A high-speed logic analysis method for image processing chips

Abstract

The present invention relates to a high-speed logic analysis method for an image processing chip, and relates to the field of image processing technology. The method comprises: obtaining color tendency test data of multiple test users, and determining color tendency characteristics of the multiple test users; obtaining color tendency test data of a current user, and determining initial color tendency characteristics of the current user; obtaining an image to be adjusted; extracting image characteristics of the image to be adjusted; determining an optimal color adjustment scheme for the image to be adjusted according to the image characteristics of the image to be adjusted, the initial color tendency characteristics of the current user, and the color tendency characteristics of multiple test users; and performing color adjustment on the image to be adjusted according to the optimal color adjustment scheme to generate an adjusted image. The method has the advantage of realizing personalized image color adjustment according to the color tendency of the user.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a high-speed logic analysis method for an image processing chip. Background Art

[0002] An image processing chip is an integrated circuit chip specially used for processing image data. It has the advantages of fast processing speed, low energy consumption and high performance. Image processing chips are widely used in computer vision, artificial intelligence, medical imaging and other fields. Image processing chips have high-performance processing capabilities and can realize high-speed processing and analysis of images. By utilizing the ability of parallel computing, it can process multiple pixels at the same time and can realize the calculation of multiple complex algorithms. For example, in the field of computer vision, image processing chips can realize tasks such as image recognition, target detection, and image segmentation. Image processing chips can achieve low energy consumption and high efficiency processing. Due to its dedicated design and fine optimization, image processing chips can minimize energy consumption and improve energy utilization efficiency when processing image data. Image processing chips also have high stability and reliability. It has been rigorously tested and verified during the design and manufacturing process, and can run stably in various working environments and application scenarios without prone to failure.

[0003] Color balance is an important technology in image processing. The color balance processing of an image can correct the color cast, oversaturation or undersaturation in the image, and can also adjust the desired color tendency according to the user's needs and settings to achieve a better display effect. For example, white balance processing can be used to set the desired white color for the image, and blue light reduction processing can reduce the blue light component in the image. However, common color balance methods often use the same adjustment method for each pixel of the entire picture. For example, when it is desired to reduce the blue light, the blue component of all pixels is often reduced, which affects the color of the entire picture. This color balance processing method cannot obtain the best display result.

[0004] Therefore, it is necessary to provide a high-speed logic analysis method for an image processing chip to achieve personalized image color adjustment according to the user's color preference. Summary of the invention

[0005] The present invention provides a high-speed logic analysis method for an image processing chip, comprising: obtaining color tendency test data of multiple test users, and determining the color tendency characteristics of the multiple test users; obtaining color tendency test data of a current user, and determining the initial color tendency characteristics of the current user; obtaining an image to be adjusted; extracting image characteristics of the image to be adjusted; determining an optimal color adjustment scheme for the image to be adjusted according to the image characteristics of the image to be adjusted, the initial color tendency characteristics of the current user, and the color tendency characteristics of the multiple test users; and performing color adjustment on the image to be adjusted according to the optimal color adjustment scheme to generate an adjusted image.

[0006] Furthermore, color tendency test data of multiple test users are obtained, including: obtaining multiple test images, wherein there are differences in color distribution and entity combination between any two test images; dividing the multiple test images into multiple test image groups based on the color distribution and entity combination of any two test images; for each of the test image groups, repeatedly sampling test images from the test image group, and obtaining color tendency test data of the test user on the sampled test images until a preset sampling stop condition is met, wherein the color tendency test data of the test user includes the color tendency test data of all sampled test images of each of the test image groups by the test user.

[0007] Furthermore, the step of obtaining the color tendency of the sampled test image by the test user includes: S11, based on a preset color adjustment scheme, performing color adjustment on the sampled test image to generate a first adjusted test image; S12, obtaining the test user's selection feedback information on the sampled test image and the first adjusted test image; S13, performing color adjustment on the first adjusted test image according to the test user's selection feedback information on the sampled test image and the first adjusted test image to generate a second adjusted test image; S1 4. Obtain the test user's selection feedback information on the first adjusted test image and the second adjusted test image; S15. According to the test user's selection feedback information on the t-1th adjusted test image and the tth adjusted test image, perform color adjustment on the tth adjusted test image to generate the t+1th adjusted test image, where t is an integer greater than 1; S16. Set t=t+1; S17. Obtain the test user's selection feedback information on the t-1th adjusted test image and the tth adjusted test image until the preset color adjustment condition is met.

