JP2023044689A - Image processing unit - Google Patents

Abstract

To provide an image processing unit capable of storing image data outside an RGB color space into the RGB color space while hardly causing discomfort according to user's preference, in the case where an RGB image generated after a digital image has been processed is outside the area of a color space.SOLUTION: An image processing unit 1 includes: an image processing section 11 executing predetermined processing on input image data to output an RGB image; a conversion section 10a for converting a pixel value of the RGB image into a signal value of image data expressed in other color space, the signal value being separated into luminance or brightness and a color component, and a correction section 10b for correcting the signal value. The pixel value of an RGB image output from the image processing section 11 includes a pixel value outside a color gamut corresponding to an RGB color space in the other color space. The correction section 10b corrects chroma while, for example, fixing the luminance or brightness and the hue of a signal value of image data after the conversion.SELECTED DRAWING: Figure 1

Description

The present invention relates to an image processing apparatus for correcting digital image data.

Conventionally, image data expressed in the YCbCr color space, such as digital video, is processed and then converted into the RGB color space on the output side. Since the YCbCr color space has a wider color space than the RGB color space, the values after conversion sometimes fall outside the color gamut of the RGB color space. In this case, the image quality will be degraded if the value outside the color gamut is simply stored in the RGB color space.

Therefore, in order to reduce changes in the brightness and hue of image data that may occur during such color conversion processing, while fixing the Y value (luminance value), CbCr (hue value) ) without changing the ratio (that is, the saturation is lowered while the hue remains the same) has been disclosed (see, for example, Patent Document 1).

Also, in a pixel of interest whose color coordinates are outside the reproducible color gamut in the color space of the output image, a plurality of color difference signals are generated while maintaining the luminance signal and maintaining the ratio of the plurality of color difference signals. An image processing apparatus that adjusts saturation while maintaining the hue of a pixel of interest is also disclosed (see, for example, Patent Document 2).

JP-A-2005-252393 Patent No. 6318497

By the way, in recent years, among machine learning, machine learning using a neural network has been applied to many fields. Especially in the field of image recognition, deep learning, which uses neural networks in a multi-layered structure, demonstrates high recognition accuracy. In multi-layered deep learning, image processing is performed using a convolutional neural network (CNN) that uses multiple convolutional layers and pooling layers for extracting input features.

In image data processing using this neural network, target image data is input to a machine learning model. Image data used in machine learning models is expressed as a collection of pixels (picture elements) partitioned by coordinates, and each pixel is assigned a pixel value (three values of RGB). In the machine learning model, all pixel values of target image data are input to input neurons and machine learning is executed. By using such a "trained model" of machine learning, "input image (arbitrary data: low-resolution image, for example) → trained model → output image (inference data: high-resolution image, for example)" An output image can be inferred from an input image.

However, when outputting an RGB image as an image after image processing using machine learning or the like, the RGB pixel values may be values outside the range of the RGB color space (values outside 0-255). In this case, if the value outside the color gamut is simply stored in the RGB color space, the RGB ratio will change, causing changes in the brightness and hue of the image data, resulting in image quality deterioration such as bleeding.

The present invention has been made in view of the above problems, and in the case where an RGB image generated after processing a digital image is outside the area of the RGB color space, the image data outside the RGB color space can be changed according to the user's preference. It is also an object of the present invention to provide an image processing device that is less likely to cause a sense of incongruity and that can be accommodated within the RGB color space.

In order to achieve the above object, an image processing apparatus according to the present invention comprises image processing means for executing predetermined processing on input image data and outputting an RGB image; conversion means for converting pixel values of an image into signal values of image data represented in another color space separated into luminance or lightness and color components; and a correcting means for correcting the signal value when it exists outside the color gamut corresponding to the RGB color space in the other color space, wherein the pixel value of the RGB image output from the image processing means is , and pixel values outside the color gamut corresponding to the RGB color space in the other color space.

In this image processing apparatus, the correcting means corrects the signal value of the image data after the conversion by adjusting the luminance or the lightness if the luminance or the lightness exceeds the range that the color gamut corresponding to the RGB color space can take. is rounded to the nearest value, and then its luminance or lightness and hue are fixed, and the saturation is corrected to move to the first boundary point of the color gamut corresponding to the RGB color space in the other color space. is preferred.

