Objective evaluation method and device for image quality, equipment, medium and program product

Abstract

The invention relates to an objective evaluation method and device for image quality, equipment, a medium and a program product, and particularly relates to the technical field of image processing. Comprising the steps that an image data set is acquired and preprocessed; the image data set comprises a subjective quality evaluation label; for each image in the preprocessed image data set, calculating an objective quality evaluation value according to a preset weight coefficient and a plurality of preset algorithms; drawing a scatter diagram by taking the objective quality evaluation value as an abscissa and the subjective quality evaluation label as an ordinate, and calculating a drop point probability of an ideal point in the scatter diagram; and when the drop point probability is less than or equal to a preset threshold value, adjusting a preset weight coefficient according to a preset rule, and recalculating the drop point probability until the drop point probability is greater than the preset threshold value, so as to calculate an objective quality evaluation value of the image according to the corresponding preset weight coefficient. The image quality objective evaluation method is used for solving the problem that the matching degree between an existing image quality objective evaluation algorithm and an artificial subjective evaluation result is not high.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to an objective evaluation method, device, equipment, medium and program product for image quality. Background Art

[0002] Image quality evaluation plays a very important role in modern science and technology, especially in visual signal processing such as image restoration, image fusion, image inpainting and biomedical imaging. The goal of image quality evaluation is to obtain an objective evaluation that is consistent with human subjective judgment, that is, to design an algorithm model to replace human beings in making a reasonable and accurate objective evaluation of image quality.

[0003] Clarity is the basis and an important part of the subjective evaluation system of image quality. With the development of image processing technology, current image clarity objective algorithms can be roughly divided into three categories, namely, objective clarity evaluation algorithms based on gradient, frequency domain and statistics. However, the above three types of image clarity objective algorithms only have a high degree of match with subjective clarity evaluation in a certain single scene, and a low degree of match in multiple scenes. It is currently impossible to fully match with manual subjective evaluation in all scenes or most scenes. Summary of the invention

[0004] In order to solve the above technical problems or at least partially solve the above technical problems, the present application provides an objective evaluation method, device, equipment, medium and program product for image quality, which can improve the matching degree between objective evaluation and manual subjective evaluation.

[0005] In order to achieve the above purpose, the technical solutions provided by the embodiments of the present application are as follows:

[0006] In a first aspect, the present application provides an objective evaluation method for image quality, comprising: obtaining an image data set and preprocessing the image data set; the image data set comprises an image, an image region of interest and a subjective quality evaluation label; for each image in the preprocessed image data set, calculating an objective quality evaluation value according to a preset weight coefficient and a plurality of preset algorithms; drawing a scatter plot with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and calculating the landing probability of an ideal point in the scatter plot; wherein the ideal point is a point whose horizontal and vertical coordinate difference is within a preset range, and the landing probability is used to indicate the degree of match between the objective quality evaluation value and the subjective quality evaluation label; when the landing probability is less than or equal to a preset threshold, adjusting the preset weight coefficient according to a preset rule, and recalculating the landing probability until the landing probability is greater than the preset threshold, so as to obtain the objective quality evaluation value of the image according to the corresponding preset weight coefficient.

[0007] As an optional implementation of the embodiment of the present application, for each image in the preprocessed image data set, an objective quality evaluation value is calculated according to a preset weight coefficient and multiple preset algorithms, including: for each image in the preprocessed image data set, multiple quality evaluation parameters are calculated according to multiple preset algorithms; according to the multiple quality evaluation parameters and the corresponding preset weight coefficients, an objective quality evaluation value is obtained by weighted summation; wherein the sum of the preset weight coefficients corresponding to the multiple quality evaluation parameters is 1.

[0008] As an optional implementation of an embodiment of the present application, calculating the landing probability of an ideal point in a scatter plot includes: after drawing the scatter plot, dividing the horizontal and vertical axes according to preset intervals to obtain multiple sub-areas of the scatter plot; counting the number of ideal points and the number of abnormal points in each sub-area; an abnormal point is a point whose horizontal and vertical coordinate difference is outside a preset range; based on the number of ideal points and the number of abnormal points, calculating the landing probability of the ideal point in each sub-area; and calculating the average landing probability based on the landing probability of the ideal points in all sub-areas as the landing probability corresponding to the scatter plot.

[0009] As an optional implementation of an embodiment of the present application, the preset rules include: when the adjustment parameter is within a preset numerical range, adjusting the preset weight coefficient according to a preset increment corresponding to the preset numerical range; wherein the adjustment parameter is calculated based on the landing probability and the last landing probability.

[0010] In a second aspect, the present application provides an objective evaluation device for image quality, the device comprising:

[0011] An acquisition module is used to acquire and preprocess an image dataset; the image dataset includes images, image regions of interest, and subjective quality evaluation labels;

[0012] A calculation module, used for calculating an objective quality evaluation value for each image in the preprocessed image data set according to a preset weight coefficient and a plurality of preset algorithms;

[0013] A drawing module is used to draw a scatter plot with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and calculate the probability of an ideal point falling in the scatter plot; wherein the ideal point is a point where the difference between the horizontal and vertical coordinates is within a preset range, and the probability of falling is used to indicate the degree of matching between the objective quality evaluation value and the subjective quality evaluation label;

[0014] The adjustment module is used to adjust the preset weight coefficient according to the preset rules when the landing probability is less than or equal to the preset threshold, and recalculate the landing probability until the landing probability is greater than the preset threshold, so as to calculate the objective quality evaluation value of the image according to the corresponding preset weight coefficient.

