Optimization processing method of digital photo frame based on image processing

Abstract

The invention relates to the technical field of image processing, in particular to a digital photo frame optimization processing method based on image processing. Pixel identification values and edge identification values are obtained by carrying out pixel distribution identification and edge distribution identification on each block of an image, the pixel and edge distribution states of the image can be accurately identified and measured, and comprehensive analysis is carried out to identify an area with a relatively high contrast adjustment requirement; by calling and analyzing the pixel distribution value and the edge distribution value of each block in the image, the area needing to be adjusted can be accurately identified and adjusted in a targeted manner, the contrast ratio of the image can be accurately optimized, and the brightness and the contrast ratio in the image can be effectively improved by adopting a local optimization mode, so that the image effect is more ideal; the smooth filter is applied after the contrast is adjusted, noise points caused by excessive sharpening can be reduced, unnatural noise points or artifacts on detail parts of the image are avoided, and it is ensured that the finally displayed image not only is scientifically optimized, but also meets the visual requirements of users.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to an optimization processing method of a digital photo frame based on image processing. Background Art

[0002] A digital photo frame is an electronic device used to display digital images or photos. Unlike traditional paper photo frames, digital photo frames can display multiple photos on an electronic screen, and users can transfer photos to the device through various methods (such as USB, SD card, Wi-Fi, etc.); users can directly operate and browse photos on the screen; Current digital photo frames usually display images based on images input by users. Some images may have inappropriate contrast, overexposure of some areas of the image (loss of bright details), and excessive dimming of some areas (invisible dark details), resulting in unclear image display and poor user viewing experience. Summary of the invention

[0003] The purpose of the present invention is to provide an optimization processing method for a digital photo frame based on image processing to solve the problems mentioned in the above background technology.

[0004] The object of the present invention can be achieved by the following technical solution: An optimization processing method for a digital photo frame based on image processing comprises the following steps: R1: There are several images stored in the digital photo frame. Select one of the images that has not been optimized and execute R2; R2: Identify the pixel distribution state of the image by comprehensively analyzing the pixels and brightness of each block in the image, and obtain the pixel identification value accordingly; R3: The edge distribution state of the image is identified by comprehensively analyzing the gradient amplitude and gradient direction of the pixel points in each block in the image, and the edge recognition value is obtained accordingly; R4: Perform weighted calculation on the pixel identification value and the edge identification value to obtain the optimization requirement value. If the optimization requirement value is greater than or equal to the set requirement threshold, execute R5; otherwise, return to R1; R5: Optimize the image based on the pixel distribution and edge distribution of the image. The specific steps are as follows: Step 1: retrieve the pixel distribution value and edge distribution value of each block in the image, and perform weighted calculation to obtain an adjustment index. When the adjustment index is greater than or equal to the set adjustment threshold, the block is recorded as an adjustment block. Step 2: Retrieve the brightness L and intensity mean D of the adjustment block, and use the set formula The adjusted contrast LD is calculated, wherein g1 and g2 are respectively the set proportional coefficients, and α is the set contrast conversion coefficient; thus, the adjusted contrast of each adjustment block can be obtained, and the adjustment block is adjusted according to the adjusted contrast; Step 3: Apply a smoothing filter to smooth the image after contrast adjustment, thereby obtaining an optimized image and saving it for subsequent display operations.

