Image quality detection method and device and electronic equipment

Abstract

The invention provides an image quality detection method and device and electronic equipment, and relates to the technical field of image processing, in particular to the application fields of computer vision, digital media processing, visual content design, e-commerce product display and the like. The specific implementation scheme is as follows: segmenting a target image to obtain a plurality of image regions; taking each image area in the plurality of image areas as a target area, and respectively obtaining quality characterization parameters for the target area under the plurality of quality detection dimensions; wherein the plurality of quality detection dimensions comprise at least one of a flocculent area detection dimension and a definition detection dimension; and obtaining an overall quality detection result for the target image based on the quality characterization parameters.

Description

Technical Field

[0001] The present disclosure relates to the field of image processing technology, and in particular to application fields such as computer vision, digital media processing, visual content design, and e-commerce product display, and specifically to an image quality detection method, device, and electronic equipment. Background Art

[0002] When designing visual content (e.g., advertising design), in order to highlight the main object, it is usually necessary to place the main object in a completely new background. Therefore, the main object in the original image needs to be cut out. However, due to the imperfection of existing cutout technology, when dealing with complex image backgrounds and details, the image quality of the cutout result may be negatively affected, making the cutout result unusable. Therefore, it is necessary to perform quality inspection on the cutout result to obtain the overall quality inspection result for the cutout result. Summary of the invention

[0003] The present disclosure provides an image quality detection method, device and electronic equipment.

[0004] According to a first aspect of the present disclosure, there is provided an image quality detection method, comprising:

[0005] Segment the target image to obtain multiple image regions;

[0006] Taking each of the multiple image regions as a target region, respectively obtaining quality characterization parameters for the target region under multiple quality detection dimensions; wherein the multiple quality detection dimensions include at least one of a flocculent region detection dimension and a clarity detection dimension;

[0007] Based on the quality characterization parameters, the overall quality detection result of the target image is obtained.

[0008] According to a second aspect of the present disclosure, there is provided an image quality detection device, comprising:

[0009] An image segmentation unit, used for segmenting a target image to obtain multiple image regions;

[0010] A parameter acquisition unit, configured to take each of the plurality of image regions as a target region, and obtain quality characterization parameters for the target region under a plurality of quality detection dimensions, respectively; wherein the plurality of quality detection dimensions include at least one of a flocculent region detection dimension and a clarity detection dimension;

[0011] The result acquisition unit is used to obtain the overall quality detection result for the target image based on the quality characterization parameter.

[0012] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0013] at least one processor;

[0014] a memory communicatively coupled to the at least one processor;

[0015] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method provided in the first aspect of the present disclosure.

[0016] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method provided according to the first aspect of the present disclosure.

[0017] According to a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program, which implements the method provided according to the first aspect of the present disclosure when executed by a processor.

[0018] The use of the present disclosure can improve the reliability of the overall quality detection results for the target image.

[0019] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

[0021] Figure 1 A flowchart of an image quality detection method provided by an embodiment of the present disclosure;

[0022] Figure 2 A diagram illustrating a target image segmentation method provided in an embodiment of the present disclosure;

[0023] Figure 3 A schematic diagram of an application scenario of an image quality detection method provided by an embodiment of the present disclosure;

[0024] Figure 4 A schematic structural block diagram of an image quality detection device provided in an embodiment of the present disclosure;

[0025] Figure 5 A schematic structural block diagram of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0026] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0027] As described in the background art, when designing visual content (e.g., advertising design), in order to highlight the main object, it is usually necessary to place the main object in a completely new background. Therefore, it is necessary to perform a cutout process on the main object in the original image. However, due to the imperfection of existing cutout technology, when processing complex image backgrounds and details, it may have a negative impact on the image quality of the cutout result, thereby rendering the cutout result unusable.

[0028] Specifically, the inventors have found that when processing complex image backgrounds and details, flocculent areas and / or poor clarity often occur, making the cutout results unusable. Currently, the quality of the cutout results is mainly tested manually, which has certain subjective factors and human bias. Therefore, the reliability of the overall quality test results of the cutout results cannot be ensured.

[0029] In view of the above problems, the embodiments of the present disclosure provide an image quality detection method, which can be applied to electronic devices. The electronic device can be a server or a terminal device. Here, the terminal device can be a workbench, a mainframe computer, a conventional computer (for example, a desktop computer, a laptop computer, a car computer, etc.), a personal digital processing unit or other similar computing devices. Figure 1 The flowchart diagram shown in the figure illustrates an image quality detection method provided by an embodiment of the present disclosure. It should be noted that although the logical order is shown in the flowchart diagram, in some cases, the steps shown or described in the flowchart can also be performed in other orders.

[0030] Step S101, segmenting the target image to obtain multiple image regions.

[0031] The target image may be a new image obtained by performing a cutout process on the main object in the original image, and then placing the main object in a new background, that is, a cutout result. Here, the main object may be an object that needs to be highlighted, and the embodiments of the present disclosure do not limit this.

[0032] Step S102: taking each of the multiple image regions as a target region, and obtaining quality characterization parameters for the target region under multiple quality detection dimensions.

[0033] The multiple quality detection dimensions include at least one of a flocculent area detection dimension and a clarity detection dimension.

[0034] Based on this, it can be understood that in the embodiment of the present disclosure, when multiple quality detection dimensions include a flocculent region detection dimension, a quality characterization parameter for the target region under the flocculent region detection dimension can be obtained, and it can be determined whether the target region includes a flocculent region; when multiple quality detection dimensions include a clarity detection dimension, a quality characterization parameter for the target region under the clarity detection dimension can be obtained, and it can be determined whether the target region has an unclear problem. The flocculent region can be a region with a loose, fibrous or clumping structure.

[0035] Step S103: obtaining an overall quality detection result for the target image based on the quality characterization parameter.

[0036] The overall quality detection result can be used to indicate whether the image quality of the target image is qualified.

[0037] The image quality detection method provided by the embodiment of the present disclosure can be used to segment the target image to obtain multiple image areas, and each of the multiple image areas is used as the target area, and quality characterization parameters for the target area under multiple quality detection dimensions are obtained respectively, and the multiple quality detection dimensions include at least one of the flocculent area detection dimension and the clarity detection dimension, and then based on the quality characterization parameters, the overall quality detection result for the target image is obtained. In this way, compared with the manual detection method in the prior art, on the one hand, subjective factors and human bias are put aside, so the reliability of the overall quality detection result for the target image can be improved; on the other hand, the target image can be relatively comprehensively and accurately detected at least from the flocculent area detection dimension and / or the clarity detection dimension, which can further improve the reliability of the overall quality detection result for the target image.

[0038] In some optional implementations, step S101, that is, "segmenting the target image to obtain multiple image regions" may include:

[0039] The target image is segmented according to a plurality of preset segmentation methods to obtain a plurality of image regions.

[0040] The multiple preset cutting methods are used to indicate that the target image is cut according to different cutting methods to obtain multiple image regions, so as to make the cutting edge of the reference region be in a non-edge position in at least one other region of the multiple image regions except the reference region. Here, the reference region is any image region of the multiple image regions.

[0041] Exemplarily, it is assumed that there is a first preset cutting method and a second preset cutting method.

