A method and system for evaluating picture quality

Abstract

The present invention discloses a method and a system for evaluating picture quality, which relate to the field of computer technology. A specific embodiment of the method includes: obtaining a target picture and the project identifier of the visual specification project to which the target picture belongs, and extracting the picture information of the target picture; calculating the metric value of a set evaluation parameter of the target picture according to the picture information; obtaining the reference value of the evaluation parameter according to the project identifier, wherein the reference value is determined by calculating the metric values of multiple reference pictures; comparing the metric value of the target picture with the reference value according to a set comparison rule to obtain the evaluation result of the target picture. This embodiment realizes the automatic evaluation of picture quality, improves the evaluation efficiency, and the evaluation result is objective and fair by extracting the picture information of the target picture, generating digital evaluation parameters, and then comparing them with the reference values of the evaluation parameters.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and particularly to a method and system for evaluating picture quality. Background Art

[0002] In the actual business of the Internet, User Interface (UI) designers are required to design a large number of pictures. In order to ensure relative consistency of the pictures designed by multiple UI designers, a series of visual specifications need to be set. For example, all parts of a picture tend to form an integral whole, the core object tends to be at the center of the picture, the color change of the picture is not abrupt, and the picture remains relatively neat, etc.

[0003] In the prior art, usually UI designers evaluate the picture quality according to the visual specifications, and after the picture design is completed, it is stored and managed by UI designers or project-related personnel.

[0004] In the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art:

[0005] The evaluation of picture quality is carried out manually, with strong subjectivity and low efficiency; the pictures are stored by UI designers or project-related personnel, and the management is decentralized, making it difficult to refer to and compare with each other. Summary of the Invention

[0006] In view of this, embodiments of the present invention provide a method and system for evaluating picture quality. By extracting the picture information of a target picture, generating digital evaluation parameters, and then comparing them with the reference values of the evaluation parameters, the automatic evaluation of picture quality is realized, the evaluation efficiency is improved, and the evaluation results are objective and fair.

[0007] To achieve the above object, according to one aspect of the embodiments of the present invention, a method for evaluating picture quality is provided.

[0008] A method for evaluating picture quality according to an embodiment of the present invention includes: obtaining a target picture and the project identifier of the visual specification project to which the target picture belongs, and extracting the picture information of the target picture; calculating the measurement value of the set evaluation parameter of the target picture according to the picture information; obtaining the reference value of the evaluation parameter according to the project identifier, where the reference value is determined by calculating the measurement values of multiple reference pictures; comparing the measurement value of the target picture with the reference value according to the set comparison rule to obtain the evaluation result of the target picture.

[0009] Optionally, the evaluation parameter includes any one or more of the number of objects, proximity, similarity, and neatness of the picture; wherein, the proximity is used to measure the distance of the core object from the center of the picture, the similarity is used to measure the color change of the picture, and the neatness is used to measure the neatness of the picture.

[0010] Optionally, when the evaluation parameter includes the number of objects, the picture information of the target picture is extracted, including: obtaining the first size information of the target picture and the second size information of the objects included in the target picture; according to the picture information, calculating the measurement value of the number of objects in the target picture, including: reducing the first size information according to a set scaling ratio to obtain the reference size information; traversing the objects included in the target picture, and counting the total number of objects in the target picture whose second size information is greater than the reference size information, and the total number of objects is the measurement value of the number of objects.

[0011] Optionally, when the evaluation parameter includes the proximity, the picture information of the target picture is extracted, including: dividing the target picture into N×M grid regions, initializing the scale of the coordinate axis according to the grid regions, and determining the center coordinates of the target picture; where N and M are integers; obtaining the pixel information of the target picture, and filtering the pixel points of the target picture according to the pixel information and a set filtering condition to obtain target pixel points; according to the picture information, calculating the measurement value of the proximity of the target picture, including: calculating the matrix center of the grid region to which the target pixel point belongs, and mapping the matrix center to a corresponding key; counting the number of the target pixel points with the same key to obtain the information amount of the grid region corresponding to the key; calculating the distance between the matrix centers of the top K grid regions with large information amounts and the center coordinates, and accumulating and summing the K distances to obtain the measurement value of the proximity; where K is an integer.

[0012] Optionally, the pixel information is the color value of the pixel point, and the filtering condition is that the absolute value of the difference between the color value of the current pixel point and the color values of the pixel points in the four directions of up, down, left, and right of the current pixel point is less than a set threshold; filtering the pixel points of the target picture according to the pixel information and the set filtering condition, including: subtracting the color value of the current pixel point from the color values of the pixel points in the four directions of up, down, left, and right of the current pixel point respectively to obtain four differences; if the absolute values of the four differences are all less than the threshold, then filter the current pixel point.

[0013] Optionally, when the evaluation parameter includes the similarity, extracting the picture information of the target picture includes: extracting the color values of multiple pixel points of the target picture; calculating a metric value of the similarity of the target picture according to the picture information, including: taking the difference between the color values of multiple pixel points of the target picture and the color values of the pixel points in its upper, lower, left, and right directions respectively, and accumulating the sum of the squares of the differences; calculating the average color value difference of the target picture according to the accumulated result, and the average color value difference is the metric value of the similarity.

[0014] Optionally, when the evaluation parameter includes the neatness, extracting the picture information of the target picture includes: obtaining the first size information of the target picture; calculating a metric value of the neatness of the target picture according to the picture information, including: scaling the target picture to a set size according to the first size information; taking the grayscale image data corresponding to the scaled target picture as a one-dimensional array, and statistically calculating the probability of occurrence of the array elements in the one-dimensional array; calculating the information entropy of the target picture according to the probability, and the information entropy is the metric value of the neatness.

[0015] Optionally, determining the reference value of the evaluation parameter according to the metric values of the evaluation parameters of multiple reference pictures includes: adding up the metric values of the evaluation parameters of the multiple reference pictures and then dividing by the number of pictures of the multiple reference pictures to obtain the reference value of the evaluation parameter. To achieve the above object, according to another aspect of the embodiments of the present invention, a picture quality evaluation system is provided.

[0016] A picture quality evaluation system according to an embodiment of the present invention includes: an extraction module, configured to obtain a target picture and the project identifier of the visual specification project to which the target picture belongs, and extract the picture information of the target picture; a calculation module, configured to calculate a metric value of a set evaluation parameter of the target picture according to the picture information; an acquisition module, configured to obtain the reference value of the evaluation parameter according to the project identifier; wherein, the reference value is determined by calculating the metric values of the evaluation parameters of multiple reference pictures; a comparison module, configured to compare the metric value of the target picture with the reference value according to a set comparison rule to obtain an evaluation result of the target picture.

[0017] Optionally, the evaluation parameter includes any one or more of the number of objects in the picture, proximity, similarity, and neatness; wherein, the proximity is used to measure the distance of the core object from the center of the picture, the similarity is used to measure the color change of the picture, and the neatness is used to measure the neatness of the picture.

