CN113344843B

CN113344843B - Image quality evaluation method, device and system

Info

Publication number: CN113344843B
Application number: CN202110383595.3A
Authority: CN
Inventors: 梁彦君
Original assignee: ThunderSoft Co Ltd
Current assignee: ThunderSoft Co Ltd
Priority date: 2021-04-09
Filing date: 2021-04-09
Publication date: 2024-04-19
Anticipated expiration: 2041-04-09
Also published as: CN113344843A

Abstract

The invention discloses an image quality evaluation method, an image quality evaluation device and an image quality evaluation system. The method comprises the following steps: determining the confirmation weight of each selected judgment dimension according to the image type and at least one image attribute of the to-be-evaluated image; comparing the to-be-evaluated image with the selected comparison image to obtain an evaluation result of the to-be-evaluated image and the comparison image in each evaluation dimension; and determining the comprehensive evaluation result of the to-be-evaluated image and the comparison image according to the evaluation result of each evaluation dimension and the confirmation weight of each evaluation dimension. The automatic image quality evaluation can be realized, the evaluation accuracy and the working efficiency are greatly improved, the evaluation requirements of different image types and different customer requirements can be met, and the universality is higher.

Description

Image quality evaluation method, device and system

Technical Field

The present invention relates to the field of computer vision, and in particular, to a method, apparatus, and system for evaluating image quality.

Background

In the process of acquiring, processing, transmitting and recording images, because imaging equipment, processing methods, transmission media, recording equipment and the like are not perfect, and the influence of factors such as object motion, noise pollution, shooting environment and the like is added, the shot images inevitably have distortion and degradation, therefore, the image quality of the images is evaluated, and parameters and processes of links such as imaging, processing, transmitting, recording and the like are regulated, so that the images with higher quality are necessary. The image quality is evaluated by subjective evaluation and objective evaluation.

The subjective evaluation takes a person as an observer, evaluates the image, and aims to truly reflect the visual perception of the person; in this way, generally, an observer compares an evaluation image to be measured with an image photographed by a contrast machine, and evaluates image quality by comparing a plurality of sets of images photographed by a plurality of sets of photographing apparatuses. The manual evaluation mode has low accuracy and poor working efficiency.

The objective evaluation reflects subjective perception of human eyes by means of a mathematical model, and gives an evaluation result based on digital calculation, for example: U.S. patent application publication No. US20190080443A1 discloses extracting a reference image by a computer, evaluating the quality of a target image by comparing the reference image with the target image; also for example: chinese patent application publication No. CN108596901a discloses a machine learning method, and uses a data model to perform image quality assessment; the existing objective evaluation mode is used for summarizing and manufacturing a reference data model through evaluation of learning experts and machines, the consumption of early-stage preparation work is large, and the reference model cannot be adapted to all client requirements due to different client requirements.

Disclosure of Invention

The present invention has been made in view of the above problems, and has as its object to provide an image quality evaluation method, apparatus and system which overcome or at least partially solve the above problems.

The embodiment of the invention provides an image quality evaluation method, which is characterized by comprising the following steps:

Determining the confirmation weight of each selected judgment dimension according to the image type and at least one image attribute of the to-be-evaluated image;

comparing the to-be-evaluated image with the selected comparison image to obtain an evaluation result of the to-be-evaluated image and the comparison image in each evaluation dimension;

and determining the comprehensive evaluation result of the to-be-evaluated image and the comparison image according to the evaluation result of each evaluation dimension and the confirmation weight of each evaluation dimension.

In some optional embodiments, the determining the validation weight of each selected evaluation dimension according to the image type and at least one image attribute of the to-be-evaluated image includes:

Determining initial weights of selected evaluation dimensions according to the image types of the evaluation images to be tested and the selected first image attributes;

And according to at least one second image attribute, adjusting the initial weight of each judgment dimension at least once to obtain the confirmation weight of each judgment dimension.

In some optional embodiments, the image attribute comprises at least one of an image scene, an image content feature, a photographing device, a photographing parameter;

the assessment dimension includes at least one of color, brightness, sharpness, noise, saturation, contrast, sharpness, dynamic range.

In some optional embodiments, comparing the to-be-evaluated image with the selected comparison image to obtain an evaluation result of the to-be-evaluated image and the comparison image in each evaluation dimension, including:

Dividing the region of the image to be evaluated, and extracting image features aiming at the divided image region or region intersection;

Comparing the extracted image features with the image features of the corresponding areas or the area crossing points of the comparison images to obtain comparison results of each corresponding area or area crossing point of the evaluation images to be tested in each evaluation dimension relative to the comparison images;

And obtaining the evaluation results of the to-be-evaluated image and the comparison image in each evaluation dimension according to the comparison result.

In some optional embodiments, when the comparison image is one, the determining, according to the evaluation result of each evaluation dimension and the confirmation weight of each evaluation dimension, the comprehensive evaluation result of the to-be-evaluated image and the comparison image includes:

And carrying out weighted calculation on the evaluation results of each evaluation dimension and the confirmation weights of the corresponding dimensions to obtain the comprehensive evaluation results of the to-be-evaluated image and the comparison image.

