A method for displaying gray scale promotion of an electrowetting display

Abstract

The application provides a gray scale promotion method of an electrowetting display, which comprises the following steps: firstly, preprocessing the input image; then, performing edge segmentation on the image to obtain image edge and non-edge regions; respectively, performing detail processing on the edge and non-edge regions; performing adaptive error diffusion processing on the image according to the image type; and performing image classification according to the image gray scale characteristics. The method performs weighted emphasis on the edge region, introduces pixel neighborhood similarity and local visual deviation into the non-edge region to retain the texture of the image smooth region, finally, fuses the gray scale image with enhanced details, and performs adaptive error diffusion according to the image classification result to obtain a 64 gray scale image suitable for the electrowetting electronic paper display. The application can make the electrowetting display with 64 gray scales reproduce an image with 256 gray scales, and the image display is well adapted to the visual perception of the human eye, and the processed image is highly consistent with the original image, and the image details and textures are restored on the electrowetting display.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of image processing and display, and particularly relates to a method for improving the display gray scale of an electrowetting display. BACKGROUND

[0002] The principle of the electrowetting electronic paper display is to change the wettability of ink on a substrate by changing the voltage between the ink and the insulating substrate, change the contact angle, make the ink deform, and produce displacement, and a successful electrowetting display is made by using this principle. Compared with liquid crystal and LCD displays, the electrowetting display has the advantages of fast response speed, reflective light emission, no need for a backlight source, low power consumption, high brightness, still viewable in strong light, no viewing angle limitation, and the like, so the electrowetting display has a wide application prospect.

[0003] As a new type of display, the electrowetting display has attracted extensive attention from domestic and foreign scholars, however, most of the researches are mainly directed to the structure of the electrowetting device, ink materials, driving systems and the like, and the research on high-quality display of the electrowetting image is still rare. The current electrowetting display is limited by the driving chip and the driving mode, resulting in that the displayable gray scale level is less than the mainstream image gray scale level. In order to realize the display of high gray scale images on the electrowetting display, the low-bit data information is directly discarded, and only the high-bit data information is retained for display, resulting in that the image appears step jump and details are lost. The existing error diffusion algorithm is a halftone technology, which diffuses the current pixel error to the adjacent unprocessed pixels, and effectively improves the display image quality by using the spatial characteristics of the human eye, but the traditional error diffusion algorithm still has the problems of edge blur, detail loss, and obvious artificial texture. SUMMARY

[0004] Therefore, the purpose of the present application is to provide an electrowetting display gray scale improvement method based on an error diffusion algorithm, so that the displayed 64 gray scale image can better restore the edge information and texture details of the 256 gray scale image, and improve the readability of the electrowetting electronic paper display.

[0005] In the scheme, first, the input image is preprocessed in gray scale, then edge segmentation is performed, the image edge and non-edge region are obtained, after detail processing, adaptive error diffusion processing is performed according to the image type, and image classification is performed according to the image gray scale characteristics. The method emphasizes the edge region by weighting, and introduces the pixel neighborhood similarity and local visual deviation in the non-edge region to retain the texture of the image smooth area. Finally, a gray scale image with enhanced details is obtained by fusion, adaptive error diffusion is performed according to the image classification result, and a 64 gray scale image suitable for electrowetting electronic paper display is obtained. The electrowetting display with 64 gray scales can reproduce an image with 256 gray scales, and the image display is well adapted to the observation of the human eye. The processed image is highly consistent with the original image, and the details and textures of the image are restored on the electrowetting display.

[0006] The technical scheme specifically adopted by the present application to solve the technical problems is:

[0007] A method for displaying gray scale enhancement of an electrowetting display,

[0008] Converting a color image into a 256-level gray scale image;

[0009] Performing sub-pixel edge segmentation processing based on Zernike orthogonal matrix on the gray scale image, dividing the image into edge region and non-edge region, weighting the edge region to emphasize the image edge, adjusting the gray scale value of the non-edge region according to the relationship between the local visual deviation of the human eye and the neighborhood similarity and visual perception error, and fusing to generate an ideal gray scale image that emphasizes the edge and highlights the details and textures;

[0010] According to the relationship among the peak gray scale, the mean gray scale and the median gray scale, the image is divided into three categories: high information image, uniform information image and low information image. After classification according to the gray scale distribution characteristics of the image to be processed, the original 256-level gray scale image is classified and mapped into a 64-level gray scale image according to the nonlinear brightness characteristic curve of the electrowetting display.

