38. In the LOG converter 38, the fine scan digital image data is processed with a higher density resolution than in the pre-scan. For example, the fine scan digital image data is converted by the LUT so as to allocate a density range of 2 in density D to 10-bit data. Is sent to the fine scan memory 42.

In the image processing section 54, the density data, color data and gradation data of the fine scan image data are converted by the color density gradation conversion means 64 in accordance with the look-up table.
The chroma data of the image signal is converted according to the matrix operation. Then, after the number of pixel data of the image data signal is increased or decreased by the electronic scaling unit 68 according to the size of the color image output to the color paper, the image signal is
It is input to the edge part extracting means 70.

The original image signal whose color tone has been converted by the color tone converting means 74 is converted into a luminance signal Y by a luminance signal generating means 76 and input to a gradient calculating means 80. In the gradient calculating means 80, the template 7
8 is used to calculate a gradient that is a vector quantity representing the magnitude and direction of the density gradient of the target pixel and its surrounding pixels.

For example, as shown in FIG. 11A, 3 × 3 image data including a target pixel and peripheral pixels is compared with FIG.
Using the 3 × 3 templates a1 to a8 as shown in FIG. 0, the product sum operation is performed as described above, and the gradient of the pixel of interest having the intensity and direction from the template corresponding to the largest product sum and the maximum product sum is obtained. Is calculated. Next, from these peripheral pixels, two adjacent pixels on both sides in a direction 90 ° (orthogonal direction) to the direction of the gradient of the pixel of interest are selected as neighboring pixels used to determine the presence or absence of connectivity, For these two adjacent pixels,
A 3 × 3 image centered on the adjacent pixel is taken, and the gradient of the adjacent pixel is calculated in the same manner. At this time,
In the case where the gradient direction of the target pixel is as shown by an arrow in FIG. 6, the one indicated by a circle is an adjacent pixel. For example,
Assuming that the direction of the 3 × 3 image data is the direction indicated by the arrow in FIG. 11B, the pixel value 1 at the upper left corner in FIG.
And the one with a pixel value of 5 in the lower right corner in the figure are two adjacent pixels.

The gradient calculating means 80 calculates a gradient for each of these adjacent pixels and finds the direction. The gradients of the target pixel and the peripheral pixels thus obtained are stored in the gradient storage unit 81. The connectivity calculating unit 82 determines the connectivity of the target pixel from the stored gradient directions of the target pixel and the peripheral pixels. When the directions of the gradients of two adjacent pixels are both within 45 ° of the direction of the gradient of the target pixel, the target pixel is determined to have connectivity.

On the other hand, directivity calculating means for each of RGB (hereinafter, referred to as “direction calculating means”)
The directionality calculating means) 84 similarly calculates gradients for the three image data of R, G, and B for the same pixel of interest using the template 78, and as shown in FIG. Is calculated.

The edge portion judging / extracting means 86 judges and extracts an edge portion from the connectivity of the pixel of interest calculated by the connectivity calculating means 82 and the directivity calculating means 84 and the directionality for each of RGB. When the pixel of interest has connectivity and the directionality of each RGB is within 45 °, the pixel of interest is determined to be an edge part, is left as an edge part, and is not so. In that case, it is discarded. Here, the directionality of each RGB is 45
Having a directivity within ° means that when focusing on a channel of interest, for example, G, both the directions of R and B are within 45 ° with respect to the direction of G, as shown in FIG.

For the pixel of interest determined to be an edge in this way, the direction of the gradient calculated by the gradient calculating means 80 is called the direction of the edge.
The size of the gradient is called edge strength. The edge strength indicates the slope of the edge, and an edge having a large strength can be said to be a sharp edge having a large slope. Therefore, using this intensity, it is possible to control the degree of sharpness enhancement and graininess suppression of an image, that is, the gain that determines the intensity. Note that the gradient intensity is set to 0 for a target pixel that is not determined to be an edge portion. For example, in the edge portion, the gain of the high frequency component of the image data is increased according to the gradient intensity to enhance the sharpness of the image, and in the flat portion (granular portion) not including the edge portion, the gain of the medium frequency component of the image data is increased. By reducing the intensity according to the intensity of the gradient, it is possible to suppress roughness or the like due to the granularity of the image.