In this image processing device, if the signal value of the image data after conversion is such that the saturation exceeds the possible range of the color gamut corresponding to the RGB color space, the saturation is rounded to the nearest value. After that, it is preferable to correct the luminance or lightness while fixing the saturation and hue, and move to the second boundary point of the color gamut corresponding to the RGB color space in the other color space.

In this image processing apparatus, the correction means corrects the luminance or lightness and saturation of the signal values of the converted image data while fixing the hue, and converts the signal values into the RGB color space in the other color space. It is preferable to move to the boundary with the corresponding color gamut.

The image processing apparatus may further comprise intermediate color computing means for obtaining an intermediate color by multiplying the color components obtained at the first boundary point and the second boundary point by weights set in advance and taking an average value. preferable.

In this image processing apparatus, the image processing means preferably uses a machine learning model to execute a predetermined inference process on the input image data and outputs an RGB image.

In order to achieve the above object, an image processing method according to the present invention provides an image processing step of executing a predetermined process on input image data to output an RGB image, and an image processing step of outputting an RGB image in the image processing step. a conversion step of converting the pixel values of the RGB image into signal values of image data represented in another color space separated into luminance or brightness and color components; and a signal of the image data after conversion in the conversion step. and a correction step of correcting the signal value when the value is outside the color gamut corresponding to the RGB color space in the other color space, the pixel of the RGB image output in the image processing step. The values are characterized in that they include pixel values outside the color gamut corresponding to the RGB color space in said other color space.

In order to achieve the above object, a program according to the present invention is a program that functions in an image processing apparatus, and includes an image processing step of executing predetermined processing on input image data and outputting an RGB image. a conversion step of converting the pixel values of the RGB image output in the image processing step into signal values of image data represented in another color space separated into luminance or lightness and color components; a correction step of correcting the signal values of the image data after conversion in the step when the signal values are outside the color gamut corresponding to the RGB color space in the other color space, wherein The pixel values of the RGB image output in the step include pixel values outside the color gamut corresponding to the RGB color space in the other color space.

Further, in order to achieve the above object, the present invention transmits input data to the image processing device or program described above, and receives and uses output data from the image processing device or program. A computer system characterized by:

In order to achieve the above object, an image processing apparatus according to the present invention comprises image processing means for executing predetermined processing on input image data and outputting an RGB image; conversion means for converting pixel values of an image into signal values of image data represented in another color space separated into luminance or lightness and color components; and correction means for correcting the signal value when the signal value exists outside the color gamut corresponding to the RGB color space in the other color space, wherein the correction means corrects the signal value of the converted image data. is moved to the first boundary point of the color gamut corresponding to the RGB color space in the other color space while the luminance or lightness and hue are fixed, and the converted image correcting the luminance or lightness of the signal value of the data while fixing its saturation and hue, moving it to a second boundary point of the color gamut corresponding to the RGB color space in the other color space, and performing the image processing; The apparatus further comprises intermediate color calculation means for obtaining an intermediate color by multiplying the color components obtained at the first boundary point and the second boundary point by a preset coefficient and taking an average value. and

An image processing apparatus according to the present invention includes an image processing unit that executes predetermined processing on input image data and outputs an RGB image, and the signal values of the image data after conversion exist outside the color gamut corresponding to the RGB color space in the other color space. and a correction unit 10b that corrects the signal value in the case where the signal value is corrected. The pixel values of the RGB image output from the image processing unit include pixel values outside the color gamut corresponding to the RGB color space in other color spaces. The correction unit corrects the signal value of the image data after conversion, for example, while fixing the luminance or lightness and hue, and corrects the saturation to convert the signal value to the first color gamut corresponding to the RGB color space in the other color space. Move to boundary point. With this configuration, according to the present invention, image data outside the RGB color space can be accommodated within the RGB color space according to the user's preference without causing a sense of discomfort.

1 is a block diagram showing the configuration of an image processing device according to an embodiment of the present invention; FIG. 4A to 4C are diagrams for explaining a correction method in the same image processing apparatus; FIG. 4 is a flow chart showing an example of an operation procedure of the same image processing apparatus; It is a functional block diagram of an image processing device according to a modification of the embodiment of the present invention.

(Embodiment)
An image processing apparatus according to an embodiment of the present invention will be described with reference to FIGS. 1 to 3. FIG. Note that the image processing apparatus according to the present embodiment has an RGB image after image processing that is outside the color gamut of the RGB color space. It performs calculations.

