Advertisement promotion AI picture automatic screening method based on online network platform

Abstract

The invention discloses an advertisement promotion AI picture automatic screening method based on an online network platform, and belongs to the technical field of image processing.The method comprises the steps that firstly, pictures of the online network platform are converted into a YUV space, and Y, U and V channel images are obtained; performing various 4-neighborhood and 8-neighborhood Laplacian operator operations on each channel image to construct a first channel detail enhancement image and a second channel detail enhancement image; respectively calculating the marginal definition of each region in the two types of images, and constructing a first marginal definition matrix and a second marginal definition matrix; processing the matrixes corresponding to the Y, U and V channel images by adopting a picture quality evaluation model to obtain a picture quality score; and finally, when the picture quality score is greater than a score threshold, marking the picture as a recommendation state. According to the invention, the problem of low picture quality evaluation precision in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to an automatic screening method for advertising promotion AI pictures based on an online network platform. Background Art

[0002] With the rapid development of the Internet advertising market, the impact of image quality on advertising effectiveness is becoming more and more significant. At present, most advertising platforms still rely on manual review and simple image quality screening techniques, which are mainly based on basic image properties such as resolution, file size, and simple color histogram analysis. However, this traditional method lacks in-depth analysis of image details and edge clarity, and cannot accurately evaluate the quality of images. Existing technologies are usually unable to effectively distinguish between images with similar resolutions but significant differences in visual quality, resulting in low-quality images frequently entering the advertising promotion system. This not only reduces the visual appeal of the advertisement, but may also affect users' click-through rate and conversion rate. Summary of the invention

[0003] In view of the above-mentioned deficiencies in the prior art, the present invention provides an automatic screening method for advertising promotion AI pictures based on an online network platform, which solves the problem of low accuracy of picture quality assessment in the prior art.

[0004] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is: an automatic screening method for advertising promotion AI pictures based on an online network platform, comprising the following steps:

[0005] S1. Convert the image on the online network platform into YUV space to obtain multiple channel images, wherein the multiple channel images include: Y channel image, U channel image and V channel image;

[0006] S2, performing a variety of 4-neighborhood and 8-neighborhood Laplacian operator operations on each channel image, respectively, to construct a first channel detail enhanced image and a second channel detail enhanced image;

[0007] S3, respectively calculating the edge clarity of each region in the first channel detail enhanced image and the second channel detail enhanced image, and constructing a first edge clarity matrix and a second edge clarity matrix;

[0008] S4, using the picture quality assessment model to process the first edge clarity matrix and the second edge clarity matrix corresponding to the Y, U and V channel images to obtain a picture quality score;

[0009] S5. When the image quality score is greater than the score threshold, mark the image as recommended.

[0010] Furthermore, the 4-neighborhood Laplacian operator in S2 includes: standard 4-neighborhood Laplacian operator , the modified 4-neighborhood Laplacian ; 8-neighborhood Laplacian operators include: Standard 8-neighborhood Laplacian operator , gradient 8-neighborhood Laplacian .

[0011] Further, S2 includes the following sub-steps:

[0012] S21, performing a plurality of 4-neighborhood Laplacian operator operations on each channel image respectively, fusing the images after the operations to obtain a first channel detail enhanced image;

[0013] S22, performing a plurality of 8-neighborhood Laplacian operator operations on each channel image respectively, fusing the images after the operations to obtain a second channel detail enhanced image.

[0014] Further, S21 includes the following sub-steps:

[0015] S211, performing standard 4-neighborhood Laplacian operator and modified 4-neighborhood Laplacian operator operations on each channel image respectively to obtain a first detail image and a second detail image;

[0016] S212, taking the absolute value of each pixel value in the first detail image, and taking the absolute value of each pixel value in the second detail image;

[0017] S213. Add the pixel values ​​of the first detail image after taking the absolute value and the second detail image after taking the absolute value at the same pixel points to obtain a first channel detail enhanced image. When the channel image in S211 is a Y channel image, the first channel detail enhanced image is a first Y channel detail enhanced image. When the channel image in S211 is a U channel image, the first channel detail enhanced image is a first U channel detail enhanced image. When the channel image in S211 is a V channel image, the first channel detail enhanced image is a first V channel detail enhanced image.