[0015] As an optional implementation of the embodiment of the present application, the calculation module is specifically used to: for each image in the preprocessed image data set, calculate multiple quality evaluation parameters according to multiple preset algorithms; according to the multiple quality evaluation parameters and the corresponding preset weight coefficients, perform weighted summation to obtain an objective quality evaluation value; wherein the sum of the preset weight coefficients corresponding to the multiple quality evaluation parameters is 1.

[0016] As an optional implementation of the embodiment of the present application, the drawing module is specifically used to: after drawing the scatter plot, divide the horizontal and vertical axes according to preset intervals to obtain multiple sub-areas of the scatter plot; count the number of ideal points and the number of abnormal points in each sub-area; the abnormal point is a point whose horizontal and vertical coordinate difference is outside a preset range; based on the number of ideal points and the number of abnormal points, calculate the landing probability of the ideal point in each sub-area; based on the landing probability of the ideal points in all sub-areas, calculate the average landing probability as the landing probability corresponding to the scatter plot.

[0017] As an optional implementation of an embodiment of the present application, the preset rules include: when the adjustment parameter is within a preset numerical range, adjusting the preset weight coefficient according to a preset increment corresponding to the preset numerical range; wherein the adjustment parameter is calculated based on the landing probability and the last landing probability.

[0018] In a third aspect, the present application provides an electronic device comprising: a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the objective image quality evaluation method as described in the first aspect or any one of its optional embodiments.

[0019] In a fourth aspect, the present application provides a computer-readable storage medium, comprising: a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, it implements the objective evaluation method of image quality as described in the first aspect or any optional embodiment thereof.

[0020] In a fifth aspect, the present application provides a computer program product, comprising: the computer program product comprises a computer program, and when the computer program runs on a computer, the computer implements the objective evaluation method of image quality as described in the first aspect or any one of its optional embodiments.

[0021] Compared with the prior art, the technical solution provided by the embodiments of the present application has the following advantages:

[0022] The disclosed embodiments provide an objective evaluation method, apparatus, device, medium and program product for image quality, wherein the method first obtains and preprocesses an image data set, wherein the image data set includes images, image regions of interest and subjective quality evaluation labels; for each image in the preprocessed image data set, an objective quality evaluation value is calculated according to a preset weight coefficient and a plurality of preset algorithms; then a scatter plot is drawn with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and the probability of an ideal point in the scatter plot whose horizontal and vertical coordinate differences are within a preset range is calculated to indicate the degree of match between the objective quality evaluation value and the subjective quality evaluation label; whereby when the probability of a landing is less than or equal to a preset threshold, the preset weight coefficient is adjusted according to a preset rule, and the probability of a landing is recalculated until the probability of a landing is greater than the preset threshold, so as to obtain the objective quality evaluation value of the image according to the corresponding preset weight coefficient. In this way, the present application performs a comprehensive evaluation of image quality through multiple preset algorithms, which is suitable for diversified scenarios, improves the scene adaptability of objective image quality evaluation, can accurately match manual subjective evaluation, and improves the accuracy of the algorithm's objective evaluation of image quality, thereby improving the degree of automation, efficiency and accuracy of image quality evaluation, reducing or replacing the work of manual evaluation of image quality, and reducing labor costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0024] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0025] Figure 1 A schematic diagram of a flow chart of an objective evaluation method for image quality provided in an embodiment of the present application;

[0026] Figure 2 A schematic diagram of a scatter plot provided in an embodiment of the present application;

[0027] Figure 3 A schematic diagram of a method flow for adjusting a preset weight coefficient provided in an embodiment of the present application;

[0028] Figure 4 A schematic diagram of the structure of an objective evaluation device for image quality provided in an embodiment of the present application;

[0029] Figure 5 A schematic diagram of the structure of an electronic device described in an embodiment of the present application. DETAILED DESCRIPTION

[0030] In order to more clearly understand the above-mentioned purposes, features and advantages of the present application, the scheme of the present application will be further described below. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0031] In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only part of the embodiments of the present application, rather than all of the embodiments.

[0032] In order to solve some or all of the technical problems existing in the related art, the embodiments of the present application provide an objective evaluation method, device, equipment, medium and program product for image quality, wherein the method first obtains and preprocesses an image data set, the image data set including an image, an image region of interest and a subjective quality evaluation label; for each image in the preprocessed image data set, an objective quality evaluation value is calculated according to a preset weight coefficient and multiple preset algorithms; then a scatter plot is drawn with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and the landing probability of an ideal point in the scatter plot whose horizontal and vertical coordinate differences are within a preset range is calculated to indicate the degree of match between the objective quality evaluation value and the subjective quality evaluation label; thereby, when the landing probability is less than or equal to a preset threshold, the preset weight coefficient is adjusted according to a preset rule, and the landing probability is recalculated until the landing probability is greater than the preset threshold, so as to calculate the objective quality evaluation value of the image according to the corresponding preset weight coefficient. In this way, the present application performs a comprehensive evaluation of image quality through multiple preset algorithms, which is suitable for diversified scenarios, improves the scene adaptability of objective image quality evaluation, can accurately match manual subjective evaluation, and improves the accuracy of the algorithm's objective evaluation of image quality, thereby improving the degree of automation, efficiency and accuracy of image quality evaluation, reducing or replacing the work of manual evaluation of image quality, and reducing labor costs.