[0042] Please combine Figure 2 According to the first preset cutting method, the target image A is cut with the horizontal median line of the target image A as the cutting line, and the image area A1 and the image area A2 can be obtained; according to the second preset cutting method, the target image A is cut with the vertical median line of the target image A as the cutting line, and the image area A3 and the image area A4 can be obtained. Among them, the cutting edges a11-a12 of the image area A1 are in non-edge positions in both the image area A3 and the image area A4; the cutting edges a11-a12 of the image area A2 are in non-edge positions in both the image area A3 and the image area A4; the cutting edges a21-a22 of the image area A3 are in non-edge positions in both the image area A1 and the image area A2; the cutting edges a21-a22 of the image area A4 are in non-edge positions in both the image area A1 and the image area A2.

[0043] Through the above method, in the embodiment of the present disclosure, the target image can be segmented according to a plurality of preset segmentation methods respectively to obtain a plurality of image regions, and it can be ensured that the cutting edge of the reference region (the reference region is any image region among the plurality of image regions) is in a non-edge position in at least one other region of the plurality of image regions except the reference region. In this way, on the one hand, by setting a plurality of preset segmentation methods, different business needs or application scenarios can be flexibly adapted; on the other hand, by ensuring that the cutting edge of the reference region is in a non-edge position in at least one other region of the plurality of image regions except the reference region, the probability of false detection due to cutting edge noise can be reduced, thereby further improving the reliability of the overall quality detection result for the target image.

[0044] In some optional implementations, the step of “respectively obtaining quality characterization parameters for the target area under multiple quality detection dimensions” in step S102 may include:

[0045] Segment the target area to obtain multiple image grids;

[0046] Taking each image grid in the multiple image grids as a target grid, obtaining a single grid quality detection result for the target grid under the target detection dimension, so as to obtain a plurality of single grid quality detection results corresponding to the multiple image grids under the target detection dimension; wherein the target detection dimension is any quality detection dimension in the multiple quality detection dimensions;

[0047] Based on multiple single-grid quality detection results corresponding one-to-one to multiple image grids in the target detection dimension, a quality characterization parameter for the target area in the target detection dimension is obtained.

[0048] In one example, the target area may be segmented according to a preset grid size to obtain a plurality of image grids.

[0049] Exemplarily, it is assumed that there is a target area B, and the size of the target area B is (w, h), that is, the width of the target area B is w pixels, and the length is h pixels.

[0050] The size of each of the multiple image grids can be preset as (p, q), that is, the width of each of the multiple image grids is p pixels and the height is q pixels. Here, p and q can be constants, for example, 10. In this way, by segmenting the target area B, m×n image grids can be obtained, specifically, m=w / p, n=h / q.

[0051] Further, in an example, the target detection dimension is a flocculent region detection dimension. After segmenting the target region to obtain a plurality of image grids, and taking each of the plurality of image grids as a target grid, “obtaining a single grid quality detection result for the target grid under the target detection dimension” may include:

[0052] Get the transparency representation value of each pixel in the target grid;

[0053] Based on the transparency representation value of each pixel in the target grid, the transparent pixel ratio of the target grid is obtained;

[0054] Based on the transparent pixel ratio of the target grid, the single-grid quality detection result of the target grid in the flocculent area detection dimension is obtained to characterize whether the target grid is a semi-transparent grid.

[0055] In a specific example, the transparency representation value of each pixel in the target grid can be obtained through the pixel parameter acquisition function provided by the image processing library (e.g., Pillow). The transparency representation value is in the interval [0, 255]. Here, 0 is used to represent a fully transparent state, and 255 is used to represent an opaque state.

[0056] Continuing with the previous example, assume that there are pixels (x, y) in the target grid, where x∈[1, p] and y∈[1, q].

[0057] After obtaining the transparency representation value of the pixel point (x, y), when the transparency representation value of the pixel point (x, y) is less than the threshold value ε, the pixel point (x, y) can be confirmed as a transparent pixel, and the transparent state of the pixel point (x, y) is represented by 1; when the transparency representation value of the pixel point (x, y) is greater than or equal to the threshold value ε, the pixel point (x, y) is confirmed as a non-transparent pixel, and the transparent state of the pixel point (x, y) is represented by 0. Wherein, ε can be a constant, for example, 200.

[0058] Finally, the transparent state a of the pixel (x, y) can be obtained. xy ∈{1,0}.

[0059] After obtaining the transparency state of each pixel in the target grid, the transparent pixel ratio of the target grid can be obtained based on the transparency state of each pixel in the target grid. This process can be characterized as:

[0060]

[0061] Among them, β ij It is used to represent the transparent pixel ratio of the target grid at position (i, j) in the m×n image grid; p is used to represent the width of the target grid; q is used to represent the height of the target grid; (x, y) is used to represent the xth horizontal and yth vertical pixel in the target grid; a xy Used to represent the transparency state of the pixel (x, y).

[0062] Obviously, β ij ∈[0,1]. Further, in the above example, ij When the value is less than the threshold λ and greater than the threshold τ, the target grid is confirmed as a semi-transparent grid, and 1 is used to represent the single-grid quality detection result of the target grid under the flocculent area detection dimension; ij When the value is greater than or equal to the threshold λ, or less than or equal to the threshold τ, the target grid is confirmed as a non-translucent grid, and the single-grid quality detection result of the target grid under the flocculent area detection dimension is represented by 0. Among them, λ can be a constant, for example, 0.9; τ can be a constant, for example, 0.2.

[0063] Finally, the single-grid quality detection result β for the target grid can be obtained. ij ∈{1,0}.

[0064] Based on the above example, “obtaining a quality characterization parameter for a target area in the target detection dimension based on a plurality of single-grid quality detection results corresponding one-to-one to a plurality of image grids in the target detection dimension” may include:

[0065] Based on multiple single-grid quality detection results corresponding to multiple image grids in the flocculent area detection dimension, the proportion of semi-transparent grids in the target area is obtained as a quality characterization parameter for the target area in the flocculent area detection dimension.

[0066] Continuing with the above example, it is assumed that when executing step S101 to segment the target image, the target image A is cut according to the first preset cutting method with the horizontal median line of the target image A as the cutting line, and image area A1 and image area A2 are obtained; according to the second preset cutting method, the target image A is cut with the vertical median line of the target image A as the cutting line, and image area A3 and image area A4 are obtained. Then, the above process of obtaining the quality characterization parameters for the target area under the flocculent area detection dimension can be characterized as:

[0067]

[0068] Among them, ω z It is used to characterize the quality characterization parameters of the target area z under the flocculent area detection dimension; m is used to characterize the number of image grids in each row of the target area z; n is used to characterize the number of image grids in each column of the target area z; (i, j) is used to characterize the i-th horizontal and j-th vertical pixel points in the target area z; β ij It is used to characterize the single-grid quality detection result of the target grid located at the position (i, j) in the m×n image grids included in the target area z under the flocculent area detection dimension.