[0018] Further, S22 includes the following sub-steps:

[0019] S221, performing standard 8-neighborhood Laplacian operator and modified 8-neighborhood Laplacian operator operations on each channel image respectively to obtain a third detail image and a fourth detail image;

[0020] S222, taking the absolute value of each pixel value in the third detail image, and taking the absolute value of each pixel value in the fourth detail image;

[0021] S223. Add the pixel values ​​of the third detail image after taking the absolute value and the fourth detail image after taking the absolute value at the same pixel points to obtain a second channel detail enhanced image. When the channel image in S221 is a Y channel image, the second channel detail enhanced image is a second Y channel detail enhanced image. When the channel image in S221 is a U channel image, the second channel detail enhanced image is a second U channel detail enhanced image. When the channel image in S221 is a V channel image, the second channel detail enhanced image is a second V channel detail enhanced image.

[0022] Furthermore, S3 includes the following sub-steps:

[0023] S31, calculating edge clarity of each region in the first channel detail enhanced image, and constructing a first edge clarity matrix;

[0024] S32, calculating edge clarity for each region in the second channel detail enhanced image, and constructing a second edge clarity matrix.

[0025] Furthermore, constructing the edge definition matrix in S31 and S32 includes the following steps:

[0026] A1. Pixels whose pixel values ​​in the channel detail enhanced image are greater than the pixel threshold are marked as edge points;

[0027] A2, dividing the channel detail enhanced image into multiple regions;

[0028] A3. In each region, connect each adjacent edge point in the horizontal direction to form multiple horizontal edge lines;

[0029] A4. In each region, connect each adjacent edge point in the vertical direction to form multiple vertical edge lines;

[0030] A5. Calculate edge definition of the region according to the plurality of horizontal edge lines and the plurality of vertical edge lines;

[0031] A6. Use the edge clarity of the region as an element to construct an edge clarity matrix.

[0032] Furthermore, the formula for calculating the edge definition of the area in A5 is: , where θ is the edge clarity of the region, N level,i is the number of edge points on the i-th horizontal edge line, N ver,i is the number of edge points on the i-th vertical edge line, i is a positive integer, L level is the number of horizontal edge lines in a region, L ver is the number of vertical edge lines in a region.

[0033] Further, the picture quality assessment model in S4 includes: a Y channel clarity feature fusion unit, a U channel clarity feature fusion unit, a V channel clarity feature fusion unit, a first convolution block, a second convolution block, a third convolution block, a fourth convolution block, a fifth convolution block, a sixth convolution block, a first fully connected layer, a second fully connected layer, and a third fully connected layer;

[0034] The input end of the Y channel definition feature fusion unit is used to input the first edge definition matrix and the second edge definition matrix corresponding to the Y channel image;

[0035] The input end of the U channel definition feature fusion unit is used to input the first edge definition matrix and the second edge definition matrix corresponding to the U channel image;

[0036] The input end of the V channel definition feature fusion unit is used to input the first edge definition matrix and the second edge definition matrix corresponding to the V channel image;

[0037] The input end of the first convolution block is connected to the output end of the Y channel clarity feature fusion unit, and its output end is connected to the input end of the fourth convolution block and the input end of the first fully connected layer respectively;

[0038] The input end of the second convolution block is connected to the output end of the U channel clarity feature fusion unit, and its output end is connected to the input end of the fifth convolution block and the input end of the first fully connected layer respectively;

[0039] The input end of the third convolution block is connected to the output end of the V channel clarity feature fusion unit, and its output end is connected to the input end of the sixth convolution block and the input end of the first fully connected layer respectively;

[0040] The input end of the second fully connected layer is connected to the output end of the fourth convolution block, the output end of the fifth convolution block, and the output end of the sixth convolution block respectively;

[0041] The input end of the third fully connected layer is connected to the output end of the first fully connected layer and the output end of the second fully connected layer respectively, and the output end of the third fully connected layer serves as the output end of the picture quality assessment model.