More specifically, up to a predetermined gradient intensity, the value increases from a value smaller than 1.0 to a value larger than 1.0 according to the gradient intensity. A look for medium / high frequency gain control, where the function that becomes constant is the gain of the medium frequency component, and the function that is similar to the gain of this medium frequency component and takes a value larger than the gain of this medium frequency component is the gain of the high frequency component. An up-table (LUT) is prepared, and the sharpness enhancement of the edge portion considers the intensity of the gradient as the intensity of the edge, and calculates the gain intensity by the gain control LUT of the middle and high frequency components according to the gradient intensity. At the non-edge portion, the gradient intensity is considered to be 0, and the gain intensity is calculated similarly. Thus, gain control is performed according to the gradient intensity of the pixel of interest. At the edge portion, the sharpness is enhanced by enhancing both the middle and high frequency components. A pixel having a larger edge strength has a stronger sharpness. At this time, the gain of the medium frequency component is made smaller than the gain of the high frequency component in order to suppress granular roughness accompanying sharpness enhancement. In the non-edge portion, particularly, the gain of this medium frequency component is set to 1.0.
By making the film smaller, the graininess can be effectively suppressed even in a film having poor graininess. At the same time, the gain of the high-frequency component is set to 1.0 in order to maintain the sharpness.
Make it bigger. At this time, a sharpness image with a sense of edge can be obtained by applying a gain slightly weaker than the gain of the edge portion. Further, the negative density of the target pixel may be calculated, and in the case of underexposure, the granularity may be more strongly suppressed.

The image data whose edge has been determined and extracted is sent to the image processing block 72. In the image processing block 72, sharpness enhancement, graininess suppression, and other image processing are performed. Thereafter, the image data that has been subjected to the image processing is input to the image data conversion unit 56, converted into image output image data, and then output from the image processing device 14 to the image recording device 16.

The image recording device 16 is a printer (printing device) for exposing a photosensitive material (printing paper) in accordance with input image data to record a latent image, and a printer for performing a predetermined process on the exposed photosensitive material. And a processor (developing device) that outputs the result. In a printer, for example, after cutting a photosensitive material into a predetermined length corresponding to a finished print, a back print is recorded, and then three types of light, R exposure, G exposure, and B exposure, according to the spectral sensitivity characteristics of the photosensitive material. The light beam is modulated in accordance with the image data output from the processing device 14 and is deflected in the main scanning direction, and the photosensitive material is conveyed in the sub-scanning direction orthogonal to the main scanning direction. A latent image is recorded by two-dimensional scanning exposure and supplied to a processor. The processor that has received the photosensitive material on which the latent image has been formed performs a predetermined wet development process such as color development, bleach-fixing, and washing with water, and then dry to obtain a finished print.
The sheets are sorted into a predetermined unit such as one film and accumulated.

The operation of the image processing apparatus and the digital photo printer according to the present invention has been basically described above, including the image processing method according to the present invention. The image processing method according to the present invention will be described below with reference to FIG. I will explain. FIG. 8 is a flowchart showing each processing of the image processing method of the present invention performed by the above-described edge part extracting means 70. In FIG. 8, in step 1, the gradient direction and intensity of the pixel of interest are calculated. In step 2, the gradient of two adjacent pixels in the direction of the gradient of the target pixel and the direction of 90 ° is calculated, and the directionality is checked. In step 3, the directions of two adjacent pixels and the pixel of interest are compared, and if both have a directionality within 45 °, it is determined that there is connectivity. In step 4, the gradient of the target pixel is calculated for each of the R, G, and B channels, and the direction is checked. Finally, in step 5, it is determined whether or not the pixel of interest is an edge based on the previously obtained connectivity and the directionality for each of RGB.

