CN115661447B - A product image adjustment method based on big data - Google Patents

Abstract

The invention relates to a product image adjusting method based on big data, which comprises the following steps: the information acquisition module of the intelligent electronic commerce platform sends a plurality of display images corresponding to the target product to each user terminal, and acquires image feedback information of each test user; the interest recognition module analyzes the interest degree of each image characteristic region and the region information quantity of the corresponding image characteristic region according to all the test users to obtain the visual priority of each image characteristic region; the parameter enhancement module determines the region processing parameters of each visual key region in the corresponding display image according to the first parameter interval, the second parameter interval and the visual priority of each visual key region; and the image processing module enhances the contrast of each visual key area in the corresponding display image according to the area processing parameters so as to obtain an effect enhanced image of the corresponding display image.

Description

A product image adjustment method based on big data

technical field

The invention relates to the field of e-commerce, in particular to a method for adjusting product images based on big data.

Background technique

E-commerce usually refers to a wide range of commercial and trade activities around the world, under the open network environment of the Internet, based on the client/server application mode, buyers and sellers conduct various business activities without meeting each other, so as to realize consumers' online shopping, It is a new type of business operation mode for online transactions and online electronic payments between merchants, as well as various business activities, transaction activities, financial activities and related comprehensive service activities.

As human beings have higher and higher requirements for the accuracy of information transmission, images, as a relatively rich and effective carrier for information transmission, have become an indispensable element in our lives. Although the human eye has a very wide field of vision when observing images, its attention The concentration range is very small, and more time is invested in the attention point. Compared with other areas, this part of the area has a greater impact on the image quality.

The scene content contained in the traditional product display image is too complicated, and the display effect of the product display image is not ideal, which makes the user's attention more distracted. The product display image is difficult to highlight the product details that consumers are more interested in, and it is difficult to attract the user's attention .

Contents of the invention

Aiming at the deficiencies of the prior art, the present invention provides a product image adjustment method based on big data, including:

The information acquisition module of the smart e-commerce platform sends several display images corresponding to the target product to the user terminals of each test user in response to the image optimization request sent by the management terminal, and obtains image feedback information of each test user for different display images;

The interest recognition module obtains the degree of interest of the test user in each image feature area in the corresponding test image according to the image feedback information analysis of the test user, and according to the interest degree of all test users in each image feature area and the area information of the corresponding image feature area Quantitative analysis to obtain the visual priority of each image feature region;

The parameter enhancement module identifies the visual key area in the corresponding display image according to the visual priority of each image feature area, and determines each of the corresponding display images according to the first parameter interval, the second parameter interval and the visual priority of each visual key area. Region processing parameters for visually critical regions;

The image processing module enhances the contrast of each key visual area in the corresponding display image according to the region processing parameters to obtain an effect-enhanced image of the corresponding display image.

According to a preferred embodiment, the image optimization request includes a device identifier, a product number, several display images of the target product, and format information of the display images; the display images are used to display the shape and structure of the target product.

According to a preferred embodiment, the image feedback information is used to characterize the gaze information of the test user on different image regions of the display image, which includes the position information of the gaze point of the image, the dwell time, the test user's gaze scanning path and the number of eye saccades.

According to a preferred embodiment, the interest identification module obtains the degree of interest of the test user in each image feature region in the corresponding test image according to the test user's image feedback information analysis including:

The interest identification module analyzes the human eye movement state of the test user according to the image feedback information of the test user to obtain the eye movement characteristics of the corresponding test user, and obtains the degree of interest of the test user to each image feature area according to the analysis of the eye movement characteristics , wherein the eye movement feature is used to characterize the corresponding test user's dwell time, eye saccade data, and scanning trajectory for each image feature area in the corresponding display image.

According to a preferred embodiment, the interest recognition module obtains the visual priority of each image feature area according to the degree of interest of all test users in each image feature area and the area information volume of the corresponding image feature area:

The interest identification module determines the area weight value of the corresponding image feature area according to the area information amount corresponding to each image feature area, wherein the area information amount is used to represent the number of product features of the target product contained in the corresponding image feature area;

The interest recognition module weights and fuses the different interests of all test users corresponding to the same image feature area according to the area weight value of each image feature area to obtain the visual priority of the corresponding image feature area.