[0033] An objective evaluation method of image quality provided in the embodiments of the present application can be implemented by an objective evaluation device or electronic device of image quality, and the electronic device includes but is not limited to a personal computer, a laptop computer, a tablet computer, a smart phone, etc. The operating system of the electronic device may include Android, a mobile operating system (iOS) developed by Apple, an operating system (Windows) developed by Microsoft Corporation of the United States, etc., and the embodiments of the present application are not limited to this. The electronic device can be operated alone to implement the present application, or it can be connected to a network and implement the present application through interactive operations with other computer devices in the network. Among them, the network in which the electronic device is located includes but is not limited to the Internet, a wide area network, a metropolitan area network, a local area network, a virtual private network (VPN) network, etc.

[0034] It should be noted that the protection scope of the objective evaluation method of image quality described in the embodiment of the present application is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing or replacing steps in the prior art based on the principles of the present application are included in the protection scope of the present application.

[0035] like Figure 1 As shown, Figure 1 1 is a flow chart of an objective evaluation method of image quality according to an embodiment of the present application. The method can be performed by an objective evaluation device for image quality, wherein the device can be implemented by software and / or hardware and can generally be integrated in. Figure 1 As shown, the method mainly includes the following steps S101 to S104:

[0036] S101, obtaining an image data set and performing preprocessing.

[0037] Preprocessing includes labeling subjective quality evaluation labels and data cleaning. Subjective quality evaluation labels include but are not limited to: subjective clarity evaluation labels, subjective brightness evaluation labels, and subjective color evaluation labels.

[0038] In some embodiments, image data is collected according to different types of scenes or dimensions, and scene types include but are not limited to: scenery, brightness, color temperature, indoors and outdoors.

[0039] After collecting image data, the image data set is labeled with a clarity label. Multiple experimenters conduct subjective evaluations on the quality of each image in the image data set. Assuming that the experimenters evaluate each image and give a quality score, the maximum and minimum values ​​in the quality score are first eliminated, and then the average value is calculated, and the average value is used as the subjective quality evaluation label of the image.

[0040] The above embodiment uses the subjective evaluation of image quality by humans as the precise value, and pre-labels the quality of each image in the image dataset. The labeled quality labels will be used as true values ​​in subsequent steps to feedback learning to obtain an objective evaluation algorithm that is highly matched with the subjective evaluation by humans.

[0041] In some embodiments, the image data set is denoted as N, and the image set includes multiple training elements, and the training elements are denoted as X. i , training element X i It can be expressed as: X i =(Image i , ROI i , Flag i ), i∈N. Each training element X i It includes image Image, image region of interest ROI and subjective quality evaluation Flag. An image Image can include multiple image regions of interest ROI. Subjective quality evaluation Flag is manually labeled after subjective evaluation of the quality of each image in the image dataset. The quality label Flag can adopt a point system or a grade system.

[0042] In some embodiments, after acquiring the image data set, the image data set is cleaned, and optionally, the annotated image data set is output to a preset algorithm model to remove abnormal data in the image data set, the abnormal data including abnormal regions of interest, abnormal quality labels, and images of abnormal sizes.

[0043] The above embodiment performs data cleaning on the image data set to remove abnormal data, thereby preventing adverse effects in the subsequent training process.

[0044] S102: For each image in the preprocessed image data set, an objective quality evaluation value is calculated according to a preset weight coefficient and a plurality of preset algorithms.

[0045] In some embodiments, when executing step S102, first, for each image in the image data set, corresponding multiple quality evaluation parameters are calculated according to multiple preset algorithms; then, based on the multiple quality evaluation parameters and the corresponding preset weight coefficients, a weighted sum is performed to obtain an objective quality evaluation value.

[0046] The preset algorithms include but are not limited to: Laplace algorithm, grayscale variance algorithm, gradient vector square function, sum of absolute difference, Robert operator, Sobel operator, FFT algorithm. The number of preset algorithms is M.

[0047] Multiple quality evaluation parameters are denoted as P j , j∈M. Quality evaluation parameter P j Includes: 1 , P2 , ..., P M .

[0048] The preset weight coefficient is denoted as α i , preset weight coefficient α i Support feedback adjustment, the formula is as follows: α i =α i +Step i , i∈N. Step i is the preset increment, which will be described in detail in the following content. Preset weight coefficient α i Includes: α 1 , α 2 , ..., α N The sum of the preset weight coefficients is 1.

[0049] Objective quality evaluation value P i The calculation formula is as follows (1):

[0050]

[0051] The above embodiment converts manual subjective evaluation into a process of dynamically adjusting preset weight coefficients by weighted fusion of objective quality evaluation parameters, so as to achieve comprehensive evaluation of image quality.

[0052] S103 , draw a scatter plot with the objective quality evaluation value as the horizontal axis and the subjective quality evaluation label as the vertical axis, and calculate the probability of the ideal point falling in the scatter plot.

