A skin texture monitoring method based on image processing

Abstract

The present invention discloses a skin texture monitoring method based on image processing, belonging to the technical field of image processing. First, the present invention counts the number of pixel values of pixel points in a skin texture image in a local area, the row where the pixel points are located, and the column where the pixel points are located, to obtain the local number, the row number, and the column number; then, according to the difference between the local number of the same pixel point and the row and column numbers, the row and column local difference degrees are calculated, and a corresponding difference image is constructed; then, abnormal pixel points are marked in the difference image, and the first and second form attention degrees are obtained according to their forms in the original image; then, the difference image is enhanced to obtain an enhanced image; finally, the enhanced image is processed by a dual-channel convolutional neural network, and combined with the form attention degrees, the skin texture deviation degree is obtained. By comparing the number of pixel values in the local area with the row and column numbers, the present invention highlights local abnormal conditions and improves the accuracy of the monitoring results.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly relates to a skin texture monitoring method based on image processing. Background Art

[0002] Traditional skin texture monitoring methods mainly rely on manual observation, which not only consumes a large amount of time and energy, but also has strong subjectivity, and there are large differences in judgments among different observers. With the development of image processing technology, automated skin texture monitoring methods have gradually emerged. Existing skin texture monitoring technologies based on image processing mostly focus on simple texture feature extraction, such as extracting features such as the direction and contrast of texture through the gray-level co-occurrence matrix. However, these methods often only consider the global features of the image and lack the excavation of local area features. In complex skin texture monitoring scenarios, due to the diversity and individual differences of skin textures, global features cannot accurately reflect local detail changes. For example, when monitoring skin aging, lesions, etc., the subtle changes in local textures are key information, but traditional methods are difficult to capture the changes in these local features, resulting in inaccurate monitoring results and the inability to detect potential skin problems in a timely manner. Summary of the Invention

[0003] Aiming at the above deficiencies in the prior art, a skin texture monitoring method based on image processing provided by the present invention solves the problem of inaccurate monitoring results existing in the prior art.

[0004] In order to achieve the above invention object, the technical solution adopted by the present invention is: A skin texture monitoring method based on image processing, comprising the following steps:

[0005] Count the number of occurrences of the pixel value of each pixel point in the local area, the row where it is located, and the column where it is located in the skin texture image to obtain the local number, the row number, and the column number;

[0006] Calculate the row local difference degree and the column local difference degree according to the differences between the local number of the same pixel point and the row number and the column number respectively, and construct a row local difference image and a column local difference image;

[0007] Mark abnormal pixel points on the row local difference image and the column local difference image, and obtain the first morphological attention degree and the second morphological attention degree based on the morphology of the abnormal pixel points on the skin texture image;

[0008] Perform difference degree enhancement processing on the row local difference image and the column local difference image respectively to obtain a row local difference enhanced image and a column local difference enhanced image;

[0009] A dual-channel convolutional neural network is used to process the row-local difference enhanced image and the column-local difference enhanced image, and the skin texture deviation is obtained based on the attention imposed by the first morphological attention and the second morphological attention.

[0010] Further, the specific process of obtaining the local quantity, the row quantity, and the column quantity is as follows: Taking a neighborhood centered on each pixel point in the skin texture image, counting the number of occurrences of the pixel value at the central pixel point in the neighborhood to obtain the local quantity, counting the number of occurrences of the pixel value of each pixel point in the row where it is located to obtain the row quantity, and counting the number of occurrences of the pixel value of each pixel point in the column where it is located to obtain the column quantity, where N is a positive integer.

[0011] Further, the specific process of constructing the row-local difference image and the column-local difference image is as follows: According to the difference between the local quantity and the row quantity of the same pixel point, calculate the row-local difference degree, and replace the original pixel value with the row-local difference degree of the same pixel point to obtain the row-local difference image;

[0012] According to the difference between the local quantity and the column quantity of the same pixel point, calculate the column-local difference degree, and replace the original pixel value with the column-local difference degree of the same pixel point to obtain the column-local difference image.

[0013] Further, the formula for calculating the row-local difference degree is: , where is the row-local difference degree of the pixel point at the i-th row and the j-th column, is the local quantity of the pixel point at the i-th row and the j-th column, is the row quantity of the pixel point at the i-th row and the j-th column, and e is the natural constant;

[0014] The formula for calculating the column-local difference degree is: , where is the column-local difference degree of the pixel point at the i-th row and the j-th column, is the column quantity of the pixel point at the i-th row and the j-th column, and i and j are positive integers.