[0069] In the above example, the transparency representation value of each pixel in the target grid can be obtained, and based on the transparency representation value of each pixel in the target grid, the transparent pixel ratio of the target grid is obtained, and then based on the transparent pixel ratio of the target grid, the single-grid quality detection result for the target grid under the flocculent region detection dimension is obtained to characterize whether the target grid is a semi-transparent grid. Thereafter, based on multiple single-grid quality detection results corresponding to multiple image grids under the flocculent region detection dimension, the semi-transparent grid ratio of the target area can be obtained as the quality representation parameter for the target area under the flocculent region detection dimension. In this way, on the one hand, the evaluation granularity of the quality representation parameter for the target area under the flocculent region detection dimension can be reduced to the pixel granularity, thereby improving the accuracy of the quality representation parameter for the target area under the flocculent region detection dimension; on the other hand, the transparency representation value is used as an important visual feature, which can effectively identify the semi-transparent grid in the target area, so as to further improve the accuracy of the quality representation parameter for the target area under the flocculent region detection dimension.

[0070] In another example, the target detection dimension is a clarity detection dimension. After segmenting the target area to obtain a plurality of image grids, and taking each of the plurality of image grids as a target grid, “obtaining a single grid quality detection result for the target grid under the target detection dimension” may include:

[0071] Calculate the clarity of the target grid;

[0072] Based on the clarity of the target grid, a single-grid quality detection result for the target grid under the clarity detection dimension is obtained to characterize whether the target grid is a non-clear grid.

[0073] In a specific example, the clarity of the target grid can be calculated by an image clarity algorithm, such as the Laplace algorithm, the Sobel algorithm, etc. The clarity of the target grid (specifically, the target grid at position (i, j) in the m×n image grids) can be represented as: γ ij ∈(0,1). Here, i∈[1,m], j∈[1,n].

[0074] Furthermore, in the above example, ij When it is less than the threshold value θ, the target grid is confirmed as a non-clear grid, and 1 is used to represent the single-grid quality detection result of the target grid under the clarity detection dimension; ij When the value is greater than or equal to the threshold value θ, the target grid is confirmed as a clear grid, and the single grid quality detection result of the target grid under the clarity detection dimension is represented by 0. θ can be a constant, for example, 0.8.

[0075] Finally, the single-grid quality detection result γ for the target grid under the clarity detection dimension can be obtained. ij ∈{1,0}.

[0076] Based on the above example, “obtaining a quality characterization parameter for a target area in the target detection dimension based on a plurality of single-grid quality detection results corresponding one-to-one to a plurality of image grids in the target detection dimension” may include:

[0077] Based on multiple single-grid quality detection results corresponding one-to-one to multiple image grids under the clarity detection dimension, the proportion of non-clear grids in the target area is obtained as a quality characterization parameter for the target area under the clarity detection dimension.

[0078] Continuing with the above example, it is assumed that when executing step S101 to segment the target image, the target image A is cut according to the first preset cutting method with the horizontal median line of the target image A as the cutting line, and image area A1 and image area A2 are obtained; according to the second preset cutting method, the target image A is cut with the vertical median line of the target image A as the cutting line, and image area A3 and image area A4 are obtained. Then, the process of quality characterization parameters for the target area under the above definition detection dimension can be characterized as:

[0079]

[0080] in, It is used to characterize the quality characterization parameters for the target area z under the definition detection dimension; m is used to characterize the number of image grids in each row of the target area z; n is used to characterize the number of image grids in each column of the target area z; (i, j) is used to characterize the i-th horizontal and j-th vertical pixel points in the target area z; γ ij It is used to characterize the single-grid quality detection result of the target grid located at position (i, j) in the m×n image grids included in the target area z under the clarity detection dimension.

[0081] In the above example, the clarity of the target grid can be calculated, and based on the clarity of the target grid, the single-grid quality detection result for the target grid under the clarity detection dimension can be obtained to characterize whether the target grid is an unclear grid. Thereafter, based on multiple single-grid quality detection results corresponding to multiple image grids under the clarity detection dimension, the proportion of unclear grids in the target area can be obtained as a quality characterization parameter for the target area under the clarity detection dimension. In this way, the evaluation granularity of the quality characterization parameter for the target area under the clarity detection dimension can be reduced to the image grid granularity, thereby improving the accuracy of the quality characterization parameter for the target area under the clarity detection dimension.

[0082] In summary, through the above method, in the embodiment of the present disclosure, the target area can be segmented to obtain multiple image grids, and each image grid in the multiple image grids is used as a target grid to obtain a single grid quality detection result for the target grid under the target detection dimension (the target detection dimension is any quality detection dimension in the multiple quality detection dimensions), so as to obtain multiple single grid quality detection results corresponding to the multiple image grids under the target detection dimension, and then based on the multiple single grid quality detection results corresponding to the multiple image grids under the target detection dimension, the quality characterization parameters for the target area under the target detection dimension are obtained. In this way, on the one hand, the evaluation granularity of the quality characterization parameters for the target area under the target detection dimension can be reduced to at least the image grid granularity, thereby improving the accuracy of the quality characterization parameters for the target area under the target detection dimension, and facilitating the location of the problem area in the later stage; on the other hand, the quality characterization parameters for the target area are obtained from multiple quality detection dimensions, which can improve the comprehensiveness and accuracy of the instruction characterization parameters for the target area, so as to further improve the reliability of the overall quality detection results for the target image.

[0083] After executing step S101 and step S102, the process may proceed to step S103, that is, obtaining an overall quality detection result for the target image based on the quality characterization parameter.

[0084] In some optional embodiments, each quality detection dimension under multiple quality detection dimensions can be used as a dimension to be detected. After obtaining the quality characterization parameter for the target area under the dimension to be detected, the quality characterization parameter for the target area under the dimension to be detected is compared with a preset parameter threshold (here, when the dimension to be detected is a flocculent area detection dimension, the preset parameter threshold can be defined as a first preset parameter threshold; when the dimension to be detected is a clarity detection dimension, the preset parameter threshold can be defined as a second preset parameter threshold) to obtain an overall quality detection result for characterizing that the image quality of the target image under the dimension to be detected is unqualified when the quality characterization parameter for the target area under the dimension to be detected is greater than the preset parameter threshold; and obtain an overall quality detection result for characterizing that the image quality of the target image under the dimension to be detected is qualified when the quality characterization parameter for the target area under the dimension to be detected is less than or equal to the preset parameter threshold.

[0085] Among them, the preset parameter threshold can be set according to actual business needs or application scenarios, which is not elaborated in the embodiments of the present disclosure.

[0086] In other optional embodiments, each quality detection dimension under multiple quality detection dimensions can be used as a dimension to be detected. After obtaining the quality characterization parameters for the target area under the dimension to be detected, the ratio of the quality characterization parameters for the target area under the dimension to be detected to the reference characterization parameters can be calculated, so that when the ratio of the quality characterization parameters for the target area under the dimension to be detected to the reference characterization parameters is greater than a preset ratio threshold (here, when the dimension to be detected is a flocculent area detection dimension, the preset ratio threshold can be defined as a first preset ratio threshold; when the dimension to be detected is a clarity detection dimension, the preset ratio threshold can be defined as a second preset ratio threshold), an overall quality detection result for characterizing that the image quality of the target image under the dimension to be detected is unqualified is obtained; when the ratio of the quality characterization parameters for the target area under the dimension to be detected to the reference characterization parameters is less than or equal to the preset ratio threshold, an overall quality detection result for characterizing that the image quality of the target image under the dimension to be detected is qualified is obtained.