[0008] Furthermore, the color tendency characteristics of the multiple test users are determined, including: for each of the test users, based on the color tendency test data of all sampled test images of each test image group for the test user, determining the color tendency characteristics of the test user for multiple color distributions and entity combinations.

[0009] Furthermore, the color tendency test data of the current user is obtained to determine the initial color tendency characteristics of the current user, including: determining the color tendency differentiation parameters of each color distribution and entity combination according to the color tendency differentiation parameters of the test user for multiple color distributions and entity combinations; determining the key color distribution and entity combination according to the color tendency differentiation parameters of each color distribution and entity combination; obtaining the color tendency test data of the current user according to the key color distribution and entity combination; determining the initial color tendency characteristics of the current user according to the color tendency test data of the current user, wherein the initial color tendency characteristics of the current user include the color tendency characteristics of the current user for the key color distribution and key entities.

[0010] Furthermore, extracting image features of the image to be adjusted includes: extracting color distribution and entity combination of the image to be adjusted.

[0011] Furthermore, based on the image features of the image to be adjusted, the initial color tendency features of the current user and the color tendency features of the multiple test users, an optimal color adjustment scheme for the image to be adjusted is determined, including: determining a reference test user from the multiple test users based on the initial color tendency features of the current user and the color tendency features of the multiple test users; obtaining the color tendency features of the reference test user; and determining the optimal color adjustment scheme for the image to be adjusted based on the image features of the image to be adjusted and the color tendency features of the reference test user.

[0012] Furthermore, a reference test user is determined from the multiple test users based on the initial color tendency characteristics of the current user and the color tendency characteristics of the multiple test users, including: for each of the test users, calculating the color tendency similarity between the test user and the current user based on the initial color tendency characteristics of the current user and the color tendency characteristics of the test user for key color distribution and entity combination; determining multiple candidate test users based on the color tendency similarity between each test user and the current user; sorting the multiple candidate test users based on the color tendency similarity between each candidate test user and the current user to obtain a sorting result; and determining a reference test user from the multiple test users based on a preset reference quantity and the sorting result.

[0013] Furthermore, based on the image features of the image to be adjusted and the color tendency features of the reference test user, an optimal color adjustment scheme for the image to be adjusted is determined, including: for each of the reference test users, predicting the color tendency of the reference test user with respect to the color distribution and entity combination of the image to be adjusted based on the color distribution and entity combination of the image to be adjusted and the color tendency features of the reference test user; and determining the optimal color adjustment scheme for the image to be adjusted based on the color tendency of each of the reference test users with respect to the color distribution and entity combination of the image to be adjusted.

[0014] Furthermore, based on the color distribution and color tendency of the entity combination of the image to be adjusted by each of the reference test users, the optimal color adjustment scheme of the image to be adjusted is determined, including: generating the optimal color adjustment scheme of the image to be adjusted by using a color adjustment model based on the color distribution and color tendency of the entity combination of the image to be adjusted by each of the reference test users, wherein the optimal color adjustment scheme of the image to be adjusted includes at least gain factors of the three channels of red, green and blue, and a saturation adjustment factor and a brightness adjustment factor of at least one color.

[0015] Compared with the prior art, the high-speed logic analysis method for an image processing chip provided in this specification has at least the following beneficial effects:

[0016] 1. This method can perform personalized color adjustment according to the color tendency characteristics of the current user, so that the adjusted image is more in line with the user's color preference and enhances the user's visual experience. Through the high-speed logic analysis method, the image features can be quickly extracted, the color adjustment scheme can be determined, and the image can be color adjusted, thereby improving the efficiency of image processing. When determining the optimal color adjustment scheme, the method comprehensively considers the image characteristics and the user's color tendency characteristics, which can make the color distribution of the adjusted image more reasonable and beautiful. The color of the image after personalized color adjustment is more in line with the user's preferences, thereby enhancing the attractiveness of the image and making the image more visually impactful. By meeting the user's personalized needs for color, this method can improve the user's satisfaction with the image processing results, thereby improving the overall user experience.