According to a preferred embodiment, the parameter enhancement module determines the area processing parameters corresponding to each visual key area in the display image according to the first parameter interval, the second parameter interval and the visual priority of each visual key area, including:

The parameter enhancement module obtains the pixel features of each visual key area to compare it with the global pixel features of the corresponding display image to obtain the first regional difference feature of each visual key area, and according to the first regional difference of each visual key area The first feature difference degree analysis corresponding to the feature obtains the first parameter interval corresponding to the key visual area;

The parameter enhancement module compares the pixel features of each visual critical area with the pixel features of its related adjacent image feature areas to obtain the second regional difference feature of each visual critical area, and according to the second regional difference of each visual critical area The second feature difference analysis corresponding to the feature obtains the second parameter interval corresponding to the key visual area;

The parameter enhancement module determines, according to the first parameter interval and the second parameter interval of each visual key area, the weight coefficient of each pixel corresponding to the visual key area in the display image whose visual priority is greater than the preset priority threshold, and according to each The pixel value of the pixel point and the weight coefficient are fused to obtain the area processing parameters corresponding to the key visual area.

According to a preferred embodiment, the analysis of the second feature difference degree corresponding to the second regional difference feature of each visual key area to obtain the second parameter interval corresponding to the visual key area includes:

The parameter enhancement module compares the difference of each feature value in the second region difference feature corresponding to each visual key area with the second difference threshold to obtain the second feature between each visual key area and its associated adjacent image feature area degree of difference, wherein the second degree of feature difference is used to characterize the local chromaticity contrast and local luminance contrast between each visual critical area and its associated adjacent image feature area;

The parameter enhancement module determines the transformable range of each feature value difference in the second area difference feature according to the minimum visual difference of the human eye and the second feature difference degree of the corresponding key visual area to obtain the second parameter interval corresponding to the key visual area.

According to a preferred embodiment, the parameter enhancement module compares the pixel features of each key visual area with the pixel features of its associated adjacent image feature areas to obtain the second regional difference features of each key visual area, including:

The parameter enhancement module establishes a corresponding first key feature matrix for the corresponding visual key area according to the pixel features of each visual key area, and establishes a corresponding second key feature matrix according to the pixel features of each adjacent image feature area related to it;

The parameter enhancement module obtains the first key neighborhood entropy of each visual key area according to the matrix variance of the first key feature matrix corresponding to each visual key area and the first matrix neighborhood entropy corresponding to each first key feature matrix, According to the matrix variance of each adjacent image feature area corresponding to each visual key area corresponding to the second key feature matrix and the second matrix neighborhood entropy analysis corresponding to each second key feature matrix, the second value of each adjacent image feature area is obtained. key neighborhood entropy, wherein the first matrix neighborhood entropy is used to characterize the weight coefficients of each feature vector in the first key feature matrix;

The parameter enhancement module projects the feature component of the first key neighborhood entropy of each visual key area and the feature component of the second key neighborhood entropy of each adjacent image feature area related to it to the feature subspace of different scales to obtain A number of eigenvalue differences between each visual critical area and its associated adjacent image feature areas, and generate a second regional difference feature corresponding to the visual key area according to the several eigenvalue differences, wherein the second regional difference feature is used It is used to characterize the pixel average difference and gray-scale aggregation difference between each visual key area and its related adjacent image feature areas.

According to a preferred embodiment, the first parameter interval is used to characterize the enhancement range of the pixel value of each key visual area compared with the corresponding display image; the second parameter interval is used to characterize the range of each key visual area compared to The range of pixel values associated with adjacent image feature regions may be enhanced.

According to a preferred embodiment, the minimum visual difference is the prior knowledge of the human body stored in the system, which represents the minimum pixel value difference perceivable by human eyes.

The present invention has the following beneficial effects:

The product image adjustment method based on big data provided by the present invention obtains the degree of interest of each test user in different image regions in the product display images by acquiring the eye movement characteristics of the test users when they watch different display images of the product, and according to each According to the analysis of the interest degree of each test user, the key visual areas in the corresponding display images are obtained, and then the contrast of each key visual area is enhanced according to the area processing parameters of each key visual area. That is to say, the present invention significantly improves the display effect of product images by identifying the visual interest areas of many users in the product images and enhancing the contrast of the visual interest regions, which is conducive to product images highlighting product details that consumers are more interested in and improving consumption. consumers' desire to purchase.