[0053] The ideal point is a point whose horizontal and vertical coordinate differences are within a preset range, and the landing point probability is used to indicate the degree of match between the objective quality evaluation value and the subjective quality evaluation label.

[0054] In some embodiments, a scatter plot is drawn for an image data set, and the objective quality evaluation value P of the image is used. i As the horizontal axis, the subjective quality evaluation label Flag i is the vertical axis, then (P i , Flag i ) is the only coordinate point in the scatter plot that corresponds to an image. By drawing a scatter plot, we can intuitively find the gap between the objective evaluation of the algorithm and the subjective evaluation of humans.

[0055] Optionally, when drawing a scatter plot, the horizontal and vertical coordinates are normalized to make the objective quality evaluation values ​​and the subjective quality evaluation labels comparable, so as to facilitate a comprehensive comparison between the objective evaluation of the algorithm and the subjective evaluation of humans.

[0056] In some embodiments, after drawing a scatter plot, the horizontal and vertical axes are first divided according to preset intervals to obtain multiple sub-areas of the scatter plot, and the number of ideal points and the number of abnormal points in each sub-area are counted, where the abnormal points are points whose horizontal and vertical coordinate differences are outside a preset range; based on the number of ideal points and the number of abnormal points, the probability of the ideal points falling in each sub-area is calculated; and then the average probability of falling is calculated based on the probability of the ideal points falling in all sub-areas as the probability of falling corresponding to the scatter plot.

[0057] For example, the preset interval is 0.2, and the horizontal axis and the vertical axis are divided into 5 segments respectively, resulting in 25 sub-areas. The horizontal axis and the vertical axis can also be divided into 10 segments or 20 segments, which is not limited in the present application. Figure 2 As shown, Figure 2 A schematic diagram of a scatter plot provided in an embodiment of the present application, the scatter plot is divided into 25 sub-areas, taking the area {(p,f)│0≤p<0.2,0≤f<0.2} as an example, the black coordinate points therein are ideal points (Ideal Point), and the gray coordinate points in the area {(p,f)│0≤p<0.2,f≥0.2} are unexpected points (Unexpected Point).

[0058] The probability of the ideal point falling in each sub-area is calculated according to the following formula (1):

[0059]

[0060] In formula (2), Ideal Points Count is the number of ideal points in each region, and Unexpected Points Count is the number of abnormal points in each region.

[0061] After calculating the probability of the ideal point falling in all regions, the average is taken to obtain the probability of the ideal point falling in the scatter plot.

[0062] The above embodiment converts the matching degree between the objective rating of the algorithm and the subjective evaluation of the human being into the probability of the ideal point falling in the coordinate system.

[0063] S104. When the landing point probability is less than or equal to the preset threshold, adjust the preset weight coefficient according to the preset rule, and recalculate the landing point probability until the landing point probability is greater than the preset threshold, so as to calculate the objective quality evaluation value of the image according to the corresponding preset weight coefficient.

[0064] The preset threshold is a preset threshold used to indicate that the objective quality evaluation value matches the subjective quality evaluation label, and an exemplary preset threshold is 95%.

[0065] Determine whether the landing probability is greater than the preset threshold. If so, it means that the objective quality evaluation value matches the subjective evaluation label. Then save the preset weight coefficient corresponding to the current objective quality evaluation value, and use this preset weight coefficient to objectively evaluate other images to obtain a quality evaluation that is highly matched with the manual subjective evaluation.

[0066] If the landing probability is less than or equal to the preset probability, it means that the objective quality evaluation value does not match the subjective evaluation label, then return to step S102 to adjust the preset weight coefficient, and re-execute step S103 to draw a scatter plot to calculate the landing probability of the ideal point in the scatter plot until the landing probability is greater than the preset threshold, indicating that the objective evaluation of the algorithm is consistent with the manual subjective evaluation.

[0067] In some embodiments, the preset rules include: when the adjustment parameter is within a preset numerical range, adjusting the preset weight coefficient according to a preset increment corresponding to the preset numerical range, wherein the adjustment parameter is calculated based on the landing probability and the last landing probability.

[0068] Optionally, the adjustment parameter is calculated based on the following formula (3):

[0069]

[0070] In formula (3), Current Ideal P represents the landing probability calculated this time; Pre Ideal P represents the last landing probability, which is the landing probability corresponding to the last adjustment of the preset weight coefficient, and can be the landing probability corresponding to the initial preset weight coefficient.

[0071] In the above embodiment, the number of preset numerical ranges may be multiple, and optionally, the preset numerical ranges include: less than the first threshold, greater than the first threshold and less than the second threshold, greater than the second threshold and less than the third threshold, and greater than the third threshold. Exemplarily, assuming that the first threshold is -0.3, the second threshold is 0, and the third threshold is 0.3, the preset numerical ranges include: ΔIdeal P<-0.3, -0.3<ΔIdeal P<0, 0<ΔIdeal P<0.3, ΔIdeal P>0.3. The above settings on the size and number of thresholds are only exemplary, and the present disclosure does not specifically limit this.