[0087] The reference characterization parameter may be a quality characterization parameter for the reference area under the dimension to be detected; the preset ratio threshold may be set according to actual business needs or application scenarios, which will not be described in detail in the embodiments of the present disclosure. In addition, it should be noted that in the embodiments of the present disclosure, the reference area may be any image area other than the target area among the multiple image areas including the target area obtained by segmenting the target image according to a preset segmentation method.

[0088] Continuing with the above example, it is assumed that when executing step S101 to segment the target image, the target image A is cut according to the first preset cutting method with the horizontal median line of the target image A as the cutting line, and image area A1 and image area A2 are obtained; according to the second preset cutting method, the target image A is cut with the vertical median line of the target image A as the cutting line, and image area A3 and image area A4 are obtained. It is further assumed that the multiple quality detection dimensions include the flocculent area detection dimension and the clarity detection dimension, then, step S102 is executed, and each of the multiple image areas (image area A1, image area A2, image area A3 and image area A4) is used as the target area, and the quality characterization parameters for the target area under the multiple quality detection dimensions obtained respectively may include:

[0089] Quality characterization parameter ω for image area A1 in flocculent area detection dimension A1 , and the quality characterization parameters for image area A1 under the definition detection dimension

[0090] Quality characterization parameter ω for image area A2 in flocculent area detection dimension A2 , and the quality characterization parameters for image area A2 under the definition detection dimension

[0091] Quality characterization parameter ω for image area A3 in flocculent area detection dimension A3 , and the quality characterization parameters for image area A3 under the definition detection dimension

[0092] Quality characterization parameter ω for image area A4 in flocculent area detection dimension A4 , and the quality characterization parameters for image area A4 under the definition detection dimension

[0093] Then, the quality characterization parameter ω of the image area A1 can be obtained under the flocculent area detection dimension. A1 When it is greater than the first preset parameter threshold μ, an overall quality detection result for characterizing that the image quality of the target image under the flocculent area detection dimension is unqualified is obtained; the quality characterization parameter ω for the image area A2 under the flocculent area detection dimension A2 When the first preset parameter threshold μ is greater than the first preset parameter threshold, an overall quality detection result is obtained for characterizing that the image quality of the target image under the flocculent area detection dimension is unqualified; the quality characterization parameter ω for the image area A3 under the flocculent area detection dimension A3When it is greater than the first preset parameter threshold μ, an overall quality detection result for characterizing that the image quality of the target image under the flocculent area detection dimension is unqualified is obtained; the quality characterization parameter ω for the image area A4 under the flocculent area detection dimension A4 When the value is greater than the first preset parameter threshold μ, an overall quality detection result is obtained for characterizing that the image quality of the target image under the flocculent region detection dimension is unqualified. The first preset parameter threshold μ can be set according to actual business needs or application scenarios, for example, it can be set to 0.05, which is not described in detail in the embodiment of the present disclosure.

[0094] The quality characterization parameter ω for the image area A1 can be calculated under the flocculent area detection dimension. A1 The quality characterization parameter ω for the image area A2 under the flocculent area detection dimension A2 When the ratio of is greater than the first preset ratio threshold ξ, an overall quality detection result for characterizing that the image quality of the target image under the flocculent area detection dimension is unqualified is obtained; the quality characterization parameter ω for the image area A2 under the flocculent area detection dimension A2 The quality characterization parameter ω for the image area A1 under the flocculent area detection dimension A1 When the ratio of is greater than the first preset ratio threshold ξ, an overall quality detection result for characterizing that the image quality of the target image under the flocculent area detection dimension is unqualified is obtained; the quality characterization parameter ω for the image area A3 under the flocculent area detection dimension is A3 The quality characterization parameter ω for image area A4 under the flocculent area detection dimension A4 When the ratio of is greater than the first preset ratio threshold ξ, an overall quality detection result for characterizing that the image quality of the target image under the flocculent area detection dimension is unqualified is obtained; the quality characterization parameter ω for the image area A4 under the flocculent area detection dimension is A4 The quality characterization parameter ω for image area A3 under the flocculent area detection dimension A3 When the ratio of is greater than the first preset ratio threshold ξ, an overall quality detection result is obtained for characterizing that the image quality of the target image under the flocculent region detection dimension is unqualified. The first preset ratio threshold ξ can be set according to actual business needs or application scenarios, for example, it can be set to 3, which is not elaborated in the embodiments of the present disclosure.

[0095] The quality characterization parameters for the image area A1 can be obtained under the definition detection dimension. When the image quality of the target image under the definition detection dimension is unqualified, the quality characterization parameter of the image area A2 under the definition detection dimension is obtained. When the image quality of the target image under the definition detection dimension is unqualified, the quality characterization parameter of the image area A3 under the definition detection dimension is obtained. When the image quality of the target image under the definition detection dimension is unqualified, the quality characterization parameter of the image area A4 under the definition detection dimension is obtained. When the value is greater than the second preset parameter threshold value ρ, an overall quality detection result is obtained for characterizing that the image quality of the target image under the definition detection dimension is unqualified. The second preset parameter threshold value ρ can be set according to actual business needs or application scenarios, for example, it can be set to 0.7, which is not described in detail in the embodiment of the present disclosure.

[0096] The quality characterization parameters for the image area A1 can be obtained under the definition detection dimension. Quality characterization parameters for image area A2 under the definition detection dimension When the ratio of is greater than the second preset ratio threshold σ, an overall quality detection result for characterizing that the image quality of the target image under the definition detection dimension is unqualified is obtained; the quality characterization parameter for the image area A2 under the definition detection dimension is Quality characterization parameters for image area A1 under the definition detection dimension When the ratio of is greater than the second preset ratio threshold σ, an overall quality detection result for characterizing that the image quality of the target image under the definition detection dimension is unqualified is obtained; the quality characterization parameter for the image area A3 under the definition detection dimension is Quality characterization parameters for image area A4 under the definition detection dimension When the ratio of is greater than the second preset ratio threshold σ, an overall quality detection result for characterizing that the image quality of the target image under the definition detection dimension is unqualified is obtained; the quality characterization parameter for the image area A4 under the definition detection dimension is Quality characterization parameters for image area A3 under the definition detection dimension When the ratio of is greater than the second preset ratio threshold σ, an overall quality detection result for characterizing that the image quality of the target image under the definition detection dimension is unqualified is obtained. The second preset ratio threshold σ can be set according to actual business needs or application scenarios, for example, it can be set to 3, which is not elaborated in the embodiment of the present disclosure.

[0097] Through the above manner, in the embodiment of the present disclosure, multiple evaluation criteria are used for evaluation in the process of obtaining the overall quality detection result for the target image based on the quality characterization parameters, thereby further ensuring the reliability of the overall quality detection result for the target image.

[0098] The following describes an integrity process of an image quality detection method provided by an embodiment of the present disclosure.

[0099] (1) Segment the target image

[0100] The target image is segmented according to a plurality of preset segmentation methods to obtain a plurality of image regions.

[0101] The multiple preset cutting methods are used to indicate that the target image is cut according to different cutting methods to obtain multiple image regions, so as to make the cutting edge of the reference region be in a non-edge position in at least one other region of the multiple image regions except the reference region. Here, the reference region is any image region of the multiple image regions.

[0102] Exemplarily, it is assumed that there is a first preset cutting method and a second preset cutting method.

