JP6928903B2 - Image processing equipment, image processing methods, and programs - Google Patents

Description

The present disclosure relates to an image processing technique for performing image processing of a color image.

Image processing that converts a photo into an illustration-like image is known. In such an illustration image processing, a process of making the edge black is performed. Alternatively, a process of emphasizing the edge is performed by a filtering process (see Patent Documents 1 and 2).

Japanese Unexamined Patent Publication No. 2001-24872 Japanese Patent No. 369574

However, if the edges are simply made black, there is a risk that the image will be clearly generated by CG (Computer Graphics). Further, in the process of emphasizing the edge by the filtering process, the edge in the subject image is emphasized, and the subject image becomes unnatural.

The present disclosure has been made in consideration of such points, and provides an image processing device, an image processing method, and a program capable of generating an illustration-like image having a more natural feeling as drawn by a human being. It is a thing.

In one aspect of the disclosure,
A conversion unit that converts a first color image into lightness information and color information other than the lightness information, and
Compressed lightness information is thinned out. A lightness compression unit that generates lightness information, and a lightness compression unit.
A reconversion unit that combines the compressed brightness information and the color information to generate a second color image, an edge extraction unit that extracts edges included in the first color image, and an edge extraction unit.
A pixel value changing unit that generates a third color image by changing the pixel value at the edge position in the second color image while maintaining similar colors.
An image processing apparatus comprising the above is provided.

When the brightness of the inner region of the edge in the second color image is higher than that of the outer region of the edge, the pixel value changing unit determines the pixel value at the position of the edge in the second color image. The pixel value may be changed to a darker shade while maintaining the similar color.
The pixel value changing unit may change the pixel value at the edge position in the second color image to a value obtained by multiplying the pixel value by a predetermined coefficient.

The edge extraction unit
A monochrome conversion unit that converts the first color image into a monochrome image,
A blur processing unit that generates a blurred image that blurs the monochrome image,
A ratio detection unit that detects the ratio between the pixel value of the blurred image and the pixel value of the monochrome image corresponding to the pixel value of the blurred image.
A ratio determination unit that determines whether or not the ratio is greater than a predetermined threshold value,
An edge output unit that outputs pixels whose ratio is determined to be larger than the threshold value as edges in the monochrome image.
May be further provided.

The blur processing unit may generate the blurred image by filtering the monochrome image with a Gaussian filter having a filter size corresponding to the outer size of the monochrome image.

The blur processing unit may generate the blurred image by filtering the monochrome image with a Gaussian filter having a filter size corresponding to the outer size of the third color image.

In one aspect of the disclosure,
A blur processing unit that generates a blurred image that blurs a monochrome image,
A ratio detection unit that detects the ratio between the pixel value of the blurred image and the pixel value of the monochrome image corresponding to the pixel value of the blurred image.
A ratio determination unit that determines whether or not the ratio is greater than a predetermined threshold value,
An edge output unit that outputs pixels whose ratio is determined to be larger than the threshold value as edges in the monochrome image.
An image processing apparatus comprising the above is provided.

In one aspect of the disclosure,
The first color image is converted into lightness information and color information other than the lightness information.
The compressed lightness information obtained by thinning out the lightness information is generated.
The compressed brightness information and the color information are combined to generate a second color image.
Extract the edges contained in the first color image and
An image processing method is provided for generating a third color image by changing the pixel value at the edge position in the second color image while maintaining similar colors.

In one aspect of the disclosure,
Generates a blurred image that blurs a monochrome image,
The ratio of the pixel value of the blurred image to the pixel value of the monochrome image corresponding to the pixel value of the blurred image is detected.
It is determined whether or not the ratio is larger than a predetermined threshold value, and it is determined.
An image processing method is provided in which pixels determined to have the ratio greater than the threshold value are output as edges in the monochrome image.