Description of drawings

Fig. 1 is a flow chart of a method for adjusting a product image based on big data provided by an exemplary embodiment.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in combination with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are exemplary only, and are not intended to limit the scope of the present invention. Also, in the following description, descriptions of well-known structures and techniques are omitted to avoid unnecessarily obscuring the concept of the present invention.

The terminology used in the present invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used herein and in the appended claims, the singular forms "a", "the", and "the" are intended to include the plural forms as well, unless the context clearly dictates otherwise. It should also be understood that the term "and/or" as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

It should be understood that although the terms first, second, third, etc. may be used in the present invention to describe various information, the information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of the present invention, first information may also be called second information, and similarly, second information may also be called first information. Depending on the context, the word "if" as used herein may be interpreted as "at" or "when" or "in response to a determination."

Referring to Fig. 1, in one embodiment, the product image adjustment method based on big data may include:

S1. The information acquisition module of the smart e-commerce platform sends several display images corresponding to the target product to the user terminals of each test user in response to the image optimization request sent by the management terminal, and obtains image feedback information of each test user on different display images.

Optionally, the image optimization request includes a device identifier, a product number, several display images of the target product, and format information of the display images; the display images are used to display the shape and structure of the target product.

Optionally, the device identifier is used to uniquely identify the management terminal; the product number is used to uniquely identify the product; and the format information is used to represent the compression format and decoding format of the corresponding display image.

Optionally, the image feedback signal is used to characterize the gaze information of the test user on different image regions of the display image, which includes the position information of the gaze point of the image, the duration of stay, the test user's line of sight scanning path and the number of eye saccades, the gaze information It is obtained by the real-time collection of the eye movement behavior of the rehabilitation user by the corresponding user terminal through its external camera device.

Optionally, the management terminal is a device with computing function, storage function and communication function used by product sellers, which includes: smart phones, desktop computers and notebook computers.

S2. The interest identification module obtains the degree of interest of the test user in each image feature area in the corresponding test image according to the image feedback information analysis of the test user, and according to the interest of all test users in each image feature area and the corresponding image feature area The visual priority of each image feature region is obtained by analyzing the amount of regional information.

Specifically, the interest identification module obtains the degree of interest of the test user in each image feature area in the corresponding test image according to the image feedback information analysis of the test user, including:

The interest identification module analyzes the human eye movement state of the test user according to the image feedback information of the test user to obtain the eye movement characteristics of the corresponding test user, and obtains the degree of interest of the test user to each image feature area according to the analysis of the eye movement characteristics , wherein the eye movement feature is used to characterize the corresponding test user's dwell time, eye saccade data, and scanning trajectory for each image feature area in the corresponding display image.

Specifically, the interest recognition module obtains the visual priority of each image feature area according to the interest degree of each test user to each image feature area and the area information amount of the corresponding image feature area:

The interest identification module determines the area weight value of the corresponding image feature area according to the area information amount corresponding to each image feature area, wherein the area information amount is used to represent the number of product features of the target product contained in the corresponding image feature area;

The interest recognition module weights and fuses the different interests of all test users corresponding to the same image feature area according to the area weight value of each image feature area to obtain the visual priority of the corresponding image feature area.

Optionally, the area weight value is used to represent the importance of the corresponding image feature area, that is, the more product features the image feature area contains, the larger the area weight value it occupies. Optionally, the visual priority is used to characterize the degree of attraction of the corresponding image feature area to human eyesight, that is, the greater the visual priority, the greater the degree of user attention to the corresponding image feature area.

S3. The parameter enhancement module identifies the visual key area in the corresponding display image according to the visual priority of each image feature area, and determines the corresponding display image according to the first parameter interval, the second parameter interval and the visual priority of each visual key area. Region processing parameters for each visually critical region in .

Optionally, the region processing parameters are used to adjust the pixels of each pixel in the corresponding visual critical region.