[0103] Please combine Figure 2 According to the first preset cutting method, the target image A is cut with the horizontal median line of the target image A as the cutting line, and the image area A1 and the image area A2 can be obtained; according to the second preset cutting method, the target image A is cut with the vertical median line of the target image A as the cutting line, and the image area A3 and the image area A4 can be obtained. Among them, the cutting edges a11-a12 of the image area A1 are in non-edge positions in both the image area A3 and the image area A4; the cutting edges a11-a12 of the image area A2 are in non-edge positions in both the image area A3 and the image area A4; the cutting edges a21-a22 of the image area A3 are in non-edge positions in both the image area A1 and the image area A2; the cutting edges a21-a22 of the image area A4 are in non-edge positions in both the image area A1 and the image area A2.

[0104] (2) Taking each of the multiple image regions as the target region, obtaining the quality characterization parameters for the target region under the flocculent region detection dimension

[0105] The target area is segmented to obtain multiple image grids.

[0106] In one example, the target area may be segmented according to a preset grid size to obtain a plurality of image grids.

[0107] For example, it is assumed that there is a target area B, and the size of the target area B is (w, h),

[0108] That is, the width of the target area B is w pixels, and the length is h pixels.

[0109] The size of each of the multiple image grids can be preset as (p, q), that is, the width of each of the multiple image grids is p pixels and the height is q pixels. Here, p and q can be constants, for example, 10. In this way, by segmenting the target area B, m×n image grids can be obtained, specifically, m=w / p, n=h / q.

[0110] Thereafter, each image grid in the multiple image grids can be used as a target grid to obtain a single grid quality detection result for the target grid under the flocculent area detection dimension, so as to obtain multiple single grid quality detection results corresponding one-to-one to the multiple image grids under the flocculent area detection dimension.

[0111] Among them, “obtaining the single-grid quality detection result for the target grid in the flocculent area detection dimension” may include:

[0112] Get the transparency representation value of each pixel in the target grid;

[0113] Based on the transparency representation value of each pixel in the target grid, the transparent pixel ratio of the target grid is obtained;

[0114] Based on the transparent pixel ratio of the target grid, the single-grid quality detection result of the target grid in the flocculent area detection dimension is obtained to characterize whether the target grid is a semi-transparent grid.

[0115] In a specific example, the transparency representation value of each pixel in the target grid can be obtained through the pixel parameter acquisition function provided by the image processing library (e.g., Pillow). The transparency representation value is in the interval [0, 255]. Here, 0 is used to represent a fully transparent state, and 255 is used to represent an opaque state.

[0116] Continuing with the previous example, assume that there are pixels (x, y) in the target grid, where x∈[1, p] and y∈[1, q].

[0117] After obtaining the transparency representation value of the pixel point (x, y), when the transparency representation value of the pixel point (x, y) is less than the threshold value ε, the pixel point (x, y) can be confirmed as a transparent pixel, and the transparent state of the pixel point (x, y) is represented by 1; when the transparency representation value of the pixel point (x, y) is greater than or equal to the threshold value ε, the pixel point (x, y) is confirmed as a non-transparent pixel, and the transparent state of the pixel point (x, y) is represented by 0. Wherein, ε can be a constant, for example, 200.

[0118] Finally, the transparent state a of the pixel (x, y) can be obtained. xy ∈{1,0}.

[0119] After obtaining the transparency state of each pixel in the target grid, the transparent pixel ratio of the target grid can be obtained based on the transparency state of each pixel in the target grid. This process can be characterized as:

[0120]

[0121] Among them, β ij It is used to represent the transparent pixel ratio of the target grid at position (i, j) in the m×n image grid; p is used to represent the width of the target grid; q is used to represent the height of the target grid; (x, y) is used to represent the xth horizontal and yth vertical pixel in the target grid; a xy Used to represent the transparency state of the pixel (x, y).

[0122] Obviously, β ij ∈[0,1]. Further, in the above example, ij When the value is less than the threshold λ and greater than the threshold τ, the target grid is confirmed as a semi-transparent grid, and 1 is used to represent the single-grid quality detection result of the target grid under the flocculent area detection dimension; ij When the value is greater than or equal to the threshold λ, or less than or equal to the threshold τ, the target grid is confirmed as a non-translucent grid, and the single-grid quality detection result of the target grid under the flocculent area detection dimension is represented by 0. Among them, λ can be a constant, for example, 0.9; τ can be a constant, for example, 0.2.

[0123] Finally, the single-grid quality detection result β for the target grid can be obtained. ij ∈{1,0}.

[0124] After obtaining multiple single-grid quality detection results corresponding to multiple image grids in the clarity detection dimension, the quality characterization parameter for the target area in the clarity detection dimension can be obtained based on the multiple single-grid quality detection results corresponding to multiple image grids in the clarity detection dimension. Specifically, based on the multiple single-grid quality detection results corresponding to multiple image grids in the flocculent area detection dimension, the proportion of semi-transparent grids in the target area can be obtained as the quality characterization parameter for the target area in the flocculent area detection dimension. This process can be characterized as:

[0125]

[0126] Among them, ω z It is used to characterize the quality characterization parameters of the target area z under the flocculent area detection dimension; m is used to characterize the number of image grids in each row of the target area z; n is used to characterize the number of image grids in each column of the target area z; (i, j) is used to characterize the i-th horizontal and j-th vertical pixel points in the target area z; β ij It is used to characterize the single-grid quality detection result of the target grid located at the position (i, j) in the m×n image grids included in the target area z under the flocculent area detection dimension.

[0127] (3) Taking each of the multiple image regions as a target region, obtaining a quality characterization parameter for the target region under the definition detection dimension

[0128] The target area is segmented to obtain multiple image grids.

[0129] In one example, the target area may be segmented according to a preset grid size to obtain a plurality of image grids.

[0130] Exemplarily, it is assumed that there is a target area B, and the size of the target area B is (w, h), that is, the width of the target area B is w pixels, and the length is h pixels.

[0131] The size of each of the multiple image grids can be preset as (p, q), that is, the width of each of the multiple image grids is p pixels and the height is q pixels. Here, p and q can be constants, for example, 10. In this way, by segmenting the target area B, m×n image grids can be obtained, specifically, m=w / p, n=h / q.

[0132] Thereafter, each image grid in the multiple image grids can be used as a target grid to obtain a single-grid quality detection result for the target grid under the clarity detection dimension, so as to obtain multiple single-grid quality detection results corresponding one-to-one to the multiple image grids under the clarity detection dimension.

[0133] Among them, “obtaining the single-grid quality detection result for the target grid under the clarity detection dimension” may include:

[0134] Calculate the clarity of the target grid;

[0135] Based on the clarity of the target grid, a single-grid quality detection result for the target grid under the clarity detection dimension is obtained to characterize whether the target grid is a non-clear grid.

[0136] In a specific example, the clarity of the target grid can be calculated by an image clarity algorithm, such as the Laplace algorithm, the Sobel algorithm, etc. The clarity of the target grid (specifically, the target grid at position (i, j) in the m×n image grids) can be represented as: γ ij ∈(0,1). Here, i∈[1,m], j∈[1,n]. Further, in the above example, ij When it is less than the threshold value θ, the target grid is confirmed as a non-clear grid, and 1 is used to represent the single-grid quality detection result of the target grid under the clarity detection dimension; ij When the value is greater than or equal to the threshold value θ, the target grid is confirmed as a clear grid, and the single grid quality detection result of the target grid under the clarity detection dimension is represented by 0. θ can be a constant, for example, 0.8.