In one aspect of the disclosure,
On the computer
A step of converting a first color image into lightness information and color information other than the lightness information, and
The step of generating compressed lightness information by thinning out the lightness information, and
A step of generating a second color image by synthesizing the compressed brightness information and the color information, and a step of extracting edges included in the first color image.
A step of generating a third color image by changing the pixel value at the edge position in the second color image while maintaining similar colors.
A computer-readable program is provided to execute.

In one aspect of the disclosure,
On the computer
Blur step to generate a blur image that blurs a monochrome image,
A step of detecting the ratio of the pixel value of the blurred image to the pixel value of the monochrome image corresponding to the pixel value of the blurred image, and
A step of determining whether or not the ratio is greater than a predetermined threshold, and
A step of extracting pixels whose ratio is determined to be larger than the threshold value as edges in the monochrome image, and
A computer-readable program is provided to execute.

According to the present disclosure, it is possible to generate an illustration-like image having a more natural feeling as if it was drawn by a human.

The figure which shows the structure of the image processing system in 1st Embodiment. The block diagram which shows the structure of the edge extraction part. The figure which shows the coefficient of the filter size [5 × 5]. The figure which shows the flowchart of the illustration processing in an image processing apparatus. The figure which shows the image about the edge extraction process. The figure which shows the image about the thinning-out process. The figure which shows the flowchart of the illustration process in 2nd Embodiment.

Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.

(First Embodiment)
The image processing apparatus according to the first embodiment changes the pixel value at the edge position of the color image to a darker hue pixel value while maintaining similar colors, based on the extracted edge position. This is an attempt to obtain a more natural-looking illustration-like image. More details will be given below.

FIG. 1 is a diagram showing a configuration of an image processing system 1 according to the first embodiment. The image processing system 1 of FIG. 1 can be built in, for example, various electronic devices (for example, a personal computer) having a built-in CPU (Central Processing Unit). More specifically, the processing operation of the image processing system 1 of FIG. 1 can be performed by the CPU executing a dedicated program. Alternatively, a dedicated hardware device that performs the processing operation of the image processing system 1 of FIG. 1 may be provided.

The image processing system 1 of FIG. 1 includes a storage device 10, a control device 12, a camera 14, an input device 16, a display device 18, a printer 20, and an image processing device 100. In addition to the system program, the storage device 10 stores various processing programs such as a face image photographing processing program and an image processing program, and data processed by each program. The control device 12 has a CPU, a RAM (Random Access Memory), and the like, reads various programs such as a system program and an image processing program stored in the storage device 10 and develops them in the RAM, and operates each part according to the programs. Is centrally controlled.

The camera 14 has an optical lens, an optical sensor such as a CCD (Charge Coupled Device), an A / D (Analog / Digital) conversion unit, and the like, and a subject image input via the optical lens at the time of shooting is photoelectric by the optical sensor. Convert to generate an analog image signal. Then, the A / D conversion unit converts the analog image signal into a digital first color image and supplies it to the control device 12. One pixel of the first color image, which is a natural image, is represented by, for example, gradation signals of three primary colors of R (red), G (green), and B (blue). More specifically, each of the RGB elements of one pixel is represented by an 8-bit gradation signal. Therefore, one pixel of the first color image is composed of 24 bits (8 bits x 3 colors).

The input device 16 includes a touch panel integrated into the panel of the display device 18, a photo type selection switch (hereinafter referred to as SW) for selecting a photo type, a shooting SW for instructing the start of shooting, and a shooting image. It has various function keys such as a print switch for instructing print output, and outputs an operation signal corresponding to the operated key to the control device 12.

The display device 18 has, for example, an LCD (Liquid Crystal Display), and various operation screens such as a guide screen for explaining a shooting procedure and a selection screen for selecting a photo type according to a control signal from the control device 12, and a camera. The image of the subject taken by 14 and the illustration-like image processed by the image processing device 100 are displayed.