[0015] Further, the specific process of obtaining the first morphological attention and the second morphological attention is as follows:

[0016] Calculate the mean value of all row-local difference degrees on the row-local difference image to obtain the first mean value;

[0017] Mark the pixel points with row-local difference degrees greater than the first mean value on the row-local difference image as the first type of abnormal pixel points;

[0018] Calculate the mean value of all column-local difference degrees on the column-local difference image to obtain the second mean value;

[0019] Mark the pixel points with column local difference degrees greater than the second mean value on the column local difference image as the second type of abnormal pixel points;

[0020] Obtain the first morphological attention degree according to the positions of the first type of abnormal pixel points on the skin texture image;

[0021] Obtain the second morphological attention degree according to the positions of the second type of abnormal pixel points on the skin texture image.

[0022] Furthermore, the specific processes of obtaining the first morphological attention degree and the second morphological attention degree are both:

[0023] Calculate the curvature for each edge pixel point of the connected region according to the connected regions formed by each abnormal pixel point on the skin texture image;

[0024] Take the normalized value of the variance of the curvatures of all edge pixel points of the connected region as the edge clutter degree;

[0025] Calculate the pixel contrast according to the difference between the pixel values on the connected region and the pixel values of other pixel points on the skin texture image;

[0026] Add the edge clutter degree and the pixel contrast to obtain the morphological feature value of the connected region;

[0027] Calculate the morphological attention degree according to the areas and morphological feature values of each connected region.

[0028] Furthermore, the formula for calculating the pixel contrast is: , where, θ n is the pixel contrast of the nth connected region, P n,avg is the mean value of the pixel values of the nth connected region, P all,avg is the mean value of the pixel values of other pixel points on the skin texture image, | | is the absolute value operation, n is a positive integer, and ε is a constant;

[0029] The formula for calculating the morphological attention degree is: , where, ζ is the morphological attention degree, E n is the area of the nth connected region, γ n is the morphological feature value of the nth connected region, and M is the number of connected regions.

[0030] Furthermore, the formula for enhancing the difference degree of the row local difference image is: , where, is the enhanced row local difference degree of the pixel point at the i-th row and j-th column, is the row local difference degree of the pixel point at the i-th row and j-th column, D R,max is the maximum row local difference degree;

[0031] The formula for enhancing the degree of difference of the column local difference image is as follows: , where is the enhanced column local difference degree of the pixel at the i-th row and j-th column, is the column local difference degree of the pixel at the i-th row and j-th column, and D C,max is the maximum column local difference degree.

[0032] Furthermore, the dual-channel convolutional neural network includes: a first image feature extraction channel, a second image feature extraction channel, a first attention application layer, a second attention application layer, a first convolutional block, a second convolutional block, a feature fusion layer, and a fully connected layer;

[0033] The input end of the first image feature extraction channel is used to input the row local difference enhanced image;

[0034] The input end of the second image feature extraction channel is used to input the column local difference enhanced image;

[0035] The first input end of the first attention application layer is connected to the output end of the first image feature extraction channel, its second input end is used to input the first morphological attention, and its output end is connected to the input end of the first convolutional block; the first input end of the second attention application layer is connected to the output end of the second image feature extraction channel, its second input end is used to input the second morphological attention, and its output end is connected to the input end of the second convolutional block; the input ends of the feature fusion layer are respectively connected to the output ends of the first convolutional block and the second convolutional block, and its output end is connected to the input end of the fully connected layer;

[0036] The output end of the fully connected layer serves as the output end of the dual-channel convolutional neural network.

[0037] Furthermore, the expression of the first attention application layer is: , where G1 is the output of the first attention application layer, ζ1 is the first morphological attention, and S1 is the output of the first image feature extraction channel;

[0038] The expression of the second attention application layer is: , where G2 is the output of the second attention application layer, ζ2 is the second morphological attention, and S2 is the output of the second image feature extraction channel.

[0039] The beneficial effects of the present invention are as follows:

[0040] 1. By statistically counting the number of occurrences of each pixel point in the local area, the row where it is located, and the column where it is located, the present invention can carefully analyze the local characteristics of the image. The present invention considers pixel points from multiple dimensions (local, row, column), captures local abnormalities in skin texture, and improves the sensitivity to local texture changes.