[0017] 2. By using multiple test images with different color distributions and entity combinations, various possible colors and image features can be covered more comprehensively, so as to obtain the color tendency of the test user more accurately. By grouping the test images and sampling the test, combined with the user's selection feedback information, the user's real color preference can be gradually approached, and the accuracy of the color tendency test can be improved. According to the color tendency test data of each test user, its unique color tendency characteristics can be determined to provide a personalized color adjustment scheme for subsequent image processing. This method can take into account the differences in color preferences between different users, so that the image processing results are more in line with the user's personalized needs. Through the preset color adjustment scheme and the gradual adjustment process, the user's color preference can be efficiently explored and the number of unnecessary tests can be reduced. At the same time, the grouping and sampling methods also reduce the test cost and improve the test efficiency.

[0018] 3. This method can accurately obtain the color tendency characteristics of the current user for key color distribution and key entities, thereby providing a more personalized color adjustment scheme. By calculating the color tendency similarity between the test user and the current user and determining the reference test user, it can be further ensured that the color adjustment scheme meets the preferences of the current user. According to the color distribution and entity combination of the image to be adjusted, and the color tendency characteristics of the reference test user, the color adjustment scheme that best suits the current user can be predicted. This method can comprehensively consider image characteristics and user preferences to ensure that color adjustment is consistent with both image content and user aesthetics. The optimal color adjustment scheme is generated through the color adjustment model, which includes gain factors for the three channels of red, green, and blue, a saturation adjustment factor of at least one color, and a brightness adjustment factor, which can accurately adjust the image color. This fine color adjustment can significantly improve the color matching of the image, making the adjusted image more in line with the user's expectations. The personalized color adjustment scheme can improve the user's satisfaction with the image processing results, thereby enhancing the user experience. The method can also continuously optimize the color adjustment scheme based on user feedback to maintain long-term user satisfaction. Through the preset color adjustment scheme and the gradual adjustment process, the color adjustment scheme that best suits the current user can be efficiently explored. At the same time, the grouping and sampling methods also reduce testing costs and improve overall processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] This specification will be further described in the form of exemplary embodiments, which will be described in detail by the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same number represents the same structure, wherein:

[0020] Figure 1 is a flow chart of a high-speed logic analysis method for an image processing chip shown in an embodiment of the present application;

[0021] Figure 2 It is a flowchart of obtaining the color preference of a test user for a sampled test image shown in one embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following briefly introduces the drawings required for describing the embodiments.

[0023] Figure 1 is a flow chart of a high-speed logic analysis method for an image processing chip shown in an embodiment of the present application, such as Figure 1 As shown, a high-speed logic analysis method for an image processing chip may include the following steps.

[0024] Step 110, obtaining color tendency test data of multiple test users, and determining color tendency characteristics of the multiple test users.

[0025] Preferably, obtaining color tendency test data of multiple test users includes:

[0026] Obtain multiple test images, where the color distribution and entity combination of any two test images are different, for example, one test image is mainly blue, while the other test image is mainly warm tones (for example, red and yellow). The difference in color distribution can include the type of color, the proportion of each color, the gradient and contrast of colors, etc. Entities can include people, animals, plants, buildings, vehicles, etc. These objects usually have clear features such as shape, color, texture, etc.;

[0027] Based on the color distribution and entity combination of any two test images, the multiple test images are divided into multiple test image groups;

[0028] For each test image group, the process of sampling test images from the test image group is repeated to obtain the test user's color tendency test data for the sampled test images until a preset sampling stop condition is met, wherein the test user's color tendency test data includes the test user's color tendency test data for all sampled test images for each test image group.

[0029] Specifically, for any two test images, the cosine similarity of the color distribution of the two test images can be calculated as the color distribution similarity, the overlap of the entity combination of the two test images can be calculated as the entity combination similarity, and the color distribution similarity and the entity combination similarity are weighted and summed as the image similarity of the two test images. A clustering algorithm (for example, a K-means clustering algorithm) is used to divide multiple test images into multiple test image groups according to the image similarity of any two test images.