The printer 20 prints out based on the image input from the control device 12. As a printing method of the printer 20, for example, a sublimation type transfer method in which a coloring material is sublimated and transferred to form an image can be applied. Alternatively, a melt transfer recording method may be applied in which the recording layer of the recording medium is transferred to the image receiving layer of the image receiving material by superimposing the recording medium and the image receiving material and applying heat to form an image. An electrophotographic method, an inkjet method, or another printing method may be applied. Moreover, you may form a silver salt photograph image.

The image processing device 100 performs image processing to convert the input first color image into an illustration-like image. The image processing device 100 includes a conversion unit 102, a brightness compression unit 104, a reconversion unit 106, an edge extraction unit 108, and a pixel value change unit 110.

The conversion unit 102 converts the first color image input from the control device 12 into lightness information and color information other than the lightness information. This brightness information is a value indicating a measure of the brightness of a color image.
For example, the conversion unit 102 converts the first color image of RGB into the image data of YCbCr.
Luminance (brightness) Y can be calculated by a conversion formula such as Y = 0.29900 × R + 0.58700 × G + 0.11400 × B. As a result, brightness information for each pixel is generated using the first color image.

Further, the conversion unit 102 converts the first color image of RGB into Cb which is color information other than the brightness information by a conversion formula of, for example, Cb = −0.16874 × R−0.33126 × G + 0.50000 × B. do. Similarly, the conversion unit 102 converts the first color image of RGB into Cr, which is color information other than the brightness information by a conversion formula of, for example, Cr = 0.50000 × R-0.41869 × G-0.0081 × B. Convert. Here, Cb is also called color difference information, and indicates the hue and saturation of the blue system. Cr is also called color difference information and indicates the hue and saturation of the red system. As a result, the color information Cb and Cr for each pixel are generated using the first color image.

The conversion unit 102 may generate brightness information and color information by a method other than converting the first color image of RGB into YCbCr. For example, the first color image may be converted into a Lab color system, an HLS color system, or the like.

The lightness compression unit 104 generates compressed lightness information obtained by thinning out the lightness information obtained by the conversion unit 102. Originally, even if the first color image has linear brightness information, the first color image is converted into, for example, five levels of brightness.

More specifically, for example, when the brightness information for each pixel obtained by the conversion unit 102 is in the range of (0 to 100), the values of the pixels in the range of (100 to 95) are thinned out. Convert to compressed lightness information (100). Similarly, the value of the pixel whose lightness information is in the range of (95 to 55) is converted into the compressed lightness information (80) after thinning out. Similarly, the values of the pixels in the range of lightness information (55 to 30) are converted into the compressed lightness information (60) after thinning, and the values of the pixels in the range of lightness information (30 to 10) are thinned out. It is converted into the compressed lightness information (25) of the above, and the value of the pixel in the range of the lightness information (10 to 0) is converted into the compressed lightness information (0) after thinning. By performing such thinning processing, conversion to a brightness-compressed image in which the difference in brightness is clarified as a whole is performed. That is, the compressed lightness information Y'here is equivalent to the pixel value for each pixel in the lightness compressed image obtained by thinning out the brightness of the luminance image.

The reconversion unit 106 generates a second color image by synthesizing the compressed lightness information obtained by the lightness compression unit 104 and the color information other than the lightness information obtained by the conversion unit 102. More specifically, the reconversion unit 106 generates a second color image using the compressed lightness information Y'obtained by the lightness compression unit 104 and the color information Cb and Cr obtained by the conversion unit 102. do. The second color image from which the brightness information is thinned out in this way is converted into an illustration-like brightness image in which the difference in brightness is clear as a whole.

The edge extraction unit 108 extracts the edges included in the first color image. The detailed configuration of the edge extraction unit 108 will be described later.

The pixel value changing unit 110 changes the pixel value at the edge position in the second color image while maintaining similar colors to generate a third color image (illustration-like image). More specifically, when the brightness of the inner region of the edge in the second color image is higher than the outer region of the edge, the pixel value changing unit 110 determines the pixel value at the position of the edge in the second color image. , Change to a darker pixel value while maintaining similar colors.