[0005] Preferably, the specific process of pixel identification is: Extract the image stored in the digital photo frame, divide the image into several blocks and record it as n, where n=1, 2, 3...N, N is a positive integer, N represents the number of blocks in the image, and n represents the sequence number of any block in the image; grayscale the image, and identify the brightness of each block and record it as Ln; identify the pixels in each block, and perform statistics to obtain the number of pixels in each block, and record it as Sn; set a standard brightness value and record it as HL; Substitute the block brightness Ln, the number of pixels Sn and the standard brightness value HL into the set formula The pixel distribution value LSn of each block is calculated, where a1 and a2 are respectively set as proportional coefficients; Compare and analyze the pixel distribution value of each block with the set distribution interval to divide the blocks corresponding to the pixel distribution value into high-degree pixel blocks, medium-degree pixel blocks and low-degree pixel blocks, count the number of high-degree pixel blocks, medium-degree pixel blocks and low-degree pixel blocks respectively, and record them as Q1, Q2 and Q3 respectively; sum the pixel distribution values ​​corresponding to the high-degree pixel blocks, medium-degree pixel blocks and low-degree pixel blocks respectively to obtain the high-degree pixel value, medium-degree pixel value and low-degree pixel value, and record them as Q4, Q5 and Q6 respectively; Substitute the pixel distribution value LSn of each block, the number of high pixel blocks Q1, the number of medium pixel blocks Q2, the number of low pixel blocks Q3, the high pixel value Q4, the medium pixel value Q5 and the low pixel value Q6 into the set formula The pixel identification value Qb is calculated, wherein b1, b2, b3, and b4 are respectively set proportional coefficients, b1>b2>0 and b3>b4>0. is the mean value of pixel distribution in each block.

[0006] Preferably, the specific process of edge identification is: 301: Obtaining the gradient magnitude and gradient direction of each pixel point in the block in the horizontal and vertical directions respectively through the gradient operators in the horizontal and vertical directions; 302: performing non-maximum suppression based on the gradient magnitude and gradient direction of each pixel in the horizontal and vertical directions to identify edge pixels and the gradient magnitude corresponding to the edge pixels; 303: Analyze the edge distribution state based on each edge pixel point in the block and its corresponding gradient amplitude to obtain an edge distribution value; 304: Compare and analyze the edge distribution value corresponding to each block with the set edge interval to divide the edge distribution value into a high edge block, a medium edge block and a low edge block, count the cumulative number of the high edge block, the medium edge block and the low edge block respectively, and record them as B1, B2, and B3 respectively; sum the edge distribution values ​​corresponding to the high edge block, the medium edge block and the low edge block respectively to obtain a high edge value, a medium edge value and a low edge value, and record them as B4, B5, and B6 respectively; 305: Substitute the edge distribution value DTn of each block, the cumulative number of high edge blocks B1, the cumulative number of medium edge blocks B2, the cumulative number of low edge blocks B3, the high edge value B4, the medium edge value B5 and the low edge value B6 into the set formula The edge recognition value Bd is calculated, where d1, d2, d3, and d4 are respectively the set proportional coefficients, d1>d2>0 and d3>d4>0. is the mean of the marginal distribution of each block.

[0007] Preferably, the gradient operators of the gradient amplitude and gradient direction of each pixel point in the block in the horizontal and vertical directions are: Formula (1); Formula (2); Formula (3); Formula (4); The above formulas (1) to (4) can be used to obtain the gradient magnitude GM and gradient direction θ of each pixel in the block.

[0008] Preferably, the specific process of non-maximum suppression is: Select any pixel point and take it as the center point; take the adjacent pixel points in the gradient direction of the center point as the comparison points; if the gradient amplitude corresponding to the center point is greater than the corresponding gradient amplitude of each comparison point, then the center point is recorded as an edge pixel point, and its corresponding gradient amplitude is retained; otherwise, the center point is recorded as a non-edge pixel point, and its corresponding gradient amplitude is set to zero, that is, the pixel is suppressed.

[0009] Preferably, the edge distribution state analysis is specifically as follows: The gradient amplitude corresponding to each edge pixel point is averaged and the intensity mean of each block is recorded as Dn; Select the edge point with the largest gradient amplitude and the edge point with the smallest gradient amplitude from the edge points in the block, and record the largest gradient amplitude and the smallest gradient amplitude as Dmax and Dmin; Assume that there is a standard edge strength value recorded as HD, and substitute it with the intensity mean Dn, the maximum gradient amplitude Dmax and the minimum gradient amplitude Dmin into the set formula to calculate The edge distribution value DTn is calculated, where h1 and h2 are the set proportional coefficients respectively.