[0011] Finally, the error is set as the difference between the ideal gray scale image and the 64-level gray scale image obtained by classification and mapping, the Floyd-Steinberg kernel is used to perform error diffusion processing on the image, and a gray scale image suitable for electrowetting electronic paper display is obtained.

[0012] Further, the process of performing sub-pixel edge segmentation processing based on Zernike orthogonal matrix is specifically:

[0013] First, the Zernike moments of 8 pixels around each pixel in the de-noised gray image are calculated, and then the edge strength factor of each pixel is calculated according to the Zernike moments, the edge strength factors are arranged in ascending order, and the first four edge strength factors k, |G|, |L| and |l2-l1| are selected as the judgment basis of the final edge detection result; and the real edge position of the pixel is calculated by using the edge strength factor, and the sub-pixel edge detection is completed.

[0014]

[0015] In the formula, (xs, ys) represents the image edge coordinates obtained by rotating the sub-pixel edge detection of the pixel, (i, j) represents the rotation origin, and theta represents the rotation angle.

[0016] The edge judgment condition of the sub-pixel edge segmentation is:

[0017]

[0018] In the formula, k, G, L, l2 and l1 are edge strength factors; k t , G t , L t and l t are judgment thresholds.

[0019] The real edge position of the pixel is calculated by using the edge strength factor, and the sub-pixel edge detection is completed, and the image is segmented into edge regions and non-edge regions.

[0020] Further, the edge regions are weighted to emphasize the image edge, and the specific method is:

[0021]

[0022] In the formula, Itf(i,j) represents the pixel gray value of the emphasized edge region, and alpha is a weighting coefficient, and f edge (i,j) represents the pixel gray value of the edge region.

[0023] Further, the non-edge region is adjusted according to the relationship between the local visual deviation of the human eye and the neighborhood similarity and the visual perception error of the non-edge region, and the specific method is:

[0024]

[0025] In the formula, M(i,j) represents the non-edge region pixel gray value after the threshold adjustment of the pixel neighborhood similarity based on the local visual deviation of the human eye, beta represents the non-edge region adjustment threshold, D(i,j) represents the neighborhood similarity of the non-edge region pixel to be processed, and delta F t (i,j) represents the human visual perception error of the non-edge region.

[0026] Furthermore, the image classification criteria are:

[0027]

[0028]

[0029] Where index1 and index2 are grayscale evaluation parameters; ω 1、 ω 2、 ω3 is the weight of the mean grayscale, median grayscale and peak grayscale, average represents the mean grayscale, mid represents the median grayscale, and peak represents the peak grayscale; index1 uses Th1 as the threshold to divide the image into high-bit information images and other images, and index2 uses Th2 as the threshold to divide the image into low-bit information images and uniform information images.

[0030] Furthermore, the down-scaling mapping method for the low-bit information image is:

[0031]

[0032] The down-order mapping method for high-bit information images is:

[0033]

[0034] The order reduction mapping method for uniform information images is:

[0035]

[0036] Where b(i, j) is the grayscale value of the image pixel after the reduction process, z1(i, j) is the ideal image pixel grayscale value that strengthens the edge and refines the texture, th represents the segmentation threshold, l represents the low-bit information image, and h represents the high-bit information image.

[0037] Furthermore, the error diffusion algorithm is used to define the image error as the difference between the ideal image with enhanced edges and refined textures and the image after image classification and reduction. The Floyd-Steinberg error diffusion kernel is then used for error diffusion processing to disperse the error of the pixel to be processed to the surrounding pixels.

[0038] The specific content of the error diffusion process is:

[0039] The difference between the ideal image emphasizing edge refinement and the image classified and reduced according to the distribution of image grayscale information is used as the error for error diffusion using the Floyd-Steinberg kernel to obtain the corrected pixel grayscale. The specific formula is:

[0040]

[0041] where b(i,j) is the image pixel after classification and down-sampling according to the distribution of image gray information, b'(i,j) is the corrected image pixel, W(i,j) is the Floyd-Steinberg error diffusion kernel, and e(i,j) is the quantization error between the ideal image and the classification and down-sampled image.