[0041] 2. The present invention constructs a difference image based on the differences between the local quantity of the same pixel point and the quantities of rows and columns. This method highlights the differences between the local area and the quantities of rows and columns, thereby measuring the uniformity of texture distribution and highlighting abnormal areas.

[0042] 3. The present invention marks abnormal pixel points on the difference image and obtains the attention degree according to their forms on the original skin texture image, which helps to more accurately identify real abnormalities.

[0043] 4. The present invention performs a difference degree enhancement process on the row and column local difference images, further highlighting the difference between abnormal textures and normal textures.

[0044] 5. The present invention uses a dual-channel convolutional neural network to process the enhanced difference image, which can make full use of the spatial information and features of the image, and combines the attention degrees imposed by the first form attention degree and the second form attention degree to more accurately calculate the skin texture deviation degree, thereby comprehensively and accurately evaluating the skin texture state and improving the accuracy of skin monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a flowchart of a skin texture monitoring method based on image processing;

[0046] Figure 2 is a schematic diagram of an example of a neighborhood;

[0047] Figure 3 is a schematic diagram of the structure of a dual-channel convolutional neural network. DETAILED DESCRIPTION OF THE INVENTION

[0048] The following describes the specific embodiments of the present invention to facilitate those skilled in the art of the present technology to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art of the present technology, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are within the scope of protection.

[0049] As Figure 1 shown, a skin texture monitoring method based on image processing includes the following steps:

[0050] Statistical the quantities of the pixel values of each pixel point in the local area, the row where it is located, and the column where it is located in the skin texture image to obtain the local quantity, the row quantity, and the column quantity;

[0051] According to the differences between the local quantity of the same pixel point and the row quantity and the column quantity respectively, calculate the row local difference degree and the column local difference degree, and construct a row local difference image and a column local difference image;

[0052] Mark abnormal pixel points on the row local difference image and the column local difference image, and obtain the first morphological attention degree and the second morphological attention degree based on the morphology of the abnormal pixel points on the skin texture image;

[0053] Perform difference degree enhancement processing on the row local difference image and the column local difference image respectively to obtain a row local difference enhanced image and a column local difference enhanced image;

[0054] Process the row local difference enhanced image and the column local difference enhanced image using a two-channel convolutional neural network, and obtain the skin texture deviation degree based on the attention degrees imposed by the first morphological attention degree and the second morphological attention degree.

[0055] In this embodiment, the specific process of obtaining the local quantity, the row quantity, and the column quantity is as follows: Taking each pixel point in the skin texture image as the center, take a neighborhood, and count the number of occurrences of the pixel value at the central pixel point in the neighborhood to obtain the local quantity, count the number of occurrences of the pixel value of each pixel point in the row where it is located to obtain the row quantity, and count the number of occurrences of the pixel value of each pixel point in the column where it is located to obtain the column quantity, where N is a positive integer.

[0056] For example, taking the pixel point at the 5th row and the 3rd column as the center, then count the number of occurrences of the pixel value at the central pixel point in the neighborhood to obtain the local quantity, then count the number of occurrences of the pixel value at the central pixel point in the 5th row, and then count the number of occurrences of the pixel value at the central pixel point in the 3rd column.

[0057] In this embodiment, the skin texture image is at a resolution, with 1920 pixel points in one row and 1080 pixel points in one column.

[0058] In this embodiment, the skin texture image can be an image collected by a dermoscope, not limited to the dermoscope image pointed out in this embodiment, and other images that can reflect the skin texture state are all acceptable.

[0059] In this embodiment, is set to or etc. When set to , it reflects the distribution gap situation of the pixel values within the range of and the pixel values in one row or one column. When set to , it reflects the situation of The distribution gap between pixel values within a range and those in a row or a column. When N is set to a smaller value, it is better to compare the pixel values in a local area with those in a row or a column, and it is more capable of extracting the features of the local area. The size of N set in the present invention is not limited, and the size of N can be adjusted according to different skin detection situations. During the specific operation process, different N values can be set to obtain the skin texture deviation degrees under different N values.