[0018] Optionally, when the evaluation parameter includes the number of objects, the extraction module is further configured to: obtain first size information of the target picture and second size information of the objects included in the target picture; the calculation module is further configured to: reduce the first size information according to a set scaling ratio to obtain reference size information; and traverse the objects included in the target picture, and count the total number of objects in the target picture whose second size information is greater than the reference size information, and the total number of objects is the measurement value of the number of objects.

[0019] Optionally, when the evaluation parameter includes the proximity, the extraction module is further configured to: divide the target picture into N×M grid regions, initialize the scale of the coordinate axes according to the grid regions, and determine the center coordinates of the target picture; where N and M are integers; and obtain pixel information of the target picture, and filter pixel points of the target picture according to the pixel information and a set filtering condition to obtain target pixel points; the calculation module is further configured to: calculate the matrix center of the grid region to which the target pixel point belongs, map the matrix center to a corresponding key; count the number of target pixel points with the same key to obtain the information amount of the grid region corresponding to the key; and calculate the distances between the matrix centers of the top K grid regions with large information amounts and the center coordinates, and accumulate and sum the K distances to obtain the measurement value of the proximity; where K is an integer.

[0020] Optionally, the pixel information is the color value of a pixel point, and the filtering condition is that the absolute value of the difference between the color value of the current pixel point and the color values of the pixel points in the four directions of up, down, left, and right of the current pixel point is less than a set threshold; the extraction module is further configured to: subtract the color value of the current pixel point from the color values of the pixel points in the four directions of up, down, left, and right of the current pixel point respectively to obtain four differences; if the absolute values of the four differences are all less than the threshold, then filter the current pixel point.

[0021] Optionally, when the evaluation parameter includes the similarity, the extraction module is further configured to: extract the color values of multiple pixel points of the target picture; the calculation module is further configured to: subtract the color values of the pixel points in the four directions of up, down, left, and right of the target picture from the color values of the multiple pixel points of the target picture respectively, and accumulate the sum of the squares of the differences; and calculate the average color value difference of the target picture according to the accumulated result, and the average color value difference is the measurement value of the similarity.

[0022] Optionally, when the evaluation parameter includes the neatness, the extraction module is further configured to: obtain first size information of the target picture; the calculation module is further configured to: scale the target picture to a set size according to the first size information; use the grayscale data corresponding to the scaled target picture as a one-dimensional array, and count the probability of occurrence of array elements in the one-dimensional array; and calculate the information entropy of the target picture according to the probability, and the information entropy is the measurement value of the neatness.

[0023] Optionally, the system further includes: a determination module, configured to sum up the measurement values of the evaluation parameters of the multiple reference pictures and then divide by the number of the multiple reference pictures to obtain the reference value of the evaluation parameter.

[0024] To achieve the above object, according to another aspect of the embodiments of the present invention, there is provided a picture quality evaluation system.

[0025] To achieve the above object, according to still another aspect of the embodiments of the present invention, there is provided an electronic device.

[0026] An electronic device according to an embodiment of the present invention includes: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement a picture quality evaluation method according to an embodiment of the present invention.

[0027] To achieve the above object, according to still another aspect of the embodiments of the present invention, there is provided a computer-readable medium.

[0028] A computer-readable medium according to an embodiment of the present invention stores a computer program thereon, and when the program is executed by a processor, it implements a picture quality evaluation method according to an embodiment of the present invention.

[0029] One of the above embodiments of the invention has the following advantages or beneficial effects: by extracting the picture information of the target picture, generating a digital evaluation parameter, and then comparing it with the reference value of the evaluation parameter, the automatic evaluation of the picture quality is realized, the evaluation efficiency is improved, and the evaluation result is objective and fair; setting the evaluation parameter of the picture according to the product requirements realizes the customized picture evaluation; comparing the second size information of each object in the picture with the reference size information to count the total number of objects that need to be concerned, realizes the digitization of the index of the number of objects, and improves the evaluation speed while ensuring the evaluation effect.

[0030] One embodiment of the above invention has the following advantages or beneficial effects: Using the sum of the distances between the matrix center of the grid area with a large amount of information and the picture center as the distance from the key object to the picture center realizes the digitization of the proximity index; by eliminating the pixels whose color values are close to those of the pixels above, below, left, and right, the outline of the key object is determined, facilitating the subsequent statistical analysis of the information volume of the grid area; using the average color value difference of all the pixels in the picture to measure the color change of the picture realizes the digitization of the similarity index; using information entropy to measure the chaos degree of the picture realizes the digitization of the neatness index; by statistically analyzing the measurement values of the evaluation parameters in the reference picture and then combining with the number of pictures to obtain the reference value, the objectivity of the evaluation result is further ensured.

[0031] The further effects of the above non-conventional optional methods will be described below in combination with specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings are used to better understand the present invention and do not unduly limit the present invention. Among them:

[0033] Figure 1 is a schematic diagram of the main steps of the picture quality evaluation method according to an embodiment of the present invention;

[0034] Figure 2 is a schematic diagram of the main process of the picture quality evaluation method according to an embodiment of the present invention;

[0035] Figure 3 is a schematic diagram of the management interface of the visual specification item list according to an embodiment of the present invention;

[0036] Figure 4 is a schematic diagram of the management interface of the main tone map of the visual specification according to an embodiment of the present invention;

[0037] Figure 5 is a schematic diagram of the display result of the reference value of the evaluation parameter according to an embodiment of the present invention;

[0038] Figure 6 is a schematic diagram of the upload picture detection and scoring interface according to an embodiment of the present invention;

[0039] Figure 7 is a schematic diagram of the evaluation result of the target picture according to an embodiment of the present invention;

[0040] Figure 8 is a schematic diagram of the management interface of the visual specification detection log table according to an embodiment of the present invention;

[0041] Figure 9 is a schematic diagram of the calculation process of the reference value of the object quantity according to an embodiment of the present invention;

[0042] Figure 10Schematic diagram of the process for calculating the reference value of proximity in an embodiment of the present invention;

[0043] Figure 11 Schematic diagram of the process for calculating the reference value of similarity in an embodiment of the present invention;

[0044] Figure 12 Schematic diagram of the process for calculating the reference value of neatness in an embodiment of the present invention;

[0045] Figure 13 Schematic diagram of the main modules of the picture quality evaluation system according to an embodiment of the present invention;

[0046] Figure 14 Exemplary system architecture diagram to which an embodiment of the present invention can be applied;

[0047] Figure 15 Schematic diagram of the structure of a computer device of an electronic device suitable for implementing an embodiment of the present invention. Detailed implementation manners

[0048] The following describes exemplary embodiments of the present invention with reference to the accompanying drawings. Various details of the embodiments of the present invention are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, descriptions of well-known functions and structures are omitted below for clarity and conciseness.