[0042] Furthermore, the expressions of the Y channel definition feature fusion unit, the U channel definition feature fusion unit and the V channel definition feature fusion unit are: , where R is the output of the Y channel definition feature fusion unit, the U channel definition feature fusion unit or the V channel definition feature fusion unit, and G 1 is the first edge definition matrix, G 2 is the second edge definition matrix, is element-wise addition, and Conv is a convolution operation.

[0043] The beneficial effects of the present invention are:

[0044] 1. The present invention converts the image into YUV space, separates brightness and color, constructs a Y channel image through brightness Y, a U channel image through blue chromaticity U, and a V channel image through red chromaticity V. The edge clarity of the image is analyzed from three aspects: brightness Y, blue chromaticity U and red chromaticity V. The edge features of the image can be captured from multiple angles and in all directions, avoiding the limitation of single channel analysis.

[0045] 2. The present invention performs multiple 4-neighborhood Laplace operator operations on each channel image, and multiple 8-neighborhood Laplace operator operations on each channel image. The 4-neighborhood Laplace operator mainly focuses on the relationship between the pixel point and its 4 adjacent pixels above, below, left and right, which can highlight the relatively small-scale details and edges in the image, and has a good enhancement effect on some subtle lines, textures and other details in the image. The 8-neighborhood Laplace operator takes into account the adjacent pixels in 8 directions around the pixel point, and can capture larger-scale image features and edge information, and has a better outlining effect on the contours of objects in the image, large-area edges, etc. By performing these two operations on each channel image at the same time, the detailed information of the image can be fully captured from different scales, so that the enhanced image retains rich subtle textures and highlights obvious edges and contours, improving the overall clarity and detail expression of the image.

[0046] 3. The present invention calculates edge clarity for the first channel detail enhanced image and the second channel detail enhanced image respectively. After detail enhancement, the edge clarity of images in different channels reflects the edge features of the images under different attributes of brightness and chromaticity. Then, the first edge clarity matrix and the second edge clarity matrix corresponding to the Y, U and V channel images are processed by the picture quality assessment model to obtain the picture quality score. The picture quality score is obtained from multiple dimensions, which improves the assessment accuracy of picture quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a flow chart of an automatic AI image screening method for advertising promotion based on an online network platform;

[0048] Figure 2 Schematic diagram of the structure of the image quality assessment model. DETAILED DESCRIPTION

[0049] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0050] like Figure 1 As shown, an automatic screening method for advertising promotion AI pictures based on an online network platform includes the following steps:

[0051] S1. Convert the image on the online network platform into YUV space to obtain multiple channel images, wherein the multiple channel images include: Y channel image, U channel image and V channel image;

[0052] S2, performing a variety of 4-neighborhood and 8-neighborhood Laplacian operator operations on each channel image, respectively, to construct a first channel detail enhanced image and a second channel detail enhanced image;

[0053] S3, respectively calculating the edge clarity of each region in the first channel detail enhanced image and the second channel detail enhanced image, and constructing a first edge clarity matrix and a second edge clarity matrix;

[0054] S4, using the picture quality assessment model to process the first edge clarity matrix and the second edge clarity matrix corresponding to the Y, U and V channel images to obtain a picture quality score;

[0055] S5. When the image quality score is greater than the score threshold, mark the image as recommended.

[0056] In this embodiment, the scoring threshold is specifically set based on experiments or experience.

[0057] In this embodiment, the 4-neighborhood Laplacian operator in S2 includes: standard 4-neighborhood Laplacian operator , the modified 4-neighborhood Laplacian ; 8-neighborhood Laplacian operators include: Standard 8-neighborhood Laplacian operator , gradient 8-neighborhood Laplacian .