[0060] For example, as shown in the neighborhood Figure 2 if the pixel value of the pixel point at the center is 120, then within the neighborhood range, 120 appears 7 times. Looking at the row (the 100th row) where the pixel point at the center is located, there are a total of 1920 pixel points in this row. Select some pixel values from them for display: 116, 119, 120, 122, 124, 125, 115, …, 118, 120, 120, 123, 125, 127, 130. Count the number of times 120 appears. When counting the local quantity, row quantity, and column quantity, the pixel value of the pixel point at the center itself is included. The pixel values in the example adopt the image grayscale values.

[0061] For a normal skin texture image, there is consistency in each local area and as a whole, while abnormal texture changes will lead to an increase in the difference between the local area and the whole. By calculating the row local difference degree and the column local difference degree, the areas with abnormal texture can be highlighted.

[0062] In this embodiment, the specific process of constructing the row local difference image and the column local difference image is as follows: Calculate the row local difference degree according to the difference between the local quantity and the row quantity of the same pixel point, and replace the original pixel value with the row local difference degree of the same pixel point to obtain the row local difference image;

[0063] Calculate the column local difference degree according to the difference between the local quantity and the column quantity of the same pixel point, and replace the original pixel value with the column local difference degree of the same pixel point to obtain the column local difference image.

[0064] In this embodiment, the difference between the local quantity and the row quantity of the same pixel point can be used as the row local difference degree, and the difference between the local quantity and the column quantity of the same pixel point can be used as the column local difference degree. More preferably, the formula for calculating the row local difference degree is: where, is the row local difference degree of the pixel point at the i-th row and the j-th column, is the local quantity of the pixel point at the i-th row and the j-th column, is the row quantity of the pixel point at the i-th row and the j-th column, and e is the natural constant;

[0065] The formula for calculating the column local difference degree is: where, is the column local difference degree of the pixel at the i-th row and j-th column, is the column quantity of the pixel at the i-th row and j-th column, where i and j are positive integers.

[0066] By calculating the difference between the local quantity of the same pixel and the row and column quantities, and constructing a difference degree formula in exponential form, the present invention can amplify the difference between the local and overall distributions of pixels. Normal skin texture has a certain regularity and consistency, while when there are lesions, aging, etc., the local texture will deviate from the overall rule. This calculation method can keenly capture these deviations and prominently reflect the differences.

[0067] At greater than , greater than When, the larger the values of the row local difference degree and the column local difference degree, while at less than , less than When, the values of the row local difference degree and the column local difference degree are between 0 and 1. Therefore, the abnormal and normal values can be significantly distinguished, facilitating the acquisition of the first form attention degree and the second form attention degree.

[0068] In this embodiment, the specific process of obtaining the first form attention degree and the second form attention degree is as follows:

[0069] Calculate the mean value of all row local difference degrees on the row local difference image to obtain the first mean value;

[0070] On the row local difference image, mark the pixels whose row local difference degree is greater than the first mean value as the first type of abnormal pixels;

[0071] Calculate the mean value of all column local difference degrees on the column local difference image to obtain the second mean value;

[0072] On the column local difference image, mark the pixels whose column local difference degree is greater than the second mean value as the second type of abnormal pixels;

[0073] According to the positions of the first type of abnormal pixels on the skin texture image, obtain the first form attention degree;

[0074] According to the positions of the second type of abnormal pixels on the skin texture image, obtain the second form attention degree.

[0075] The present invention determines the threshold by calculating the mean values of the local difference images of rows and columns, marks the pixel points greater than the mean value as abnormal pixel points, and applies statistical principles. In image processing, the mean value can reflect the central tendency of data, and the pixel points greater than the mean value can be regarded as abnormal points deviating from the normal distribution. This screening method accurately locates the pixel points with texture abnormalities at the data level.

[0076] The abnormalities of skin texture are not evenly distributed but concentrated in certain specific areas. The abnormal pixel points screened by the above method can better capture these abnormal areas. For example, when wrinkles or skin spots appear on the skin, the texture in the corresponding area will show differences in local and overall comparisons, and be marked as abnormal pixel points, and then the morphological attention obtained can reflect the distribution pattern of these abnormal areas.

[0077] In this embodiment, the specific processes of obtaining the first morphological attention and the second morphological attention are both:

[0078] According to the connected regions formed by each abnormal pixel point on the skin texture image, calculate the curvature for each edge pixel point of the connected region;

[0079] Take the normalized value of the variance of the curvatures of all edge pixel points of the connected region as the edge clutter degree;

[0080] According to the difference between the pixel values on the connected region and the pixel values of other pixel points on the skin texture image, calculate the pixel contrast;

[0081] Add the edge clutter degree and the pixel contrast to obtain the morphological feature value of the connected region;

[0082] Calculate the morphological attention according to the area and morphological feature value of each connected region.