In some optional embodiments, when the comparison image is more than one, determining the comprehensive evaluation result of the to-be-evaluated image and the comparison image according to the evaluation result of each evaluation dimension and the confirmation weight of each evaluation dimension includes:

weighting calculation is carried out on the evaluation results of each evaluation dimension and the confirmation weights of the corresponding dimension for each comparison image to respectively obtain the evaluation results of the to-be-evaluated image and each comparison image, and the average value of the evaluation results of the to-be-evaluated image and each comparison image is determined to obtain the comprehensive evaluation results of the to-be-evaluated image and the comparison images; or (b)

Weighting calculation is carried out on the evaluation results of each evaluation dimension and the confirmation weights of the corresponding dimensions aiming at each comparison image, and the evaluation results of the to-be-evaluated image and each comparison image are respectively obtained and used as comprehensive evaluation results of the to-be-evaluated image and the comparison image; or (b)

Determining the average value of the evaluation results of the image to be evaluated and each comparison image according to each evaluation dimension to obtain a comprehensive evaluation result of each evaluation dimension, and carrying out weighted calculation on the comprehensive evaluation result of each evaluation dimension and the confirmation weight of the corresponding dimension to obtain the comprehensive evaluation result of the image to be evaluated and the comparison image; or (b)

And determining the average value of the evaluation results of the image to be evaluated and each comparison image according to each evaluation dimension to obtain the comprehensive evaluation result of each evaluation dimension, wherein the comprehensive evaluation result is used as the comprehensive evaluation result of the image to be evaluated and the comparison image.

In some alternative embodiments, the method further comprises:

and adding evaluation description information based on the comprehensive evaluation results of the evaluation image and the comparison image, and providing the comprehensive evaluation results and the evaluation description information for a user.

The embodiment of the invention also provides an image quality evaluation device, which comprises:

The weight determining module is used for determining the confirmation weight of each selected evaluation dimension according to the image type and at least one image attribute of the evaluation image to be tested;

the first evaluation module is used for comparing the to-be-evaluated image with the selected comparison image to obtain an evaluation result of the to-be-evaluated image and the comparison image in each evaluation dimension;

and the second evaluation module is used for determining the comprehensive evaluation result of the to-be-evaluated image and the comparison image according to the evaluation result of each evaluation dimension and the confirmation weight of each evaluation dimension.

In some optional embodiments, the weight determining module is specifically configured to:

In some optional embodiments, the first evaluation module is specifically configured to:

When the comparison image is one:

Weighting calculation is carried out on the evaluation results of all the evaluation dimensions and the confirmation weights of the corresponding dimensions, so that the comprehensive evaluation results of the to-be-evaluated image and the comparison image are obtained;

when more than one of the comparison images is present:

In some alternative embodiments, the apparatus further comprises:

And the result output module is used for adding the evaluation description information based on the comprehensive evaluation result of the evaluation image and the comparison image and providing the comprehensive evaluation result and the evaluation description information for a user.

The embodiment of the invention also provides an image quality evaluation method, which is characterized by comprising the following steps:

determining initial weights of selected evaluation dimensions according to the image types and the image scenes of the evaluation images to be tested;

According to the image content of the to-be-tested evaluation image, adjusting the initial weight of each evaluation dimension to obtain the confirmation weight of each evaluation dimension;

In some alternative embodiments, before obtaining the validation weight of each evaluation dimension, the method further includes:

And adjusting the initial weight after the first adjustment at least once according to at least one of the type of a camera of shooting equipment used when shooting the evaluation image to be detected and/or the gender of the person and the age of the person included in the image.

The weight determining module is used for determining the initial weight of each selected evaluation dimension according to the image type and the image scene of the evaluation image to be tested; according to the image content of the to-be-tested evaluation image, adjusting the initial weight of each evaluation dimension to obtain the confirmation weight of each evaluation dimension;

The embodiment of the invention also provides an image quality evaluation system, which comprises: the device comprises a testing machine, at least one comparison machine and the image quality evaluation device;

the image quality evaluation device is used for acquiring an evaluation image to be tested from the testing machine and acquiring a comparison image from the comparison machine; and evaluating the obtained evaluation image to be tested and the obtained comparison image to obtain a comprehensive evaluation result of the evaluation image to be tested and the comparison image.

The embodiment of the invention also provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions realize the image quality evaluation method when being executed by a processor.

The embodiment of the invention also provides electronic equipment, which comprises: the image quality evaluation system comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the image quality evaluation method when executing the program.

The technical scheme provided by the embodiment of the invention has the beneficial effects that at least:

Determining the confirmation weight of each evaluation dimension according to the image type and the image attribute of the to-be-evaluated image, comparing the to-be-evaluated image with the comparison image in each evaluation dimension, and obtaining the comprehensive evaluation result of the to-be-evaluated image and the comparison image based on the determined confirmation weight and the evaluation result of each evaluation dimension; from weight determination to image comparison and evaluation result acquisition, each stage can be automatically realized, so that human resource investment is greatly reduced, and the working efficiency and accuracy of image evaluation are greatly improved; the scheme does not need a large amount of learning sample data to manufacture a data model, so that the workload of early-stage preparation work is greatly reduced, different evaluation dimensions and weights can be selected for evaluation according to different requirements, the evaluation dimensions and weights can be selectively determined based on image types and image attributes, namely, different evaluation dimensions can be used for different evaluation images to be tested, and different confirmation weights are used in each evaluation dimension, so that the evaluation requirements of different types and different customer requirements are met, and various different evaluation requirements can be adapted.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.

The technical scheme of the invention is further described in detail through the drawings and the embodiments.

Drawings

The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:

FIG. 1 is a flowchart of an image quality evaluation method according to an embodiment of the present invention;

FIG. 2 is a diagram showing an example of an image storage format according to a first embodiment of the present invention;

FIG. 3 is a diagram showing an example of naming codes for image storage in accordance with a first embodiment of the present invention;

FIG. 4 is a flowchart of an image quality evaluation method according to a second embodiment of the present invention;

FIG. 5 is a flowchart of a first weight confirmation in a second embodiment of the present invention;

FIG. 6 is a flowchart of a second weight confirmation in a second embodiment of the present invention;

FIG. 7 is a diagram showing a contrast focus principle in a second embodiment of the present invention;

FIG. 8 is a diagram illustrating an RGB color model according to a second embodiment of the present invention;

FIG. 9 is a diagram showing an example of images with different contrast ratios according to the second embodiment of the present invention;

FIG. 10 is an exemplary diagram of an image histogram in a second embodiment of the invention;

FIG. 11 is a schematic diagram of an image quality evaluation apparatus according to an embodiment of the present invention;

FIG. 12 is a schematic view of another image quality evaluation apparatus according to an embodiment of the present invention;

fig. 13 is a schematic structural diagram of an image quality evaluation system according to an embodiment of the present invention.