First, each processing unit provided in the image processing apparatus 1 will be described with reference to FIG. The image processing apparatus 1 includes a control section 10, an image processing section 11, a storage section 12, a communication section 13, a display section 14, an operation section 15 and a reading section 16, as shown in FIG. Although the image processing apparatus 1 and the operation of the image processing apparatus 1 will be described below assuming one computer, the processing may be distributed among a plurality of computers.

The control unit 10 uses a processor such as a CPU and a memory to control components of the image processing apparatus 1 to realize various functions. In this embodiment, the control unit 10 performs image correction processing (color conversion processing) by executing an image processing program 1P stored in the storage unit 12, for example. That is, the control unit 10 performs a predetermined conversion on input target image data (target data), and performs calculations for correcting conversion values outside the color gamut.

The image processing unit 11 uses a processor such as a GPU or a dedicated circuit and a memory, and executes image processing according to control instructions from the control unit 10 . Note that the control unit 10 and the image processing unit 11 are configured as one piece of hardware (SoC: System on a Chip) in which a processor such as a CPU or GPU, a memory, a storage unit 12 and a communication unit 13 are integrated. good too.

The storage unit 12 uses a hard disk or flash memory. The storage unit 12 stores an image processing program 1P, image data 1D to be displayed on the display unit 14, and the like. This image data 1D includes digital video expressed in the RGB color space or the YCbCr color space, and is, for example, image data obtained from a network via the communication unit 13 or image data read from the reading unit 16.

The communication unit 13 is a communication module that realizes communication connection to a communication network such as the Internet. For the communication unit 13, for example, a network card, a wireless communication device, or a carrier communication module is used.

The display unit 14 uses a liquid crystal panel, an organic EL (Electro Luminescence) display, or the like. The display unit 14 can display an image by processing in the image processing unit 11 according to instructions from the control unit 10 . The display unit 14 displays image data in accordance with general RGB format data for each pixel forming the screen. Normally, when a digital image in the YCbCr color space is displayed on the display unit 14, it is displayed after being converted into the sRGB (standard RGB) color space. This is because an image displayed on the display unit 14 of a general information display device such as a personal computer can only be displayed in RGB format data.

Digital image data in the RGB color space is expressed as a collection of pixels (picture elements) partitioned by coordinates, and each pixel is given a pixel value (three values of color information RGB). A pixel value can be represented by (R, G, B) where R is the luminance value of the R channel, G is the luminance value of the G channel, and B is the luminance value of the B channel at a certain pixel position. Normally, each (R, G, B) value is represented by 8 bits (256 steps), etc., but the possible range for each (R, G, B) is 0.0 to 1.0 (upper limit 1.0 and lower limit 0.0 ) can also be normalized to the range

The operating unit 15 includes a user interface such as a keyboard or mouse. A physical button provided on the housing may be used, or a software button or the like displayed on the display unit 14 may be used. The operation unit 15 notifies the control unit 10 of operation information by the user.

The reading unit 16 uses, for example, a disk drive, and can read the image processing program 2P stored in the recording medium 2 (optical disk, magnetic disk, magneto-optical disk, etc.) using an optical disk or the like. Further, the reading unit 16 may be used when reading image data of a moving image or a still image of a color image to be displayed on the display unit 14 from the recording medium 2 or the like. The image processing program 1P stored in the storage unit 12 is acquired from the network via the communication unit 13, or the image processing program 2P read from the recording medium 2 by the reading unit 16 is read by the control unit 10 from the storage unit. 12 may be replicated.

Note that the brightness component in the description of the present embodiment is a component related to luminance, lightness, density, etc. of image data, and is typically a luminance component Y in YCbCr. In addition to Y, there are L in the Lab color space, V in the HSV color space, and L in the HLS color space. YUV or YPbPr may be used instead of YCbCr. Also, the color component is typically the color difference component CbCr in YCbCr and the hue and saturation in HSV. Furthermore, hue is a color that characterizes colors, such as red, yellow, green, and blue, and the color changes depending on the difference in the wavelength of light. Saturation (chroma) is one of the three attributes of color (hue, lightness, and saturation), and refers to the strength and degree of vividness of a color. Even if the hue and brightness are the same, the higher the saturation, the clearer the image will look.