[0030] The preset sampling stop condition may be that the difference value of the color tendency of the test images sampled by the test user from the test image group is less than a preset difference value threshold or the number of the sampled test images is greater than a preset sampling number.

[0031] For example, the difference value of the color tendency of the test images sampled by the test user for the test image group can be calculated according to the following formula:

[0032] ;

[0033] in, is the difference value of the color tendency of the sampled test images of the f-th test image group by the e-th test user, is the color tendency of the i-th test image sampled by the e-th test user from the f-th test image group, is the color tendency of the jth test image sampled by the eth test user from the fth test image group, is the total number of sampled test images of the f-th test image group, is the cosine similarity between the color tendency of the i-th test image sampled by the e-th test user from the f-th test image group and the color tendency of the j-th test image sampled by the e-th test user from the f-th test image group.

[0034] Figure 2 FIG. 1 is a flowchart of obtaining the color preference of a test user for a sampled test image in an embodiment of the present application. Figure 2 As shown, as an example, obtaining the color preference of the test user for the sampled test image includes:

[0035] S11, based on a preset color adjustment scheme, performing color adjustment on the sampled test image to generate a first adjusted test image;

[0036] S12, obtaining the test user's selection feedback information on the sampled test image and the test image after the first adjustment;

[0037] S13, according to the test user's selection feedback information on the sampled test image and the first adjusted test image, color adjustment is performed on the first adjusted test image to generate a second adjusted test image;

[0038] S14, obtaining the test user's selection feedback information on the first adjusted test image and the second adjusted test image;

[0039] S15, performing color adjustment on the t-th adjusted test image according to the test user's selection feedback information on the t-1th adjusted test image and the tth adjusted test image to generate a t+1th adjusted test image, where t is an integer greater than 1;

[0040] S16, let t=t+1;

[0041] S17, obtaining the test user's selection feedback information on the test image after the t-1th adjustment and the test image after the tth adjustment, until the preset color adjustment condition is met.

[0042] Specifically, the preset color adjustment scheme may include preset gain factors of three channels, namely, red, green and blue, and a saturation adjustment factor and a brightness adjustment factor of at least one color.

[0043] A test multiple-choice question can be generated based on the sampled test image and the corresponding test image after the first adjustment, instructing the test user to select the image with better appearance between the sampled test image and the corresponding test image after the first adjustment, and the selection feedback information may include the image with better appearance selected by the test user and the time taken by the test user to make the selection.

[0044] It is necessary to analyze the selection result of the test user, that is, whether the test user selected the sampled test image or the first adjusted test image as the image with better visual perception. At the same time, the time taken by the test user to make the selection can also be considered. Although this is not a factor that directly determines the color adjustment, it may reflect the user's sensitivity and preference for the difference between the two images. According to the test user's selection, the direction of color adjustment can be determined. For example, if the test user selects the first adjusted test image, it means that the current color adjustment scheme (including the gain factor, saturation adjustment factor and brightness adjustment factor of the three channels of red, green and blue) is moving in the right direction. If the test user selects the sampled test image, the direction or amplitude of the color adjustment scheme may need to be adjusted. According to the determined color adjustment direction, the color adjustment parameters of the first adjusted test image are adjusted. This includes adjusting the gain factors of the three channels of red, green and blue, and the saturation adjustment factor and brightness adjustment factor of at least one color. The amplitude of the adjustment can be determined according to the intensity of the test user's selection feedback information. For example, if the test user clearly selects an image in a certain direction, the color adjustment parameters may need to be adjusted significantly; if the test user is hesitant or takes a long time to select, the color adjustment parameters may need to be adjusted slightly. Using the adjusted color adjustment parameters, the color of the test image after the first adjustment is adjusted again to generate a test image after the second adjustment.

[0045] The image generation model can be used to perform color adjustment on the first adjusted test image based on the test user's selection feedback information on the sampled test image and the first adjusted test image to generate a second adjusted test image, wherein the image generation model can be a Generative Adversarial Networks (GAN) model.