In the above embodiment, the pixel of interest is the center,
In order to determine the presence / absence of connectivity with respect to image data of 3 × 3 pixels having eight surrounding pixels as peripheral pixels, using eight types of 3 × 3 templates respectively indicating eight directions shown in FIG. As a method of taking the neighboring pixels used in the above, two neighboring pixels on both sides in the direction of 90 ° in the direction of the gradient of the pixel of interest are selected from the peripheral pixels, and the connectivity determination condition, that is, the connectivity determination condition The angle between the direction of the gradient of the adjacent pixels on both sides thereof and the direction of the gradient of the pixel of interest is both within 45 °, the determination condition of the edge portion, that is, connectivity as a condition to be the edge portion, Although the case of selecting that the directionality for each RGB has a directionality within 45 ° has been described as a representative example, the present invention is not limited to this, and the pixel of interest and peripheral pixels and connectivity are not limited to this. Any template may be used for determining the presence / absence of neighboring pixels, and any template may be used as long as it corresponds to the method of taking the target pixel and surrounding pixels. As long as the determination of the portion is possible, the determination condition of the connectivity and the determination condition of the edge portion may be set in any manner.

For example, the size and shape of the pixel of interest and its surrounding pixels used in the present invention are not particularly limited, and may be 5 × 5, 7 × 7, 9 × 9, etc. in addition to 3 × 3 pixels. It may be a square pixel, a non-square rectangular pixel such as 3 × 5, 3 × 7, or 5 × 9, or a pixel of any shape. Therefore, the surrounding pixels are also 14 pixels in addition to the 8 pixels around the pixel of interest.
The image may be viewed over a wider pixel range such as pixels, 20 pixels, and 24 pixels. The size, shape, and number of the templates used in the present invention are not particularly limited. In addition to the 3 × 3 square matrix, 5 × 5 or 7 × 7 depending on the target pixel used and its surrounding pixels. And 9x
Even a square matrix such as 9 can be 3 × 5, 3 × 7 or 5 × 9
Or a non-square rectangular matrix, or a template of another shape. The number of templates is not limited to the eight types indicating eight directions, and may be more or less.

There are no particular restrictions on the manner in which neighboring pixels are determined for determining the presence or absence of connectivity. The size and shape of the neighboring pixels are not particularly limited, and two adjacent pixels on both sides orthogonal to the gradient direction of the target pixel may be used. It is also possible to view several peripheral pixels at the same time. For example, pixels other than the above-described square pixels, non-square rectangular pixels, and pixels of any shape except for the pixel of interest around the pixel of interest may be used. With the pixel as the center, there are 1 × 3 pixels, 1 × 5 pixels, 3 × 5 pixels, 3 × 7 pixels, and 5 × 5 pixels as shown in FIG. 12 which are more present in the direction orthogonal to the direction (gradient) of the target pixel. It is preferable that the pixel be a pixel obtained by removing a central pixel from a rectangular pixel such as a × 9 pixel. The reason for this is that the edge of the image has a certain length, and therefore, it is better to widen it in the length direction of the edge, that is, in the direction orthogonal to the direction of the pixel of interest, in order to check the connectivity. When the direction of the edge of the target pixel is oblique, a plurality of pixels are taken in an oblique direction orthogonal to the direction.
When pixels are taken, there is an arrangement of 4 pixels between the arrangement of 5 diagonal pixels, but it is preferable to take all 23 pixels including all of them as neighboring pixels.