[0042] Further, the neighborhood of the pixel to be processed in the non-edge region of the image is set as m x n, and the neighborhood weighted average gray value Avg(i,j) based on the visual deviation of the human eye is:

[0043]

[0044] where F t (i,j) represents the gray value of the pixel to be processed in the non-edge region, and S(m,n) is a spatial weight function.

[0045] The human visual perception error in the non-edge region is represented as the average gray difference between the pixel to be processed and the neighborhood pixels:

[0046]

[0047] The neighborhood similarity D(i,j) of the pixel to be processed is calculated according to the following formula:

[0048]

[0049] where represents the deviation between the current pixel and the pixels in the adjacent region; h is a control factor for retaining detail smoothing noise.

[0050] Further, the distribution of the image gray information is evaluated, and the evaluation parameter is:

[0051]

[0052]

[0053] where index1 and index2 are both gray evaluation parameters; ω 1、 ω 2、 ω3 are the weights of the mean gray, the median gray, and the peak gray, respectively, average represents the mean gray, mid represents the median gray, and peak represents the peak gray.

[0054] Thus, the judgment criterion of the image gray information concentrated region is established: Th1 is taken as the criterion of the gray information concentrated in the high gray scale part, when index1 is greater than or equal to Th1, the image is classified as a high-bit information image; when index1 is less than Th1, Th2 is taken as the criterion of the gray information concentrated in the low gray scale part and the gray information uniformly distributed in each gray scale, when index2 is greater than or equal to Th2, the image is classified as a low-bit information image; when index2 is less than Th2, the image is classified as a uniform information image.

[0055] Compared with the prior art, the beneficial effects of the present application and the preferred solutions thereof at least include:

[0056] When the input image gray scale level does not match the displayable gray scale level of the electrowetting display, the adaptive error diffusion processing according to the image gray information distribution characteristics proposed by the present application can better restore the edge profile and image texture details of the original image on the electrowetting display, and improve the readability of the electrowetting display. Compared with the existing algorithm, the objective evaluation indexes of the present algorithm are all improved by different magnitudes, among which the average value of the image peak signal-to-noise ratio PSNR can reach 45.6343 dB, the average value of the structural similarity SSIM with the original image is improved to 0.9834, the average value of the overall image quality UQI is improved to 0.9994, the average value of the mean absolute error MAE with the original image is reduced to 0.58251, and the average value of the mean square error MSE with the original image is reduced to 1.02100. BRIEF DESCRIPTION OF DRAWINGS

[0057] The present application will be further described below in conjunction with the accompanying drawings and specific embodiments:

[0058] Figure 1 The present application is an embodiment of the overall flowchart. DETAILED DESCRIPTION

[0059] In order to make the features and advantages of the present patent more apparent, the following embodiments are specifically described, and the detailed description is as follows:

[0060] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present application. Unless otherwise indicated, all technical and scientific terms used in the present description have the same meaning as commonly understood by those skilled in the art to which the present application belongs.

[0061] It should be noted that the terms used herein are only for the purpose of describing specific embodiments, and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form, and in addition, it should be understood that when the terms "comprise" and / or "include" are used in the present description, they indicate the presence of a feature, step, operation, device, component and / or their combination.

[0062] As Figure 1 shown, the embodiment provides an error diffusion algorithm-based method for displaying gray scale of an electrowetting display, comprising the following steps:

[0063] Step S1: obtaining a color image and converting it into a gray scale image;

[0064] Step S2: performing improved sub-pixel edge detection processing on the gray scale image based on Zernike orthogonal matrix, dividing the image into edge regions and non-edge regions, assigning weights to the edge regions to emphasize the image edges, adjusting the gray scale values of the non-edge regions according to the relationship between the local visual deviation of the human eye and the neighborhood similarity and the visual perception error, and fusing to generate an ideal gray scale image that emphasizes edges and highlights details and textures;

[0065] Step S3: according to the relationship among the peak gray scale, the mean gray scale, and the median gray scale, the image can be divided into three categories: high-bit information image, uniform information image, and low-bit information image. After classification according to the gray scale distribution characteristics of the image to be processed, the original 256 gray scale gray scale image is classified and mapped into a 64 gray scale image according to the nonlinear luminance characteristic curve of electrowetting;

[0066] Step S4: setting the error as the difference between the ideal gray scale image and the 64 gray scale image obtained by classification and mapping, performing error diffusion processing on the image using the Floyd-Steinberg kernel, and obtaining a gray scale image suitable for electrowetting electronic paper display.