[0046] The test user's selection feedback information on the first adjusted test image and the second adjusted test image is similar to the test user's selection feedback information on the sampled test image and the first adjusted test image, and will not be repeated here.

[0047] The preset color adjustment condition may be that the time taken by the test user to make selections corresponding to the continuous multiple-choice test questions is greater than a time threshold.

[0048] Preferably, determining the color tendency characteristics of multiple test users includes:

[0049] For each test user, the color tendency characteristics of the test user for multiple color distributions and entity combinations are determined based on the color tendency test data of all sampled test images of each test image group by the test user. The color tendency characteristics of the test user for multiple color distributions and entity combinations may include the gain factors of the three channels of red, green and blue, and the saturation adjustment factor and brightness adjustment factor of at least one color of the test user for each color distribution and entity combination.

[0050] The color tendency characteristics of the test user for multiple color distributions and entity combinations can be determined through a first feature extraction model based on the color tendency test data of all sampled test images of each test image group by the test user, wherein the first feature extraction model can be a Convolutional Neural Network (CNN) model.

[0051] Step 120, obtaining the color tendency test data of the current user, and determining the initial color tendency characteristics of the current user.

[0052] Specifically include:

[0053] According to the color tendency characteristics of the test users for the multiple color distributions and entity combinations, the color tendency differentiation parameters of each color distribution and entity combination are determined;

[0054] Determine the key color distribution and entity combination according to the color tendency differentiation parameters of each color distribution and entity combination;

[0055] Obtain the color preference test data of the current user based on key color distribution and entity combination;

[0056] According to the color tendency test data of the current user, the initial color tendency characteristics of the current user are determined, wherein the initial color tendency characteristics of the current user include the color tendency characteristics of the current user for key color distribution and key entities.

[0057] Specifically, the color tendency differentiation parameter can be calculated according to the following formula:

[0058] ;

[0059] in, is the color tendency differentiation parameter of the kth color distribution and entity combination, is the color preference of the g-th test user for the k-th color distribution and entity combination, is the color preference of the hth test user for the kth color distribution and entity combination, is the total number of test users, It is the cosine similarity between the color tendency of the g-th test user for the k-th color distribution and entity combination and the color tendency of the h-th test user for the k-th color distribution and entity combination.

[0060] For example, a color distribution and entity combination whose color tendency differentiation parameter is greater than a color tendency differentiation parameter threshold value may be used as a key color distribution and entity combination.

[0061] The initial color tendency feature of the current user may include gain factors of the three channels of red, green and blue for each key color distribution and entity combination, and a saturation adjustment factor and a brightness adjustment factor of at least one color of the current user.

[0062] The initial color tendency feature of the current user may be determined according to the color tendency test data of the current user through a second feature extraction model, wherein the second feature extraction model may be a convolutional neural network (CNN) model.

[0063] Step 130: Obtain the image to be adjusted.

[0064] Step 140: extracting image features of the image to be adjusted.

[0065] Specifically include:

[0066] Extract the color distribution and entity combination of the image to be adjusted.

[0067] Step 150 , determining an optimal color adjustment scheme for the image to be adjusted according to the image characteristics of the image to be adjusted, the initial color tendency characteristics of the current user, and the color tendency characteristics of multiple test users.

[0068] Specifically include:

[0069] Determining a reference test user from the multiple test users according to the initial color tendency characteristics of the current user and the color tendency characteristics of the multiple test users;

[0070] Obtain the color preference characteristics of the reference test user;

[0071] An optimal color adjustment scheme for the image to be adjusted is determined based on the image features of the image to be adjusted and the color tendency features of the reference test user.

[0072] Preferably, determining a reference test user from a plurality of test users according to the initial color tendency feature of the current user and the color tendency features of a plurality of test users comprises:

[0073] For each test user, the color tendency similarity between the test user and the current user is calculated based on the initial color tendency characteristics of the current user and the color tendency characteristics of the test user for key color distribution and entity combination;

[0074] Determine multiple candidate test users according to the color tendency similarity between each test user and the current user, for example, determine the test users whose color tendency similarity is greater than the color tendency similarity threshold as candidate test users;

[0075] According to the color tendency similarity between each candidate test user and the current user, multiple candidate test users are sorted to obtain a sorting result. Specifically, the multiple candidate test users are sorted from large to small according to the color tendency similarity;

[0076] According to the preset reference quantity and the ranking result, a reference test user is determined from multiple test users. For example, according to the preset reference quantity, a test user with a higher ranking is selected as a reference test user.