[0072] In the above embodiment, different preset value ranges correspond to different preset increments. For example, as shown in Table 1:

[0073] Table 1

[0074] Preset value range <![CDATA[Preset increment (Step i )]]> Less than the first threshold First preset increment Greater than the first threshold and less than the second threshold Second preset increment Greater than the second threshold and less than the third threshold Third preset increment Greater than the third threshold Fourth preset increment

[0075] For example, see Table 2:

[0076] Table 2

[0077] Preset value range Preset Increment ΔIdeal P<-0.3 -0.2 -0.3<ΔIdeal P<0 -0.1 0<ΔIdeal P<0.3 0.1 ΔIdeal P>0.3 0.2

[0078] For example, Figure 3 As shown in the figure, after calculating the landing probability, it is determined whether the landing probability is greater than the preset threshold. If so, multiple preset algorithms corresponding to preset weight coefficients are saved; if not, the adjustment parameter ΔIdeal P is judged respectively. If ΔIdeal P < 0 and ΔIdeal P < -0.3, the preset increment Step i =-0.2; if ΔIdeal P>-0.3 and ΔIdeal P<0, then the preset increment Step i =-0.1; if ΔIdeal P>0 and ΔIdeal P<0.3, the preset increment Step i =0.1; if ΔIdeal P>0 and ΔIdeal P>0.3, the preset increment Step i =0.2.

[0079] Optionally, the preset weight coefficient and the corresponding preset increment after each adjustment are stored to record the dynamic adjustment process. When the landing probability is greater than the preset threshold, the corresponding preset weight coefficient and preset increment are output, which can be used as the optimal weight coefficient for objective evaluation of other image qualities.

[0080] In the above embodiment, feedback learning is achieved by dynamically adjusting the preset weight coefficient through preset rules, so that the objective quality evaluation value obtained by the algorithm matches the subjective quality evaluation label. Subsequently, the objective quality evaluation values ​​of other images can be accurately calculated using such preset weight coefficients.

[0081] In summary, the embodiment of the present application provides an objective evaluation method for image quality, which first obtains and preprocesses an image data set, wherein the image data set includes an image, an image region of interest, and a subjective quality evaluation label; for each image in the preprocessed image data set, an objective quality evaluation value is calculated according to a preset weight coefficient and multiple preset algorithms; then a scatter plot is drawn with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and the probability of the ideal point in the scatter plot whose horizontal and vertical coordinate difference is within a preset range is calculated to indicate the degree of match between the objective quality evaluation value and the subjective quality evaluation label; thus, when the probability of the landing point is less than or equal to the preset threshold, the preset weight coefficient is adjusted according to the preset rule, and the probability of the landing point is recalculated until the probability of the landing point is greater than the preset threshold, so as to calculate the objective quality evaluation value of the image according to the corresponding preset weight coefficient. In this way, the present application comprehensively evaluates the image quality through multiple preset algorithms, is suitable for diversified scenes, improves the scene adaptability of the objective evaluation of image quality, can accurately match with the manual subjective evaluation, and improves the accuracy of the algorithm's objective evaluation of image quality, thereby improving the automation, efficiency and accuracy of image quality evaluation, reducing or replacing the work of manual evaluation of image quality, and reducing labor costs. In addition, it can be applied to camera debugging. Through algorithms, the quality of images taken by the camera can be accurately evaluated, thus realizing automated debugging of the camera.

[0082] like Figure 4 As shown, Figure 4 An objective evaluation device for image quality provided in an embodiment of the present application includes:

[0083] The acquisition module 401 is used to acquire and preprocess an image data set; the image data set includes an image, an image region of interest, and a subjective quality evaluation label;

[0084] A calculation module 402, for calculating an objective quality evaluation value for each image in the preprocessed image data set according to a preset weight coefficient and a plurality of preset algorithms;

[0085] A drawing module 403 is used to draw a scatter plot with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and calculate the probability of an ideal point falling in the scatter plot; wherein the ideal point is a point where the difference between the horizontal and vertical coordinates is within a preset range, and the probability of falling is used to indicate the degree of matching between the objective quality evaluation value and the subjective quality evaluation label;

[0086] The adjustment module 404 is used to adjust the preset weight coefficient according to the preset rule when the landing probability is less than or equal to the preset threshold, and recalculate the landing probability until the landing probability is greater than the preset threshold, so as to calculate the objective quality evaluation value of the image according to the corresponding preset weight coefficient.

[0087] As an optional implementation of the embodiment of the present application, the calculation module 402 is specifically used to: for each image in the preprocessed image data set, calculate multiple quality evaluation parameters according to multiple preset algorithms; according to the multiple quality evaluation parameters and the corresponding preset weight coefficients, perform weighted summation to obtain an objective quality evaluation value; wherein the sum of the preset weight coefficients corresponding to the multiple quality evaluation parameters is 1.

[0088] As an optional implementation of the embodiment of the present application, the drawing module 403 is specifically used to: after drawing the scatter plot, divide the horizontal and vertical axes according to preset intervals to obtain multiple sub-areas of the scatter plot; count the number of ideal points and the number of abnormal points in each sub-area; the abnormal point is a point whose horizontal and vertical coordinate difference is outside the preset range; according to the number of ideal points and the number of abnormal points, calculate the landing probability of the ideal point in each sub-area; according to the landing probability of the ideal points in all sub-areas, calculate the average landing probability as the landing probability corresponding to the scatter plot.