Next, a method for correcting an RGB image after image processing in the image processing apparatus 1 according to this embodiment will be described. Specifically, the control unit 10 of the image processing apparatus 1 includes a conversion unit 10a, a correction unit 10b, and a neutral color calculation unit 10c.

The conversion unit 10a converts the pixel values of the RGB image output from the image processing unit 11 into signal values of image data represented in another color space separated into luminance or lightness and color components. In the present embodiment, another color space will be described as a YCbCr color space represented by a luminance component Y and a color difference component CbCr for the sake of convenience. Normally, image data composed of YCbCr signals can reproduce a wider range than the sRGB color space. Therefore, even if the color coordinate values after conversion are outside the color gamut of the RGB color space (that is, outside the range of 0.0 to 1.0), correction processing can be performed using the signal values of the YCbCr color space. Note that other color spaces are not limited to this, and may be YUV, YPbPr, HSV, HLS, HSB, Lab color space, or CIE Luv color space in addition to YCbCr color space.

The conversion unit 10a converts pixel values of RGB image data represented in the RGB color space generated after image processing into pixel values of image data represented in the YCbCr color space based on a predetermined calculation. Specifically, upon receiving a correction command for RGB image data after image processing from a user via the operation unit 15 or the like, the conversion unit 10a converts the designated image data from the RGB color space to the YCbCr color space. At this time, the conversion unit 10a performs, for example, the calculation shown in Equation 1 below.

Here, the pixel value of the RGB color space of the image data before conversion is (R, G, B) (where R is the luminance value of the R channel, G is the luminance value of the G channel, and B is the luminance value of the B channel) , the signal value in the YCbCr color space after conversion using Equation 1 is (YCbCr) (where Y is the luminance component and Cb and Cr are the chrominance components). The signal value Y in the YCbCr color space is in the range of 0.0≦Y≦1.0, and the formula difference signals Cb and Cr are in the range of −0.5≦Cb and Cr≦0.5.

If the signal value of the image data after conversion in the conversion unit 10a exists outside the color gamut corresponding to the RGB color space in another color space (here, the YCbCr color space), the correction unit 10b corrects the signal value is corrected to obtain the corrected signal value.

Here, various correction methods in the correction unit 10b will be described with reference to FIG. FIG. 2 is a cross-sectional view of a specific hue in the YCbCr space where the horizontal axis is saturation and the vertical axis is luminance. Although this figure is expressed in black and white for explanation, it actually has a predetermined hue (that is, the angle formed by Cb and Cr in the YCbCr space) and is surrounded by a triangle indicated by gradation. The triangular area is the color gamut corresponding to the RGB color space in the YCbCr space, and the edge of this triangular area is the boundary line. In the YCbCr space, the luminance on the vertical axis is the Y value, and the saturation on the horizontal axis can be obtained by the square root of the sum of the squares of the color difference components Cb and Cr.

As a first correction method, as indicated by an arrow L1 in FIG. 2(a), the correction unit 10b converts the signal value S1 is corrected for saturation while its luminance or lightness and hue are fixed, and the closest point (first boundary Move to point B1). This correction method can be said to be brightness-priority correction because it corrects saturation while maintaining luminance or brightness. An arrow L4 in FIG. 2(b) also indicates a similar correction method.

As a second correction method, as indicated by an arrow L2 in FIG. 2A, the correction unit 10b converts the signal value S1 of the image data converted into another color space (here, the YCbCr color space) by the conversion unit 10a. is corrected for luminance or lightness while keeping its saturation and hue fixed, and moved to the nearest point (second boundary point B2) of the color gamut corresponding to the RGB color space in another color space (YCbCr) Let This correction can be said to be saturation-priority correction because it corrects luminance or lightness while maintaining saturation. An arrow L6 in FIG. 2(c) also indicates a similar correction method.

As an extension of the second correction method, as indicated by arrow L3 in FIG. (In the case of YCbCr, it varies for each hue)), the saturation of S2 is clipped and rounded (rounded to the nearest value). As an extension of the first correction method, as indicated by the arrow L5 in FIG. It may also clip and round the luminance of S3 (round the luminance or brightness to the nearest value) when it exceeds the case 0 to 1). In such a case, the correction unit 10b corrects the luminance or lightness and saturation of the signal value of the converted image data while fixing the hue, and converts the signal value to the RGB color space in another color space (YCbCr). It is moved to the boundary with the corresponding color gamut. Although not described in detail, the conversion unit 10a also has a function of converting the signal values of the boundary points B1 and B2 in the YCbCr color space into pixel values in the RGB color space based on a predetermined arithmetic expression. The display unit 14 displays the corrected image data according to the RGB format data.