Detailed Description

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

In order to solve the problems that in the prior art, the accuracy and the working efficiency of manual image quality evaluation are low, the evaluation workload is large and different customer demands cannot be met by means of a mathematical model, and the like, the embodiment of the invention provides an image quality evaluation method, which is characterized in that different weights are set according to different image types and different customer demands, and the image quality evaluation is performed based on the weights, so that the automatic intelligent image quality evaluation can be realized, a large amount of preliminary preparation work is not needed, various different evaluation demands can be adapted, and the evaluation accuracy and the working efficiency are high.

Example 1

An embodiment of the present invention provides an image quality evaluation method, the flow of which is shown in fig. 1, comprising the following steps:

Step S101: and determining the confirmation weight of each selected judgment dimension according to the image type and at least one image attribute of the to-be-tested evaluation image.

In this step, the image type is identified, which of the evaluation dimensions are used for evaluation can be determined for different image types, and different weights are used for different image types in each of the evaluation dimensions in the comprehensive evaluation. In addition, one or more weight adjustments can be performed according to the image attributes to determine the final validation weight for comprehensive evaluation.

The image type can comprise at least one of the types of original image type images, algorithm type images and the like; the image attribute comprises at least one of an image scene, image content characteristics, shooting equipment and shooting parameters; the evaluation dimension includes at least one of color, brightness, sharpness, noise, saturation, contrast, sharpness, dynamic range, although not limited to these image types, image attributes, and evaluation dimensions, and may be increased or decreased as desired in a particular application. That is, the dimension of the image judgment may be increased according to the requirement, and the adjustment number of the weight confirmation of the judgment may be increased according to the requirement, for example: the confirmation is performed one or more times according to scene division or according to the sex of the person, according to age, or the like.

Specifically, according to the image type of the evaluation image to be tested and the selected first image attribute, determining the initial weight of each selected evaluation dimension; and according to at least one second image attribute, adjusting the initial weight of each judgment dimension at least once to obtain the confirmation weight of each judgment dimension. For example: the first image attribute may select an image scene and the second image attribute may select image content.

Step S102: and comparing the to-be-evaluated image with the selected comparison image to obtain an evaluation result of the to-be-evaluated image and the comparison image in each evaluation dimension.

In the step, the evaluation image to be tested shot by the testing machine is compared with the comparison image shot by the comparison machine or a plurality of comparison machines. When in comparison, the image to be evaluated can be partitioned, and the partitioned areas can be compared, or the characteristics at the cross points of the areas can be compared. Specifically, region division is carried out on the image to be evaluated, and image feature extraction is carried out on the divided image region or region intersection; comparing the extracted image features with the image features of the corresponding areas or the area crossing points of the comparison images to obtain comparison results of each corresponding area or area crossing point of the evaluation images to be tested in each evaluation dimension relative to the comparison images; and obtaining the evaluation results of the to-be-evaluated image and the comparison image in each evaluation dimension according to the comparison result.

When the images are compared, different modes can be selected for comparison aiming at different judgment dimensions. For example, the color or saturation can be compared by adopting an RGB color model; the brightness is compared, and a histogram mode can be adopted; contrast ratio for sharpness may be performed using a contrast ratio histogram. The comparison of noise may be performed by calculating the signal-to-noise ratio (Signal Noise Ratio, SNR) in the image.

Step S103: and determining the comprehensive evaluation result of the evaluation image to be tested and the comparison image according to the evaluation result of each evaluation dimension and the confirmation weight of each evaluation dimension.

After the evaluation results of the to-be-evaluated image and the comparison image in each evaluation dimension are obtained, comprehensive processing can be performed according to the confirmation weights of each evaluation dimension, so that a comprehensive evaluation result is obtained. In order to obtain a better and accurate evaluation result, images shot by a plurality of comparison machines are selected as comparison images to be compared with the evaluation image to be tested, so that different modes are also available when the evaluation results of all evaluation dimensions are comprehensively sorted.

And when the comparison image is one, weighting calculation is carried out on the evaluation result of each evaluation dimension and the confirmation weight of the corresponding dimension, so as to obtain the comprehensive evaluation result of the evaluation image to be tested and the comparison image. Namely, for the case of a comparison image, an evaluation report can be formed based on the comprehensive evaluation result obtained after the weighting calculation; alternatively, the evaluation result of each evaluation dimension and the confirmation weight of the corresponding dimension may be directly used as the comprehensive evaluation result without performing the weighted calculation to form an evaluation report.

When more than one comparison image is provided, the evaluation results of each comparison image can be firstly arranged, and then the evaluation results of the comparison images are synthesized to obtain a comprehensive evaluation result, so that an evaluation report is formed; the evaluation results of all the evaluation dimensions can be sorted, and comprehensive evaluation results are obtained by integrating the evaluation results of all the evaluation dimensions to form an evaluation report; of course, the comprehensive processing is not required, and the evaluation results of each evaluation dimension of each comparison image and the corresponding confirmation weight are directly used as the comprehensive evaluation results to form an evaluation report. The comprehensive evaluation result of the evaluation image to be tested and the comparison image is obtained by adopting any one or more of the following modes:

mode one:

And for each comparison image, carrying out weighted calculation on the evaluation result of each evaluation dimension and the confirmation weight of the corresponding evaluation dimension to respectively obtain the evaluation result of the to-be-evaluated image and each comparison image, and determining the average value of the evaluation results of the to-be-evaluated image and each comparison image to obtain the comprehensive evaluation result of the to-be-evaluated image and the comparison image.