[0049] Figure 1 Schematic diagram of the main steps of the picture quality evaluation method according to an embodiment of the present invention. As Figure 1 shown, the picture quality evaluation method of the embodiment of the present invention is implemented by a picture quality evaluation system and mainly includes the following steps:

[0050] Step S101: Obtain a target picture and the project identifier of the visual specification project to which the target picture belongs, and extract the picture information of the target picture. Herein, the target picture is the picture to be evaluated. The user submits a target picture for a certain visual specification project to initiate a picture evaluation request to the picture quality evaluation system. After receiving the picture evaluation request, the picture quality evaluation system obtains the target picture submitted by the user and the project identifier of the visual specification project. Then, it extracts the picture information from the target picture.

[0051] Step S102: Calculate the metric value of the set evaluation parameter of the target image according to the image information. Here, the evaluation parameter is an index for evaluating the image quality, which can be any one or more of the number of objects, proximity, similarity, and neatness in the image. The number of objects is the total number of objects contained in the image. Proximity is used to measure the distance of the core object from the center of the image. Similarity is used to measure the color change of the image. Neatness is used to measure the neatness of the image.

[0052] Different evaluation parameters require different image information. For example, when the evaluation parameter is the number of objects, the image information is the first size information of the target image and the second size information of the objects contained in the target image. When the evaluation parameter is proximity, the image information is the center coordinates of the target image and the target pixel point. When the evaluation parameter is similarity, the image information is the color values of multiple pixel points of the target image. When the evaluation parameter is neatness, the image information is the first size information of the target image.

[0053] Step S103: Obtain the reference value of the evaluation parameter according to the project identifier. Here, the reference value is determined by calculating the metric values of the evaluation parameters of multiple reference images. Specifically, calculate the metric values of the evaluation parameters of multiple reference images, sum up all the metric values, and then divide by the number of reference images to obtain the reference value of the evaluation parameter. The project identifiers of each visual specification item and the reference values of the evaluation parameters are pre-saved. Subsequently, based on the project identifier, the reference value of the evaluation parameter of the corresponding visual specification item can be obtained.

[0054] Step S104: Compare the metric value of the target image with the reference value according to the set comparison rule to obtain the evaluation result of the target image. Compare the size of the metric value of the number of objects in the target image with the reference value. If the deviation of the metric value from the reference value is greater than the set threshold, it indicates that the number of objects in the target image does not meet the evaluation standard. Compare the size of the metric values of the proximity, similarity, and neatness of the target image with their respective reference values. If the metric value is greater than the set multiple of the corresponding reference value, it indicates that the proximity, similarity, and neatness of the target image do not meet the evaluation standard. Thus, the automatic evaluation of the image quality is realized.

[0055] Figure 2 It is a schematic diagram of the main process of the image quality evaluation method according to an embodiment of the present invention. As Figure 2 shown, the image quality evaluation method according to an embodiment of the present invention mainly includes the following steps:

[0056] Step S201: Construct a visual specification file. The visual specification file includes a visual specification item table, a visual specification keynote chart, and a visual specification detection log table. The visual specification item table is used to record various parameters of the visual specification. The visual specification keynote chart is used to store the reference pictures uploaded for the visual specification items. The visual specification detection log table is used to record the evaluation history of the quality evaluation of the target pictures. The visual specification file will be described in detail below with reference to Tables 1 - 3.

[0057] Table 1 is the visual specification item table of the embodiment. This table includes three data fields: field name, field type, and description. Among them, project_id is the project identifier of the visual specification project; project_config is the visual specification evaluation benchmark, which can be in json format and includes four fields: the benchmark value of the number of objects group_number, the benchmark value of proximity center_distance, the benchmark value of similarity variance, and the benchmark value of neatness entropy.

[0058] Table 1

[0059]

[0060]

[0061] Table 2 is the visual specification keynote chart of the embodiment. This table includes three data fields: field name, field type, and description. Among them, referpic_id is the reference picture identifier; project_id is the project identifier of the visual specification project to which the reference picture belongs, and referpic_path is the storage path of the reference picture (i.e., the keynote picture). The reference picture is used to calculate the four fields of project_config in the visual specification item table.

[0062] Table 2

[0063] Field Name Field Type Description referpic_id int(11) unsigned project_id int(11) Visual Specification Project to Which It Belongs referpic_path varchar(255) Storage Path of the Main Tone Picture creator varchar(255) Creator created_at datetime Record Creation Time

[0064] Table 3 is the visual specification detection log table of the embodiment. This table includes three data fields: field name, field type, and description. Among them, targetpic_id is the target picture identifier; project_id is the project identifier of the visual specification project to which the target picture belongs; the platform refers to the platform that uses the picture quality evaluation method of this embodiment to evaluate the picture quality.

[0065] Table 3

[0066] Field Name Field Type Description targetpic_id int(11) unsigned project_id int(11) Visual Specification Project to Which It Belongs name varchar(255) Picture Description of the Target Picture targetpic_path varchar(255) Storage Path of the Target Picture check_result longtext Evaluation Result login varchar(255) User Who Logs in to the Platform for Evaluation created_at datetime Record Creation Time

[0067] Step S202: Establish a visual specification item table management interface according to the visual specification document, so that the user can upload the reference pictures belonging to the current visual specification item through the visual specification item table management interface. The user clicks the [Main Tone Map Management] button in the operation field of the visual specification item table management interface (see Figure 3 ), and enters the visual specification main tone map management interface (see Figure 4 ). Through this interface, the reference pictures belonging to the current visual specification item can be uploaded.

[0068] Figure 3 is a schematic diagram of the visual specification item table management interface of the embodiment of the present invention. As shown in Figure 3 , the visual specification item table management interface of the embodiment of the present invention may include fields such as id, visual specification item name, visual specification main tone calculation result, creator, record creation time, and operation. Among them, id is the project identifier of the current visual specification item.

[0069] The operation field can implement functions such as main tone map management, automatic calculation of reference parameters, upload picture detection scoring, editing, copying, and deleting. The editing function is used to allow the user to re-upload pictures, and the copying function is used to copy this record for modification to generate new data. These two functions are optimization functions and can be deleted. In addition, this interface can also implement the function of adding new data. The user clicks on the corresponding position of this interface to add a new visual specification item.

[0070] Figure 4 is a schematic diagram of the visual specification main tone map management interface of the embodiment of the present invention. As shown in Figure 4 , the visual specification main tone map management interface of the embodiment of the present invention may include fields such as id, affiliated project, main tone map storage path, creator, record creation time, and operation. Among them, id is the reference picture identifier. For example, the ids of pictures 1-3 are 4, 5, and 6 respectively.

[0071] The operation field can implement functions such as editing, copying, and deleting, and these functions are optimization functions and can be deleted. In addition, this interface can also implement the function of adding new data. The user clicks on the corresponding position of this interface to upload a new reference picture.

[0072] Step S203: After the automatic calculation of reference parameters function of the visual specification item table management interface is triggered, obtain the reference pictures belonging to the current visual specification item, and calculate the reference values of the evaluation parameters. Among them, the evaluation parameters are the number of objects group_number, proximity center_distance, similarity variance, and regularity entropy of the picture.