[0083] In the present invention, a connected region is that each abnormal pixel point is adjacent and connected in spatial position to form a continuous region.

[0084] The curvature reflects the degree of bending of the edge, and the variance normalized value reflects the overall change of the edge. In image processing, this method can accurately capture the contour features of the abnormal areas of skin texture. At the same time, calculating the pixel contrast measures the difference between the abnormal area and the surrounding area from the aspect of pixel difference, and the combination of the two comprehensively describes the morphological features of the connected region.

[0085] In this embodiment, the formula for calculating the pixel contrast is: , where θ n is the pixel contrast of the nth connected region, P n,avg is the mean value of the pixel values of the nth connected region, P all,avgis the mean value of the pixel values of other pixel points on the skin texture image, || is the absolute value operation, n is a positive integer, and ε is a constant.

[0086] In the formula for calculating the pixel contrast in the present invention, other pixel points on the skin texture image refer to pixel points except for each abnormal pixel point.

[0087] The constant ε is used to avoid making the denominator zero.

[0088] The present invention obtains the pixel contrast by finding the difference between the mean value of the pixels in the connected region and the mean value of other pixels, and normalizing the difference, so as to measure the degree of difference between the connected region and other parts of the skin texture image.

[0089] In this embodiment, the mean value of the morphological feature values of each connected region can be used as the morphological attention degree. More preferably, the formula for calculating the morphological attention degree is: , where ζ is the morphological attention degree, E n is the area of the nth connected region, γ n is the morphological feature value of the nth connected region, and M is the number of connected regions.

[0090] The present invention uses the area of each connected region as a weight to calculate the centroid of each connected region, which is convenient for measuring the overall morphological situation. In this way, the connected regions with a relatively large area proportion can obtain a larger proportion. Compared with the mean value, the centroid can better reflect the concentration trend of the abnormal distribution of the skin texture and highlight the dominant role of the large-area abnormal region in the overall morphology.

[0091] In this embodiment, the formula for enhancing the difference degree of the row local difference image is: , where is the enhanced row local difference degree of the pixel point at the i-th row and j-th column, is the row local difference degree of the pixel point at the i-th row and j-th column, and D R,max is the maximum row local difference degree;

[0092] The formula for enhancing the difference degree of the column local difference image is: , where is the enhanced column local difference degree of the pixel point at the i-th row and j-th column, is the column local difference degree of the pixel point at the i-th row and j-th column, and D C,max is the maximum column local difference degree.

[0093] In the present invention, by performing a square operation on the local difference degrees of rows and columns and utilizing the characteristics of the square function, the original difference degree values can be effectively amplified. For the pixel points with a relatively large original difference degree (corresponding to the abnormal skin texture area), the enhanced difference degree will increase significantly; while for the pixel points with a small difference degree (normal texture area), the change is relatively small. In this way, the gap between the difference degrees of the normal and abnormal areas is further widened, making the abnormal texture part more prominent in the image and facilitating the subsequent accurate identification of texture abnormalities. For example, in a local area, the occurrence number of the pixel value at the center point is 5, while the occurrence number of this pixel value in the row where it is located is only 1, then the local row difference degree is e 4 ; in a local area, the occurrence number of the pixel value at the center point is 5, while the occurrence number of this pixel value in the row where it is located is 100, then the local row difference degree is e -95 , and then by squaring e 4 and e -95 , the gap between the two is further widened. Normalization is to reduce the data volume processed by the dual-channel convolutional neural network and at the same time make each value on the same scale.

[0094] As Figure 3 shown, the dual-channel convolutional neural network includes: a first image feature extraction channel, a second image feature extraction channel, a first attention application layer, a second attention application layer, a first convolutional block, a second convolutional block, a feature fusion layer, and a fully connected layer;

[0095] The input end of the first image feature extraction channel is used to input the row local difference enhanced image;

[0096] The input end of the second image feature extraction channel is used to input the column local difference enhanced image;

[0097] The first input end of the first attention application layer is connected to the output end of the first image feature extraction channel, its second input end is used to input the first morphological attention, and its output end is connected to the input end of the first convolutional block; the first input end of the second attention application layer is connected to the output end of the second image feature extraction channel, its second input end is used to input the second morphological attention, and its output end is connected to the input end of the second convolutional block; the input ends of the feature fusion layer are respectively connected to the output ends of the first convolutional block and the second convolutional block, and its output end is connected to the input end of the fully connected layer;

[0098] The output end of the fully connected layer serves as the output end of the dual-channel convolutional neural network.