[0077] Specifically, the color tendency similarity between the candidate test user and the current user can be calculated according to the following formula:

[0078] ;

[0079] in, is the color tendency similarity between the i-th candidate test user and the current user, is the color preference feature of the i-th candidate test user for key color distribution and entity combination, The color preference characteristics of the current user for key color distribution and entity combination, is the cosine similarity between the color tendency characteristics of the key color distribution and entity combination of the i-th candidate test user and the color tendency characteristics of the key color distribution and entity combination of the current user, is the total number of key color distributions and entity combinations.

[0080] Preferably, determining the optimal color adjustment scheme for the image to be adjusted according to the image features of the image to be adjusted and the color tendency features of the reference test user includes:

[0081] For each reference test user, predicting the color tendency of the reference test user with respect to the color distribution and entity combination of the image to be adjusted according to the color distribution and entity combination of the image to be adjusted and the color tendency characteristics of the reference test user;

[0082] According to each reference test user's color distribution and color tendency of entity combination of the image to be adjusted, the optimal color adjustment scheme of the image to be adjusted is determined.

[0083] Specifically, the color tendency of the reference test user with respect to the color distribution and entity combination of the image to be adjusted can be predicted by a tendency prediction model based on the color distribution and entity combination of the image to be adjusted and the color tendency characteristics of the reference test user, wherein the tendency prediction model can be a Convolutional Neural Network (CNN) model.

[0084] Preferably, according to the color distribution and color tendency of entity combination of each reference test user on the image to be adjusted, determining the optimal color adjustment scheme of the image to be adjusted includes:

[0085] The color adjustment model is used to generate an optimal color adjustment scheme for the image to be adjusted according to the color distribution and color tendency of the entity combination of each reference test user for the image to be adjusted, wherein the optimal color adjustment scheme for the image to be adjusted includes at least gain factors of the three channels of red, green and blue, and a saturation adjustment factor and a brightness adjustment factor of at least one color, wherein the color adjustment model can be a convolutional neural network (CNN) model.

[0086] Step 160 , performing color adjustment on the image to be adjusted according to the optimal color adjustment solution to generate an adjusted image.

[0087] Specifically, the image to be adjusted can be color adjusted according to the optimal color adjustment scheme based on the image generation model to generate an adjusted image, wherein the image generation model can be a Generative Adversarial Networks (GAN) model.

[0088] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A high-speed logic analysis method for an image processing chip, characterized in that: include: Acquire color tendency test data of multiple test users, and determine color tendency characteristics of the multiple test users; Acquire color tendency test data of the current user to determine the initial color tendency characteristics of the current user; Obtain the image to be adjusted; Extracting image features of the image to be adjusted; Determining an optimal color adjustment scheme for the image to be adjusted according to the image features of the image to be adjusted, the initial color tendency features of the current user, and the color tendency features of the multiple test users; According to the optimal color adjustment scheme, color adjustment is performed on the image to be adjusted to generate an adjusted image; Wherein, obtaining the color tendency test data of the current user and determining the initial color tendency characteristics of the current user include: Determining a color tendency differentiation parameter of each color distribution and entity combination according to the color tendency characteristics of the test user for the plurality of color distributions and entity combinations; Determine the key color distribution and entity combination according to the color tendency differentiation parameters of each color distribution and entity combination; According to the key color distribution and entity combination, obtaining the color tendency test data of the current user; According to the color tendency test data of the current user, the initial color tendency characteristics of the current user are determined, wherein the initial color tendency characteristics of the current user include the color tendency characteristics of the current user for key color distribution and key entities.