[0010] Beneficial effects of the present invention: 1. By performing pixel distribution recognition and edge distribution recognition on each block of the image to obtain pixel recognition values ​​and edge recognition values, the pixel and edge distribution states of the image can be accurately recognized and measured, and a comprehensive analysis will be performed to identify areas with high contrast adjustment requirements, which will help to optimize the image in a targeted manner and improve the visual effect; it provides strong data support for subsequent image processing steps to ensure that the adjustment of the image is more scientific and reasonable; 2. By retrieving and analyzing the pixel distribution values ​​and edge distribution values ​​of each block in the image, the area that needs to be adjusted can be accurately identified and adjusted in a targeted manner, which can accurately optimize the contrast of the image; effectively avoid unnecessary adjustments to ensure that the details of key areas in the image are retained and enhanced; the use of local optimization methods can effectively improve the brightness and contrast in the image, making the image effect more ideal; applying a smoothing filter (such as Gaussian blur) after contrast adjustment can reduce the noise caused by over-sharpening, help maintain the natural smoothness of the image, and avoid unnatural noise or artifacts in the details of the image, ensuring that the final displayed image is not only scientifically optimized, but also meets the user's visual needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The present invention will be further described below in conjunction with the accompanying drawings.

[0012] Figure 1 It is a schematic flow chart of the method of the present invention. DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0014] See also Figure 1As shown, the present invention is an optimization processing method for a digital photo frame based on image processing, comprising the following steps: R1: There are several images stored in the digital photo frame. Select one of the images that has not been optimized and execute R2; R2: Obtain pixel identification values ​​by performing pixel distribution on the image, specifically: Extract the image stored in the digital photo frame, divide the image into several blocks and record it as n, where n=1, 2, 3...N, N is a positive integer, N represents the number of blocks in the image, and n represents the sequence number of any block in the image; grayscale the image, and identify the brightness of each block and record it as Ln; identify the pixels in each block, and perform statistics to obtain the number of pixels in each block, and record it as Sn; set a standard brightness value and record it as HL; it should be noted that the standard brightness value usually refers to an ideal brightness level, which is designed to make the image best suited to the user's visual perception under standard conditions; By setting the formula The pixel distribution value LSn of each block is calculated, where a1 and a2 are respectively set proportional coefficients; it can be seen from the formula that when the brightness of the block is greater and the number of pixels in the block is greater, it means that many pixels are concentrated in the high-brightness block, which means that the possibility of overexposure of the image is greater, and the pixel distribution value is greater; when the brightness of the block is smaller and the number of pixels in the block is greater, it means that many pixels are concentrated in the low-brightness block, which means that the possibility of over-brightness of the image is greater, and the pixel distribution value is greater; when the brightness is closer to the standard brightness value, the pixel distribution value of the block is smaller; Compare and analyze the pixel distribution value of each block with the set distribution interval. When the pixel distribution value is greater than the maximum value in the set distribution interval, it means that the block corresponding to the pixel distribution value has a relatively large possibility of overexposure and overdarkness, and the area is recorded as a high pixel block; when the pixel distribution value is greater than the set distribution interval, the area is recorded as a medium pixel block; when the pixel distribution value is less than the minimum value in the set distribution interval, the area is recorded as a low pixel block; count the number of high pixel blocks, medium pixel blocks and low pixel blocks respectively, and record them as Q1, Q2 and Q3 respectively; sum the pixel distribution values ​​corresponding to the high pixel blocks, medium pixel blocks and low pixel blocks respectively to obtain the high pixel value, medium pixel value and low pixel value, and record them as Q4, Q5 and Q6 respectively; use the set formula The pixel identification value Qb is calculated, wherein b1, b2, b3, and b4 are respectively set proportional coefficients, b1>b2>0 and b3>b4>0. is the mean value of pixel distribution of each block; from the formula, when the pixel distribution value of the block is larger, the number of highly distributed blocks and moderately distributed blocks is larger, which means that the pixels are more concentrated in the exposed or dim area, and the pixel recognition value is larger; when the difference of the mean value of pixel distribution of each block is larger, the pixel recognition value is larger; R3: Obtain edge recognition value by performing edge distribution recognition on the image, specifically: The gradient magnitude and gradient direction of each pixel in the horizontal and vertical directions are calculated by using gradient operators in the horizontal (Gx) and vertical (Gy) directions (such as the Sobel operator). The commonly used gradient operators are as follows: Formula (1); Formula (2); Formula (3); Formula (4); The above formulas (1) to (4) can be used to obtain the gradient amplitude GM and gradient direction θ of each pixel in the block; select any one of the pixel points and use it as the center point; select the adjacent pixel points in the gradient direction of the center point as the comparison points; if the gradient amplitude corresponding to the center point is greater than the gradient amplitude corresponding