[0137] Finally, the single-grid quality detection result γ for the target grid under the clarity detection dimension can be obtained. ij ∈{1,0}.

[0138] After obtaining multiple single-grid quality detection results corresponding to multiple image grids in the clarity detection dimension, the non-clear grid ratio of the target area can be obtained based on the multiple single-grid quality detection results corresponding to multiple image grids in the clarity detection dimension, as a quality characterization parameter for the target area in the clarity detection dimension. This process can be characterized as:

[0139]

[0140] in, It is used to characterize the quality characterization parameters for the target area z under the definition detection dimension; m is used to characterize the number of image grids in each row of the target area z; n is used to characterize the number of image grids in each column of the target area z; (i, j) is used to characterize the i-th horizontal and j-th vertical pixel points in the target area z; γ ij It is used to characterize the single-grid quality detection result of the target grid located at position (i, j) in the m×n image grids included in the target area z under the clarity detection dimension.

[0141] (4) Obtain the overall quality detection result for the target image

[0142] In some optional embodiments, each quality detection dimension in the flocculent region detection dimension and the clarity detection dimension can be used as a dimension to be detected. After obtaining the quality characterization parameter for the target area under the dimension to be detected, the quality characterization parameter for the target area under the dimension to be detected is compared with a preset parameter threshold (here, when the dimension to be detected is the flocculent region detection dimension, the preset parameter threshold can be defined as a first preset parameter threshold; when the dimension to be detected is the clarity detection dimension, the preset parameter threshold can be defined as a second preset parameter threshold) to obtain an overall quality detection result for characterizing that the image quality of the target image under the dimension to be detected is unqualified when the quality characterization parameter for the target area under the dimension to be detected is greater than the preset parameter threshold; and obtain an overall quality detection result for characterizing that the image quality of the target image under the dimension to be detected is qualified when the quality characterization parameter for the target area under the dimension to be detected is less than or equal to the preset parameter threshold.

[0143] Among them, the preset parameter threshold can be set according to actual business needs or application scenarios, which is not elaborated in the embodiments of the present disclosure.

[0144] In other optional embodiments, each quality detection dimension under multiple quality detection dimensions can be used as a dimension to be detected. After obtaining the quality characterization parameters for the target area under the dimension to be detected, the ratio of the quality characterization parameters for the target area under the dimension to be detected to the reference characterization parameters can be calculated, so that when the ratio of the quality characterization parameters for the target area under the dimension to be detected to the reference characterization parameters is greater than a preset ratio threshold (here, when the dimension to be detected is a flocculent area detection dimension, the preset ratio threshold can be defined as a first preset ratio threshold; when the dimension to be detected is a clarity detection dimension, the preset ratio threshold can be defined as a second preset ratio threshold), an overall quality detection result for characterizing that the image quality of the target image under the dimension to be detected is unqualified is obtained; when the ratio of the quality characterization parameters for the target area under the dimension to be detected to the reference characterization parameters is less than or equal to the preset ratio threshold, an overall quality detection result for characterizing that the image quality of the target image under the dimension to be detected is qualified is obtained.

[0145] The reference characterization parameter may be a quality characterization parameter for the reference area under the dimension to be detected; the preset ratio threshold may be set according to actual business needs or application scenarios, which will not be described in detail in the embodiments of the present disclosure. In addition, it should be noted that in the embodiments of the present disclosure, the reference area may be any image area other than the target area among the multiple image areas including the target area obtained by segmenting the target image according to a preset segmentation method.

[0146] Continuing with the above example, through steps (1), (2) and (3), the quality characterization parameters for the target area under multiple quality detection dimensions obtained may include:

[0147] Quality characterization parameter ω for image area A1 in flocculent area detection dimension A1 , and the quality characterization parameters for image area A1 under the definition detection dimension

[0148] Quality characterization parameter ω for image area A2 in flocculent area detection dimension A2 , and the quality characterization parameters for image area A2 under the definition detection dimension

[0149] Quality characterization parameter ω for image area A3 in flocculent area detection dimension A3 , and the quality characterization parameters for image area A3 under the definition detection dimension

[0150] Quality characterization parameter ω for image area A4 in flocculent area detection dimension A4 , and the quality characterization parameters for image area A4 under the definition detection dimension

[0151] Then, the quality characterization parameter ω of the image area A1 can be obtained under the flocculent area detection dimension. A1 When it is greater than the first preset parameter threshold μ, an overall quality detection result for characterizing that the image quality of the target image under the flocculent area detection dimension is unqualified is obtained; the quality characterization parameter ω for the image area A2 under the flocculent area detection dimension A2 When the first preset parameter threshold μ is greater than the first preset parameter threshold, an overall quality detection result is obtained for characterizing that the image quality of the target image under the flocculent area detection dimension is unqualified; the quality characterization parameter ω for the image area A3 under the flocculent area detection dimension A3 When it is greater than the first preset parameter threshold μ, an overall quality detection result for characterizing that the image quality of the target image under the flocculent area detection dimension is unqualified is obtained; the quality characterization parameter ω for the image area A4 under the flocculent area detection dimension A4 When the value is greater than the first preset parameter threshold μ, an overall quality detection result is obtained for characterizing that the image quality of the target image under the flocculent region detection dimension is unqualified. The first preset parameter threshold μ can be set according to actual business needs or application scenarios, for example, it can be set to 0.05, which is not described in detail in the embodiment of the present disclosure.

[0152] The quality characterization parameter ω for the image area A1 can be calculated under the flocculent area detection dimension. A1The quality characterization parameter ω for the image area A2 under the flocculent area detection dimension A2 When the ratio of is greater than the first preset ratio threshold ξ, an overall quality detection result for characterizing that the image quality of the target image under the flocculent area detection dimension is unqualified is obtained; the quality characterization parameter ω for the image area A2 under the flocculent area detection dimension A2 The quality characterization parameter ω for the image area A1 under the flocculent area detection dimension A1 When the ratio of is greater than the first preset ratio threshold ξ, an overall quality detection result for characterizing that the image quality of the target image under the flocculent area detection dimension is unqualified is obtained; the quality characterization parameter ω for the image area A3 under the flocculent area detection dimension is A3 The quality characterization parameter ω for image area A4 under the flocculent area detection dimension A4 When the ratio of is greater than the first preset ratio threshold ξ, an overall quality detection result for characterizing that the image quality of the target image under the flocculent area detection dimension is unqualified is obtained; the quality characterization parameter ω for the image area A4 under the flocculent area detection dimension is A4 The quality characterization parameter ω for image area A3 under the flocculent area detection dimension A3 When the ratio of is greater than the first preset ratio threshold ξ, an overall quality detection result is obtained for characterizing that the image quality of the target image under the flocculent region detection dimension is unqualified. The first preset ratio threshold ξ can be set according to actual business needs or application scenarios, for example, it can be set to 3, which is not elaborated in the embodiments of the present disclosure.