[0099] In this embodiment, both of the two image feature extraction channels are: CNN neural networks.

[0100] In this embodiment, the feature fusion layer adds the output of the first convolutional block and the output of the second convolutional block element by element to obtain the fused feature.

[0101] The expression of the first attention application layer is: , where G1 is the output of the first attention application layer, ζ1 is the first morphological attention, and S1 is the output of the first image feature extraction channel;

[0102] The expression of the second attention application layer is: , where G2 is the output of the second attention application layer, ζ2 is the second morphological attention, and S2 is the output of the second image feature extraction channel.

[0103] The present invention uses two independent image feature extraction channels to process the row and column local difference images respectively, which can capture the texture difference features from two dimensions, avoid information omission caused by a single channel, and improve the integrity and accuracy of feature extraction.

[0104] The present invention combines the first and second morphological attentions through the first and second attention application layers, enabling the network to enhance the attention to important features when processing images. In the present invention, the larger the first and second morphological attentions, the greater the possibility of the existence of abnormal regions in the skin texture image. Multiplying the morphological attention by the output of the image feature extraction channel focuses more on important texture difference features and suppresses the interference of irrelevant or secondary features, which is beneficial to improving the recognition accuracy of the network for texture abnormalities.

[0105] The present invention can carefully analyze the local characteristics of the image by counting the number of occurrences of each pixel point in the local area, the row where it is located, and the column where it is located. The present invention considers pixel points from multiple dimensions (local, row, column), captures local abnormalities in skin texture, and improves the sensitivity to local texture changes.

[0106] The present invention constructs a difference image based on the difference between the local quantity and the row and column quantities of the same pixel point. This method highlights the differences between the local area and the row quantity as well as the column quantity, thereby measuring the uniformity of texture distribution and highlighting abnormal regions.

[0107] The present invention marks abnormal pixel points on the difference image and obtains the attention according to their morphology on the original skin texture image, which helps to more accurately identify real abnormalities.

[0108] The present invention performs a difference enhancement process on the row and column local difference images, further highlighting the difference between abnormal textures and normal textures.

[0109] The present invention uses a two-channel convolutional neural network to process the enhanced difference image, which can make full use of the spatial information and features of the image, and combines the attention applied by the first morphological attention and the second morphological attention to more accurately calculate the skin texture deviation degree, thereby comprehensively and accurately evaluating the skin texture state.

[0110] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A skin texture monitoring method based on image processing, characterized in that, It includes the following steps: Count the number of occurrences of the pixel value of each pixel point in the skin texture image in the local area, the row where it is located, and the column where it is located to obtain the local quantity, row quantity, and column quantity; Calculate the row local difference degree and column local difference degree according to the differences between the local quantity of the same pixel point and the row quantity and column quantity respectively, and construct a row local difference image and a column local difference image; Mark abnormal pixel points on the row local difference image and the column local difference image, and obtain the first morphological attention degree and the second morphological attention degree based on the morphology of the abnormal pixel points on the skin texture image; Perform difference degree enhancement processing on the row local difference image and the column local difference image respectively to obtain a row local difference enhanced image and a column local difference enhanced image; Use a two-channel convolutional neural network to process the row local difference enhanced image and the column local difference enhanced image, and obtain the skin texture deviation degree based on the attention degrees applied by the first morphological attention degree and the second morphological attention degree; Calculate the mean value of all row local difference degrees on the row local difference image to obtain the first mean value; Mark the pixel points with row local difference degrees greater than the first mean value on the row local difference image as the first type of abnormal pixel points; Calculate the mean value of all column local difference degrees on the column local difference image to obtain the second mean value; Mark the pixel points with column local difference degrees greater than the second mean value on the column local difference image as the second type of abnormal pixel points; Obtain the first morphological attention degree according to the positions of the first type of abnormal pixel points on the skin texture image; Obtain the second morphological attention degree according to the positions of the second type of abnormal pixel points on the skin texture image; According to the connected regions formed by each abnormal pixel point on the skin texture image, calculate the curvature of each edge pixel point of the connected region; Take the normalized value of the variance of the curvatures of all edge pixel points of the connected region as the edge clutter degree; Calculate the pixel contrast according to the difference between the pixel value on the connected region and the pixel values of other pixel points on the skin texture image: , where θ n is the pixel contrast of the nth connected region, P n,avg is the mean value of the pixel values of the nth connected region, P all,avg is the mean value of the pixel values of other pixel points on the skin texture image, | | represents the absolute value operation, n is a positive integer, and ε is a constant; Add the edge clutter degree to the pixel contrast to obtain the morphological feature value of the connected region; Calculate the morphological attention degree according to the area and morphological eigenvalue of each connected region: , where ζ is the morphological attention degree, and E n is the area of the nth connected region, and γ n is the morphological eigenvalue of the nth connected region, and M is the number of connected regions.