[0067] Specifically, the Zernike moments of 8 pixels around each pixel of the denoised gray scale image are calculated, and then the edge intensity factor of each pixel is calculated according to the Zernike moments. The edge intensity factors are arranged in order from small to large, and the first four edge intensity factors k, |G|, |L|, and |l2-l1| are selected as the judgment basis for the final edge detection result. The corrected edge judgment condition is:

[0068]

[0069] where k, G, L, l2, and l1 are edge intensity factors. k t , G t , L t , and l t are judgment thresholds. The real edge position of the pixel is calculated using the edge intensity factor, and sub-pixel edge detection is completed.

[0070]

[0071] In the formula, (xs, ys) represents the image edge coordinates obtained by sub-pixel edge detection after Zernike matrix rotation of the pixel, (i, j) represents the rotation origin, and theta represents the rotation angle.

[0072] The improved sub-pixel edge detection algorithm based on the Zernike orthogonal matrix can effectively and finely segment the edge details of the image, effectively improve the edge detection precision, and maintain good contour continuity, thereby laying a good foundation for subsequent processing.

[0073] According to the sub-pixel edge detection result, the image is segmented, and the gray value of the edge region pixel is weighted to emphasize the image edge.

[0074]

[0075] In the formula, Itf(i,j) represents the gray value of the edge region pixel after emphasis; alpha is a weighting coefficient, which is set to 0.005 in the embodiment; f edge (i,j) represents the gray value of the edge region pixel.

[0076] In the embodiment, the neighborhood of the non-edge region pixel to be processed in the image is set to 3*3, and the neighborhood weighted average gray value Avg(i,j) based on the human eye visual deviation is:

[0077]

[0078] In the formula, F t (i,j) represents the gray value of the non-edge region pixel to be processed, and S(m,n) is a spatial weight function. Since the sensitivity of the human eye in the horizontal and vertical directions is higher than that in the diagonal direction, the weight of the pixel to be processed itself is taken as 1, the weight of the pixel in the diagonal direction is taken as 0.1035, and the weight of the pixel in the horizontal and vertical directions is taken as 0.1465:

[0079]

[0080] The neighborhood similarity D(i,j) of the pixel to be processed is calculated according to the following formula:

[0081]

[0082] In the formula, m*n represents the neighborhood size, which is set to 3*3 in the embodiment. represents the deviation between the current pixel and the pixels in the adjacent region; and h is a control factor for retaining detail smoothing noise, which is set to 0.4 in the embodiment.

[0083] The human eye visual perception error of the non-edge region, i.e., the average gray difference between the current pixel to be processed and the neighborhood pixels:

[0084]

[0085] The non-edge region adopts a pixel neighborhood similarity adjustment threshold based on the local visual deviation of the human eye to highlight the image texture details:

[0086]

[0087] In the formula, M(i,j) represents the pixel gray value of the non-edge region after the pixel neighborhood similarity adjustment threshold based on the local visual deviation of the human eye, and β represents the adjustment threshold of the non-edge region, which is set to 0.1 in this embodiment.

[0088] In this embodiment, the image gray information distribution is evaluated, and the evaluation parameters are:

[0089]

[0090]

[0091] In the formula, index1 and index2 are both gray evaluation parameters. ω 1、 ω 2、 ω3 are the weights of the mean gray, the median gray, and the peak gray, respectively, average represents the mean gray, mid represents the median gray, and peak represents the peak gray.

[0092] In this embodiment, the image gray information concentration region judgment criterion is that Th1 is used as the criterion for the concentration of gray information in the high gray level part, and when index1 is greater than or equal to Th1, the image is classified as a high-bit information image. When index1 is less than Th1, Th2 is used as the criterion for distinguishing whether the gray information is concentrated in the low gray level part or uniformly distributed in each gray level, and when index2 is greater than or equal to Th2, the image is classified as a low-bit information image; when index2 is less than Th2, the image is classified as a uniform information image.

[0093] The electro-wetting display has a nonlinear characteristic in terms of the relative brightness of the displayed input image data, and the relative brightness changes rapidly in the 0-50 gray scale and the 200-255 gray scale. Therefore, adaptive down-sampling processing is used for images with different gray information concentration regions.