[0073] After the user clicks the

Automatically Calculate Benchmark Parameters

[0074] Step S204: Display the benchmark values of the evaluation parameters on the Visual Specification Project Table Management Interface and store them in the visual specification file. In the embodiment, the benchmark values of the evaluation parameters are displayed in the visual specification keynote calculation result field on the Visual Specification Project Table Management Interface. At the same time, the benchmark values of the evaluation parameters are packaged into a specified data format, such as json format, and then stored in the project_config field of the visual specification project table.

[0075] Figure 5 is a schematic diagram of the display result of the benchmark values of the evaluation parameters in the embodiment of the present invention. As Figure 5 shown, after the processing of step S203, the benchmark values of each evaluation parameter for testing this visual specification item are: the number of objects group_number = 8, the proximity center_distance = 100, the similarity variance = 6284, and the neatness entropy = 7.1172646166933.

[0076] After calculating the benchmark values of the evaluation parameters, the target image can be uploaded for the current visual specification project, and the image quality evaluation and detection can be performed.

[0077] Step S205: When the upload image detection and scoring function on the Visual Specification Project Table Management Interface is triggered, display the upload image detection and scoring interface so that the user can upload the target image through the upload image detection and scoring interface. When the user clicks the

Upload Image Detection and Scoring

[0078] Figure 6 is a schematic diagram of the upload image detection and scoring interface in the embodiment of the present invention. As Figure 6 shown, the upload image detection and scoring interface in the embodiment of the present invention includes a picture description, a select file button, a submit button, and a cancel button. Among them, the picture description is used for the user to mark the purpose of the picture, such as the XX project YY advertisement carousel picture, which is convenient for the user to find their own picture. When the user clicks the

Select File

[0079] Step S206: Obtain the target image uploaded by the user and the project identifier of the visual specification project to which the target image belongs, and calculate the measurement value of the evaluation parameter of the target image. The measurement value of the evaluation parameter here is the measurement value of the number of objects group_number, proximity center_distance, similarity variance, and neatness entropy. The specific calculation process of the measurement value is shown in the following description about Figures 9 to 12 .

[0080] Step S207: According to the project identifier, obtain the reference value of the evaluation parameter of the corresponding visual specification project. According to the project identifier, take out the project_config field of the corresponding visual specification project table and perform json decoding to obtain the reference values of the number of objects group_number, proximity center_distance, similarity variance, and neatness entropy.

[0081] Step S208: Compare the measurement value of the evaluation parameter of the target image with the reference value according to the set comparison rule to obtain the evaluation result of the target image. The comparison rule is set according to each evaluation parameter and actual requirements. For the number of objects group_number, the comparison rule can be that the measurement value deviates from the reference value by more than the set threshold; for the proximity center_distance, similarity variance, and neatness entropy, the comparison rule can be that the measurement value is greater than a set multiple of the corresponding reference value. It should be noted that the multiples set for the reference values of different evaluation parameters can be the same or different. The following gives examples for each evaluation parameter respectively.

[0082] Compare the measurement value of the number of objects group_number of the target image with the reference value. If the measurement value is greater than the reference value and the number is greater than 2, then the number of objects group_number of the target image does not meet the standard, and the user can be advised to appropriately reduce the number of larger objects (such as objects with a height or width greater than 1 / 10 of the image) on the target image; if the measurement value is less than the reference value and the number is greater than 2, then the number of objects group_number of the target image does not meet the standard, and the user can be advised to appropriately increase the number of larger objects on the target image; if the measurement value deviates (including greater than and less than) from the reference value by less than or equal to 2, then the number of objects group_number of the target image meets the standard.

[0083] Compare the measured value of the proximity center_distance of the target image with the reference value. The smaller the value of center_distance, the closer the core object is to the center of the image. If the measured value is greater than 1.1 times the reference value, the proximity center_distance of the target image does not meet the standard, and the user can be advised to move the core object of the target image closer to the center of the image; if the measured value is less than or equal to 1.1 times the reference value, the proximity center_distance of the target image meets the standard.

[0084] Compare the measured value of the similarity variance of the target image with the reference value. The smaller the value of variance, the slower the color change of the image. If the measured value is greater than 1.1 times the reference value, the similarity variance of the target image does not meet the standard, and the user can be prompted that the gradient color of the target image is too abrupt and needs to be modified; if the measured value is less than or equal to 1.1 times the reference value, the similarity variance of the target image meets the standard.

[0085] Compare the measured value of the neatness entropy of the target image with the reference value. The smaller the value of entropy, the neater the image. If the measured value is greater than 1.1 times the reference value, the neatness entropy of the target image does not meet the standard, and the user can be prompted that the objects on the target image are not neat enough and need to be modified; if the measured value is less than or equal to 1.1 times the reference value, the neatness entropy of the target image meets the standard.

[0086] Figure 7 It is a schematic diagram of the evaluation result of the target image in the embodiment of the present invention. As Figure 7 shown, according to step S206 - step S208, calculate the measured values of the evaluation parameters of a certain target image (i.e., Figure 7 the score of the uploaded image), and then compare it with the reference value of the corresponding evaluation parameter (i.e., Figure 7 the reference score), and the evaluation, suggestion and description of the target image can be obtained. As Figure 7 can be seen, the number of image objects in the target image does not meet the standard, and the user can be advised to appropriately reduce the number of larger objects in the image. It should be noted that the image object is an example of the objects included in the image.

[0087] Step S208: Store the evaluation result in the visual specification file. Retain the evaluation result of each target image in the visual specification file for the user to compare and reference with each other. In the embodiment, the evaluation result is stored in the check_result field of the visual specification detection log table. By designing the evaluation parameters of the image quality and digitizing the evaluation parameters through the above steps, the automatic evaluation of the image quality is realized on the premise of following the visual specification.

[0088] Figure 8 It is a schematic diagram of the management interface of the visual specification detection log table according to an embodiment of the present invention. As Figure 8 shown, the management interface of the visual specification detection log table according to an embodiment of the present invention may include fields such as id, affiliated project, picture description, uploaded picture, evaluation result, logged-in user, record creation time, and operation. Among them, the id is the target picture identifier. For example, the ids of pictures 4 and 5 are 6 and 5 respectively. The uploaded picture field is the target picture uploaded by the user for quality evaluation.

[0089] From Figure 8 it can be seen that both pictures 4 and 5 are target pictures uploaded by the user and belong to the project of testing visual specifications. Among them, the number of picture objects and proximity of picture 4 do not meet the standards, while the similarity and neatness meet the standards; the number of picture objects of picture 5 does not meet the standards, while the proximity, similarity, and neatness meet the standards.

[0090] The above-mentioned picture quality evaluation method of the embodiment can be used for the review and evaluation of UI design works, and can also be customized by the platform to provide customized picture quality evaluation for enterprises. For example, the benchmark value of the evaluation parameters can be customized according to the requirements of the business product line, so that the design drawings of the business product line are based on a unified evaluation benchmark, and pictures with similar scores are retained to obtain consistency. For example, if the business product line is related to political and legal products, the benchmark value is customized based on political and legal requirements, and the obtained pictures will be relatively serious; if the business product line is related to animation products, the benchmark value is customized based on animation requirements, and the obtained pictures will be more active.