2. The method for monitoring skin texture based on image processing according to claim 1, wherein The specific process of obtaining the local quantity, the row quantity, and the column quantity is as follows: Taking each pixel point in the skin texture image as the center, take a neighborhood, and count the number of occurrences of the pixel value at the central pixel point in the neighborhood to obtain the local quantity. Count the number of occurrences of the pixel value of each pixel point in the row where it is located to obtain the row quantity, and count the number of occurrences of the pixel value of each pixel point in the column where it is located to obtain the column quantity, where N is a positive integer.

3. The method for monitoring skin texture based on image processing according to claim 1, wherein The specific process of constructing the row local difference image and the column local difference image is as follows: calculate the row local difference degree according to the difference between the local quantity of the same pixel point and the row quantity, and replace the original pixel value with the row local difference degree of the same pixel point to obtain the row local difference image; Calculate the column local difference degree according to the difference between the local quantity of the same pixel point and the column quantity, and replace the original pixel value with the column local difference degree of the same pixel point to obtain the column local difference image.

4. The method for monitoring skin texture based on image processing according to claim 1, wherein The formula for calculating the row local difference degree is as follows: , where is the row local difference degree of the pixel at the j-th column of the i-th row, is the local quantity of the pixel at the j-th column of the i-th row, is the row quantity of the pixel at the j-th column of the i-th row, and e is the natural constant; The formula for calculating the local column difference is as follows: , where is the local column difference of the pixel at the i-th row and j-th column, is the number of columns of the pixel at the i-th row and j-th column, and i and j are positive integers.

5. The method for monitoring skin texture based on image processing according to claim 1, wherein The formula for enhancing the degree of difference of the line local difference image is as follows: , where is the enhanced line local difference degree of the pixel at the j-th column of the i-th row, is the line local difference degree of the pixel at the j-th column of the i-th row, and D R,max is the maximum line local difference degree, and i and j are positive integers; The formula for enhancing the degree of difference of the column local difference image is as follows: , where is the column local difference degree of the pixel at the i-th row and j-th column, and D C,max is the maximum column local difference degree.

6. The skin texture monitoring method based on image processing according to claim 1, characterized in that The two-channel convolutional neural network includes: a first image feature extraction channel, a second image feature extraction channel, a first attention application layer, a second attention application layer, a first convolutional block, a second convolutional block, a feature fusion layer, and a fully connected layer; The input end of the first image feature extraction channel is used to input the row local difference enhanced image; The input end of the second image feature extraction channel is used to input the column local difference enhanced image; The first input end of the first attention application layer is connected to the output end of the first image feature extraction channel, its second input end is used to input the first morphological attention degree, and its output end is connected to the input end of the first convolutional block; The first input end of the second attention application layer is connected to the output end of the second image feature extraction channel, its second input end is used to input the second form attention, and its output end is connected to the input end of the second convolutional block; The input end of the feature fusion layer is respectively connected to the output end of the first convolutional block and the output end of the second convolutional block, and its output end is connected to the input end of the fully connected layer; The output end of the fully connected layer serves as the output end of the dual-channel convolutional neural network.

7. The method for monitoring skin texture based on image processing according to claim 6, wherein The expression of the first attention application layer is as follows: , where G1 is the output of the first attention application layer, ζ1 is the first form attention, and S1 is the output of the first image feature extraction channel; The expression of the second attention application layer is as follows: , where G2 is the output of the second attention application layer, ζ2 is the second form attention, and S2 is the output of the second image feature extraction channel.

Images