[0089] As an optional implementation of an embodiment of the present application, the preset rules include: when the adjustment parameter is within a preset numerical range, adjusting the preset weight coefficient according to a preset increment corresponding to the preset numerical range; wherein the adjustment parameter is calculated based on the landing probability and the last landing probability.

[0090] In summary, the embodiment of the present application provides an objective evaluation device for image quality, which first obtains and preprocesses an image data set, wherein the image data set includes an image, an image region of interest, and a subjective quality evaluation label; for each image in the preprocessed image data set, an objective quality evaluation value is calculated according to a preset weight coefficient and a plurality of preset algorithms; then a scatter plot is drawn with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and the probability of the ideal point in the scatter plot whose horizontal and vertical coordinate difference is within a preset range is calculated to indicate the degree of match between the objective quality evaluation value and the subjective quality evaluation label; thus, when the probability of the landing point is less than or equal to the preset threshold, the preset weight coefficient is adjusted according to the preset rule, and the probability of the landing point is recalculated until the probability of the landing point is greater than the preset threshold, so as to calculate the objective quality evaluation value of the image according to the corresponding preset weight coefficient. In this way, the present application comprehensively evaluates the image quality through a plurality of preset algorithms, is applicable to diversified scenarios, improves the scene adaptability of the objective evaluation of image quality, can accurately match the manual subjective evaluation, and improves the accuracy of the algorithm's objective evaluation of image quality, thereby improving the degree of automation, efficiency and accuracy of image quality evaluation, reducing or replacing the work of manual evaluation of image quality, and reducing labor costs.

[0091] The specific definition of the objective evaluation device for image quality can be found in the definition of the objective evaluation method for image quality above, which will not be repeated here. Each module in the above-mentioned objective evaluation device for image quality can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0092] In one embodiment, the present application provides an electronic device, which may be a terminal, and its internal structure diagram may be as follows: Figure 5 As shown. The electronic device includes a processor, a memory, a communication interface, a display screen and an input device connected by a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the electronic device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, an operator network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a method for detecting a jam is implemented. The display screen of the electronic device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the electronic device can be a touch layer covered on the display screen, or a button, a trackball or a touchpad provided on the housing of the electronic device, or an external keyboard, touchpad or mouse, etc.

[0093] Those skilled in the art will understand that Figure 5 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the electronic device to which the scheme of the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0094] In one embodiment, the objective evaluation device for image quality provided by the present application can be implemented in the form of a computer program. The computer program can be used in Figure 5 The memory of the electronic device may store various program modules constituting the objective evaluation device of the image quality, for example, Figure 4 The illustrated acquisition module 401, calculation and debugging module 402, drawing module 403 and adjustment module 404. The computer program composed of various program modules enables the processor to execute the steps of the objective evaluation method of image quality of various embodiments of the present application described in this specification.

[0095] For example, Figure 5 The electronic device shown can be Figure 4 The acquisition module 401 in the objective evaluation device for image quality shown in the figure acquires an image data set and performs preprocessing; the electronic device may use the simulation calculation module 402 to calculate the objective quality evaluation value for each image in the preprocessed image data set according to a preset weight coefficient and a plurality of preset algorithms; the electronic device may use the drawing module 403 to draw a scatter plot with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and calculate the landing probability of the ideal point in the scatter plot; the electronic device may use the adjustment module 404 to adjust the preset weight coefficient according to the preset rule when the landing probability is less than or equal to the preset threshold, and recalculate the landing probability until the landing probability is greater than the preset threshold, so as to calculate the objective quality evaluation value of the image according to the corresponding preset weight coefficient.

[0096] In one embodiment, the present application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0097] An image data set is obtained and preprocessed; the image data set includes images, image regions of interest, and subjective quality evaluation labels; for each image in the preprocessed image data set, an objective quality evaluation value is calculated according to a preset weight coefficient and a plurality of preset algorithms; a scatter plot is drawn with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and the probability of an ideal point falling in the scatter plot is calculated; wherein the ideal point is a point whose horizontal and vertical coordinate differences are within a preset range, and the probability of a landing is used to indicate the degree of matching between the objective quality evaluation value and the subjective quality evaluation label; when the probability of a landing is less than or equal to a preset threshold, the preset weight coefficient is adjusted according to a preset rule, and the probability of a landing is recalculated until the probability of a landing is greater than the preset threshold, so as to calculate the objective quality evaluation value of the image according to the corresponding preset weight coefficient.

[0098] In one embodiment, when the processor executes the computer program, the following steps are also implemented: for each image in the preprocessed image data set, according to a preset weight coefficient and multiple preset algorithms, an objective quality evaluation value is calculated, including: for each image in the preprocessed image data set, multiple quality evaluation parameters are calculated according to multiple preset algorithms; according to the multiple quality evaluation parameters and the corresponding preset weight coefficients, the objective quality evaluation value is obtained by weighted summation; wherein the sum of the preset weight coefficients corresponding to the multiple quality evaluation parameters is 1.