[0091] The calculation process of the benchmark value of each evaluation parameter in the embodiment and the measured value of each evaluation parameter in each picture will be described in detail below.

[0092] Figure 9 It is a schematic diagram of the calculation process of the benchmark value of the object number according to an embodiment of the present invention. As Figure 9 shown, the benchmark value of the object number according to an embodiment of the present invention is obtained by processing the benchmark pictures, and the specific calculation process includes the following steps:

[0093] Step S901: Initialize totalNumber and picCount. Among them, totalNumber represents the number of objects that meet the set conditions (that is, need to be concerned) in all benchmark pictures; picCount is the number of benchmark pictures affiliated to the current visual specification project. In the embodiment, both totalNumber and picCount are initialized to 0.

[0094] Step S902: Determine whether the traversal of the reference image is completed. If not, execute Step S903; if completed, execute Step S907. In the embodiment, Python (a cross-platform computer programming language) and the OpenCV library are used to identify the number of objects in the image. The OpenCV library is a cross-platform computer vision library distributed under the BSD license (open source).

[0095] Step S903: Obtain the first size information of the current reference image and initialize the reference size information according to the set scaling ratio. Among them, the first size information includes the first width and the first height; the reference size information includes the reference width maxWidth and the reference height maxHeight, which can be calculated by the following formula:

[0096]

[0097] A picture usually contains many objects, but generally, the human eye only focuses on 5-7 objects. Therefore, the scaling ratio can be set based on experience, such as 1 / 9, 1 / 10, 1 / 11, etc. to obtain the reference size information, and the objects larger than the reference size information are regarded as the objects that need to be focused on, while the objects less than or equal to the reference size information are filtered out.

[0098] Step S904: Detect the object contours of the objects contained in the current reference image and determine the second size information of each object. Among them, the second size information includes the second width and the second height. After converting the current reference image into a grayscale image, then convert the grayscale image into a binary image, detect the object contours in the binary image, and determine the width and height of each object. Among them, a binary image refers to an image with only two gray levels, that is, the gray value of any pixel point in the image is 0 or 255, representing black and white respectively.

[0099] It should be noted that in the embodiment, the execution order of Step S903 and Step S904 is not limited, and Step S904 can be executed first and then Step S903.

[0100] Step S905: Initialize the counter realCount and traverse the objects contained in the current reference image. The counter realCount represents the number of objects that meet the set conditions in the current reference image. In the embodiment, the counter realCount is initialized to 0.

[0101] Step S906: Use the counter realCount to count the total number of objects in the current reference picture whose second width is greater than the reference width or the second height is greater than the reference height, and accumulate it to totalNumber, then execute Step S902. If in the current reference picture, the second width of the current object is greater than the reference width or the second height is greater than the reference height, then set realCount = realCount + 1, and this quantity is the measurement value of the number of objects group_numbe in the current reference picture. totalNumber can be calculated using the following formula:

[0102] totalNumber = totalNumber + realCount Formula 2 Step S907: Divide totalNumber by picCount to perform a division operation to obtain the reference value of the number of objects. After traversing the reference pictures, the reference value of the number of objects group_numbe can be calculated, and this reference value can be calculated using the following formula:

[0103] Reference value of group_numbe = totalNumber / picCount Formula 3

[0104] In an optional embodiment, the calculation method of the measurement value of the number of objects group_numbe in the target picture is the same as that of the measurement value of the number of objects group_numbe in the current reference picture. Specifically, see Steps S903 to S906. The total number of objects obtained in Step S906 is the measurement value of the number of objects group_numbe. It should be noted that this measurement value is obtained based on processing the target picture.

[0105] Figure 10 This is a schematic diagram of the calculation process of the proximity reference value in the embodiment of the present invention. As Figure 10 shown, the proximity reference value in the embodiment of the present invention is obtained based on processing the reference picture, and the specific calculation process includes the following steps:

[0106] Step S1001: Initialize distanceTotal, infoArr, gzDistanceX, gzDistanceY, and xNumber. distanceTotal represents the sum of the index distances, infoArr is used to store the reserved pixel points (i.e., target pixel points), gzDistanceX is the ruler that divides the x-axis into N parts (i.e., the scale of the x-axis), gzDistanceY is the ruler that divides the y-axis into M parts (i.e., the scale of the y-axis), and xNumber represents the scale variable. Among them, N and M are integers greater than 1.

[0107] In the embodiment, distanceTotal is initialized to 0; xNumber is initialized to 10; both N and M are 10. At this time, gzDistanceX is a ruler that divides the x-axis into 10 parts, and gzDistanceY is a ruler that divides the y-axis into 10 parts. The two can be calculated using the following formula:

[0108]

[0109] Step S1002: Determine whether the traversal of the reference image is completed. If it is not completed, execute Step S1003; if it is completed, execute Step S1011.

[0110] Step S1003: Divide the current reference image into a grid area of N×M, and determine the center coordinates of the current reference image according to the initialized gzDistanceX and gzDistanceY. Still taking N and M both being 10 as an example, divide the current reference image into a 10×10 grid area. Among them, the x-axis coordinates are 0-9, the y-axis coordinates are 0-9, and there are a total of 100 grids, so the center coordinates are (4.5, 4.5).

[0111] Step S1004: Obtain the pixel information of the current reference image. In the embodiment, the RGB color values of all pixel points of the current reference image are obtained.

[0112] Step S1005: Determine whether the traversal of the x-axis coordinates of the current reference image is completed. If it is not completed, execute Step S1006; if it is completed, execute Step S1009.

[0113] Step S1006: Determine whether the traversal of the y-axis coordinates of the current reference image is completed. If it is not completed, execute Step S1007; if it is completed, execute Step S1005.

[0114] Step S1007: According to the pixel information and the set filtering conditions, detect whether the current pixel point needs to be discarded. If it needs to be discarded, execute Step S1006; if it does not need to be discarded, execute Step S1008. This step is used to eliminate those pixel points that are close to the upper, lower, left, and right, and obtain the contour of the core object. Among them, the filtering condition is that the absolute value of the difference between the color value of the current pixel point and the color values of the pixel points in the four directions of the upper, lower, left, and right of the current pixel point is less than the set threshold.

[0115] In the embodiment, the R color value, G color value, and B color value of the current pixel are respectively compared with the R color values, G color values, and B color values of the pixels in the four directions of up, down, left, and right of it. If the absolute values of the differences between the two are all less than the threshold (such as 30), then the current pixel is discarded. It should be noted that if there is no pixel in a certain direction of the current pixel, it is also considered that the absolute value of the difference between the two is less than the threshold.