The intermediate color calculator 10c obtains an intermediate color by multiplying the color components obtained at the first boundary point B1 and the second boundary point B2 with weights set in advance and taking an average value. As a result, it is possible to determine whether brightness is given priority or saturation is given priority according to the user's preference. Note that the first boundary point B1 and the second boundary point B2 are within the color gamut of the RGB color space, so the intermediate colors always fit within the RGB color space.

Next, an example of the operation procedure in the correction unit 10b of the image processing apparatus 1 according to the present embodiment will be described with reference to the flowchart shown in FIG. First, the user selects image data to be subjected to image processing via the operation unit 15 or the like, and image processing is performed via the image processing unit 11 (S31). A neural network may be used for image processing here.

Next, the RGB image after the image processing is selected, and the saturation priority correction or the brightness priority correction is selected via the operation unit 15 (S32). If the saturation priority is selected (Yes in S32), the correction unit 10b fixes the signal value of the image data after conversion to the YCbCr color space by the first correction method described above, and fixes the luminance and hue. In this state, the saturation is corrected and moved to the first boundary point of the color gamut corresponding to the RGB color space in the YCbCr color space (S33). On the other hand, when brightness priority is selected (No in S32), the correction unit 10b uses the above-described second correction method, that is, the signal value of the converted image data is fixed in saturation and hue. In this state, the luminance is corrected and moved to the second boundary point of the color gamut corresponding to the RGB color space in the YCbCr color space (S34).

As described above, in the image processing apparatus 1 according to the present embodiment, the pixel values of the RGB image output from the image processing unit 11 are outside the color gamut corresponding to the RGB color space in another color space. contains pixel values that are The correction unit 10b executes the above-described first and second correction methods according to the user's preference such as brightness priority or saturation priority. With this configuration, even if the RGB image generated after the processing of the digital image is outside the area of the RGB color space, the image data outside the RGB color space can be adjusted to suit the user's taste without causing a sense of incongruity. can be accommodated inside. As a result, it is possible to prevent the problem that the brightness and hue of the image data are changed and the image quality is deteriorated such as bleeding. In addition, the user can select a preferred image correction method such as brightness priority or saturation priority according to the user's needs.

(Modification)
A modification of the image processing apparatus 1 according to the embodiment of the present invention will be described with reference to FIG. In this modification, as shown in FIG. 4, the image processing unit 11 uses a machine learning model 111 to perform predetermined inference processing on image data input from the input unit 110, and outputs an RGB image. Output from the unit 112 . The machine learning model 111, for example, when using a trained model, performs image data optimization processing (for example, image resolution enhancement, noise removal, etc.) based on already learned parameters.

Next, the image processing function of the image processing apparatus 1 according to this modification will be described. The control unit 10 of the image processing device 1 includes an inference processing execution unit 101 . This inference processing execution unit 101 performs image processing based on the image processing program 1P stored in the storage unit 12 . That is, the inference processing execution unit 101 uses a machine learning model to execute predetermined inference processing on input target image data. The inference processing execution unit 101 also functions as an image processing execution unit that inputs image data, which is input data, to the input unit 110 based on the user's operation using the operation unit 15 .

The input unit 110 of the image processing unit 11 receives input of target image data to be inferred. The inference processing execution unit 101 outputs inference target image data to the machine learning model 111 . The type of each image data can be determined, for example, by reading information or an identifier in the header portion of the image data. Image data processed by the machine learning model 111 is input to the output unit 112 . In this modified example, the image data output from the output unit 112 is an RGB image, and the pixel values of the RGB image are outside the range of the RGB color space. In order to allow pixel values after image processing to fall outside the range of the RGB color space, a number of bits exceeding the original number of bits may be assigned to each color component. Finally, the output image from the output unit 112 is output to the conversion unit 10a.

The present invention is not limited to the configuration of the above-described embodiment, and various modifications are possible without changing the gist of the invention. Further, in order to achieve the object of the present invention, the present invention is implemented as an image processing method having steps of characteristic constituent means included in the image processing apparatus 1, or as a program including those characteristic steps. can also The program can be distributed not only in a ROM or the like, but also via a recording medium such as a USB memory or a communication network. Also, in the description of the above embodiment, an example in which each function of the image processing apparatus 1 is realized by a program was described, but it may be realized by hardware.