[0153] The quality characterization parameters for the image area A1 can be obtained under the definition detection dimension. When the image quality of the target image under the definition detection dimension is unqualified, the quality characterization parameter of the image area A2 under the definition detection dimension is obtained. When the image quality of the target image under the definition detection dimension is unqualified, the quality characterization parameter of the image area A3 under the definition detection dimension is obtained. When the image quality of the target image under the definition detection dimension is unqualified, the quality characterization parameter of the image area A4 under the definition detection dimension is obtained. When the value is greater than the second preset parameter threshold value ρ, an overall quality detection result is obtained for characterizing that the image quality of the target image under the definition detection dimension is unqualified. The second preset parameter threshold value ρ can be set according to actual business needs or application scenarios, for example, it can be set to 0.7, which is not described in detail in the embodiment of the present disclosure.

[0154] The quality characterization parameters for the image area A1 can be obtained under the definition detection dimension. Quality characterization parameters for image area A2 under the definition detection dimension When the ratio of is greater than the second preset ratio threshold σ, an overall quality detection result for characterizing that the image quality of the target image under the definition detection dimension is unqualified is obtained; the quality characterization parameter for the image area A2 under the definition detection dimension is Quality characterization parameters for image area A1 under the definition detection dimension When the ratio of is greater than the second preset ratio threshold σ, an overall quality detection result for characterizing that the image quality of the target image under the definition detection dimension is unqualified is obtained; the quality characterization parameter for the image area A3 under the definition detection dimension is Quality characterization parameters for image area A4 under the definition detection dimension When the ratio of is greater than the second preset ratio threshold σ, an overall quality detection result for characterizing that the image quality of the target image under the definition detection dimension is unqualified is obtained; the quality characterization parameter for the image area A4 under the definition detection dimension is Quality characterization parameters for image area A3 under the definition detection dimension When the ratio of is greater than the second preset ratio threshold σ, an overall quality detection result for characterizing that the image quality of the target image under the definition detection dimension is unqualified is obtained. The second preset ratio threshold σ can be set according to actual business needs or application scenarios, for example, it can be set to 3, which is not elaborated in the embodiment of the present disclosure.

[0155] See also Figure 3 , is a schematic diagram of an application scenario of an image quality detection method provided by an embodiment of the present disclosure. As mentioned above, the image quality detection method provided by an embodiment of the present disclosure is applied to an electronic device. Among them, the electronic device can be a server or a terminal device. Here, the terminal device can be a workbench, a mainframe computer, a conventional computer (for example, a desktop computer, a laptop computer, a car computer, etc.), a personal digital processing or other similar computing device.

[0156] Among them, electronic equipment is used for:

[0157] Segment the target image to obtain multiple image regions;

[0158] Taking each of the multiple image regions as a target region, respectively obtaining quality characterization parameters for the target region under multiple quality detection dimensions; wherein the multiple quality detection dimensions include at least one of a flocculent region detection dimension and a clarity detection dimension;

[0159] Based on the quality characterization parameters, the overall quality detection result of the target image is obtained.

[0160] It should be noted that in the embodiments of the present disclosure, Figure 3 The scene diagram shown is only illustrative and not restrictive. Those skilled in the art can Figure 3 Various obvious changes and / or substitutions are made to the examples, and the obtained technical solutions still fall within the scope of the disclosure of the embodiments of the present disclosure.

[0161] In order to better implement the image quality detection method, the embodiment of the present disclosure also provides an image quality detection device, which can be integrated into an electronic device. The electronic device can be a server or a terminal device. Here, the terminal device can be a workbench, a large computer, a conventional computer (for example, a desktop computer, a laptop computer, a car computer, etc.), a personal digital processing or other similar computing device. Figure 4 The schematic structural block diagram shown is used to illustrate an image quality detection device 400 provided in the disclosed embodiment.

[0162] The image quality detection device 400 comprises:

[0163] An image segmentation unit 401 is used to segment a target image to obtain multiple image regions;

[0164] The parameter acquisition unit 402 is used to take each image area of ​​the multiple image areas as a target area, and obtain quality characterization parameters for the target area under multiple quality detection dimensions respectively; wherein the multiple quality detection dimensions include at least one of a flocculent area detection dimension and a clarity detection dimension;

[0165] The result acquisition unit 403 is used to obtain the overall quality detection result for the target image based on the quality characterization parameter.

[0166] In some optional implementations, the parameter acquisition unit 402 is used to:

[0167] Segment the target area to obtain multiple image grids;

[0168] Taking each image grid in the multiple image grids as a target grid, obtaining a single grid quality detection result for the target grid under the target detection dimension, so as to obtain a plurality of single grid quality detection results corresponding to the multiple image grids under the target detection dimension; wherein the target detection dimension is any quality detection dimension in the multiple quality detection dimensions;

[0169] Based on multiple single-grid quality detection results corresponding one-to-one to multiple image grids in the target detection dimension, a quality characterization parameter for the target area in the target detection dimension is obtained.

[0170] In some optional implementations, the target detection dimension is a flocculent region detection dimension; the parameter acquisition unit 402 is used to:

[0171] Get the transparency representation value of each pixel in the target grid;

[0172] Based on the transparency representation value of each pixel in the target grid, the transparent pixel ratio of the target grid is obtained;

[0173] Based on the transparent pixel ratio of the target grid, the single-grid quality detection result of the target grid in the flocculent area detection dimension is obtained to characterize whether the target grid is a semi-transparent grid.

[0174] In some optional implementations, the parameter acquisition unit 402 is used to:

[0175] Based on multiple single-grid quality detection results corresponding to multiple image grids in the flocculent area detection dimension, the proportion of semi-transparent grids in the target area is obtained as a quality characterization parameter for the target area in the flocculent area detection dimension.

[0176] In some optional implementations, the target detection dimension is a clarity detection dimension; and the parameter acquisition unit 402 is used to:

[0177] Calculate the clarity of the target grid;

[0178] Based on the clarity of the target grid, a single-grid quality detection result for the target grid under the clarity detection dimension is obtained to characterize whether the target grid is a non-clear grid.

[0179] In some optional implementations, the parameter acquisition unit 402 is used to:

[0180] Based on multiple single-grid quality detection results corresponding one-to-one to multiple image grids under the clarity detection dimension, the proportion of non-clear grids in the target area is obtained as a quality characterization parameter for the target area under the clarity detection dimension.

[0181] In some optional implementations, the target image is segmented to obtain multiple image regions, including:

[0182] The target image is segmented according to a plurality of preset segmentation methods to obtain a plurality of image regions; wherein a cutting edge of a reference region is located at a non-edge position in at least one other region of the plurality of image regions except the reference region; and the reference region is any image region of the plurality of image regions.

[0183] In the embodiment of the present disclosure, the specific functions and examples of each unit in the image quality detection device 400 can refer to the relevant descriptions of the corresponding steps in the aforementioned image quality detection method embodiment, and will not be repeated here.

[0184] In the technical solution disclosed herein, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0185] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0186] Figure 5 A schematic block diagram of an example electronic device 500 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as vehicle-mounted computing devices, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0187] like Figure 5 As shown, the electronic device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 to a random access memory (RAM) 503. In RAM 503, various programs and data required for the operation of the electronic device 500 can also be stored. The computing unit 501, ROM 502 and RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0188] A number of components in the electronic device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of renderers, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the electronic device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0189] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSP), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as the image quality detection method. For example, in some embodiments, the image quality detection method may be implemented as a computer software program, which is tangibly included in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the computing unit 501, one or more steps in the image quality detection method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured as an image quality detection method in any other appropriate manner (eg, by means of firmware).