Specifically, the parameter enhancement module determines, according to the first parameter interval, the second parameter interval and the visual priority of each visual key area, the area processing parameters corresponding to each visual key area in the display image, including:

The parameter enhancement module obtains the pixel features of each visual key area to compare it with the global pixel features of the corresponding display image to obtain the first regional difference feature of each visual key area, and according to the first regional difference of each visual key area The first feature difference analysis corresponding to the feature obtains the first parameter interval corresponding to the key visual area, wherein the first area difference feature is used to characterize the pixel average difference between each key visual area and the corresponding display image and Gray-scale aggregation difference; the first feature difference is used to characterize the global chromaticity contrast and global brightness contrast between each key visual area and the corresponding display image;

The parameter enhancement module compares the pixel features of each visual critical area with the pixel features of its related adjacent image feature areas to obtain the second regional difference feature of each visual critical area, and according to the second regional difference of each visual critical area The second feature difference analysis corresponding to the feature obtains the second parameter interval corresponding to the key visual area;

The parameter enhancement module determines, according to the first parameter interval and the second parameter interval of each visual key area, the weight coefficient of each pixel corresponding to the visual key area in the display image whose visual priority is greater than the preset priority threshold, and according to each The pixel value of the pixel point and the weight coefficient are fused to obtain the area processing parameters corresponding to the key visual area.

Optionally, the preset priority threshold is a value preset by the system for judging whether the corresponding image characteristic area is more attractive to human sight.

Specifically, according to the second feature difference analysis corresponding to the second regional difference feature of each visual key area, the second parameter interval of the corresponding visual key area includes:

The parameter enhancement module compares the difference of each feature value in the second region difference feature corresponding to each visual key area with the second difference threshold to obtain the second feature between each visual key area and its associated adjacent image feature area degree of difference, wherein the second degree of feature difference is used to characterize the local chromaticity contrast and local luminance contrast between each visual critical area and its associated adjacent image feature area;

The parameter enhancement module determines the transformable range of each feature value difference in the second area difference feature according to the minimum visual difference of the human eye and the second feature difference degree of the corresponding key visual area to obtain the second parameter interval corresponding to the key visual area.

Optionally, the minimum visual difference is human prior knowledge pre-stored in the system, which represents the minimum pixel value difference perceivable by human eyes.

Specifically, the parameter enhancement module compares the pixel features of each visual critical area with the pixel features of its related adjacent image feature areas to obtain the second regional difference features of each visual key area, including:

The parameter enhancement module establishes a corresponding first key feature matrix for the corresponding visual key area according to the pixel features of each visual key area, and establishes a corresponding second key feature matrix according to the pixel features of each adjacent image feature area related to it;

The parameter enhancement module obtains the first key neighborhood entropy of each visual key area according to the matrix variance of the first key feature matrix corresponding to each visual key area and the first matrix neighborhood entropy corresponding to each first key feature matrix, According to the matrix variance of each adjacent image feature area corresponding to each visual key area corresponding to the second key feature matrix and the second matrix neighborhood entropy analysis corresponding to each second key feature matrix, the second value of each adjacent image feature area is obtained. key neighborhood entropy, wherein the first matrix neighborhood entropy is used to characterize the weight coefficients of each feature vector in the first key feature matrix;

The parameter enhancement module projects the feature component of the first key neighborhood entropy of each visual key area and the feature component of the second key neighborhood entropy of each adjacent image feature area related to it to the feature subspace of different scales to obtain A number of eigenvalue differences between each visual critical area and its associated adjacent image feature areas, and generate a second regional difference feature corresponding to the visual key area according to the several eigenvalue differences, wherein the second regional difference feature is used It is used to characterize the pixel average difference and gray-scale aggregation difference between each visual key area and its related adjacent image feature areas.

Optionally, the first parameter interval is used to characterize the enhancement range of the pixel value of each visual key area compared with the corresponding display image; the second parameter interval is used to characterize the range of each visual key area compared to its related The range of pixel values of adjacent image feature regions can be enhanced.

Optionally, the second matrix neighborhood entropy is used to characterize the weight coefficients of each feature vector in the second key feature matrix; the first key neighborhood entropy is used to characterize the pixel discretization of each pixel point in the corresponding visual key area degree; the second key neighborhood entropy is used to characterize the pixel dispersion of each pixel in the adjacent image feature area corresponding to the visual key area.

S4. The image processing module enhances the contrast of each key visual area in the corresponding display image according to the region processing parameters to obtain an effect-enhanced image of the corresponding display image.