Further, in the present invention, as a condition for judging that there is connectivity, the gradient direction of the target pixel is compared with the gradient direction of the neighboring pixels of the target pixel, and a predetermined number of neighboring pixels are determined. Although the direction of the pixel matches the predetermined condition, it is preferable that 50% or more of the directions of the neighboring pixels are within 45 ° of the direction of the target pixel. That is, the ratio of the predetermined number of pixels in which the directionality of the gradient in the neighboring pixels for determining the connectivity matches the target pixel under the predetermined condition should be at least 50%, more preferably 70% or more, and still more preferably. Is 80
% Or more, most preferably 90% or more, and of course 1%
Although it may be 00%, it can be changed according to various conditions as described later. Further, the predetermined condition of the direction of the neighboring pixels is preferably within 45 °, more preferably 3 °.
It is preferable that the angle be within 0 °, and of course, it may be 0 ° (perfect match) or within any angle within the range of 0 ° to 45 °. Preferably have a certain width. Also in this regard, 4
It is preferable to be within 5 ° or 30 °. This predetermined condition can be changed according to various conditions described later. Of course, as described above, one of the extremely preferable connectivity determination conditions is that all the gradient directions of neighboring pixels for determining connectivity such as two adjacent pixels on both sides of the target pixel are within 45 °. .

The directional condition for each RGB, which is one of the conditions for determining an edge portion to be determined to be an edge portion, is exactly the same as the above-described predetermined condition for directionality of neighboring pixels for connectivity determination. Although the angle may be within any angle within the range of 0 ° to 45 °, the directionality within 45 ° of the above embodiment is preferable, and the directionality within 30 ° is more preferable. The condition of this directionality can also be changed according to various conditions described later. Further, a predetermined condition of the directivity of neighboring pixels for determining connectivity and R
The directional condition for each GB is the same as within 45 ° in the above embodiment, but may be different.

Here, as described above, the size (shape) and number of the template, the method of taking neighboring pixels, the condition for determining connectivity, that is, the prescribed condition of directivity of neighboring pixels and the prescribed condition of directivity in neighboring pixels. The ratio of the pixels satisfying the condition and the condition of the directionality for each RGB can be variously changed.
For example, the document type (high-sensitivity negative film, low-sensitivity negative film, reversal film, reflection document, image captured by a digital camera, etc.), document size (240 (A
PS), 110, 135, 120 and 220 (Browney), print magnification, print size, camera type, and exposure conditions (under, over) during shooting. At least one
Preferably, it is determined according to the following. In addition, it goes without saying that the method of determining the various conditions according to at least one of these items is not limited. Here, among these items, for example, the print magnification and the document type are
This greatly affects conditions such as how to take neighboring pixels, conditions for determining connectivity, and conditions for directionality for each of RGB. However, when the print magnification changes, the way to take neighboring pixels changes mainly. It is preferable not to change the original type, for example, when the film speed changes, these conditions are mainly changed, that is, the conditions are stricter in the high-speed film than in the low-speed film, and the way of taking the neighboring pixels is not changed. Is more preferred. Furthermore, it is preferable that, under the exposure conditions at the time of photographing, in the case of under-exposure, these conditions are mainly reduced and the way of taking the neighboring pixels is not changed.

FIG. 9 shows the effect of this embodiment in comparison with the conventional one. First, FIG. 9A shows a conventional case in which the directionality of each of RGB is not determined, and one of adjacent pixels is determined to be within the allowable range (within 45 °). In this case, a large number of white particles representing granular portions remain in addition to the white line which is the original edge portion. In contrast,
In (b), the directionality of each RGB is not seen, but both of the adjacent pixels are within the allowable range (within 45 °). According to this, the determination of the edge portion becomes more accurate, and the number of particles is considerably reduced. (C) is the same as the conventional case for the adjacent pixels, in which one of the adjacent pixels is within the allowable range. In addition, the direction of each RGB is determined. . In other words, this is the case where both channels other than the channel of interest have an acceptable directionality. Again, considerable particles have been removed.

FIG. 9D shows the case of the present embodiment described above in which (b) and (c) are combined.
That is, with respect to adjacent pixels, both are within tolerance,
In addition, the directionality of each RGB (both channels other than the channel of interest are within the allowable range) is also viewed. In this case, the edge portion determination becomes more accurate, and almost no particles remain. Further, (e) shows that both of the adjacent pixels have the same direction, and that of RGB also has the same direction except for the channel of interest. The effect is almost the same as that of FIG.