[0058] The Laplacian operator is essentially a second-order derivative operator, which is used to detect sudden changes in pixel values ​​in an image, i.e., edges. The standard 4-neighborhood Laplacian operator only considers adjacent pixels in the four directions of up, down, left, and right, and is more sensitive to edges in other directions. It can highlight simple lines and texture details in the image and enhance small-scale edge features. The modified 4-neighborhood Laplacian operator increases the weight of the diagonal direction, which can detect edges more comprehensively, especially better capture oblique details and edges, and help enrich the detailed information of the image.

[0059] Compared with the 4-neighborhood Laplacian operator, the 8-neighborhood Laplacian operator considers pixel information in more directions and can detect image edges and features on a larger scale. The standard 8-neighborhood Laplacian operator has a good outlining effect on the overall contours of objects in the image and the edges of large areas; the gradient 8-neighborhood Laplacian operator is more sensitive to areas with large changes in pixel values ​​through weight setting, which helps to detect high-contrast edges in the image, realize multi-scale image analysis, and mine image structural information from different levels.

[0060] In this embodiment, S2 includes the following sub-steps:

[0061] S21, performing a plurality of 4-neighborhood Laplacian operator operations on each channel image respectively, fusing the images after the operations to obtain a first channel detail enhanced image;

[0062] S22, performing a plurality of 8-neighborhood Laplacian operator operations on each channel image respectively, fusing the images after the operations to obtain a second channel detail enhanced image.

[0063] In this embodiment, S21 includes the following sub-steps:

[0064] S211, performing standard 4-neighborhood Laplacian operator and modified 4-neighborhood Laplacian operator operations on each channel image respectively to obtain a first detail image and a second detail image;

[0065] In S211, when a standard 4-neighborhood Laplace operator operation is performed on the channel image, the result after the operation is used to replace the original pixel value in the channel image to obtain a first detail image; when a modified 4-neighborhood Laplace operator operation is performed on the channel image, the result after the operation is used to replace the original pixel value in the channel image to obtain a second detail image;

[0066] S212, taking the absolute value of each pixel value in the first detail image, and taking the absolute value of each pixel value in the second detail image;

[0067] S213. Add the pixel values ​​of the first detail image after taking the absolute value and the second detail image after taking the absolute value at the same pixel points to obtain a first channel detail enhanced image. When the channel image in S211 is a Y channel image, the first channel detail enhanced image is a first Y channel detail enhanced image. When the channel image in S211 is a U channel image, the first channel detail enhanced image is a first U channel detail enhanced image. When the channel image in S211 is a V channel image, the first channel detail enhanced image is a first V channel detail enhanced image.

[0068] The present invention uses standard 4-neighborhood and modified 4-neighborhood Laplace operator operations respectively, the former is sensitive to horizontal and vertical edges, and the latter can take into account oblique details. The combination of the two can capture detail information in the channel image from different angles, so that the generated first and second detail images contain rich image features. For example, in advertising images, the product texture and contour details can be better displayed. The present invention takes the absolute value of the pixel values ​​of the first and second detail images to avoid the loss of detail information due to the mutual cancellation of positive and negative values ​​of the Laplace operation results, thereby ensuring the stability and integrity of the detail features and making the image detail enhancement effect more reliable. The present invention adds the corresponding pixel values ​​of the two images after taking the absolute value, integrates the detail information obtained by the two operator operations, further enhances the detail effect, and makes the generated channel detail enhanced image more prominent in details in all directions.

[0069] In this embodiment, S22 includes the following sub-steps:

[0070] S221, performing standard 8-neighborhood Laplacian operator and modified 8-neighborhood Laplacian operator operations on each channel image respectively to obtain a third detail image and a fourth detail image;

[0071] S222, taking the absolute value of each pixel value in the third detail image, and taking the absolute value of each pixel value in the fourth detail image;

[0072] S223. Add the pixel values ​​of the third detail image after taking the absolute value and the fourth detail image after taking the absolute value at the same pixel points to obtain a second channel detail enhanced image. When the channel image in S221 is a Y channel image, the second channel detail enhanced image is a second Y channel detail enhanced image. When the channel image in S221 is a U channel image, the second channel detail enhanced image is a second U channel detail enhanced image. When the channel image in S221 is a V channel image, the second channel detail enhanced image is a second V channel detail enhanced image.