[0094] The down-sampling mapping mode for a low-bit information image is:

[0095]

[0096] The down-sampling mapping mode for a high-bit information image is:

[0097]

[0098] The down-sampling mapping mode for a uniform information image is:

[0099]

[0100] Where b(i, j) is the grayscale value of the image pixel after the reduction process, and z1(i, j) is the ideal image pixel grayscale value that strengthens the edge and refines the texture, that is, , z(i,j) represents the grayscale of the original pixel, th represents the segmentation threshold, l represents the low-bit information image, and h represents the high-bit information image.

[0101] In this embodiment, the specific content of the error diffusion process in step S4 is:

[0102] The difference between the ideal image emphasizing edge refinement and the image processed by classification and reduction based on the distribution of image grayscale information is used as the error to perform error diffusion with the Floyd-Steinberg kernel to obtain the corrected pixel grayscale. The specific formula is:

[0103]

[0104] Where b(i, j) is the image pixel that is classified and reduced according to the distribution of image grayscale information, b'(i, j) is the corrected image pixel, W(i, j) is the Floyd-Steinberg error diffusion kernel, and e(i, j) is the quantization error between the ideal image and the reduced-order image.

[0105] The electrowetting display grayscale enhancement algorithm based on the error diffusion algorithm can effectively improve the graininess and directional texture caused by image degradation, and effectively restore the edge information and texture details of the input image. Because electrowetting displays use reflective light emission, the overall color tone of the image displayed is darker than that of active light-emitting liquid crystal displays. Therefore, the image processing results for electrowetting displays appear brighter than the original image when displayed on an LCD.

[0106] This algorithm effectively improves the edge details and image texture of the image displayed by the electrowetting display, and enhances the visual experience of the electrowetting display.

[0107] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other manner. Any person skilled in the art may utilize the above-disclosed technical content to modify or modify the present invention into equivalent embodiments. However, any simple modifications, equivalent variations, and modifications to the above embodiments that do not depart from the technical content of the present invention and are based on the technical essence of the present invention remain within the scope of protection of the present invention.

[0108] The patent is not limited to the above best mode, anyone can draw other various forms of an electrowetting display display gray scale lifting method under the inspiration of the patent, any equivalent changes and modifications made in the patent application scope shall belong to the scope of the patent.

Claims

1. A method for improving grayscale of an electrowetting display, characterized by: Convert the color image into a 256-level grayscale image; performing sub-pixel edge segmentation processing based on Zernike orthogonal moments on the grayscale image, dividing the image into edge regions and non-edge regions, assigning weights to the edge regions to emphasize image edges, adjusting the grayscale values ​​of the non-edge regions based on the relationship between local visual deviation of the human eye and neighborhood similarity and visual perception error, and fusing the results to generate an ideal grayscale image that emphasizes edges and highlights detailed textures; Based on the relationship between peak grayscale, mean grayscale, and median grayscale, the image is divided into three categories: high-level information image, uniform information image, and low-level information image. After classification based on the grayscale distribution characteristics of the image to be processed, the original 256-level grayscale image is classified and down-mapped into a 64-level grayscale image in combination with the electrowetting nonlinear brightness characteristic curve. Finally, the error is set as the difference between the ideal grayscale image and the 64 grayscale image of the classification map, and the image is subjected to error diffusion processing using the Floyd-Steinberg kernel to obtain a grayscale image suitable for electrowetting electronic paper display. The process of performing sub-pixel edge segmentation based on Zernike orthogonal moments is specifically as follows: First, the Zernike moments of the eight pixels surrounding each pixel in the denoised grayscale image are calculated. Then, the edge strength factor of each pixel is calculated based on the Zernike moments. The edge strength factors are arranged in ascending order, and the first four edge strength factors k, |G|, |L|, and |l2-l1| are selected as the basis for judging the final edge detection result. The edge strength factors are then used to calculate the true edge position of the pixel to complete sub-pixel edge detection. Where (xs, ys) represents the image edge coordinates obtained by sub-pixel edge detection after the pixel is rotated by Zernike moment, (i, j) represents the rotation origin, and theta represents the rotation angle; The edge judgment conditions for sub-pixel edge segmentation are: Where, k, G, L, l2, l1 are edge strength factors; k t , G t , L t and l t is the judgment threshold; The edge intensity factor is used to calculate the true edge position of the pixel, complete sub-pixel edge detection, and segment the image into edge and non-edge areas; The down-order mapping method for low-bit information images is: The down-order mapping method for high-bit information images is: The order reduction mapping method for uniform information images is: Where b(i, j) is the grayscale value of the image pixel after the reduction process, z1(i, j) is the ideal image pixel grayscale value that strengthens the edge and refines the texture, th represents the segmentation threshold, l represents the low-bit information image, and h represents the high-bit information image.