As described above, according to the present embodiment, both of the adjacent pixels are within the allowable range and the directionality of each RGB is viewed, so that the edge portion can be extracted with higher accuracy than in the related art. became.

Although the image processing method and apparatus of the present invention have been described in detail above, the present invention is not limited to the above examples, and various improvements and modifications can be made without departing from the gist of the present invention. Of course, you may.

[0068]

As described above, according to the present invention, it is possible to extract an edge portion more accurately than in the past, and to obtain a high quality output image.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of a digital photo printer to which an image processing apparatus that performs an image processing method of the present invention is applied.

2A is a schematic perspective view for explaining one embodiment of a carrier mounted on the digital photo printer shown in FIG. 1, and FIG. 2B is a digital photo printer shown in FIG. 1 is a conceptual diagram of an embodiment of the image sensor of FIG.

FIG. 3 is a block diagram of an embodiment of an image processing apparatus that performs the image processing method of the present invention for the digital photo printer shown in FIG.

FIG. 4 is a block diagram of an embodiment of an image processing unit of a fine scan image data processing unit of the image processing apparatus shown in FIG.

FIG. 5 is a block diagram showing an outline of an edge portion extracting means shown in FIG. 4;

FIGS. 6A to 6D are explanatory diagrams each showing a target pixel and its adjacent pixels.

FIGS. 7A to 7C show RGs according to the present embodiment.
It is explanatory drawing which shows the directionality determination for every B.

FIG. 8 is a flowchart illustrating an outline of an edge part extracting method according to the present embodiment.

FIGS. 9A and 9B are photographs instead of drawings showing the effects of the present embodiment in comparison with the conventional case, in which FIG. 9A is the conventional case and FIGS. 9B to 9E are the present embodiment.

FIGS. 10 (a1) to (a8) are explanatory diagrams showing an example of a template used for edge portion extraction.

FIG. 11A is an explanatory diagram illustrating an example of a 3 × 3 image including a target pixel and peripheral pixels thereof, and FIG.
FIG. 3A is an explanatory diagram illustrating a direction of a gradient of a pixel of interest in FIG.

FIG. 12 is an explanatory diagram illustrating an example of a neighboring pixel selected for connectivity determination with respect to a target pixel in the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Digital photo printer 12 Scanner 14 (Image) processing device 16 Image recording device 18 Operation system 18a Keyboard 18b Mouse 20 Monitor 22 Light source 24 Variable aperture 26 Diffusion box 28 Carrier 30 Imaging lens unit 32 Image sensor 32R, 32G, 32B Line CCD Sensor 34 A / D converter 36 Scanner correction unit 38 LOG converter 40 Prescan (frame) memory 42 Finescan (frame) memory 44 Prescan data processing unit 46 Fine scan data processing unit 48 Condition setting unit 50, 54 Image data Conversion unit 58 Setup unit 60 Key correction unit 62 Parameter integration unit 64 Color density gradation conversion means 66 Saturation conversion means 68 Electronic scaling (digital magnification conversion) means 70 Edge part extraction means 72 Processing block 74 the color tone conversion means 76 the luminance signal generating means 78 template 80 gradient calculating means 81 gradient storing means 82 connected with calculation means 84 (per RGB) direction calculation means 86 edge determination and extraction means

────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] April 28, 1999 (1999.4.2
8)

[Procedure amendment 1]

[Document name to be amended] Drawing

[Correction target item name] Fig. 9

[Correction method] Change

[Correction contents]

FIG. 9

Claims (12)