[0073] The standard 8-neighborhood and modified 8-neighborhood Laplace operators take into account the information in 8 directions around the pixel, and can more comprehensively detect edges and details in all directions. The combination of the two can extract rich and complete edge and detail information in the channel image. For example, in the landscape advertisement picture, the outline and texture details of the scenery such as mountains and rivers can be better restored. Taking the absolute value of the pixel values ​​of the third and fourth detail images prevents the problem of loss of detail information due to the mutual cancellation of positive and negative values ​​in the Laplace operation results. It ensures that the detail features of the image in different areas can be effectively retained, improves the reliability of detail enhancement, and allows the subtleties in the image to be clearly presented. The present invention adds the corresponding pixel values ​​of the two images after taking the absolute value, integrates the detail information obtained by the two 8-neighborhood operator operations, and further enhances the image details. The generated second channel detail enhanced image highlights the image contour.

[0074] In this embodiment, S3 includes the following sub-steps:

[0075] S31, calculating edge clarity of each region in the first channel detail enhanced image, and constructing a first edge clarity matrix;

[0076] S32, calculating edge clarity for each region in the second channel detail enhanced image, and constructing a second edge clarity matrix.

[0077] In this embodiment, the first channel detail enhanced image and the second channel detail enhanced image can be divided into Regions, the first channel detail enhanced image and the second channel detail enhanced image are divided into M-1 times horizontally and equidistantly, and the first channel detail enhanced image and the second channel detail enhanced image are divided into L-1 times vertically and equidistantly, and the sizes of the first edge definition matrix and the second edge definition matrix constructed are: .

[0078] In this embodiment, constructing the edge definition matrix in S31 and S32 includes the following steps:

[0079] A1. Pixels whose pixel values ​​in the channel detail enhanced image are greater than the pixel threshold are marked as edge points;

[0080] A2, dividing the channel detail enhanced image into multiple regions;

[0081] A3. In each region, connect each adjacent edge point in the horizontal direction to form multiple horizontal edge lines;

[0082] A4. In each region, connect each adjacent edge point in the vertical direction to form multiple vertical edge lines;

[0083] A5. Calculate edge definition of the region according to the plurality of horizontal edge lines and the plurality of vertical edge lines;

[0084] A6. Use the edge clarity of the region as an element to construct an edge clarity matrix.

[0085] The pixel threshold is specifically set according to experiments or experience. In this embodiment, the pixel threshold can be set to the pixel mean value in the channel detail enhanced image.

[0086] In this embodiment, the formula for calculating the edge definition of the area in A5 is: , where θ is the edge clarity of the region, N level,i is the number of edge points on the i-th horizontal edge line, N ver,i is the number of edge points on the i-th vertical edge line, i is a positive integer, L levelis the number of horizontal edge lines in a region, L ver is the number of vertical edge lines in a region.

[0087] The present invention marks pixel points whose pixel values ​​are greater than the pixel threshold as edge points, marking the edge position of the image. Steps A2-A4 further clearly measure the width of the edge lines in the horizontal and vertical directions by dividing the area and connecting adjacent edge points. The present invention takes into account the number of edge lines in the horizontal and vertical directions and the number of edge points on each line, and calculates the edge clarity of the area from multiple dimensions. In this way, the distribution of texture of the image in an area and the refinement of the texture can be fully reflected. The greater the edge clarity, the finer the texture, and the better the workmanship of the picture.