[0116] Step S1008: Calculate the matrix center of the grid area to which the current pixel belongs, map the matrix center to the corresponding key, and execute step S1006. Assume that the matrix center of the grid area to which the current pixel (x, y) belongs is represented by (infoX, infoY), then the matrix center (infoX, infoY) can be calculated using the following formula:

[0117]

[0118] For the convenience of subsequent sorting, it is necessary to map the matrix center (infoX, infoY) to a one-dimensional dimension to obtain the corresponding key. The mapping process can be represented by the following formula:

[0119] key = infoX * xNumber + infoY Formula 6

[0120] Step S1009: Count the number of target pixels with the same key to obtain the information amount of the grid area corresponding to the key. Among them, the target pixel is the pixel retained in step S1007. The key of each matrix center is unique, and there is the following relationship between the key and the matrix center (infoX, infoY):

[0121]

[0122] Then the information amount of each grid area can be represented by the following formula:

[0123] infoArr[key] = infoArr[key] + 1 Formula 8

[0124] Step S1010: Calculate the distances between the matrix centers of the top K grid regions with large information amounts and the center coordinates, sum up the K distances to obtain distanceTotal, and execute Step S1002. After Step S1009, the information amounts of each of the 100 grid regions can be obtained. Then, sort infoArr in descending order according to the information amount. Initialize distance = 0 and calcCount = 0. Traverse infoArr, restore the key to the matrix center (infoX, infoY), calculate its distance from the center coordinates (4.5, 4.5), accumulate it to distance, and set calcCount = calcCount + 1.

[0125] After that, determine whether calcCount is greater than K (K is an integer greater than 1, such as 10). If calcCount is less than or equal to K, accumulate distanceTotal; if calcCount is greater than K, stop accumulating. Among them, distanceTotal can be expressed by the following formula:

[0126] distanceTotal = distanceTotal + distance Formula 9 In this embodiment, the number of contours falling in each grid region is called the information amount of each grid region. The grid region with the largest information amount is more suitable if it is closer to the center of the picture. Therefore, the sum of the relative distances between the 10 grid regions with the largest information amounts and the center of the picture can be used as a measure of proximity (i.e., distanceTotal).

[0127] Step S1011: Divide distanceTotal by picCount to perform a division operation to obtain the reference value of proximity. After traversing the reference picture, the reference value of the proximity center_distance can be calculated. This reference value can be expressed by the following formula:

[0128] Reference value of center_distance = distanceTotal / picCount Formula 10

[0129] In an optional embodiment, the calculation method of the proximity measure value center_distance in the target picture is the same as that in the current reference picture. Specifically, see Steps S1003 to S1010. The distanceTotal obtained in Step S1010 is the proximity measure value of center_distance. It should be noted that this measure value is obtained based on the processing of the target picture.

[0130] Figure 11It is a schematic diagram of the calculation process of the reference value of similarity in the embodiments of the present invention. As Figure 11 shown, the reference value of similarity in the embodiments of the present invention is obtained based on processing the reference picture, and the specific calculation process includes the following steps:

[0131] Step S1101: Determine whether the traversal of the reference picture is completed. If not, execute Step S1102; if completed, execute Step S1104.

[0132] Step S1102: Extract the color values of each pixel point of the current reference picture, and subtract the color values of the pixel points in the four directions of up, down, left, and right from the color values of each pixel point respectively to obtain the differences. In the embodiment, extract the R color value, G color value, and B color value of each pixel point of the current reference picture, and subtract the R color value, G color value, and B color value of the pixel points in the four directions of up, down, left, and right from the R color value, G color value, and B color value of each pixel point respectively.

[0133] Step S1103: Divide the sum of the squares of the differences after accumulation by the total number of pixel points to obtain the average color value difference of the current reference picture, and then execute Step S1101. This average color difference is the measurement value of similarity variance in the current reference picture. The total number of pixel points is the number of all pixel points in the current reference picture.

[0134] Step S1104: After accumulating and summing up the average color value differences of all the reference pictures, divide by the number of pictures of the reference pictures to obtain the reference value of similarity.

[0135] In an optional embodiment, the calculation method of the measurement value of similarity variance in the target picture is the same as that of the measurement value of similarity variance in the current reference picture. Specifically, see Steps S1102 to S1103, and the average color value difference obtained in Step S1103 is the measurement value of similarity variance. It should be noted that this measurement value is obtained based on processing the target picture.

[0136] Figure 12 It is a schematic diagram of the calculation process of the reference value of neatness in the embodiments of the present invention. As Figure 12 shown, the reference value of neatness in the embodiments of the present invention is obtained based on processing the reference picture, and the information entropy of the picture is used as the evaluation criterion for neatness. The specific calculation process includes the following steps:

[0137] Step S1201: Determine whether the traversal of the reference picture is completed. If not, execute Step S1202; if completed, execute Step S1205.

[0138] Step S1202: Obtain the first size information of the current reference picture and initialize changeWidth and changeHeight. The first size information is the width and height of the current reference picture. In practical applications, the picture sizes are uncertain, some are large and some are small. If the picture is too large, it will lead to a large amount of calculation and the computer cannot handle it. Therefore, changeWidth can be used to scale the picture to a unified width. In the embodiment, changeWidth = 300, changeHeight = round(height / 300), where height is the height of the current reference picture.

[0139] Step S1203: Scale the current reference picture to changeWidth and changeHeight. For pictures with smaller sizes, the computer can directly process them without scaling. Therefore, in a preferred embodiment, it can first be determined whether the picture width exceeds a threshold (such as 300). If it exceeds, it is scaled down to the threshold; then it is determined whether the picture height exceeds the threshold (such as 300). If it exceeds, it is scaled down to the threshold to prevent the computer from being unable to handle it.

[0140] Step S1204: Take the grayscale image data corresponding to the scaled current reference picture as a one-dimensional array, count the probabilities of the array elements appearing in the one-dimensional array, calculate the information entropy of the current reference picture, and execute step S1201. Information entropy is a measure of the chaos or disorder of matter. The more chaotic the matter, the greater the information entropy.

[0141] In this step, the scaled current reference picture is first converted into a grayscale image. There are 256 grayscale levels in the grayscale image, that is, the grayscale value of any pixel point in the image is one of 0 to 255. Then, the grayscale image data is used as a one-dimensional array to calculate the information entropy H of the current reference picture. The calculation formula is as follows:

[0142]

[0143] where p i is the probability of the i-th element appearing in the one-dimensional array. For example, for the one-dimensional array (1, 2, 1, 3, 1, 4, 5), a total of 7 numbers, the number of times 1 appears is 3, so the probability of 1 appearing is 3 / 7.

[0144] Step S1205: After accumulating and summing up the information entropies of all the reference pictures, divide by the number of pictures of the reference pictures to obtain the reference value of neatness.

[0145] In an optional embodiment, the method for calculating the measure value of the neatness entropy in the target picture is the same as that in the current reference picture. Specifically, see steps S1202 to S1204, and the information entropy obtained in step S1204 is the measure value of the neatness entropy. It should be noted that this measure value is obtained based on the processing of the target picture.

[0146] As can be seen from the picture quality evaluation method of the embodiments of the present invention, by extracting the picture information of the target picture, generating digital evaluation parameters, and then comparing them with the reference values of the evaluation parameters, the automatic evaluation of the picture quality is realized, the evaluation efficiency is improved, and the evaluation result is objective and fair.