2. The method for improving grayscale of an electrowetting display according to claim 1, wherein: The step of assigning weights to edge regions to emphasize image edges specifically includes: Where, Itf(i,j) represents the grayscale value of the edge area pixel after emphasis, α is the weighting coefficient, and f edge (i, j) represents the grayscale value of the pixel in the edge area.

3. The method for improving grayscale of an electrowetting display according to claim 1, wherein: The grayscale value of the non-edge area is adjusted according to the local visual deviation of the human eye and the relationship between the neighborhood similarity and the visual perception error as follows: Where M(i,j) represents the grayscale value of the pixel in the non-edge area after adjusting the threshold of the pixel neighborhood similarity based on the local visual deviation of the human eye, β represents the adjustment threshold of the non-edge area, D(i,j) represents the neighborhood similarity of the pixel to be processed in the non-edge area, ∆F t (i, j) represents the human visual perception error in the non-edge area; F t (i, j) represents the grayscale value of the pixel to be processed in the non-edge area.

4. The method for improving grayscale of an electrowetting display according to claim 1, wherein: The image classification criteria are: Where index1 and index2 are grayscale evaluation parameters; ω 1、 ω 2、 ω3 is the weight of the mean grayscale, median grayscale and peak grayscale, average represents the mean grayscale, mid represents the median grayscale, and peak represents the peak grayscale; Index1 divides the image into high-information image and other images with Th1 as the threshold, and index2 divides the image into low-information image and uniform information image with Th2 as the threshold.

5. The method for improving grayscale of an electrowetting display according to claim 1, wherein: The error diffusion algorithm is used to define the image error as the difference between the ideal image with enhanced edges and refined textures and the image after image classification and reduction. The Floyd-Steinberg error diffusion kernel is then used for error diffusion processing to disperse the error of the pixel to be processed to the surrounding pixels.

6. The method for improving grayscale of an electrowetting display according to claim 5, wherein: The specific content of the error diffusion process is: The difference between the ideal image emphasizing edge refinement and the image classified and reduced according to the distribution of image grayscale information is used as the error for error diffusion using the Floyd-Steinberg kernel to obtain the corrected pixel grayscale. The specific formula is: Where b(i, j) is the image pixel that is classified and reduced according to the distribution of image grayscale information, b'(i, j) is the corrected image pixel, W(i, j) is the Floyd-Steinberg error diffusion kernel, and e(i, j) is the quantization error between the ideal image and the classified and reduced image.

7. The method for improving grayscale of an electrowetting display according to claim 3, wherein: The neighborhood of the pixels to be processed in the non-edge area of ​​the image is set to m×n, and the weighted average grayscale value Avg(i,j) of the neighborhood based on human visual deviation is: Among them, S(m,n) is the spatial weight function; The human visual perception error in the non-edge area is expressed as the average grayscale difference between the current pixel to be processed and the neighboring pixels: The calculation formula of the neighborhood similarity D(i,j) of the pixel to be processed is: in; Represents the deviation between the current pixel and the pixels in the adjacent area; h is a control factor used to retain details and smooth noise.

8. The method for improving grayscale of an electrowetting display according to claim 1, wherein: Evaluate the distribution of image grayscale information, and the evaluation parameters are: Where index1 and index2 are grayscale evaluation parameters; ω 1、 ω 2、 ω3 is the weight of the mean grayscale, median grayscale and peak grayscale, average represents the mean grayscale, mid represents the median grayscale, and peak represents the peak grayscale; Therefore, the judgment standard for the area where the grayscale information of the image is concentrated is established: Th1 is used as the standard for the grayscale information to be concentrated in the high grayscale part. When index1 is greater than or equal to Th1, the image is classified as a high-bit information image; when index1 is less than Th1, Th2 is used as the standard for distinguishing between the grayscale information concentrated in the low grayscale part and the grayscale information evenly distributed in each grayscale. When index2 is greater than or equal to Th2, the image is classified as a low-bit information image; when index2 is less than Th2, the image is classified as a uniform information image.

Images