to each comparison point, then the center point is recorded as an edge pixel point, and its corresponding gradient amplitude is retained; otherwise (if the gradient amplitude corresponding to the center point is less than the gradient amplitude of any one of the comparison points), then the center point is recorded as a non-edge pixel point, and its corresponding gradient amplitude is set to zero, that is, the pixel is suppressed; after non-maximum suppression, only those pixels with gradient amplitudes of local maximum values ​​are retained as edge pixels, and the gradient amplitudes of other pixels are set to 0, thereby removing inconspicuous edges or noise; It should be noted that two adjacent pixels are selected as comparison points along the gradient direction of the center point. The specific selection method is as follows: there are 4 main directions, namely; 0° direction (parallel to the horizontal direction), 45° direction (parallel to the diagonal direction, i.e., upper left to lower right), 90° direction (parallel to the vertical direction) and 135° direction (parallel to the diagonal direction, i.e., lower left to upper right); the gradient direction of the center point is compared with the 4 main directions, and the main direction with the smallest difference is selected as the reference direction of the center point, and two adjacent pixels are selected along the reference direction as comparison points; for example, if the gradient direction of a pixel point is 50° and matches the 45° direction (parallel to the diagonal direction, i.e., upper left to lower right), the adjacent pixels of the diagonal direction (upper left and lower right) of the pixel point are used as comparison points, and the three are compared to obtain edge pixels; Thus, the edge pixels in each block can be obtained, and the gradient amplitudes corresponding to each edge pixel point are averaged to obtain the intensity average of each block, which is recorded as Dn. It should be noted that a higher intensity average indicates that the edges in the block are generally stronger or more obvious, indicating that the contrast of the block part is higher and the edge is clearer; the edge point with the largest gradient amplitude and the edge point with the smallest gradient amplitude are selected from the edge points in the block, and the largest gradient amplitude and the smallest gradient amplitude are recorded as Dmax and Dmin; it should be noted that the difference between the largest gradient amplitude and the smallest gradient amplitude reflects the degree of discreteness or distribution breadth of the edge intensity in the block; the larger the difference, the greater the change in edge intensity, that is, there are both strong edges and weak edges in the block, which means that the contrast of the block is greater; It is assumed that there is a standard edge strength value, which is recorded as HD. It should be noted that the standard edge strength value refers to an ideal edge strength level, which is designed to make the image best suited to the user's visual perception under standard conditions. The set formula is used to calculate The edge distribution value DTn is calculated, where h1 and h2 are respectively the set proportional coefficients; thus, the edge distribution value of each block can be obtained; from the formula, when the mean intensity value of the block is larger (or smaller), the difference between the maximum gradient amplitude and the minimum gradient amplitude in the block is larger, indicating that the edge details of the block are stronger (or weaker), the greater the need for contrast adjustment, and the larger the edge distribution value; when the mean intensity value in the block is closer to the standard edge intensity value, it means that the need for contrast adjustment of the block is smaller, and the edge distribution value is smaller; Compare and analyze the edge distribution value corresponding to each block with the set edge interval. When the edge distribution value is greater than the maximum value in the set edge interval, a high edge block is accumulated; when the edge distribution value is in the set edge interval, a medium edge block is accumulated; when the edge distribution value is less than the minimum value in the set edge interval, a low edge block is accumulated; the cumulative number of high edge blocks, medium edge blocks and low edge blocks is counted respectively, and they are recorded as B1, B2, and B3 respectively; the edge distribution values ​​corresponding to the high edge blocks, medium edge blocks and low edge blocks are summed up to obtain the high edge value, medium edge value and low edge value, and they are recorded as B4, B5, and B6 respectively; use the set formula The edge recognition value Bd is calculated, where d1, d2, d3, and d4 are respectively the set proportional coefficients, d1>d2>0 and d3>d4>0. is the edge distribution mean of each block; from this, the pixel distribution value corresponding to each block can be obtained; R4: Perform a comprehensive analysis based on the pixel identification value and edge identification value of the image to determine whether optimization processing is needed, specifically: The pixel identification value and the edge identification value are weighted to obtain the optimization requirement value, and the optimization requirement value is compared and analyzed with the set requirement threshold. If the optimization requirement value is greater than or equal to the set requirement threshold, it means that the image needs to be contrast optimized to present a better visual effect; then R5 is executed; otherwise, it means that the image does not need to be optimized, and then R1 is returned; By performing pixel distribution recognition and edge distribution recognition on each block of the image to obtain pixel recognition values ​​and edge recognition values, it is possible to accurately identify and measure the pixel and edge distribution status of the image, and conduct a comprehensive analysis to identify areas with higher contrast adjustment requirements, which helps to optimize the image in a targeted manner and improve the visual effect; it provides strong data support for subsequent image processing steps to ensure that the adjustment of the image is more scientific and reasonable.