[0147] Figure 13 It is a schematic diagram of the main modules of the picture quality evaluation system according to the embodiments of the present invention. As Figure 13 shown, the picture quality evaluation system 1300 according to the embodiments of the present invention mainly includes:

[0148] An extraction module 1301, configured to obtain a target picture and the project identifier of the visual specification project to which the target picture belongs, and extract the picture information of the target picture. Among them, the target picture is the picture to be evaluated. The user submits a target picture for a certain visual specification project to initiate a picture evaluation request to the picture quality evaluation system. After receiving the picture evaluation request, the picture quality evaluation system obtains the target picture submitted by the user and the project identifier of the visual specification project. Then, it extracts the picture information from the target picture.

[0149] A calculation module 1302, configured to calculate the measure value of the set evaluation parameter of the target picture according to the picture information. Among them, the evaluation parameter is the index for evaluating the picture quality, and can be any one or more of the number of objects, proximity, similarity, and neatness of the picture. The number of objects is the total number of objects included in the picture, proximity is used to measure the distance of the core object to the center of the picture, similarity is used to measure the color change of the picture, and neatness is used to measure the neatness of the picture.

[0150] Different evaluation parameters require different picture information. For example, when the evaluation parameter is the number of objects, the picture information is the first size information of the target picture and the second size information of the objects included in the target picture; when the evaluation parameter is proximity, the picture information is the center coordinates of the target picture and the target pixel point; when the evaluation parameter is similarity, the picture information is the color values of multiple pixel points of the target picture; when the evaluation parameter is neatness, the picture information is the first size information of the target picture.

[0151] An acquisition module 1303 is configured to obtain a reference value of the evaluation parameter according to the project identifier. The reference value is determined by calculating measurement values of the evaluation parameters of multiple reference pictures. Specifically, calculate the measurement values of the evaluation parameters of multiple reference pictures, sum up all the measurement values and divide by the number of reference pictures to obtain the reference value of the evaluation parameter. The project identifiers of each visual specification item and the reference values of the evaluation parameters are pre-saved. Subsequently, based on the project identifier, the reference value of the evaluation parameter corresponding to the visual specification item can be obtained.

[0152] A comparison module 1304 is configured to compare the measurement value of the target picture with the reference value according to a set comparison rule to obtain an evaluation result of the target picture. Compare the measurement value of the number of objects in the target picture with the reference value. If the measurement value deviates from the reference value by more than a set threshold, it indicates that the number of objects in the target picture does not meet the evaluation criteria. Compare the measurement values of the proximity, similarity, and neatness of the target picture with their respective reference values. If the measurement value is greater than a set multiple of the corresponding reference value, it indicates that the proximity, similarity, and neatness of the target picture do not meet the evaluation criteria. Thus, the automated evaluation of picture quality is realized.

[0153] In addition, the picture quality evaluation system 1300 according to an embodiment of the present invention may further include: a determination module configured to sum up the measurement values of the evaluation parameters of multiple reference pictures and divide by the number of reference pictures to obtain the reference value of the evaluation parameter.

[0154] As can be seen from the above description, by extracting the picture information of the target picture, generating digital evaluation parameters, and then comparing with the reference value of the evaluation parameter, the automated evaluation of picture quality is realized, improving the evaluation efficiency, and the evaluation result is objective and fair.

[0155] Figure 14 An exemplary system architecture 1400 is shown to which the picture quality evaluation method or the picture quality evaluation system according to an embodiment of the present invention can be applied.

[0156] As Figure 14 shown, the system architecture 1400 may include terminal devices 1401, 1402, 1403, a network 1404, and a server 1405. The network 1404 is used to provide a medium for a communication link between the terminal devices 1401, 1402, 1403 and the server 1405. The network 1404 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0157] Users can interact with the server 1405 through the network 1404 using the terminal devices 1401, 1402, and 1403 to receive or send messages, etc. Various communication client applications can be installed on the terminal devices 1401, 1402, and 1403, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0158] The terminal devices 1401, 1402, and 1403 can be various electronic devices with a display screen and supporting web browsing, including but not limited to smartphones, tablets, laptop computers, and desktop computers, etc.

[0159] The server 1405 can be a server that provides various services, such as a background management server for processing the target pictures uploaded by the users using the terminal devices 1401, 1402, and 1403. The background management server can obtain the target pictures, calculate the measurement value of the processing evaluation parameters of the target pictures, compare the measurement value with the reference value, and feedback the comparison result to the terminal devices.

[0160] It should be noted that the picture quality evaluation method provided by the embodiments of the present application is generally executed by the server 1405. Correspondingly, the picture quality evaluation system is generally set in the server 1405.

[0161] It should be understood that Figure 14 the numbers of the terminal devices, networks, and servers in

[0162] According to the embodiments of the present invention, the present invention also provides an electronic device and a computer-readable medium.

[0163] The electronic device of the present invention includes: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement a picture quality evaluation method according to the embodiments of the present invention.

[0164] The computer-readable medium of the present invention has a computer program stored thereon, and when the program is executed by a processor, it implements a picture quality evaluation method according to the embodiments of the present invention.

[0165] Next, refer to Figure 15 , which shows a schematic structural diagram of a computer system 1500 suitable for implementing the electronic device of the embodiments of the present invention. Figure 15 The shown electronic device is only an example and should not bring any limitation to the functions and usage scope of the embodiments of the present invention.

[0166] As Figure 15As shown, computer system 1500 includes a central processing unit (CPU) 1501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1502 or a program loaded from a storage section 1508 into a random access memory (RAM) 1503. In the RAM 1503, various programs and data required for the operation of the computer system 1500 are also stored. The CPU 1501, ROM 1502, and RAM 1503 are connected to each other via a bus 1504. An input / output (I / O) interface 1505 is also connected to the bus 1504.

[0167] The following components are connected to the I / O interface 1505: an input section 1506 including a keyboard, a mouse, etc.; an output section 1507 including such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage section 1508 including a hard disk, etc.; and a communication section 1509 including a network interface card such as a LAN card, a modem, etc. The communication section 1509 performs communication processing via a network such as the Internet. A drive 1510 is also connected to the I / O interface 1505 as needed. A removable medium 1511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1510 as needed so that a computer program read from it can be installed into the storage section 1508 as needed.

[0168] Specifically, according to the embodiments disclosed in the present invention, the process described in the above main step diagram can be implemented as a computer software program. For example, the embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program codes for performing the method shown in the main step diagram. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 1509, and / or installed from the removable medium 1511. When the computer program is executed by the central processing unit (CPU) 1501, the above functions defined in the system of the present invention are executed.

[0169] It should be noted that the computer-readable medium shown in the present invention can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of a computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on a computer-readable medium can be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0170] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram can represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or by a combination of dedicated hardware and computer instructions.