The product image adjustment method based on big data provided by the present invention obtains the degree of interest of each test user in different image regions in the product display images by acquiring the eye movement characteristics of the test users when they watch different display images of the product, and according to each According to the analysis of the interest degree of each test user, the key visual areas in the corresponding display images are obtained, and then the contrast of each key visual area is enhanced according to the area processing parameters of each key visual area. That is to say, the present invention significantly improves the display effect of product images by identifying the visual interest areas of many users in the product images and enhancing the contrast of the visual interest regions, which is conducive to product images highlighting product details that consumers are more interested in and improving consumption. consumers' desire to purchase.

In one embodiment, the big data-based product image adjustment system for implementing the method of the present invention includes a management terminal, a user terminal and a smart e-commerce platform. The smart e-commerce platform has communication connections with the management terminal and the user terminal respectively. The user terminal is a device with computing function, storage function and communication function used by product consumers, which includes: smart phones, desktop computers and notebook computers.

The smart e-commerce platform includes an information acquisition module, an interest identification module, a parameter enhancement module and an image processing module.

The information acquisition module is used to send several display images corresponding to the target product to the user terminals of each test user in response to the image optimization request sent by the management terminal, and obtain image feedback information of each test user for different display images.

The interest recognition module is used to obtain the degree of interest of the test user in each image feature region in the corresponding test image according to the image feedback information analysis of the test user, and according to the interest degree of all test users to each image feature region and the corresponding image feature region The visual priority of each image feature region is obtained by analyzing the amount of regional information.

The parameter enhancement module is used to identify the visual key area in the corresponding display image according to the visual priority of each image feature area, and determine the corresponding display image according to the first parameter interval, the second parameter interval and the visual priority of each visual key area Region processing parameters for each visually critical region in .

The image processing module is configured to enhance the contrast of each key visual region in the corresponding display image according to the region processing parameters to obtain an effect-enhanced image of the corresponding display image.

Additionally, while specific functions are discussed above with reference to specific modules, it should be noted that the functionality of various modules discussed herein may be divided into multiple modules, and/or at least some of the functionality of multiple modules may be combined into a single module. Additionally, a specific module discussed herein performing an action includes the specific module itself performing the action, or alternatively the specific module calls or otherwise accesses another component or module that performs the action (or performs the action in conjunction with the specific module). Accordingly, the particular module performing the action may include the particular module performing the action itself and/or another module that the particular module performing the action invokes or otherwise accesses.

It is to be understood that although the terms first, second, third etc. may be used herein to describe various devices, elements, components or elements, these devices, elements, components or elements should not be limited by these terms. These terms are only used to distinguish one device, component, component or element from another device, component, component or element.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the present invention. within the scope of protection.

Claims (7)