For example, the pixel value changing unit 110 changes the pixel value at the edge position in the second color image to a value obtained by multiplying the pixel value by a predetermined coefficient. More specifically, the pixel value changing unit 110 converts the pixel value at the edge position obtained by the edge extracting unit 108 into 1/2 or 1/4 to darken the edge color. In this case, the RGB gradation value for each pixel is changed to a value obtained by multiplying the RGB gradation value for each pixel by a predetermined coefficient (for example, 1/2 or 1/4).

This makes the original shade darker while maintaining the original shade of the edge. For example, if the original color is pink, an edge with a darker pink is generated. Also, for example, if the original color is green, an edge with darker green is generated. In this way, by generating edges while retaining the original color, it is possible to obtain an image that looks as if it was painted with paint. That is, an illustrated image that is difficult to recognize generated by CG can be obtained. In the pixel value changing unit 110, when the edge color is darkened, the value of the coefficient to be multiplied by the RGB gradation value for each pixel may be different for each RGB element.

Next, a detailed configuration of the edge extraction unit 108 will be described with reference to FIG. FIG. 2 is a block diagram showing the configuration of the edge extraction unit 108. As shown in FIG. 2, the edge extraction unit 108 includes a monochrome conversion unit 112, a blur processing unit 114, a ratio detection unit 116, a ratio determination unit 118, and an edge output unit 120.

The monochrome conversion unit 112 converts the first color image into a monochrome image. The monochrome conversion unit 112 converts the first color image of RGB into the information of the brightness (brightness) Y according to the above-mentioned conversion formula. A monochrome image is equivalent to luminance information.

The blur processing unit 114 generates a blurred image in which a monochrome image is blurred. More specifically, the blur processing unit 114 generates a blurred image by filtering the monochrome image with, for example, a Gaussian filter having a filter size corresponding to the outer size of the monochrome image.

The ratio detection unit 116 detects the ratio between the pixel value of the blurred image obtained by the blur processing unit 114 and the pixel value of the monochrome image corresponding to the pixel value of the blurred image. More specifically, the ratio is calculated by the formula of ratio = pixel of blurred image / pixel value of monochrome image.

The edge output unit 120 outputs pixels in which the ratio detected by the ratio detection unit 116 is determined to be larger than a predetermined threshold value as an edge in a monochrome image. For example, when a region having a ratio of 0.9 or more is output as an edge, the edge can be detected when the brightness of the inner region of the edge is higher than that of the outer region of the edge. As a result, for example, in a human image, an edge can be set on the outer portion along the contour of the human face, and there is no possibility that an edge that bites into the face is extracted. Therefore, there is no problem that the impression of the human face is changed by unnatural edge extraction.

Here, as the filter size of the Gaussian filter used in the blur processing unit 114 increases, the edge output unit 120 can output even smaller (weaker) edges. On the other hand, the smaller the filter size of the Gaussian filter, the smaller the edge cannot be extracted by the edge output unit 120. A small (weak) edge is a contour line whose brightness changes relatively slowly, and a large (strong) edge is a contour line whose brightness changes sharply.

Further, if an image having a small area is extracted to a small edge, the image becomes complicated, so it is desirable to extract only a large edge. On the other hand, in the case of an image having a large area, the reproducibility of the image is improved by extracting even a small edge. Therefore, in the present embodiment, the filter size is changed according to the short side size of the image data. That is, as the length of the short side of the image data becomes longer, the filter size of the Gaussian filter in the blur processing unit 114 is increased.

In this way, by changing the filter size of the Gaussian filter used in the blur processing unit 114, the size of the extracted edge is changed. Further, by changing the filter size of the Gaussian filter, the width of the edge output by the edge output unit 120 is changed.
Specifically, as the filter size increases, the width of the edge output by the edge output unit 120 increases.