[0171] The modules involved in the embodiments of the present invention can be implemented in software or in hardware. The described modules can also be provided in a processor. For example, it can be described as: a processor includes an extraction module, a calculation module, an acquisition module, and a comparison module. Among them, the names of these modules do not constitute a limitation to the module itself in some cases. For example, the extraction module can also be described as "a module for acquiring a target image and the project identifier of the visual specification project to which the target image belongs, and extracting the image information of the target image".

[0172] As another aspect, the present invention also provides a computer-readable medium, which can be included in the device described in the above embodiments; or can exist alone without being assembled into the device. The above computer-readable medium carries one or more programs. When the one or more programs are executed by the device, the device includes: acquiring a target image and the project identifier of the visual specification project to which the target image belongs, and extracting the image information of the target image; calculating a measurement value of a set evaluation parameter of the target image according to the image information; acquiring a reference value of the evaluation parameter according to the project identifier; wherein, the reference value is determined by calculating the measurement values of the evaluation parameters of multiple reference images; comparing the measurement value of the target image with the reference value according to a set comparison rule to obtain an evaluation result of the target image.

[0173] According to the technical solution of the embodiments of the present invention, by extracting the image information of the target image, generating a digital evaluation parameter, and then comparing it with the reference value of the evaluation parameter, the automatic evaluation of the image quality is realized, the evaluation efficiency is improved, and the evaluation result is objective and fair.

[0174] The above products can execute the methods provided in the embodiments of the present invention, and have the corresponding functional modules and beneficial effects of the executed methods. For the technical details not described in detail in this embodiment, reference can be made to the methods provided in the embodiments of the present invention.

[0175] The above specific embodiments do not constitute a limitation to the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for evaluating picture quality, characterized in that, Including: Obtain a target image and the project identifier of the visual specification project to which the target image belongs, and extract the image information of the target image; According to the image information, calculate the metric value of the set evaluation parameter of the target image; wherein, the evaluation parameter includes any one or more of the number of objects, proximity, similarity, and neatness of the image; the proximity is used to measure the distance of the core object to the center of the image, the similarity is used to measure the color change of the image, and the neatness is used to measure the neatness of the image; According to the project identifier, obtain the benchmark value of the evaluation parameter; wherein, the benchmark value is determined by calculating the metric values of the evaluation parameters of multiple benchmark images; Compare the metric value of the target image with the benchmark value according to the set comparison rule to obtain the evaluation result of the target image; Wherein, when the evaluation parameter includes the neatness, extracting the image information of the target image includes: obtaining the first size information of the target image; Calculating the metric value of the neatness of the target image according to the image information includes: scaling the target image to a set size according to the first size information; taking the grayscale image data corresponding to the scaled target image as a one-dimensional array, and counting the probability of the occurrence of array elements in the one-dimensional array; calculating the information entropy of the target image according to the probability, and the information entropy is the metric value of the neatness.

2. The method according to claim 1, wherein When the evaluation parameter includes the number of objects, extracting the image information of the target image includes: Obtaining the first size information of the target image and the second size information of the objects included in the target image; Calculating the metric value of the number of objects of the target image according to the image information includes: Reducing the first size information according to a set scaling ratio to obtain the benchmark size information; Traverse the objects included in the target image, and count the total number of objects in the target image whose second size information is greater than the benchmark size information, and the total number of objects is the metric value of the number of objects.

3. The method according to claim 1, wherein When the evaluation parameter includes the proximity, extracting the image information of the target image includes: Dividing the target image into N×M grid regions, initializing the scale of the coordinate axis according to the grid regions, and determining the center coordinates of the target image; wherein, N and M are integers; Obtain the pixel information of the target image, and filter the pixel points of the target image according to the pixel information and a set filtering condition to obtain target pixel points; Calculating the metric value of the proximity of the target image according to the image information includes: Calculating the matrix center of the grid region to which the target pixel point belongs, and mapping the matrix center to a corresponding key; Counting the number of the target pixel points with the same key to obtain the information amount of the grid region corresponding to the key; Calculating the distance between the matrix centers of the top K grid regions with large information amounts and the center coordinates, and accumulating and summing the K distances to obtain the metric value of the proximity; wherein, K is an integer.

4. The method according to claim 3, characterized in that The pixel information is the color value of a pixel point, and the filtering condition is that the absolute values of the differences between the color value of the current pixel point and the color values of the pixel points in the four directions of up, down, left, and right of the current pixel point are all less than a set threshold value; Filtering the pixel points of the target picture according to the pixel information and the set filtering condition includes: Taking the difference between the color value of the current pixel point and the color values of the pixel points in the four directions of up, down, left, and right of the current pixel point respectively to obtain four differences; If the absolute values of the four differences are all less than the threshold value, then filter the current pixel point.

5. The method according to claim 1, characterized in that, When the evaluation parameter includes the similarity, extracting the picture information of the target picture includes: Extracting the color values of multiple pixel points of the target picture; Calculating the measurement value of the similarity of the target picture according to the picture information includes: Taking the difference between the color values of multiple pixel points of the target picture and the color values of the pixel points in the four directions of up, down, left, and right of them respectively, and accumulating the sum of the squares of the differences; Calculating the average color value difference of the target picture according to the accumulated result, and the average color value difference is the measurement value of the similarity.

6. The method according to any one of claims 1 to 5, characterized in that, Determining the reference value of the evaluation parameter according to the measurement values of the evaluation parameters of multiple reference pictures includes: After accumulating and summing the measurement values of the evaluation parameters of the multiple reference pictures, dividing by the number of pictures of the multiple reference pictures to obtain the reference value of the evaluation parameter.

7. An image quality evaluation system, characterized in that Includes: An extraction module, configured to obtain a target picture and the project identifier of the visual specification project to which the target picture belongs, and extract the picture information of the target picture; A calculation module, configured to calculate the measurement value of the set evaluation parameter of the target picture according to the picture information; wherein, the evaluation parameter includes any one or more of the number of objects, proximity, similarity, and neatness of the picture; the proximity is used to measure the distance of the core object to the center of the picture, the similarity is used to measure the color change of the picture, and the neatness is used to measure the neatness of the picture; An acquisition module, configured to obtain the reference value of the evaluation parameter according to the project identifier; wherein, the reference value is determined by calculating the measurement values of the evaluation parameters of multiple reference pictures; A comparison module, configured to compare the measurement value of the target picture with the reference value according to a set comparison rule to obtain the evaluation result of the target picture; Wherein, when the evaluation parameter includes the neatness in the calculation module, extracting the picture information of the target picture includes: obtaining the first size information of the target picture; The acquisition module calculates the measurement value of the neatness of the target picture according to the picture information, including: scaling the target picture to a set size according to the first size information; taking the grayscale image data corresponding to the scaled target picture as a one-dimensional array, and counting the probability of the occurrence of array elements in the one-dimensional array; calculating the information entropy of the target picture according to the probability, and the information entropy is the measurement value of the neatness.

8. An electronic device, characterized in that, Includes: One or more processors; A storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-6.

9. A computer-readable medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, the method according to any one of claims 1-6 is implemented.

Images