[0099] In one embodiment, the processor further implements the following steps when executing the computer program: calculating the landing probability of an ideal point in a scatter plot, including: after drawing the scatter plot, dividing the horizontal and vertical axes according to preset intervals to obtain multiple sub-areas of the scatter plot; counting the number of ideal points and the number of abnormal points in each sub-area; an abnormal point is a point whose horizontal and vertical coordinate difference is outside a preset range; calculating the landing probability of an ideal point in each sub-area based on the number of ideal points and the number of abnormal points; and calculating the average landing probability based on the landing probability of ideal points in all sub-areas as the landing probability corresponding to the scatter plot.

[0100] In one embodiment, the processor also implements the following steps when executing the computer program: the preset rules include: when the adjustment parameter is within a preset numerical range, adjusting the preset weight coefficient according to a preset increment corresponding to the preset numerical range; wherein the adjustment parameter is calculated based on the landing point probability and the last landing point probability.

[0101] When the processor in the electronic device provided by the present application executes a computer program, an image data set is obtained and preprocessed, and the image data set includes an image, an image region of interest, and a subjective quality evaluation label; for each image in the preprocessed image data set, an objective quality evaluation value is calculated according to a preset weight coefficient and a plurality of preset algorithms; then a scatter plot is drawn with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and the probability of the landing point of an ideal point in the scatter plot whose horizontal and vertical coordinate difference is within a preset range is calculated to indicate the degree of matching between the objective quality evaluation value and the subjective quality evaluation label; thus, when the landing point probability is less than or equal to a preset threshold, the preset weight coefficient is adjusted according to a preset rule, and the landing point probability is recalculated until the landing point probability is greater than the preset threshold, so as to calculate the objective quality evaluation value of the image according to the corresponding preset weight coefficient. In this way, the present application comprehensively evaluates the image quality through a plurality of preset algorithms, is applicable to diversified scenarios, improves the scene adaptability of the objective evaluation of image quality, can accurately match with the manual subjective evaluation, and improves the accuracy of the algorithm's objective evaluation of image quality, thereby improving the automation, efficiency and accuracy of the image quality evaluation, reducing or replacing the work of manual evaluation of image quality, and reducing labor costs.

[0102] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by the computer program, the following steps are implemented:

[0103] An image data set is obtained and preprocessed; the image data set includes images, image regions of interest, and subjective quality evaluation labels; for each image in the preprocessed image data set, an objective quality evaluation value is calculated according to a preset weight coefficient and a plurality of preset algorithms; a scatter plot is drawn with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and the probability of an ideal point falling in the scatter plot is calculated; wherein the ideal point is a point whose horizontal and vertical coordinate differences are within a preset range, and the probability of a landing is used to indicate the degree of matching between the objective quality evaluation value and the subjective quality evaluation label; when the probability of a landing is less than or equal to a preset threshold, the preset weight coefficient is adjusted according to a preset rule, and the probability of a landing is recalculated until the probability of a landing is greater than the preset threshold, so as to calculate the objective quality evaluation value of the image according to the corresponding preset weight coefficient.

[0104] In one embodiment, when the computer program executes the computer program, the following steps are also implemented: for each image in the preprocessed image data set, according to a preset weight coefficient and multiple preset algorithms, an objective quality evaluation value is calculated, including: for each image in the preprocessed image data set, multiple quality evaluation parameters are calculated according to multiple preset algorithms; according to the multiple quality evaluation parameters and the corresponding preset weight coefficients, the objective quality evaluation value is obtained by weighted summation; wherein the sum of the preset weight coefficients corresponding to the multiple quality evaluation parameters is 1.

[0105] In one embodiment, when the computer program is executed, the following steps are also implemented: calculating the landing probability of an ideal point in a scatter plot, including: after drawing the scatter plot, dividing the horizontal and vertical axes according to preset intervals to obtain multiple sub-areas of the scatter plot; counting the number of ideal points and the number of abnormal points in each sub-area; an abnormal point is a point whose horizontal and vertical coordinate difference is outside a preset range; calculating the landing probability of an ideal point in each sub-area based on the number of ideal points and the number of abnormal points; calculating the average landing probability based on the landing probability of ideal points in all sub-areas as the landing probability corresponding to the scatter plot.

[0106] In one embodiment, when the computer program executes the computer program, the following steps are also implemented: the preset rules include: when the adjustment parameter is within a preset numerical range, adjusting the preset weight coefficient according to a preset increment corresponding to the preset numerical range; wherein the adjustment parameter is calculated based on the landing probability and the last landing probability.

[0107] When the computer program in the computer-readable storage medium provided by the present application executes the computer program, an image data set is obtained and preprocessed, and the image data set includes an image, an image region of interest, and a subjective quality evaluation label; for each image in the preprocessed image data set, an objective quality evaluation value is calculated according to a preset weight coefficient and a plurality of preset algorithms; then a scatter plot is drawn with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and the probability of the ideal point in the scatter plot whose horizontal and vertical coordinate difference is within a preset range is calculated to indicate the degree of match between the objective quality evaluation value and the subjective quality evaluation label; thus, when the probability of the landing point is less than or equal to the preset threshold, the preset weight coefficient is adjusted according to the preset rule, and the probability of the landing point is recalculated until the probability of the landing point is greater than the preset threshold, so as to calculate the objective quality evaluation value of the image according to the corresponding preset weight coefficient. In this way, the present application comprehensively evaluates the image quality through multiple preset algorithms, is suitable for diversified scenes, improves the scene adaptability of the objective evaluation of image quality, can accurately match with the manual subjective evaluation, and improves the accuracy of the algorithm's objective evaluation of image quality, thereby improving the automation, efficiency and accuracy of the image quality evaluation, reducing or replacing the work of manual evaluation of image quality, and reducing labor costs.