As described above, in a monochrome image, a portion where the brightness changes significantly is a strong edge, and a portion where the brightness changes gently is a weak edge. Therefore, when comparing the strong edge and the weak edge, the difference value between the corresponding pixel values between the blurred image and the monochrome image obtained by the blur processing unit 114 is larger in the strong edge. The larger the difference value, the larger the width of the edge output from the edge output unit 120. Thus, the width of the edges is thicker for strong edges than for weak edges. As a result, the strong edge whose brightness changes abruptly becomes a thick line, and the weak edge whose brightness changes gently becomes a thin line, and the edge width continuously changes according to the brightness difference. As described above, in the present embodiment, the thickness of the contour line can be changed according to the strength of the edge, so that an image as if drawn with paint can be obtained.

Next, the coefficient of the Gaussian filter in the blur processing unit 114 will be described with reference to FIG. FIG. 3 is a diagram showing the coefficients of the filter size [5 × 5]. As shown in FIG. 3, the coefficient f (x, y) is calculated according to the Gaussian distribution function represented by the following equation (1).

Here, (x, y) indicates the distance from the pixel of interest. However, in reality, the coefficients are normalized so that the sum of all the coefficients is 1. More specifically, the coefficients are normalized according to Eq. (2).

The filter size [N × N] is calculated using the values on the short sides of the input image. More specifically, it is calculated by the formula: length of one side of the filter = length of the short side / specified value. Here, the length of the short side is the shorter of the lengths of the vertical side and the horizontal side in the input image. Therefore, N = the length of one side of the filter / the size of the pixel is calculated.

Next, the flow of the illustration process will be described with reference to FIGS. 4 to 6. FIG. 4 is a diagram showing a flowchart of the illustration process in the image processing device 100. Although FIG. 4 describes an example in which the edge extraction process and the lightness thinning process are performed in parallel, one of the processes may be performed before the other process.

FIG. 5 is a diagram showing an image related to the edge extraction process. FIG. 5A is a first color image of a rose flower taken by the camera 14, FIG. 5B is a monochrome image converted by the monochrome conversion unit 112, and FIG. 5C is a monochrome image. It is a blur image obtained by the blur processing unit 114, and FIG. 5D is an image in which the edge obtained by the edge output unit 120 is clearly shown in black.

FIG. 6 is a diagram showing an image related to the thinning process. FIG. 6A is a first color image of the rose flower taken by the camera 14, and FIG. 6B is an image (luminance image) showing the brightness information converted by the conversion unit 102. FIG. 6 (c) is an image (compressed brightness image) showing the compressed brightness information obtained by the brightness compression unit 104, and FIG. 6 (d) is a second color image obtained by the reconversion unit 106. , FIG. 6E is a third color image obtained by the pixel value changing unit 110.

First, the flow of the edge extraction process will be described. As shown in FIG. 4, the first color images (FIGS. 5A and 6A) of the rose flower taken by the camera 14 are transmitted to the monochrome conversion unit 112 and the conversion unit 102 via the control device 12. Is input to (step S500).

Next, the monochrome conversion unit 112 converts the first color image into a monochrome image (FIG. 5 (b)) according to the conversion formula described above (step S502). As a result, the first color image is converted into a monochrome image having a brightness Y of 0 to 100.

Next, the blur processing unit 114 calculates the filter size based on the length of the short side of the monochrome image, and calculates the coefficient of the Gaussian distribution function. Subsequently, all the pixels of the monochrome image are filtered to generate a blurred image (FIG. 5 (c)) (step S504).

Next, the ratio detection unit 116 detects the ratio between the pixel value of the blurred image and the pixel value of the monochrome image corresponding to the pixel value of the blurred image (step S506). Then, the edge output unit 120 outputs a pixel whose ratio detected by the ratio detection unit 116 is larger than a predetermined threshold value as an edge in the monochrome image (step S508). Here, a pixel having a ratio of 0.7 or more and less than 1.0 is defined as an edge. In this case, the pixel value of the pixel corresponding to the edge is set to 0 (corresponding to black in FIG. 5 (d)), and the pixel value of the other pixels is set to 100 (corresponding to white in FIG. 5 (d)).