[0190] Various embodiments of the systems and techniques described above herein may be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0191] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0192] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a RAM, a ROM, an erasable programmable read-only memory (EPROM) or a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0193] To provide interaction with a user, the systems and techniques described herein may be implemented on a computer having: a rendering device (e.g., a cathode ray tube (CRT) renderer or a liquid crystal display (LCD)) for rendering information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other kinds of devices are also used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0194] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: Local Area Network (LAN), Wide Area Network (WAN), and the Internet.

[0195] The computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server can be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0196] The embodiment of the present disclosure also provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to execute the image quality detection method.

[0197] The embodiment of the present disclosure also provides a computer program product, including a computer program, which implements the image quality detection method when executed by a processor.

[0198] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in the present disclosure can be executed in parallel, sequentially or in a different order, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, and this document is not limited here. In addition, in the present disclosure, relational terms such as "first", "second", "third", etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. In addition, "multiple" in the present disclosure can be understood as at least two.

[0199] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A method for detecting image quality, comprising: Segment the target image to obtain multiple image regions; Taking each image area of ​​the multiple image areas as a target area, respectively obtaining quality characterization parameters for the target area under multiple quality detection dimensions; wherein the multiple quality detection dimensions include at least one of a flocculent area detection dimension and a clarity detection dimension; Based on the quality characterization parameter, an overall quality detection result for the target image is obtained.

2. The method according to claim 1, wherein: The obtaining of quality characterization parameters for the target area under multiple quality detection dimensions respectively includes: Segmenting the target area to obtain a plurality of image grids; Taking each image grid in the plurality of image grids as a target grid, obtaining a single-grid quality detection result for the target grid under a target detection dimension, so as to obtain a plurality of single-grid quality detection results corresponding one-to-one to the plurality of image grids under the target detection dimension; wherein the target detection dimension is any quality detection dimension in the plurality of quality detection dimensions; Based on a plurality of single-grid quality detection results corresponding one-to-one to the plurality of image grids under the target detection dimension, a quality characterization parameter for the target area under the target detection dimension is obtained.

3. The method according to claim 2, wherein: The target detection dimension is a flocculent region detection dimension; and obtaining a single-grid quality detection result for the target grid under the target detection dimension includes: Obtaining a transparency representation value of each pixel in the target grid; Based on the transparency representation value of each pixel in the target grid, obtaining the transparent pixel ratio of the target grid; Based on the transparent pixel ratio of the target grid, a single-grid quality detection result for the target grid under the flocculent region detection dimension is obtained to characterize whether the target grid is a semi-transparent grid.

4. The method according to claim 3, wherein: The obtaining of the quality characterization parameter for the target area under the target detection dimension based on the multiple single-grid quality detection results corresponding one-to-one to the multiple image grids under the target detection dimension includes: Based on multiple single-grid quality detection results corresponding one-to-one to the multiple image grids in the flocculent area detection dimension, the semi-transparent grid ratio of the target area is obtained as a quality characterization parameter for the target area in the flocculent area detection dimension.

5. The method according to claim 2, wherein: The target detection dimension is a clarity detection dimension; The obtaining of a single-grid quality detection result for the target grid under the target detection dimension includes: Calculating the clarity of the target grid; Based on the clarity of the target grid, a single-grid quality detection result for the target grid under the clarity detection dimension is obtained to indicate whether the target grid is a non-clear grid.

6. The method according to claim 5, wherein: The obtaining of the quality characterization parameter for the target area under the target detection dimension based on the multiple single-grid quality detection results corresponding one-to-one to the multiple image grids under the target detection dimension includes: Based on multiple single-grid quality detection results corresponding one-to-one to the multiple image grids under the clarity detection dimension, the proportion of non-clear grids in the target area is obtained as a quality characterization parameter for the target area under the clarity detection dimension.

7. The method according to claim 1, wherein: The target image is segmented to obtain multiple image regions, including: The target image is segmented according to a plurality of preset segmentation methods to obtain a plurality of image regions; wherein a cutting edge of a reference region is located at a non-edge position in at least one other region of the plurality of image regions except the reference region; and the reference region is any image region of the plurality of image regions.

8. An image quality detection device, comprising: An image segmentation unit, used for segmenting a target image to obtain multiple image regions; A parameter acquisition unit, configured to take each of the plurality of image regions as a target region, and obtain quality characterization parameters for the target region under a plurality of quality detection dimensions, respectively; wherein the plurality of quality detection dimensions include at least one of a flocculent region detection dimension and a clarity detection dimension; The result acquisition unit is used to obtain an overall quality detection result for the target image based on the quality characterization parameter.

9. The device according to claim 8, wherein: The parameter acquisition unit is used for: Segmenting the target area to obtain a plurality of image grids; Taking each image grid in the plurality of image grids as a target grid, obtaining a single-grid quality detection result for the target grid under a target detection dimension, so as to obtain a plurality of single-grid quality detection results corresponding one-to-one to the plurality of image grids under the target detection dimension; wherein the target detection dimension is any quality detection dimension in the plurality of quality detection dimensions; Based on a plurality of single-grid quality detection results corresponding one-to-one to the plurality of image grids under the target detection dimension, a quality characterization parameter for the target area under the target detection dimension is obtained.

10. The device according to claim 9, wherein: The target detection dimension is a flocculent region detection dimension; and the parameter acquisition unit is used for: Obtaining a transparency representation value of each pixel in the target grid; Based on the transparency representation value of each pixel in the target grid, obtaining the transparent pixel ratio of the target grid; Based on the transparent pixel ratio of the target grid, a single-grid quality detection result for the target grid under the flocculent region detection dimension is obtained to characterize whether the target grid is a semi-transparent grid.

11. The device according to claim 10, wherein: The parameter acquisition unit is used for: Based on multiple single-grid quality detection results corresponding one-to-one to the multiple image grids in the flocculent area detection dimension, the semi-transparent grid ratio of the target area is obtained as a quality characterization parameter for the target area in the flocculent area detection dimension.

12. The device according to claim 9, wherein: The target detection dimension is a clarity detection dimension; and the parameter acquisition unit is used for: Calculating the clarity of the target grid; Based on the clarity of the target grid, a single-grid quality detection result for the target grid under the clarity detection dimension is obtained to indicate whether the target grid is a non-clear grid.

13. The device according to claim 12, wherein: The parameter acquisition unit is used for: Based on multiple single-grid quality detection results corresponding one-to-one to the multiple image grids under the clarity detection dimension, the proportion of non-clear grids in the target area is obtained as a quality characterization parameter for the target area under the clarity detection dimension.

14. The device according to claim 8, wherein: The target image is segmented to obtain multiple image regions, including: The target image is segmented according to a plurality of preset segmentation methods to obtain a plurality of image regions; wherein a cutting edge of a reference region is located at a non-edge position in at least one other region of the plurality of image regions except the reference region; and the reference region is any image region of the plurality of image regions.

15. An electronic device, comprising: at least one processor; a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the method according to any one of claims 1 to 7.

16. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 7.

17. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 7.

Images