[0015] R5: Optimize the image based on the pixel distribution and edge distribution of the image to present a better visual effect; the specific steps are: Step 1: retrieve the pixel distribution value and edge distribution value of each block in the image, and perform weighted calculation to obtain an adjustment index. When the adjustment index is greater than or equal to the set adjustment threshold, it means that the block needs to be adjusted and optimized, and the block is recorded as an adjustment block. Step 2: Get the brightness L and intensity mean D of the adjustment block, and use the set formula The adjusted contrast LD is calculated, where g1 and g2 are respectively the set proportional coefficients, and α is the set contrast conversion coefficient; thus, the adjusted contrast of each adjustment block can be obtained, and the adjustment block is adjusted according to the adjusted contrast; the specific adjustment method is that if the adjusted contrast is 14, the contrast of the adjustment block is increased by 14; if the adjusted contrast is -14, the contrast of the adjustment block is reduced by 14; Step 3: Apply a smoothing filter (such as Gaussian blur) to smooth the image after contrast adjustment to reduce noise caused by over-sharpening. This will result in an optimized image and save it for subsequent display operations. By retrieving and analyzing the pixel distribution values ​​and edge distribution values ​​of each block in the image, the areas that need to be adjusted can be accurately identified and adjusted in a targeted manner, which can accurately optimize the contrast of the image; effectively avoid unnecessary adjustments to ensure that the details of key areas in the image are retained and enhanced; the use of local optimization methods can effectively improve the brightness and contrast in the image, making the image effect more ideal; applying a smoothing filter (such as Gaussian blur) after contrast adjustment can reduce the noise caused by over-sharpening, help maintain the natural smoothness of the image, and avoid unnatural noise or artifacts in the details of the image, ensuring that the final displayed image is not only scientifically optimized, but also meets the user's visual needs.

[0016] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. An optimization processing method for a digital photo frame based on image processing, characterized in that: The following steps are involved: R1: There are several images stored in the digital photo frame. Select one of the images that has not been optimized and execute R2; R2: Identify the pixel distribution state of the image by comprehensively analyzing the pixels and brightness of each block in the image, and obtain the pixel identification value accordingly; R3: The edge distribution state of the image is identified by comprehensively analyzing the gradient amplitude and gradient direction of the pixel points in each block in the image, and the edge recognition value is obtained accordingly; R4: Perform weighted calculation on the pixel identification value and the edge identification value to obtain the optimization requirement value. If the optimization requirement value is greater than or equal to the set requirement threshold, execute R5; otherwise, return to R1; R5: Optimize the image based on the pixel distribution and edge distribution of the image. The specific steps are as follows: Step 1: retrieve the pixel distribution value and edge distribution value of each block in the image, and perform weighted calculation to obtain an adjustment index. When the adjustment index is greater than or equal to the set adjustment threshold, the block is recorded as an adjustment block. Step 2: retrieve the average brightness and intensity of the adjustment block, normalize them and obtain their values, analyze the values ​​to obtain the adjustment contrast, thereby obtaining the adjustment contrast of each adjustment block, and adjust the adjustment block according to the adjustment contrast; Step 3: Apply a smoothing filter to smooth the image after contrast adjustment, thereby obtaining an optimized image and saving it for subsequent display operations.