Next, the flow of the thinning process will be described. As shown in FIG. 4, the conversion unit 102 converts the first color image (FIG. 6 (a)) input from the control device 12 into the brightness information (FIG. 6 (b)) Y and the color information other than the brightness information. Convert (step S602).

Next, the lightness compression unit 104 generates compressed lightness information Y'by thinning out the lightness information Y (step S604). Here, the lightness information Y (FIG. 6B) having a value range of 0 to 100 is thinned out to generate compressed lightness information Y'having any value of 100, 80, 60, 25, and 0. (Fig. 6 (c)).

Next, the reconversion unit 106 synthesizes the compressed lightness information Y'obtained by the lightness compression unit 104 and the color information Cb and Cr obtained by the conversion unit 102 to form a second color image (FIG. 6 (FIG. 6). d))) is generated (step S606). Here, each pixel is reconverted into a second color image having an RGB gradation value.

Next, the pixel value changing unit 110 changes the pixel value at the edge position in the second color image while maintaining the similar colors to generate the third color image (FIG. 6 (e)) (step S608). ). Here, the value of each of the RGB elements corresponding to the position of the edge output by the edge output unit 120 is halved. That is, the RGB gradation values of the second color image corresponding to the edge positions (pixels indicating the pixel value 100) output by the edge output unit 120 are each halved. Subsequently, the pixel value changing unit 110 outputs the third color image to the display device 18 and the printer 20 via the control device 12, and ends the entire process.

In this way, in the edge extraction process, the width of a strong edge is output thicker by detecting the ratio between the pixel value of the blurred image and the pixel value of the monochrome image corresponding to the pixel value of the blurred image. Is possible. Further, in the lightness thinning process, after the first color image is converted into the lightness information, the compressed lightness information obtained by thinning out the lightness information is generated, and the compressed lightness information is reconverted into the second color-image to obtain the whole image. It is possible to obtain a second color image in which the difference in brightness is clear. Further, by darkening the edge color of the second color image, a third color image (illustrated image) as if drawn with paint can be obtained. In this case, it is possible to obtain an illustrated image with a more natural feeling by making the edge thicker.

As described above, according to the first embodiment, since the pixel values at the positions of the extracted edges are changed while maintaining the similar colors, an illustration-like image with a more natural feeling as if drawn with paint. Can be generated. Further, since the width of the edge is variably adjusted depending on whether it is a strong edge or a weak edge, the width of the edge can be optimized according to the surrounding image, and an image without a sense of discomfort can be obtained.
(Second Embodiment)

In the first embodiment described above, the blur processing unit 114 sets the filter size according to the outer size of the monochrome image, but other conditions for setting the filter size can be considered. For example, it differs by setting the filter size according to the outer size of the third color image which is the output result of the printer 20. Hereinafter, parts different from the above-described first embodiment will be described. Since the configuration of the image processing system 1 is the same as the configuration shown in FIGS. 1 to 3, the description thereof will be omitted.

FIG. 7 is a diagram showing a flowchart of the illustration process according to the second embodiment. The image processing apparatus according to the second embodiment has the same configuration as that of FIG. 1, and only the illustration processing is different from the first embodiment. In FIG. 7, the same step numbers are assigned to the processes equivalent to those in FIG. 4, and the differences will be mainly described below. As shown in FIG. 7, when starting the illustration process, the control device 12 acquires the output size of the image set in the printer 20 and outputs it to the blur processing unit 114 (step S701).

The blur processing unit 114 blurs a monochrome image and generates a blurred image based on a Gaussian filter having a filter size set according to the length of the short side of the output size of the image set in the printer 20. (Step S704). In this way, the blur processing unit 114 sets the filter size of the Gaussian filter according to the outer size of the above-mentioned third color image.

Next, the pixel value changing unit 110 further changes the size of the third color image generated by the pixel value changing unit 110 to the outer size of the third color image, that is, the output size of the printer 20, and the display device 18 and The printer 20 is output to the printer 20 via the control device 12 (step S710), and the entire process is completed.