According to the method, firstly, the evaluation results of the to-be-evaluated image and each comparison image are sorted, for example, when two comparison images 1 and 2 exist, the evaluation results of the to-be-evaluated image and each comparison image 1 in each evaluation dimension and the confirmation weights of the corresponding evaluation dimensions can be subjected to weighted calculation to obtain the evaluation result 1 of the to-be-evaluated image and the comparison image 1, and similarly, the evaluation results of the to-be-evaluated image and each comparison image 2 in each evaluation dimension and the confirmation weights of the corresponding evaluation dimensions are subjected to weighted calculation to obtain the evaluation result 2 of the to-be-evaluated image and the comparison image 2, and then the average value of the evaluation result 1 and the evaluation result 2 is used as the comprehensive evaluation result of the to-be-evaluated image and the comparison image.

Mode two:

And for each comparison image, weighting calculation is carried out on the evaluation result of each evaluation dimension and the confirmation weight of the corresponding dimension, and the evaluation result of the to-be-evaluated image and each comparison image is respectively obtained and used as the comprehensive evaluation result of the to-be-evaluated image and the comparison image.

According to the method, firstly, the evaluation results of the to-be-evaluated image and each comparison image are sorted, for example, when two comparison images 1 and 2 exist, the evaluation results of the to-be-evaluated image and each comparison image 1 in each evaluation dimension and the confirmation weights of the corresponding evaluation dimensions can be subjected to weighted calculation to obtain the evaluation result 1 of the to-be-evaluated image and the comparison image 1, and similarly, the evaluation results of the to-be-evaluated image and each comparison image 2 in each evaluation dimension and the confirmation weights of the corresponding evaluation dimensions are subjected to weighted calculation to obtain the evaluation result 2 of the to-be-evaluated image and the comparison image 2, and then the evaluation result 1 and the evaluation result 2 are used as comprehensive evaluation results of the to-be-evaluated image and the comparison image.

Mode three:

And determining the average value of the evaluation results of the image to be evaluated and each comparison image according to each evaluation dimension to obtain a comprehensive evaluation result of each evaluation dimension, and carrying out weighted calculation on the comprehensive evaluation result of each evaluation dimension and the confirmation weight of the corresponding dimension to obtain the comprehensive evaluation result of the image to be evaluated and the comparison image.

According to the method, firstly, the evaluation results of the images to be evaluated and each comparison image in each evaluation dimension are sorted, for example, when two comparison images 1 and 2 exist, comparison is carried out in the evaluation dimensions 1, 2 and 3 respectively, the average value 1 of the evaluation results of the images to be evaluated and the comparison images 1 and 2 in the evaluation dimension 1 can be calculated, the average value 2 of the evaluation results of the images to be evaluated and the comparison images 1 and 2 in the evaluation dimension 2 is calculated, the average value 3 of the evaluation results of the images to be evaluated and the comparison images 1 and 2 in the evaluation dimension 3 is calculated, and the average value 1, the average value 2 and the average value 3 are weighted with the confirmation weights of the corresponding evaluation dimensions to obtain the comprehensive evaluation results of the images to be evaluated and the comparison images.

Mode four:

According to the method, firstly, the evaluation results of the images to be evaluated and each comparison image in each evaluation dimension are sorted, for example, when two comparison images 1 and 2 exist, comparison is carried out in the evaluation dimensions 1, 2 and 3 respectively, the average value 1 of the evaluation results of the images to be evaluated and the comparison images 1 and 2 in the evaluation dimension 1 can be calculated, the average value 2 of the evaluation results of the images to be evaluated and the comparison images 1 and 2 in the evaluation dimension 2 can be calculated, the average value 3 of the evaluation results of the images to be evaluated and the comparison images 1 and 2 in the evaluation dimension 3 can be calculated, and the average values 1, 2 and 3 can be used as the comprehensive evaluation results of the images to be evaluated and the comparison images.

In some alternative embodiments, the method further comprises:

And adding evaluation description information based on the comprehensive evaluation results of the evaluation image and the comparison image, and providing the comprehensive evaluation results and the evaluation description information for the user. The method can form an evaluation report comprising comprehensive evaluation results and evaluation description information, wherein the evaluation description information can be description information with reddish color, slightly better definition than a comparison machine, larger noise and the like.

The method of the embodiment can evaluate the image quality of the comparison between a tester and one (or more) comparison machines, and the evaluation dimension can comprise definition, color, brightness, noise and other categories, and the evaluation dimension extended from the method comprises contrast, dynamic range, saturation, sharpness and the like. Taking one testing machine and two comparison machines as examples, related images can be stored in the electronic equipment according to the form shown in fig. 2 and 3, as shown in fig. 2, images shot by the testing machine, the comparison machine 1 and the comparison machine 2 are respectively stored in folders with folder names of the testing machine, the comparison machine 1 and the comparison machine 2, the images in each file can be named and numbered according to scenes, and as shown in fig. 2, the corresponding images in the folders of the testing machine, the comparison machine 1 and the comparison machine 2 use the same named and numbered, and when in comparison, the images with the same named and numbered in each file are extracted for comparison. In this way, the extraction and comparison are convenient, and of course, other named storage organization modes can be adopted for storage, and only a corresponding relation is required to be established, so that the corresponding images can be extracted for comparison.