2. The optimization processing method of the digital photo frame based on image processing according to claim 1, characterized in that: The specific process of pixel recognition is: Extract the image stored in the digital photo frame, process the image in grayscale, and identify the brightness of each block; identify the pixels in each block, and perform statistics to obtain the number of pixels in each block; It is assumed that there is a standard brightness value; The brightness, number of pixels and standard brightness value of the block are normalized and their values ​​are taken, and the pixel distribution value of each block is obtained by numerical analysis; Compare and analyze the pixel distribution value of each block with the set distribution interval to divide the blocks corresponding to the pixel distribution value into high-degree pixel blocks, medium-degree pixel blocks and low-degree pixel blocks, and count the number of high-degree pixel blocks, medium-degree pixel blocks and low-degree pixel blocks respectively; The pixel distribution values ​​corresponding to the high-degree pixel block, the medium-degree pixel block and the low-degree pixel block are respectively summed up to obtain the high-degree pixel value, the medium-degree pixel value and the low-degree pixel value; The pixel distribution value of each block, the number of high-degree pixel blocks, the number of medium-degree pixel blocks, the number of low-degree pixel blocks, the high-degree pixel value, the medium-degree pixel value and the low-degree pixel value are calculated and analyzed in a formula to obtain the pixel identification value.

3. The optimization processing method of the digital photo frame based on image processing according to claim 1, characterized in that: The specific process of edge recognition is: 301: Obtaining the gradient magnitude and gradient direction of each pixel point in the block in the horizontal and vertical directions respectively through the gradient operators in the horizontal and vertical directions; 302: performing non-maximum suppression based on the gradient magnitude and gradient direction of each pixel in the horizontal and vertical directions to identify edge pixels and the gradient magnitude corresponding to the edge pixels; 303: Analyze the edge distribution state based on each edge pixel point in the block and its corresponding gradient amplitude to obtain an edge distribution value; 304: Compare and analyze the edge distribution value corresponding to each block with the set edge interval to divide the edge distribution value into a high edge block, a medium edge block and a low edge block, count the cumulative number of the high edge block, the medium edge block and the low edge block respectively, and sum the edge distribution values ​​corresponding to the high edge block, the medium edge block and the low edge block respectively to obtain a high edge value, a medium edge value and a low edge value; 305: Formulate and calculate the edge distribution value of each block, the cumulative number of high edge blocks, the cumulative number of medium edge blocks, the cumulative number of low edge blocks, the high edge value, the medium edge value and the low edge value to obtain an edge recognition value.

4. The optimization processing method of the digital photo frame based on image processing according to claim 3 is characterized in that: The gradient operator of the gradient amplitude and gradient direction of each pixel in the block in the horizontal and vertical directions is: Formula (1): Formula (2): Formula (3): Formula (4): The above formulas (1) to (4) can be used to obtain the gradient magnitude GM and gradient direction θ of each pixel in the block.

5. The optimization processing method of the digital photo frame based on image processing according to claim 4, characterized in that: The specific process of non-maximum suppression is: Select any pixel point and take it as the center point; take the adjacent pixel points in the gradient direction of the center point as the comparison points; if the gradient amplitude corresponding to the center point is greater than the corresponding gradient amplitude of each comparison point, then the center point is recorded as an edge pixel point, and its corresponding gradient amplitude is retained; otherwise, the center point is recorded as a non-edge pixel point, and its corresponding gradient amplitude is set to zero, that is, the pixel is suppressed.

6. The optimization processing method of the digital photo frame based on image processing according to claim 5, characterized in that: The specific analysis of marginal distribution state is as follows: The gradient amplitude corresponding to each edge pixel point is averaged to obtain the intensity mean of each block; Select the edge point with the largest gradient amplitude and the edge point with the smallest gradient amplitude from the edge points in the block; It is assumed that there is a standard edge intensity value, and the edge distribution value is obtained by formulating and analyzing it with the intensity mean, the maximum gradient amplitude and the minimum gradient amplitude.

Images