As described above, in the second embodiment, since the filter size is set according to the outer size of the third color image, that is, the output size of the printer 20, the edge of the size and width suitable for the outer size of the third color image. Can be generated. As a result, even if the outer size of the third color image changes variously, it is possible to generate an illustration-like image as if it was actually drawn by a human.

As described above, according to the second embodiment, the third color is set in order to set the filter size according to the outer size of the third color image generated by the pixel value changing unit 110, that is, the output size of the printer 20. Edge extraction that is optimal for the outer size of the image can be performed.

In the first and second embodiments described above, a Gaussian filter is used for the blurring process, but the filter used for the blurring process is not necessarily limited to the Gaussian filter, and for example, a Laplacian filter may be used. However, in the first and second embodiments, the filter is used for the blurring process instead of directly extracting the edge by the filter. Therefore, as long as the filter can be blurred and the filter size can be changed, the specific filter format does not matter.

At least a part of the image processing apparatus described in the above-described embodiment may be configured by hardware or software. When configured by software, a program that realizes at least a part of the functions of the image processing device may be stored in a recording medium such as a flexible disk or a CD-ROM, read by a computer, and executed. The recording medium is not limited to a removable one such as a magnetic disk or an optical disk, and may be a fixed recording medium such as a hard disk device or a memory.

Further, a program that realizes at least a part of the functions of the image processing device may be distributed via a communication line (including wireless communication) such as the Internet. Further, the program may be encrypted, modulated, compressed, and distributed via a wired line or wireless line such as the Internet, or stored in a recording medium.

The aspects of the present disclosure are not limited to the individual embodiments described above, but also include various modifications that can be conceived by those skilled in the art, and the effects of the present disclosure are not limited to the contents described above. That is, various additions, changes and partial deletions are possible without departing from the conceptual idea and purpose of the present disclosure derived from the contents defined in the claims and their equivalents.

100: Image processing device, 102: Conversion unit, 104: Brightness compression unit, 106: Re-conversion unit, 108: Edge extraction unit, 110: Pixel value change unit, 112: Monochrome conversion unit, 114: Blur processing unit, 116: Ratio detection unit, 118: Ratio determination unit, 120: Edge output unit

Claims (3)

A blur processing unit that generates a blur image that blurs the monochrome image of the first color image,
A ratio detection unit that detects the ratio between the pixel value of the blurred image and the pixel value of the monochrome image corresponding to the pixel value of the blurred image.
A ratio determination unit that determines whether or not the ratio is greater than a predetermined threshold value,
An edge output unit that outputs pixels whose ratio is determined to be larger than the threshold value as edges in the monochrome image.
A pixel value changing unit that generates a third color image by changing the pixel value at the position of the edge in the second color image in which at least the brightness of the first color image is compressed while maintaining similar colors.
An image processing device comprising. Generates a blurred image that blurs the monochrome image of the first color image,
The ratio of the pixel value of the blurred image to the pixel value of the monochrome image corresponding to the pixel value of the blurred image is detected.
It is determined whether or not the ratio is larger than a predetermined threshold value, and it is determined.
Pixels determined to have the ratio greater than the threshold value are output as edges in the monochrome image.
An image processing method for generating a third color image by changing the pixel value at the position of the edge in the second color image in which at least the brightness of the first color image is compressed while maintaining similar colors. On the computer
Steps to generate a blurred image that blurs the monochrome image of the first color image,
A step of detecting the ratio of the pixel value of the blurred image to the pixel value of the monochrome image corresponding to the pixel value of the blurred image, and
A step of determining whether or not the ratio is greater than a predetermined threshold, and
A step of extracting pixels whose ratio is determined to be larger than the threshold value as edges in the monochrome image, and
A step of generating a third color image by changing the pixel value at the position of the edge in the second color image in which at least the brightness of the first color image is compressed while maintaining similar colors .
A computer-readable program that runs.

Images