[0108] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media that include computer-usable program code.

[0109] In several embodiments provided by the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0110] In the present application, the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0111] In this application, memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0112] In this application, computer-readable media includes permanent and non-permanent, removable and non-removable storage media. Storage media can be implemented by any method or technology to store information, and the information can be computer-readable instructions, data structures, modules of programs or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. According to the definition in this article, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0113] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0114] The above is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An objective evaluation method for image quality, characterized in that: include: Acquire an image data set and perform preprocessing; the image data set includes images, image regions of interest, and subjective quality evaluation labels; For each image in the preprocessed image data set, an objective quality evaluation value is calculated according to a preset weight coefficient and a plurality of preset algorithms; A scatter plot is drawn with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and the probability of an ideal point in the scatter plot falling is calculated; wherein the ideal point is a point whose horizontal and vertical coordinate difference is within a preset range, and the probability of falling is used to indicate the degree of matching between the objective quality evaluation value and the subjective quality evaluation label; When the landing point probability is less than or equal to the preset threshold, the preset weight coefficient is adjusted according to the preset rule, and the landing point probability is recalculated until the landing point probability is greater than the preset threshold, so as to obtain the objective quality evaluation value of the image according to the corresponding preset weight coefficient.

2. The method according to claim 1, characterized in that The objective quality evaluation value is calculated for each image in the preprocessed image data set according to a preset weight coefficient and a plurality of preset algorithms, including: For each image in the preprocessed image data set, multiple quality evaluation parameters are calculated according to the multiple preset algorithms; Obtaining the objective quality evaluation value by weighted summation according to the multiple quality evaluation parameters and corresponding preset weight coefficients; The sum of the preset weight coefficients corresponding to the multiple quality evaluation parameters is 1.

3. The method according to claim 1, characterized in that Calculating the probability of an ideal point falling in the scatter plot includes: After drawing the scatter plot, dividing the horizontal and vertical axes according to preset intervals to obtain a plurality of sub-areas of the scatter plot; Counting the number of the ideal points and the number of abnormal points in each sub-region; the abnormal points are points whose horizontal and vertical coordinate differences are outside the preset range; Calculate the probability of the ideal point falling in each sub-region according to the number of the ideal points and the number of the abnormal points; The average landing probability is calculated according to the landing probability of the ideal points in all sub-areas to serve as the landing probability corresponding to the scatter plot.

4. The method according to claim 1, characterized in that: The preset rules include: When the adjustment parameter is within a preset value range, adjusting the preset weight coefficient according to a preset increment corresponding to the preset value range; The adjustment parameter is calculated based on the landing probability and the last landing probability.

5. An objective evaluation device for image quality, characterized in that: include: An acquisition module is used to acquire and preprocess an image data set; the image data set includes images, image regions of interest, and subjective quality evaluation labels; A calculation module, used for calculating an objective quality evaluation value for each image in the preprocessed image data set according to a preset weight coefficient and a plurality of preset algorithms; A drawing module, used to draw a scatter plot with the objective quality evaluation value as the horizontal coordinate and the subjective quality evaluation label as the vertical coordinate, and calculate the probability of an ideal point in the scatter plot falling; wherein the ideal point is a point whose horizontal and vertical coordinate difference is within a preset range, and the probability of falling is used to indicate the degree of matching between the objective quality evaluation value and the subjective quality evaluation label; The adjustment module is used to adjust the preset weight coefficient according to the preset rules when the landing probability is less than or equal to the preset threshold, and recalculate the landing probability until the landing probability is greater than the preset threshold, so as to calculate the objective quality evaluation value of the image according to the corresponding preset weight coefficient.

6. The device according to claim 5, characterized in that The computing module is specifically used for: For each image in the preprocessed image data set, multiple quality evaluation parameters are calculated according to the multiple preset algorithms; Obtaining the objective quality evaluation value by weighted summation according to the multiple quality evaluation parameters and corresponding preset weight coefficients; The sum of the preset weight coefficients corresponding to the multiple quality evaluation parameters is 1.

7. The device according to claim 5, characterized in that The drawing module is specifically used for: After drawing the scatter plot, dividing the horizontal and vertical axes according to preset intervals to obtain a plurality of sub-areas of the scatter plot; Counting the number of ideal points and the number of abnormal points in each sub-region; The abnormal point is a point whose horizontal and vertical coordinate difference is outside the preset range; Calculate the probability of the ideal point falling in each sub-region according to the number of the ideal points and the number of the abnormal points; The average landing probability is calculated according to the landing probability of the ideal points in all sub-areas to serve as the landing probability corresponding to the scatter plot.

8. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the objective evaluation method for image quality according to any one of claims 1 to 4.

9. A computer-readable storage medium, characterized in that: include: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the objective evaluation method for image quality according to any one of claims 1 to 4 is implemented.

10. A computer program product, characterized in that include: The computer program product comprises a computer program, and when the computer program is executed on a computer, the computer is caused to implement the objective evaluation method for image quality according to any one of claims 1 to 4.

Images