[0088] like Figure 2 As shown, the picture quality assessment model in S4 includes: a Y channel clarity feature fusion unit, a U channel clarity feature fusion unit, a V channel clarity feature fusion unit, a first convolution block, a second convolution block, a third convolution block, a fourth convolution block, a fifth convolution block, a sixth convolution block, a first fully connected layer, a second fully connected layer, and a third fully connected layer;

[0089] The input end of the Y channel definition feature fusion unit is used to input the first edge definition matrix and the second edge definition matrix corresponding to the Y channel image;

[0090] The input end of the U channel definition feature fusion unit is used to input the first edge definition matrix and the second edge definition matrix corresponding to the U channel image;

[0091] The input end of the V channel definition feature fusion unit is used to input the first edge definition matrix and the second edge definition matrix corresponding to the V channel image;

[0092] The input end of the first convolution block is connected to the output end of the Y channel clarity feature fusion unit, and its output end is connected to the input end of the fourth convolution block and the input end of the first fully connected layer respectively;

[0093] The input end of the second convolution block is connected to the output end of the U channel clarity feature fusion unit, and its output end is connected to the input end of the fifth convolution block and the input end of the first fully connected layer respectively;

[0094] The input end of the third convolution block is connected to the output end of the V channel clarity feature fusion unit, and its output end is connected to the input end of the sixth convolution block and the input end of the first fully connected layer respectively;

[0095] The input end of the second fully connected layer is connected to the output end of the fourth convolution block, the output end of the fifth convolution block, and the output end of the sixth convolution block respectively;

[0096] The input end of the third fully connected layer is connected to the output end of the first fully connected layer and the output end of the second fully connected layer respectively, and the output end of the third fully connected layer serves as the output end of the picture quality assessment model.

[0097] In the YUV color space, the Y channel represents brightness, and the U and V channels represent chrominance. Each channel definition feature fusion unit processes the features of the corresponding channel respectively, fuses the features corresponding to the first edge definition matrix and the second edge definition matrix of the channel, and then uses the first layer of convolution blocks: the first convolution block, the second convolution block, and the third convolution block to extract shallow features, and then uses the second convolution block: the fourth convolution block, the fifth convolution block, and the sixth convolution block to extract deep features. The first fully connected layer is used to integrate all shallow features, and the second fully connected layer is used to integrate all deep features, so as to achieve comprehensive shallow and deep features, obtain image quality scores, and improve the accuracy of image quality assessment.

[0098] In this embodiment, the expressions of the Y channel definition feature fusion unit, the U channel definition feature fusion unit and the V channel definition feature fusion unit are: , where R is the output of the Y channel definition feature fusion unit, the U channel definition feature fusion unit or the V channel definition feature fusion unit, and G 1 is the first edge definition matrix, G 2 is the second edge definition matrix, is element-wise addition, and Conv is a convolution operation.

[0099] In this embodiment, the convolution block includes: a convolution layer, a batch normalization layer and an activation function layer.

[0100] In this embodiment, the third fully connected layer is used to output the picture quality score.

[0101] The first edge clarity matrix and the second edge clarity matrix reflect the edge clarity information of the channel image from different angles. The features in the matrix are extracted through the convolution operation (Conv). Then, the features of the two matrices after convolution are fused by element addition operation to achieve feature complementarity.

[0102] The present invention converts an image into a YUV space, separates brightness and color, constructs a Y channel image through brightness Y, constructs a U channel image through blue chromaticity U, and constructs a V channel image through red chromaticity V. The edge clarity of the image is analyzed from three aspects of brightness Y, blue chromaticity U and red chromaticity V. The edge features of the image can be captured from multiple angles and in all directions, avoiding the limitation of single channel analysis.

[0103] The present invention performs multiple 4-neighborhood Laplace operator operations on each channel image, and performs multiple 8-neighborhood Laplace operator operations on each channel image. The 4-neighborhood Laplace operator mainly focuses on the relationship between the pixel point and its 4 neighboring pixels above, below, left and right, which can highlight the relatively small-scale details and edges in the image, and has a good enhancement effect on some subtle lines, textures and other details in the image. The 8-neighborhood Laplace operator takes into account the adjacent pixels in 8 directions around the pixel point, and can capture larger-scale image features and edge information, and has a better outlining effect on the contours of objects in the image, large-area edges, etc. By performing these two operations on each channel image at the same time, the detailed information of the image can be fully captured from different scales, so that the enhanced image retains rich subtle textures and highlights obvious edges and contours, improving the overall clarity and detail expression of the image.