1. A product image adjustment method based on big data, characterized in that the method comprises: The information acquisition module of the smart e-commerce platform sends several display images corresponding to the target product to the user terminals of each test user in response to the image optimization request sent by the management terminal, and obtains image feedback information of each test user for different display images; The interest recognition module obtains the degree of interest of the test user in each image feature area in the corresponding test image according to the image feedback information analysis of the test user, and according to the interest degree of all test users in each image feature area and the area information of the corresponding image feature area Quantitative analysis to obtain the visual priority of each image feature region; The parameter enhancement module identifies the visual key area in the corresponding display image according to the visual priority of each image feature area, and determines each of the corresponding display images according to the first parameter interval, the second parameter interval and the visual priority of each visual key area. Region processing parameters for visually critical regions; The parameter enhancement module determines, according to the first parameter interval, the second parameter interval and the visual priority of each visual key area, the area processing parameters corresponding to each visual key area in the display image, including: the parameter enhancement module acquires the pixel features of each visual key area By comparing it with the global pixel feature of the corresponding display image to obtain the first regional difference feature of each visual key area, and according to the first feature difference analysis corresponding to the first regional difference feature of each visual key area to obtain the corresponding visual The first parameter interval of the key area; the parameter enhancement module compares the pixel features of each visual key area with the pixel features of its associated adjacent image feature area to obtain the second regional difference feature of each visual key area, and according to each The second feature difference analysis corresponding to the second regional difference feature of the visual key area obtains the second parameter interval corresponding to the visual key area; the parameter enhancement module determines the corresponding display according to the first parameter interval and the second parameter interval of each visual key area The weight coefficient of each pixel of the visual key area in the image whose visual priority is greater than the preset priority threshold, and the area processing parameters corresponding to the visual key area are obtained by fusing the pixel value and the weight coefficient of each pixel; The parameter enhancement module compares the pixel features of each visual key area with the pixel features of its related adjacent image feature areas to obtain the second regional difference feature of each visual key area. The corresponding first key feature matrix is established for the corresponding visual key area, and the corresponding second key feature matrix is established according to the pixel features of each adjacent image feature area related to it; the parameter enhancement module is based on the first key feature matrix corresponding to each visual key area. The matrix variance of the key feature matrix and the first matrix neighborhood entropy corresponding to each first key feature matrix are analyzed to obtain the first key neighborhood entropy of each visual key area, according to each adjacent image feature corresponding to each visual key area The matrix variance of the area corresponding to the second key feature matrix and the second matrix neighborhood entropy corresponding to each second key feature matrix are analyzed to obtain the second key neighborhood entropy of each adjacent image feature area, wherein the first matrix adjacent The domain entropy is used to characterize the weight coefficients of each feature vector in the first key feature matrix; the parameter enhancement module combines the feature components of the first key neighborhood entropy of each visual key area and the second key feature area of each adjacent image feature area related to it The feature components of key neighborhood entropy are projected into feature subspaces of different scales to obtain several eigenvalue differences between each visual key region and its associated adjacent image feature regions, and generate corresponding visual key values based on several eigenvalue differences. The second regional difference feature of the region, wherein the second regional difference feature is used to characterize the pixel average difference and the gray-scale aggregation difference between each visually critical region and its associated adjacent image feature region; According to the second feature difference analysis corresponding to the second regional difference feature of each visual key area, the second parameter interval of the corresponding visual key area is obtained by the second parameter interval of the corresponding visual key area. The difference of each feature value in is compared with the second difference threshold to obtain the second feature difference between each key visual area and its associated adjacent image feature area, wherein the second feature difference is used to characterize each Local chromaticity contrast and local luminance contrast between a key vision area and its associated adjacent image feature areas; the parameter enhancement module determines the second area difference feature according to the minimum visual difference of the human eye and the second feature difference degree of the corresponding key vision area The transformable range of each eigenvalue difference in is to obtain the second parameter interval corresponding to the visual critical area; The image processing module enhances the contrast of each key visual area in the corresponding display image according to the region processing parameters to obtain an effect-enhanced image of the corresponding display image. 2. The method according to claim 1, wherein the image optimization request includes a device identifier, a product number, several display images of the target product, and format information of the display images; the display images are used to optimize the target product The shape structure is displayed. 3. The method according to claim 2, wherein the interest recognition module obtains the visual priority of each image feature area according to the degree of interest of all test users in each image feature area and the regional information volume analysis of the corresponding image feature area Degrees include: The interest identification module determines the area weight value of the corresponding image feature area according to the area information amount corresponding to each image feature area, wherein the area information amount is used to represent the number of product features of the target product contained in the corresponding image feature area; The interest recognition module weights and fuses the different interests of all test users corresponding to the same image feature area according to the area weight value of each image feature area to obtain the visual priority of the corresponding image feature area. 4. The method according to claim 3, wherein the image feedback signal is used to characterize the gaze information of the test user on different image regions of the display image, which includes the position information of the image gaze point, the length of stay, the sight line of the test user Scan path and number of saccades. 5. The method according to claim 4, wherein the interest recognition module obtains the degree of interest of the test user in the corresponding test image to each image feature region according to the image feedback information analysis of the test user: The interest identification module analyzes the human eye movement state of the test user according to the image feedback information of the test user to obtain the eye movement characteristics of the corresponding test user, and obtains the degree of interest of the test user to each image feature area according to the analysis of the eye movement characteristics , wherein the eye movement feature is used to characterize the corresponding test user's dwell time, eye saccade data, and scanning trajectory for each image feature area in the corresponding display image. 6. The method according to claim 5, wherein the first parameter interval is used to characterize the enhancement range of each key visual area compared to the pixel value of the corresponding display image; the second parameter interval is used for Pixel values characterizing each visually critical region compared to its associated adjacent image feature regions may enhance range. 7. The method according to claim 6, wherein the minimum visual difference is the prior knowledge of the human body stored in the system, which represents the minimum pixel value difference perceivable by human eyes.