In the method of the embodiment, the confirmation weight of each evaluation dimension is determined according to the image type and the image attribute of the to-be-evaluated image, and after the to-be-evaluated image and the comparison image are compared in each evaluation dimension, the comprehensive evaluation result of the to-be-evaluated image and the comparison image is obtained based on the determined confirmation weight and the evaluation result of each evaluation dimension; from weight determination to image comparison and evaluation result acquisition, each stage can be automatically realized, so that human resource investment is greatly reduced, and the working efficiency and accuracy of image evaluation are greatly improved; the scheme does not need a large amount of learning sample data to manufacture a data model, so that the workload of early-stage preparation work is greatly reduced, different evaluation dimensions and weights can be selected for evaluation according to different requirements, the evaluation dimensions and weights can be selectively determined based on image types and image attributes, namely, different evaluation dimensions can be used for different evaluation images to be tested, and different confirmation weights are used in each evaluation dimension, so that the evaluation requirements of different types and different customer requirements are met, and various different evaluation requirements can be adapted.

Example two

The second embodiment of the present invention provides a specific implementation process of an image quality evaluation method, a flow of which is shown in fig. 4, including the following steps:

step S201: and determining the initial weight of each selected evaluation dimension according to the image type and the image scene of the evaluation image to be tested.

In this embodiment, the description is made with an example of performing weight validation twice, and the practical application is not limited to two times.

Referring to fig. 5, the image type is determined first, and is an original image or an algorithm image. At the time of evaluation, it is determined whether an algorithm class (beauty/HDR/night scene) or an original image class is used for weight confirmation of each subsequent evaluation dimension. In this embodiment, the original image is used for illustration, and the method for evaluating the algorithm image is similar, except that the weight setting of each evaluation dimension may be different.

Then, the first weight confirmation is performed, at this time, it is required to confirm that the image scene, for example, the scene in the image is a day or night scene, and the image is a person or no person, so that it is confirmed that the obtained image scene may be one of a day person, a day no person, a night person or a night no person. And then determining the initial weight of each judgment dimension according to the image scene, namely the proportion of each judgment dimension in judgment.

The first weight confirmation is performed according to the daytime or the night, with or without people, and the first weight confirmation is performed in a large direction according to different large scenes: for example, night scenes and person scenes, the face brightness weight of the person can be emphasized in judgment; the overall color of the picture can be emphasized by the unmanned scene in the daytime during the judgment, and the weight of the overall brightness can be increased; this weight may be a fixed weight acknowledgement.

Step S202: and adjusting the initial weight of each judgment dimension according to the image content of the to-be-tested evaluation image to obtain the confirmation weight of each judgment dimension.

After the first weight confirmation is carried out, the initial weight of each judgment dimension is obtained, and the initial weight can be adjusted once or multiple times. Taking the one-time adjustment as an example, referring to fig. 6, the second weight confirmation is performed, and at this time, it is necessary to identify the content in the image and perform the weight adjustment according to the image content. For example, the image content is identified as building/landscape, animal/character, object/text, or other (such as artistic element, ink painting, etc.), and the second image confirmation is performed according to the identified image content, that is, the weight adjustment is performed after the identification is completed, and when the image content is different, the weight of each judgment dimension may be different. Of course, the image content is not limited to the listed ones. The secondary weight validation may be modified accordingly based on customer demand or primary direction objectives. If the customer hopes that the scenery color saturation is more bright, the judgment weight of the scenery color module is increased.

Step S203: and comparing the to-be-evaluated image with the selected comparison image to obtain an evaluation result of the to-be-evaluated image and the comparison image in each evaluation dimension.

In this embodiment, when comparing images to obtain an evaluation result, the evaluation result is taken as an example to describe the comparison score, and in practical application, the comparison score is not limited, different evaluation parameters can be adopted, and a corresponding evaluation result can be output.

And after the weight confirmation of each judgment dimension is finished, starting to carry out the comparison scoring of the images. The score for each scene comparison may be set to a range of 1-10 or 1-100 scores.

Taking color dimension as an example, pixel point RGB information of an evaluation image to be compared and a comparison image can be extracted, and the extraction mode can correspondingly distinguish people/unmanned scenes: aiming at an unmanned scene, the image can be equally divided into N (preferably N is more than or equal to 16) modules, RGB information of pixel points is extracted at equal-division intersection points, RGB numerical comparison is carried out, color dimension scores are obtained, and overall image color evaluation results are comprehensively obtained. Accordingly, a judgment description of color dimension, such as color cast, can be obtained.

For a scene with a person, the RGB information of the pixel points can be extracted correspondingly from the face of the person, and the RGB information of the pixel points can be extracted from other parts of the image. And carrying out RGB numerical comparison on the pixel points extracted from the two parts to obtain the score of the color dimension, and comprehensively obtaining the overall color evaluation result of the image. From this, a judgment description of the color dimension, such as overall color cast or partial color cast, can also be obtained.

The contrast of the saturation and brightness dimension is similar to the color dimension, and the pixel point RGB information of the evaluation image to be compared and the pixel point RGB information of the contrast image are extracted for comparison. Taking the judgment of brightness dimension as an example, if all the extracted G values of the evaluation image to be tested are larger than the G value of the contrast image, the brightness of the evaluation image to be tested is judged to be larger than the brightness of the contrast image.

The above-mentioned comparison and judgment of saturation dimension can be carried out by referring to or directly importing RGB color model, the judgment of brightness dimension can be carried out by adopting image histogram or G value in RGB, the higher a certain point on the histogram is, the more points under the brightness, and in addition, the judgment of contrast can also be carried out by adopting image histogram.