[0104] The present invention calculates edge clarity for the first channel detail enhanced image and the second channel detail enhanced image respectively. After detail enhancement, the edge clarity of images in different channels reflects the edge features of the images under different attributes of brightness and chromaticity. Then, a picture quality assessment model is used to process the first edge clarity matrix and the second edge clarity matrix corresponding to the Y, U and V channel images to obtain a picture quality score. The picture quality score is obtained from multiple dimensions, thereby improving the assessment accuracy of picture quality.

[0105] The above are only 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 in the protection scope of the present invention.

Claims

1. An automatic screening method for advertising promotion AI pictures based on an online network platform, characterized in that: The following steps are involved: S1. Convert the image on the online network platform into YUV space to obtain multiple channel images, wherein the multiple channel images include: Y channel image, U channel image and V channel image; S2, performing a variety of 4-neighborhood and 8-neighborhood Laplacian operator operations on each channel image, respectively, to construct a first channel detail enhanced image and a second channel detail enhanced image; S3, respectively calculating the edge clarity of each region in the first channel detail enhanced image and the second channel detail enhanced image, and constructing a first edge clarity matrix and a second edge clarity matrix; S4, using the picture quality assessment model to process the first edge clarity matrix and the second edge clarity matrix corresponding to the Y, U and V channel images to obtain a picture quality score; S5. When the image quality score is greater than the score threshold, mark the image as recommended.

2. The method for automatically screening advertising promotion AI pictures based on an online network platform according to claim 1 is characterized in that: The 4-neighborhood Laplacian operator in S2 includes: standard 4-neighborhood Laplacian operator , the modified 4-neighborhood Laplacian ; 8-neighborhood Laplacian operators include: Standard 8-neighborhood Laplacian operator , gradient 8-neighborhood Laplacian .

3. The method for automatically screening advertising promotion AI pictures based on an online network platform according to claim 2 is characterized in that: The S2 comprises the following sub-steps: S21, performing a plurality of 4-neighborhood Laplacian operator operations on each channel image respectively, fusing the images after the operations to obtain a first channel detail enhanced image; S22, performing a plurality of 8-neighborhood Laplacian operator operations on each channel image respectively, fusing the images after the operations to obtain a second channel detail enhanced image.

4. The method for automatically screening advertising promotion AI pictures based on an online network platform according to claim 3 is characterized in that: The S21 comprises the following sub-steps: S211, performing standard 4-neighborhood Laplacian operator and modified 4-neighborhood Laplacian operator operations on each channel image respectively to obtain a first detail image and a second detail image; S212, taking the absolute value of each pixel value in the first detail image, and taking the absolute value of each pixel value in the second detail image; S213. Add the pixel values ​​of the first detail image after taking the absolute value and the second detail image after taking the absolute value at the same pixel points to obtain a first channel detail enhanced image. When the channel image in S211 is a Y channel image, the first channel detail enhanced image is a first Y channel detail enhanced image. When the channel image in S211 is a U channel image, the first channel detail enhanced image is a first U channel detail enhanced image. When the channel image in S211 is a V channel image, the first channel detail enhanced image is a first V channel detail enhanced image.

5. The method for automatically screening advertising promotion AI pictures based on an online network platform according to claim 3 is characterized in that: The S22 comprises the following sub-steps: S221, performing standard 8-neighborhood Laplacian operator and modified 8-neighborhood Laplacian operator operations on each channel image respectively to obtain a third detail image and a fourth detail image; S222, taking the absolute value of each pixel value in the third detail image, and taking the absolute value of each pixel value in the fourth detail image; S223. Add the pixel values ​​of the third detail image after taking the absolute value and the fourth detail image after taking the absolute value at the same pixel points to obtain a second channel detail enhanced image. When the channel image in S221 is a Y channel image, the second channel detail enhanced image is a second Y channel detail enhanced image. When the channel image in S221 is a U channel image, the second channel detail enhanced image is a second U channel detail enhanced image. When the channel image in S221 is a V channel image, the second channel detail enhanced image is a second V channel detail enhanced image.