Furthermore, the present invention can also be implemented as a computer system that transmits input data to an image processing device or computer program and receives and uses output data from the image processing device or computer program. A device used in this system is an image processing device having a display unit and a communication unit or an information processing device capable of transmitting and receiving information to and from a computer.

1 image processing device 10 control section 10a conversion section (conversion means)
10b correction unit (correction means)
10c Intermediate color calculation unit (intermediate color calculation means)
REFERENCE SIGNS LIST 11 image processing unit 12 storage unit 13 communication unit 14 display unit 15 operation unit 16 reading unit 101 inference processing execution unit 111 machine learning model
B1 first boundary point
B2 second boundary point

In order to achieve the above object, an image processing apparatus according to the present invention comprises image processing means for executing predetermined processing on input image data and outputting an RGB image; conversion means for converting pixel values of an image into signal values of image data represented in another color space separated into luminance or lightness and color components; and a correcting means for correcting the signal value when it exists outside the color gamut corresponding to the RGB color space in the other color space, wherein the pixel value of the RGB image output from the image processing means is , including pixel values that are outside the color gamut corresponding to the RGB color space in the other color space, and the correcting means corrects the signal values of the converted image data so that the saturation corresponds to the RGB color space If the color gamut exceeds the range that can be taken, after rounding the saturation to the nearest value, correct the brightness or lightness while fixing the saturation and hue, and in the other color space It is characterized by moving to the second boundary point of the color gamut corresponding to the RGB color space .

In order to achieve the above object, an image processing apparatus according to the present invention comprises image processing means for executing predetermined processing on input image data and outputting an RGB image; conversion means for converting pixel values of an image into signal values of image data represented in another color space separated into luminance or lightness and color components; and a correcting means for correcting the signal value when it exists outside the color gamut corresponding to the RGB color space in the other color space, wherein the pixel value of the RGB image output from the image processing means is , including pixel values outside the color gamut corresponding to the RGB color space in the other color space, and the correcting means corrects the signal values of the converted image data so that the luminance or lightness corresponds to the RGB color space If the color gamut exceeds the range that can be taken, after rounding the brightness or brightness to the nearest value, the saturation is corrected while the brightness or brightness and hue are fixed, and the other colors space to the first boundary point of the color gamut corresponding to the RGB color space, and the correcting means adjusts the signal value of the converted image data to the range that the color gamut corresponding to the RGB color space can take in saturation. If it exceeds from , round the saturation to the nearest value, then correct the luminance or lightness while fixing the saturation and hue, and correspond to the RGB color space in the other color space The image processing device further weights the color components obtained at the first boundary point and the second boundary point by a preset coefficient to obtain an average value. It is characterized by comprising intermediate color calculation means for obtaining the intermediate color by taking the

In order to achieve the above object, an image processing method according to the present invention provides an image processing step of executing a predetermined process on input image data to output an RGB image, and an image processing step of outputting an RGB image in the image processing step. a conversion step of converting the pixel values of the RGB image into signal values of image data represented in another color space separated into luminance or brightness and color components; and a signal of the image data after conversion in the conversion step. and a correction step of correcting the signal value when the value is outside the color gamut corresponding to the RGB color space in the other color space, the pixel of the RGB image output in the image processing step. The value includes a pixel value outside the color gamut corresponding to the RGB color space in the other color space, and in the correction step, the signal value of the converted image data is converted to the RGB color space with saturation. If the corresponding color gamut exceeds the range that can be taken, after rounding the saturation to the nearest value, while fixing the saturation and hue, correct the brightness or lightness, and the other colors It is characterized by moving to the second boundary point of the color gamut corresponding to the RGB color space in space .

In order to achieve the above object, an image processing apparatus according to the present invention comprises image processing means for executing predetermined processing on input image data and outputting an RGB image; conversion means for converting pixel values of an image into signal values of image data represented in another color space separated into luminance or lightness and color components; and correction means for correcting the signal value when it exists outside the color gamut corresponding to the RGB color space in the other color space , wherein the correction means comprises the conversion The signal values of the subsequent image data are moved to the first boundary point of the color gamut corresponding to the RGB color space in the other color space while fixing the luminance or lightness and hue, correcting the saturation. , and a second boundary point of the color gamut corresponding to the RGB color space in the other color space by correcting the luminance or lightness of the signal value of the converted image data while fixing the saturation and hue. , and the image processing device further weights the color components obtained at the first boundary point and the second boundary point by a preset coefficient and takes an average value to obtain an intermediate color. It is characterized by comprising a calculation means.