The contrast judgment of definition dimension can be performed on the whole image, the comparison of partial areas in the image can be performed, the contrast of definition can be performed by adopting a contrast histogram, as shown in fig. 7, the contrast focusing basic principle is performed on the eye area in the image, the position of black and white boundary of the black eye bead and white eye ball in the eye can be seen in the leftmost column in fig. 7, the second column on the left is an AF comparison range, the third column is the calculated inverse difference in the comparison range, the rightmost column is a contrast histogram, and the definition of the lowest one in the 7 images of the eye part in the left column is the best as can be seen in fig. 3.

The contrast judgment of noise dimension can be compared by calculating the signal-to-noise ratio SNR in the image. Generally, a flat area or a dark area in an image is selected as a calculation sample area, an algorithm formula is introduced in the calculation, and the SNR value can be calculated by using the existing algorithm.

Step S204: and determining the comprehensive evaluation result of the evaluation image to be tested and the comparison image according to the evaluation result of each evaluation dimension and the confirmation weight of each evaluation dimension.

The arrangement of the evaluation results is described with reference to the correlation in step S103 in the first embodiment. Taking a testing machine, a lower limit machine and an upper limit machine as examples, comparing an evaluation image to be tested shot by the testing machine with a comparison image 1 shot by the lower limit machine to obtain comparison scores of the evaluation image to be tested and the comparison image 1 in each evaluation dimension, comparing the evaluation image to be tested shot by the testing machine with a comparison image 2 shot by the lower limit machine to obtain comparison scores of the evaluation image to be tested and the comparison image 2 in each evaluation dimension, and then adopting at least one mode in step S103 to carry out summarizing and sorting. In the averaging process, a weighted average may also be performed, for example, the weighted ratio of the comparison scores of the testing machine and the upper limit machine to the comparison scores of the testing machine and the lower limit machine is 6:4.

Of course, the comprehensive scores may not be calculated, and the comparison scores of the to-be-measured evaluation image and the comparison images in each evaluation dimension may be summarized together to form an evaluation report together with the confirmation weights of each evaluation dimension. Language descriptions may also be added to the assessment report, for example: the whole picture is redder, the definition is slightly worse than the upper limit machine and is slightly better than the lower limit machine, the brightness is uniform with the effect of the contrast machine, the whole noise is more, and the like.

In some optional embodiments, before the confirmation weight of each evaluation dimension is obtained in step S202, at least one adjustment may be further performed, and the initial weight after the first adjustment may be adjusted at least one time according to at least one of a camera type of a photographing device used when the image to be evaluated is photographed and/or a person gender and a person age included in the image, and the confirmation weight of each evaluation dimension is obtained after the adjustment. Such as: when the weight adjustment is confirmed, the judgment of the evaluation lens, such as a wide-angle lens, a main shooting lens, a macro lens, a front shooting lens, a tele shooting lens and the like, can be added, and the judgment weight is correspondingly and pertinently modified according to the characteristics of the lens.

The RGB color model in the embodiment of the present invention may be shown in fig. 8, where the space of the RGB color model is a unit cube, each point in the region of the cube corresponds to a different color, that is, each point from the origin may correspond to a different color, and is represented by a vector from the origin to the point, three coordinate values are respectively the ratio of three colors of red (R), green (G), and blue (B), black is fixed at the origin, and white is fixed at the points (1, 1). In digital systems this unit space is discretized, typically with each component represented by an 8-bit integer, so that each pixel requires a 24-bit representation. When the color model is used for judgment, an RGB color model can be used, and the judgment can be performed based on YUV and HSV color models.

The contrast focusing principle mentioned in the embodiment of the invention is briefly described as follows: the most clear point of the general image is the point with the largest contrast, the camera drives the lens, the focusing point is changed along the axis pointing to the shot object, the image is acquired on each focusing point, the image acquired on each focusing point is digitalized, similar to point-by-point scanning, the digitalized image is actually an integer matrix and is transmitted to the image processor, then the inverse quantity is calculated, the focusing point with the largest contrast is screened out by contrast, the lens is driven, the focus is placed on the focusing point with the largest inverse quantity, namely the correct focus is obtained, and whether focusing is completed is determined according to the value with the largest inverse quantity. When reflected on the screen of the electronic equipment, the process from blurring to clear to blurring and finally clear is a process of 'pulling bellows'. This determination can achieve very high focusing accuracy, as well as practical use. This focusing technique is called contrast focus. The contrast focus process is essentially a simple maximization process, and is a relatively simple matter to implement by programming, the basic aim of which is to: focusing is accomplished with a minimum number of samples.

The contrast and histogram mentioned in the embodiments of the present invention are briefly described as follows: a histogram may describe the case of image contrast, which is a measure of the difference in brightness between a bright area and a dark area in a scene of an image. A broad histogram may reflect that an image has a higher contrast, whereas a narrower histogram reflects that an image has a lower contrast. This difference in contrast may be due to a combination of lighting conditions and other factors. The image photographed in the foggy weather has low contrast; images taken under some intense light have a high contrast. The left half is an example of a low contrast image and the right half is an example of a high contrast image as shown in fig. 9.

An image histogram is a histogram for representing the distribution of luminance in an image, and is given by sharing several pixels at a certain luminance or a certain range of luminance in the image, i.e. counting the number of pixels at a certain luminance in a single image. As shown in fig. 10, the horizontal axis represents the luminance value of 0-255, the vertical axis represents the number of pixels corresponding to the luminance in the image, the left side of the histogram is pure black, the right side is pure white, and the peak of the histogram shown in fig. 10 is at the middle left position, which indicates that there are many dark gray or dark color parts in the picture. .

Based on the same inventive concept, the embodiments of the present invention further provide an image quality evaluation device, which may be provided in an electronic apparatus, for example: terminal equipment, server, etc., the structure of which is shown in fig. 11, comprising:

The weight determining module 11 is configured to determine a confirmation weight of each selected evaluation dimension according to the image type and at least one image attribute of the evaluation image to be tested.