6. The method for automatically screening advertising promotion AI pictures based on an online network platform according to claim 1, characterized in that: The S3 comprises the following sub-steps: S31, calculating edge clarity of each region in the first channel detail enhanced image, and constructing a first edge clarity matrix; S32, calculating edge clarity for each region in the second channel detail enhanced image, and constructing a second edge clarity matrix.

7. The method for automatically screening advertising promotion AI pictures based on an online network platform according to claim 6 is characterized in that: The edge definition matrix constructed in S31 and S32 includes the following steps: A1. Pixels whose pixel values ​​in the channel detail enhanced image are greater than the pixel threshold are marked as edge points; A2, dividing the channel detail enhanced image into multiple regions; A3. In each region, connect each adjacent edge point in the horizontal direction to form multiple horizontal edge lines; A4. In each region, connect each adjacent edge point in the vertical direction to form multiple vertical edge lines; A5. Calculate edge definition of the region according to the plurality of horizontal edge lines and the plurality of vertical edge lines; A6. Use the edge clarity of the region as an element to construct an edge clarity matrix.

8. The method for automatically screening advertising promotion AI pictures based on an online network platform according to claim 7 is characterized in that: The formula for calculating the edge definition of the area in A5 is: , where θ is the edge clarity of the region, N level,i is the number of edge points on the i-th horizontal edge line, N ver,i is the number of edge points on the i-th vertical edge line, i is a positive integer, L level is the number of horizontal edge lines in a region, L ver is the number of vertical edge lines in a region.

9. The method for automatically screening advertising promotion AI pictures based on an online network platform according to claim 1, characterized in that: The picture quality assessment model in S4 includes: a Y channel clarity feature fusion unit, a U channel clarity feature fusion unit, a V channel clarity feature fusion unit, a first convolution block, a second convolution block, a third convolution block, a fourth convolution block, a fifth convolution block, a sixth convolution block, a first fully connected layer, a second fully connected layer and a third fully connected layer; The input end of the Y channel definition feature fusion unit is used to input the first edge definition matrix and the second edge definition matrix corresponding to the Y channel image; The input end of the U channel definition feature fusion unit is used to input the first edge definition matrix and the second edge definition matrix corresponding to the U channel image; The input end of the V channel definition feature fusion unit is used to input a first edge definition matrix and a second edge definition matrix corresponding to the V channel image; The input end of the first convolution block is connected to the output end of the Y channel clarity feature fusion unit, and its output end is connected to the input end of the fourth convolution block and the input end of the first fully connected layer respectively; The input end of the second convolution block is connected to the output end of the U channel clarity feature fusion unit, and the output end thereof is respectively connected to the input end of the fifth convolution block and the input end of the first fully connected layer; The input end of the third convolution block is connected to the output end of the V channel clarity feature fusion unit, and its output end is connected to the input end of the sixth convolution block and the input end of the first fully connected layer respectively; The input end of the second fully connected layer is respectively connected to the output end of the fourth convolution block, the output end of the fifth convolution block, and the output end of the sixth convolution block; The input end of the third fully connected layer is connected to the output end of the first fully connected layer and the output end of the second fully connected layer respectively, and the output end thereof serves as the output end of the picture quality assessment model.

10. The method for automatically screening advertising promotion AI pictures based on an online network platform according to claim 9, characterized in that: The expressions of the Y channel definition feature fusion unit, the U channel definition feature fusion unit and the V channel definition feature fusion unit are: , where R is the output of the Y channel definition feature fusion unit, the U channel definition feature fusion unit or the V channel definition feature fusion unit, G1 is the first edge definition matrix, G2 is the second edge definition matrix, is element-wise addition, and Conv is a convolution operation.

Images