[Claims] 1. An image processing method for performing predetermined image processing on a color image signal, comprising: calculating a pixel value of a target pixel of an image formed by the color image signal to be processed and pixel values of peripheral pixels thereof A gradient representing the direction and intensity of the edge of the target pixel and its surrounding pixels is calculated and stored. From the stored direction of the gradient, the connectivity between the target pixel and its surrounding pixels is calculated, and each R is calculated. , G, and B channels, the gradient of the pixel of interest is calculated to calculate the directionality of the pixel of interest for each of RGB, and from the connectivity and the directionality of each of the RGB, the pixel of interest is By determining whether or not it is an edge portion,
An image processing method comprising extracting an edge portion. 2. Computing the connectivity between the target pixel and its peripheral pixels by comparing the direction of the gradient of the target pixel with the direction of the gradient of neighboring pixels of the target pixel. 2. The image processing method according to claim 1, wherein the connection is established when the directions of the number of pixels match predetermined conditions. 3. The case where the directions of a predetermined number of pixels among the neighboring pixels match a predetermined condition is that 50% or more of the directions of the neighboring pixels are within 45 ° of the direction of the target pixel. 3. The image processing method according to claim 2, wherein: 4. The 1 × 3 pixel, the 1 × 5 pixel, the 3 × 5 pixel, and the 3 × 7 pixel which are present in the direction perpendicular to the direction of the gradient of the pixel of interest with the pixel of interest as the center, The image processing method according to claim 2, wherein the pixel is a pixel obtained by removing a center pixel from one of a pixel and a 5 × 9 pixel. 5. A method for taking the neighboring pixels, a predetermined number of the pixels whose directions match the predetermined conditions, and a condition for determining the connectivity of pixels including the predetermined conditions where the directions of the predetermined number of pixels match. 5. The image processing method according to claim 2, wherein the image processing method is determined according to at least one of a document type, a document size, a print magnification, a camera type, and exposure conditions. 6. When calculating the connectivity between the target pixel and its surrounding pixels, the gradient direction of the target pixel and the gradient directions of adjacent pixels on both sides forming a 90 ° direction are determined by the gradient of the target pixel. 2. The image processing method according to claim 1, wherein the connection is established when the angles formed by the directions are within 45 degrees. 7. An image processing apparatus for performing predetermined image processing on a color image signal, comprising: determining a target pixel of an image formed by the color image signal to be processed and pixel values of peripheral pixels thereof Means for calculating a gradient representing the direction and intensity of an edge of the pixel of interest and its surrounding pixels; means for storing the calculated gradient; and, from the direction of the stored gradient, the pixel of interest and its surrounding pixels Means for calculating the connectivity of each of the R, G, and B channels, and means for calculating the gradient of the pixel of interest for each of the R, G, and B channels, thereby calculating the directionality of each of the RGB of the pixel of interest. Means for determining whether or not the pixel of interest is an edge portion from the directionality of each of the RGB, thereby performing edge portion extraction. Processing apparatus. 8. The means for calculating the connectivity between the target pixel and its surrounding pixels includes: a direction of a gradient of the target pixel;
The method according to claim 7, wherein the directions of gradients of neighboring pixels of the target pixel are compared, and if the directions of a predetermined number of pixels among the neighboring pixels match a predetermined condition, it is determined that there is connectivity. Image processing device. 9. The case where the direction of a predetermined number of pixels among the neighboring pixels matches a predetermined condition means that 50% or more of the directions of the neighboring pixels are within 45 ° of the direction of the pixel of interest. The image processing apparatus according to claim 8, wherein: 10. The 1 × 3 pixel, 1 × 5 pixel, 3 × 5 pixel, which extends in a direction orthogonal to the direction of the gradient of the pixel of interest with the pixel of interest as a center.
The image processing apparatus according to claim 8, wherein the pixel is a pixel obtained by removing a central pixel from one of 3 × 7 pixels and 5 × 9 pixels. 11. A method for obtaining the neighboring pixels, a predetermined number of the pixels whose directions match the predetermined conditions, and a condition for determining the connectivity of pixels including the predetermined conditions where the directions of the predetermined number of pixels match. The image processing apparatus according to claim 8, wherein the image processing apparatus is determined according to at least one of a document type, a document size, a print magnification, a camera type, and exposure conditions. 12. The means for calculating the connectivity between the pixel of interest and its surrounding pixels comprises: 8. The image processing apparatus according to claim 7, wherein the connection is established when the angles formed by the directions of the gradients are both within 45 degrees.