Claims (10)

image processing means for executing predetermined processing on input image data and outputting an RGB image;
conversion means for converting the pixel values of the RGB image output from the image processing means into signal values of image data represented in another color space separated into luminance or lightness and color components;
correction means for correcting the signal value when the signal value of the image data after conversion in the conversion means exists outside the color gamut corresponding to the RGB color space in the other color space;
The image processing apparatus, wherein the pixel values of the RGB image output from the image processing means include pixel values outside the color gamut corresponding to the RGB color space in the other color space. The correcting means rounds the signal value of the converted image data to the nearest value if the luminance or the lightness exceeds the range that the color gamut corresponding to the RGB color space can take. After that, while fixing the luminance or lightness and hue, the saturation is corrected and moved to the first boundary point of the color gamut corresponding to the RGB color space in the other color space. The image processing apparatus according to claim 1. The correction means rounds the signal value of the converted image data to the nearest value if the saturation exceeds the range that the color gamut corresponding to the RGB color space can take. , fixing the saturation and hue, correcting the luminance or lightness, and moving to the second boundary point of the color gamut corresponding to the RGB color space in the other color space. The described image processing device. The correcting means corrects the luminance or lightness and saturation of the signal value of the image data after the conversion while fixing the hue, and corrects the color gamut corresponding to the RGB color space in the other color space. 2. The image processing apparatus according to claim 1, wherein the image is moved to a boundary. 3. A neutral color calculation means for obtaining a neutral color by multiplying the color components obtained at the first boundary point and the second boundary point by a preset coefficient and taking an average value. 1. The image processing apparatus according to 1. 6. The image processing means according to any one of claims 1 to 5, wherein the image processing means executes a predetermined inference process on the input image data using a machine learning model and outputs an RGB image. 10. The image processing device according to claim 1. an image processing step of performing predetermined processing on input image data and outputting an RGB image;
a conversion step of converting the pixel values of the RGB image output in the image processing step into signal values of image data represented in another color space separated into luminance or lightness and color components;
a correction step of correcting the signal values of the image data after conversion in the conversion step when the signal values are outside the color gamut corresponding to the RGB color space in the other color space;
The image processing method, wherein the pixel values of the RGB image output in the image processing step include pixel values outside the color gamut corresponding to the RGB color space in the other color space. A program that functions in an image processing device,
an image processing step of performing predetermined processing on input image data and outputting an RGB image;
a conversion step of converting the pixel values of the RGB image output in the image processing step into signal values of image data represented in another color space separated into luminance or lightness and color components;
a correction step of correcting the signal values of the image data after conversion in the conversion step when the signal values are outside the color gamut corresponding to the RGB color space in the other color space;
The program, wherein the pixel values of the RGB image output in the image processing step include pixel values outside the color gamut corresponding to the RGB color space in the other color space. Sending input data to the image processing device according to any one of claims 1 to 6 or the program according to claim 8, and receiving and using output data from these image processing devices or programs, A computer system characterized by: image processing means for executing predetermined processing on input image data and outputting an RGB image;
conversion means for converting the pixel values of the RGB image output from the image processing means into signal values of image data represented in another color space separated into luminance or lightness and color components;
correction means for correcting the signal value when the signal value of the image data after conversion in the conversion means exists outside the color gamut corresponding to the RGB color space in the other color space;
The correcting means corrects the saturation of the signal value of the converted image data while fixing the luminance or lightness and hue of the converted image data, thereby correcting the color gamut corresponding to the RGB color space in the other color space. moving to the first boundary point, and correcting the luminance or lightness of the signal values of the converted image data while fixing the saturation and hue to correspond to the RGB color space in the other color space; Move to the second boundary point of the gamut,
The image processing device further comprises:
An image processing characterized by comprising intermediate color calculation means for obtaining an intermediate color by multiplying the color components obtained at the first boundary point and the second boundary point by a preset coefficient and obtaining an average value. Device.

Images