The first evaluation module 12 is configured to compare the to-be-evaluated image with the selected comparison image, so as to obtain an evaluation result of the to-be-evaluated image and the comparison image in each evaluation dimension.

And the second evaluation module 13 is used for determining the comprehensive evaluation result of the to-be-evaluated image and the comparison image according to the evaluation result of each evaluation dimension and the confirmation weight of each evaluation dimension.

Optionally, the weight determining module 11 is specifically configured to determine an initial weight of each selected evaluation dimension according to an image type of the evaluation image to be tested and the selected first image attribute; and according to at least one second image attribute, adjusting the initial weight of each judgment dimension at least once to obtain the confirmation weight of each judgment dimension.

Optionally, the first evaluation module 12 is specifically configured to divide a region of an image to be evaluated, and extract image features of the divided image region or region intersection; comparing the extracted image features with the image features of the corresponding areas or the area crossing points of the comparison images to obtain comparison results of each corresponding area or area crossing point of the evaluation images to be tested in each evaluation dimension relative to the comparison images; and obtaining the evaluation results of the to-be-evaluated image and the comparison image in each evaluation dimension according to the comparison result.

Optionally, the first evaluation module 12 is specifically configured to:

When the comparison image is one:

when more than one of the comparison images is present:

In some alternative embodiments, the apparatus further comprises:

and a result output module 14 for adding the evaluation description information based on the comprehensive evaluation result of the to-be-evaluated image and the comparison image, and providing the comprehensive evaluation result and the evaluation description information to the user.

The embodiment of the invention also provides another image quality evaluation device, the structure of which is shown in fig. 12, comprising:

the weight determining module 21 is configured to determine an initial weight of each selected evaluation dimension according to an image type and an image scene of the evaluation image to be tested; and adjusting the initial weight of each judgment dimension according to the image content of the to-be-tested evaluation image to obtain the confirmation weight of each judgment dimension.

The first evaluation module 22 is configured to compare the image to be evaluated with the selected comparison image, so as to obtain an evaluation result of the image to be evaluated and the comparison image in each evaluation dimension.

And the second evaluation module 23 is configured to determine a comprehensive evaluation result of the to-be-evaluated image and the comparison image according to the evaluation result of each evaluation dimension and the confirmation weight of each evaluation dimension.

The weight determining module 21 is specifically configured to perform at least one adjustment on the initial weight after the first adjustment according to at least one of a camera type of a shooting device used when the evaluation image to be tested is shot and/or a gender of a person and an age of the person included in the image.

Based on the same inventive concept, an embodiment of the present invention further provides an image quality evaluation system, where the system structure is as shown in fig. 13, and includes: a testing machine 1, at least one comparing machine 2 and an image quality evaluating device 3;

an image quality evaluation device 3 for acquiring an evaluation image to be tested from the testing machine 1 and acquiring a comparison image from the comparison machine 2; and evaluating the obtained evaluation image to be tested and the obtained comparison image to obtain a comprehensive evaluation result of the evaluation image to be tested and the comparison image.

The embodiment of the invention also provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the image quality evaluation method is realized when the computer executable instructions are executed by a processor.

The embodiment of the invention also provides electronic equipment, which comprises: the image quality evaluation system comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor realizes the image quality evaluation method when executing the computer program.

The specific manner in which the various modules perform the operations in relation to the systems and apparatus of the embodiments described above have been described in detail in relation to the embodiments of the method and will not be described in detail herein.

In the prior art, subjective image quality evaluation is generally performed by using a comparison machine to perform image quality evaluation manually and manually comparing multiple groups of pictures of multiple shooting devices. When the machine learning mode is adopted for evaluation, a large amount of sample data needs to be summarized to manufacture a reference data model, the consumption of the early-stage preparation work is large, the reference model can not be adapted to all client requirements, and the application range is narrow. The method of the invention can replace the artificial judgment mechanism, quantify the judgment mode, avoid the inaccuracy of judgment caused by long artificial image viewing time, improve the working efficiency and reduce a great deal of time and labor cost required by artificial image viewing judgment.

The image quality evaluation method, the system and the device can compare the image effects of a plurality of comparison machines and give scores for users to refer to or directly use. No extensive expert or machine is required for standard modeling. All the judgment standards are quantized, the image quality can be objectively judged by using an algorithm and image information, aesthetic fatigue generated after overlong human looking is replaced, and errors can be generated in judgment. In addition, the image may be divided into areas, and the image quality effect may be compared by quantitative evaluation in a plurality of dimensions from a part of the pixel points or a part of the image area.

When the image quality evaluation method, the system and the device evaluate the image quality, the image picture is halved mainly through Exchangeable image file (exchange IMAGE FILE EXIF) information of the image, pixel point information of crossing points of all areas or images of a certain area are intercepted to perform evaluation comparison, and the algorithm, the existing functions and principles can be combined to perform calculation comparison. EXIF refers to detailed information of the image, which may include image size, exposure time, ISO rate, etc., and this scheme may also be extended to separate analysis of each small item, such as to analyze brightness conditions of all images, i.e. to extract histogram information of all images for contrast analysis, and to analyze noise conditions of all images, i.e. to extract flat area or dark area pictures of all images for SNR calculation contrast. More refined dimensions such as focusing, sharpness, saturation and the like can be extracted and analyzed independently, and the method is not limited to overall analysis and evaluation.

Unless specifically stated otherwise, terms such as processing, computing, calculating, determining, displaying, or the like, may refer to an action and/or process of one or more processing or computing systems, or similar devices, that manipulates and transforms data represented as physical (e.g., electronic) quantities within the processing system's registers or memories into other data similarly represented as physical quantities within the processing system's memories, registers or other such information storage, transmission or display devices. Information and signals may be represented using any of a variety of different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips that may be referenced throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.