2. A high-speed logic analysis method for an image processing chip according to claim 1, characterized in that: Obtain color preference test data from multiple test users, including: Acquire multiple test images, wherein color distribution and entity combination of any two test images are different; Based on the color distribution and entity combination of any two test images, the plurality of test images are divided into a plurality of test image groups; For each of the test image groups, the process of sampling test images from the test image group is repeated to obtain the test user's color tendency test data for the sampled test images until a preset sampling stop condition is met, wherein the test user's color tendency test data includes the test user's color tendency test data for all sampled test images for each of the test image groups.

3. A high-speed logic analysis method for an image processing chip according to claim 2, characterized in that: The step of obtaining the color preference of the test user for the sampled test image includes: S11, based on a preset color adjustment scheme, performing color adjustment on the sampled test image to generate a first adjusted test image; S12, obtaining selection feedback information of the test user on the sampled test image and the first adjusted test image; S13, performing color adjustment on the first adjusted test image according to the test user's selection feedback information on the sampled test image and the first adjusted test image to generate a second adjusted test image; S14, obtaining selection feedback information of the test user on the first adjusted test image and the second adjusted test image; S15, performing color adjustment on the t-th adjusted test image according to the test user's selection feedback information on the t-1th adjusted test image and the tth adjusted test image to generate a t+1th adjusted test image, where t is an integer greater than 1; S16, let t=t+1; S17, obtaining the test user's selection feedback information on the test image after the t-1th adjustment and the test image after the tth adjustment, until the preset color adjustment condition is met.

4. A high-speed logic analysis method for an image processing chip according to claim 2 or 3, characterized in that: Determining the color tendency characteristics of the multiple test users includes: For each of the test users, the color tendency characteristics of the test user for a plurality of color distributions and entity combinations are determined based on the color tendency test data of the test user for all sampled test images of each test image group.

5. A high-speed logic analysis method for an image processing chip according to claim 4, characterized in that: Extracting image features of the image to be adjusted includes: The color distribution and entity combination of the image to be adjusted are extracted.

6. A high-speed logic analysis method for an image processing chip according to claim 5, characterized in that: Determining an optimal color adjustment scheme for the image to be adjusted according to the image features of the image to be adjusted, the initial color tendency features of the current user, and the color tendency features of the multiple test users, includes: Determining a reference test user from the multiple test users according to the initial color tendency feature of the current user and the color tendency features of the multiple test users; Obtaining the color tendency characteristics of the reference test user; An optimal color adjustment scheme for the image to be adjusted is determined according to the image features of the image to be adjusted and the color tendency features of the reference test user.

7. A high-speed logic analysis method for an image processing chip according to claim 6, characterized in that: Determining a reference test user from the multiple test users according to the initial color tendency feature of the current user and the color tendency features of the multiple test users includes: For each of the test users, calculating the color tendency similarity between the test user and the current user according to the initial color tendency characteristics of the current user and the color tendency characteristics of the test user for key color distribution and entity combination; Determine multiple candidate test users based on the color tendency similarity between each test user and the current user; Sorting the plurality of candidate test users according to the color tendency similarity between each candidate test user and the current user, and obtaining a sorting result; According to a preset reference quantity and the ranking result, a reference test user is determined from the multiple test users.

8. A high-speed logic analysis method for an image processing chip according to claim 7, characterized in that: Determining an optimal color adjustment scheme for the image to be adjusted according to the image features of the image to be adjusted and the color tendency features of the reference test user includes: For each of the reference test users, predicting the color tendency of the reference test user for the color distribution and entity combination of the image to be adjusted based on the color distribution and entity combination of the image to be adjusted and the color tendency characteristics of the reference test user; An optimal color adjustment scheme for the image to be adjusted is determined according to the color distribution and color tendency of the entity combination of the image to be adjusted by each reference test user.

9. A high-speed logic analysis method for an image processing chip according to claim 8, characterized in that: Determining the optimal color adjustment scheme for the image to be adjusted according to the color distribution and the color tendency of the entity combination of the image to be adjusted by each of the reference test users, including: An optimal color adjustment scheme for the image to be adjusted is generated through a color adjustment model according to the color distribution and color tendency of the entity combination of each reference test user for the image to be adjusted, wherein the optimal color adjustment scheme for the image to be adjusted includes at least gain factors for the three channels of red, green and blue, and a saturation adjustment factor and a brightness adjustment factor for at least one color.

Images