It should be understood that the specific order or hierarchy of steps in the processes disclosed are examples of exemplary approaches. Based on design preferences, it is understood that the specific order or hierarchy of steps in the processes may be rearranged without departing from the scope of the present disclosure. The accompanying method claims present elements of the various steps in a sample order, and are not meant to be limited to the specific order or hierarchy presented.

In the foregoing detailed description, various features are grouped together in a single embodiment for the purpose of streamlining the disclosure. This method of disclosure is not to be interpreted as reflecting an intention that the claimed embodiments of the subject matter require more features than are expressly recited in each claim. Rather, as the following claims reflect, invention lies in less than all features of a single disclosed embodiment. Thus the following claims are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate preferred embodiment of this invention.

Those of skill would further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present disclosure.

The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. The ASIC may reside in a user terminal. The processor and the storage medium may reside as discrete components in a user terminal.

For a software implementation, the techniques described in this disclosure may be implemented with modules (e.g., procedures, functions, and so on) that perform the functions described herein. These software codes may be stored in memory units and executed by processors. The memory unit may be implemented within the processor or external to the processor, in which case it can be communicatively coupled to the processor via various means as is known in the art.

The foregoing description includes examples of one or more embodiments. It is, of course, not possible to describe every conceivable combination of components or methodologies for purposes of describing the aforementioned embodiments, but one of ordinary skill in the art may recognize that many further combinations and permutations of various embodiments are possible. Accordingly, the embodiments described herein are intended to embrace all such alterations, modifications and variations that fall within the scope of the appended claims. Furthermore, as used in the specification or claims, the term "comprising" is intended to be inclusive in a manner similar to the term "comprising," as interpreted when employed as a transitional word in a claim. Furthermore, any use of the term "or" in the specification of the claims is intended to mean "non-exclusive or".

Claims

1. An image quality evaluation method, comprising:

Selecting an evaluation dimension according to the image type of the evaluation image to be tested;

According to at least one second image attribute, adjusting the initial weight of each judgment dimension at least once to obtain the confirmation weight of each judgment dimension;

the image attribute comprises at least one of an image scene, image content characteristics, shooting equipment and shooting parameters;

Comparing the to-be-evaluated image with the selected comparison image to obtain an evaluation result of the to-be-evaluated image and the comparison image in each evaluation dimension; the evaluation dimension includes at least one of color, brightness, sharpness, noise, saturation, contrast, sharpness, dynamic range;

2. The method of claim 1, wherein comparing the image to be evaluated with the selected comparison image to obtain an evaluation result of the image to be evaluated and the comparison image in each evaluation dimension, comprises:

3. The method of claim 1, wherein when the comparison image is one, the determining the comprehensive evaluation result of the to-be-evaluated image and the comparison image according to the evaluation result of each evaluation dimension and the confirmation weight of each evaluation dimension includes:

4. The method of claim 1, wherein when the comparison image is more than one, the determining the comprehensive evaluation result of the to-be-evaluated image and the comparison image according to the evaluation result of each evaluation dimension and the confirmation weight of each evaluation dimension comprises:

5. The method of any one of claims 1-4, further comprising:

6. An image quality evaluation device, comprising:

The weight determining module is used for selecting an evaluation dimension according to the image type of the evaluation image to be tested; determining initial weights of selected evaluation dimensions according to the image types of the evaluation images to be tested and the selected first image attributes; according to at least one second image attribute, the image attribute comprises at least one of an image scene, image content characteristics, shooting equipment and shooting parameters; at least one time of adjustment is carried out on the initial weight of each judgment dimension to obtain the confirmation weight of each judgment dimension;

The first evaluation module is used for comparing the to-be-evaluated image with the selected comparison image to obtain an evaluation result of the to-be-evaluated image and the comparison image in each evaluation dimension, wherein the evaluation dimensions comprise at least one of color, brightness, definition, noise, saturation, contrast, sharpness and dynamic range;

7. The apparatus as recited in claim 6, further comprising:

8. An image quality evaluation method, comprising:

adjusting the initial weight of each evaluation dimension according to the image content of the evaluation image to be tested, and adjusting the initial weight after the initial adjustment at least once according to the type of a camera of shooting equipment used when the evaluation image to be tested is shot and/or at least one of the sex of the person and the age of the person included in the image to obtain the confirmation weight of each evaluation dimension;

9. An image quality evaluation device, comprising:

The weight determining module is used for selecting an evaluation dimension according to the image type of the evaluation image to be tested; determining initial weights of selected evaluation dimensions according to the image types and the image scenes of the evaluation images to be tested; adjusting the initial weight of each evaluation dimension according to the image content of the evaluation image to be tested, and adjusting the initial weight after the initial adjustment at least once according to the type of a camera of shooting equipment used when the evaluation image to be tested is shot and/or at least one of the sex of the person and the age of the person included in the image to obtain the confirmation weight of each evaluation dimension;

The first evaluation module is used for comparing the to-be-evaluated image with the selected comparison image to obtain an evaluation result of the to-be-evaluated image and the comparison image in each evaluation dimension; the evaluation dimension includes at least one of color, brightness, sharpness, noise, saturation, contrast, sharpness, dynamic range;

10. An image quality evaluation system, comprising: a testing machine, at least one contrast machine and an image quality assessment device according to any one of claims 6 to 7, 9;

11. A computer storage medium having stored therein computer executable instructions which when executed by a processor implement the image quality assessment method of any one of claims 1-5, 8.

12. An electronic device, comprising: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the image quality assessment method according to any one of claims 1-5, 8 when the program is executed.