CN116132697A - Image blocking effect detection method, system, equipment and storage medium - Google Patents
Image blocking effect detection method, system, equipment and storage medium Download PDFInfo
- Publication number
- CN116132697A CN116132697A CN202310019966.9A CN202310019966A CN116132697A CN 116132697 A CN116132697 A CN 116132697A CN 202310019966 A CN202310019966 A CN 202310019966A CN 116132697 A CN116132697 A CN 116132697A
- Authority
- CN
- China
- Prior art keywords
- pixel
- block
- coding block
- weight coefficient
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000000903 blocking effect Effects 0.000 title claims abstract description 110
- 238000001514 detection method Methods 0.000 title claims abstract description 89
- 230000035945 sensitivity Effects 0.000 claims abstract description 18
- 238000013139 quantization Methods 0.000 claims description 48
- 238000004364 calculation method Methods 0.000 claims description 23
- 238000004590 computer program Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 74
- 238000000034 method Methods 0.000 description 18
- 230000008447 perception Effects 0.000 description 10
- 230000008569 process Effects 0.000 description 8
- 230000005540 biological transmission Effects 0.000 description 5
- 230000006835 compression Effects 0.000 description 5
- 238000007906 compression Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000000873 masking effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000016776 visual perception Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/85—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression
- H04N19/86—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression involving reduction of coding artifacts, e.g. of blockiness
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/10—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding
- H04N19/169—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding
- H04N19/17—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object
- H04N19/176—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using adaptive coding characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the adaptive coding the unit being an image region, e.g. an object the region being a block, e.g. a macroblock
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N19/00—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals
- H04N19/85—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression
- H04N19/86—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression involving reduction of coding artifacts, e.g. of blockiness
- H04N19/865—Methods or arrangements for coding, decoding, compressing or decompressing digital video signals using pre-processing or post-processing specially adapted for video compression involving reduction of coding artifacts, e.g. of blockiness with detection of the former encoding block subdivision in decompressed video
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The embodiment of the application discloses an image blocking effect detection method, an image blocking effect detection system, image blocking effect detection equipment and a storage medium. According to the technical scheme provided by the embodiment of the application, the internal gradient and the edge gradient of the coding block are determined by reading the coding block of the video image, and the initial block effect information of the coding block is determined based on the internal gradient and the edge gradient; determining pixel discrete information of the coding block according to pixel values of all pixel points in the coding block, and determining a picture content weight coefficient of the coding block based on the pixel discrete information; the blockiness intensity of the encoded block is calculated based on the initial blockiness information and the picture content weight coefficient. By adopting the technical means, the block effect sensitivity degree of human eyes to the current coding block is determined through the pixel discrete information of the coding block, and then the block effect intensity is calculated by combining the pixel discrete information and the initial block effect information of the coding block, so that the block effect intensity of the image coding block can be accurately detected by considering the sensitivity degree of human eyes to the block effect, and the detection error of the block effect is reduced.
Description
Technical Field
The embodiment of the application relates to the technical field of image compression, in particular to an image blocking effect detection method, an image blocking effect detection system, an image blocking effect detection device and a storage medium.
Background
The blocking effect is a case where pixel errors generated during quantization in block-level DCT transform coding exhibit discontinuities at block edges in video coding scenes. By accurately detecting the blocking effect intensity of different positions of the video image, the algorithm intensity of a deblocking algorithm can be adaptively adjusted according to the blocking effect intensity of different positions, so that the blocking effect of each position can be adaptively removed, and the image quality of the video image is improved.
However, the conventional block effect detection method adopts a uniform detection mode for different coding blocks of a video image, and because the sensitivity degree of human eyes to the block effect can correspondingly change along with different image contents, the fixed detection mode is difficult to accurately detect the block effect intensity of different positions of the video image. And further affects the blocking effect removal results at different positions of the video image, resulting in image quality deviation of the video image.
Disclosure of Invention
The embodiment of the application provides an image blocking effect detection method, an image blocking effect detection system, image blocking effect detection equipment and a storage medium, which can accurately detect the blocking effect intensity of an image coding block and solve the technical problem that errors exist in blocking effect intensity detection of different positions of a video image.
In a first aspect, an embodiment of the present application provides an image blocking effect detection method, including:
reading a coding block of a video image, determining an internal gradient and an edge gradient of the coding block, and determining initial block effect information of the coding block based on the internal gradient and the edge gradient;
determining pixel discrete information of the coding block according to pixel values of all pixel points in the coding block, and determining a picture content weight coefficient of the coding block based on the pixel discrete information;
the blockiness intensity of the encoded block is calculated based on the initial blockiness information and the picture content weight coefficient.
In a second aspect, embodiments of the present application provide an image blocking effect detection system, including:
an information determination module configured to read a coding block of the video image, determine an internal gradient and an edge gradient of the coding block, and determine initial blockiness information of the coding block based on the internal gradient and the edge gradient;
the coefficient determining module is configured to determine pixel discrete information of the coding block according to pixel values of all pixel points in the coding block, and determine a picture content weight coefficient of the coding block based on the pixel discrete information;
and a calculation module configured to calculate a blockiness intensity of the encoded block based on the initial blockiness information and the picture content weight coefficient.
In a third aspect, an embodiment of the present application provides an image blocking effect detection apparatus, including:
a memory and one or more processors;
the memory is configured to store one or more programs;
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the image blocking artifact detection method as described in the first aspect.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium storing computer-executable instructions that, when executed by a computer processor, are configured to perform the image blocking detection method according to the first aspect.
In a fifth aspect, embodiments of the present application provide a computer program product comprising instructions which, when executed on a computer or processor, cause the computer or processor to perform the image blocking detection method according to the first aspect.
According to the embodiment of the application, the internal gradient and the edge gradient of the coding block are determined by reading the coding block of the video image, and the initial block effect information of the coding block is determined based on the internal gradient and the edge gradient; determining pixel discrete information of the coding block according to pixel values of all pixel points in the coding block, and determining a picture content weight coefficient of the coding block based on the pixel discrete information; the blockiness intensity of the encoded block is calculated based on the initial blockiness information and the picture content weight coefficient. By adopting the technical means, the block effect sensitivity degree of human eyes to the current coding block is determined through the pixel discrete information of the coding block, and then the block effect intensity is calculated by combining the pixel discrete information and the initial block effect information of the coding block, so that the block effect intensity of the image coding block can be accurately detected by considering the sensitivity degree of human eyes to the block effect, and the detection error of the block effect is reduced. Therefore, the blocking effect is removed, the blocking effect removing effect can be improved, and the image quality of the video image is improved.
Drawings
Fig. 1 is a flowchart of an image blocking effect detection method provided in an embodiment of the present application;
FIG. 2 is a flow chart of the calculation of initial blockiness information in an embodiment of the present application;
FIG. 3 is a flow chart of determining pixel discrete information in an embodiment of the present application;
fig. 4 is a schematic structural diagram of an image blocking effect detection system according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an image blocking effect detection apparatus according to an embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present application more apparent, the following detailed description of specific embodiments thereof is given with reference to the accompanying drawings. It is to be understood that the specific embodiments described herein are merely illustrative of the application and not limiting thereof. It should be further noted that, for convenience of description, only some, but not all of the matters related to the present application are shown in the accompanying drawings. Before discussing exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although a flowchart depicts operations (or steps) as a sequential process, many of the operations can be performed in parallel, concurrently, or at the same time. Furthermore, the order of the operations may be rearranged. The process may be terminated when its operations are completed, but may have additional steps not included in the figures. The processes may correspond to methods, functions, procedures, subroutines, and the like.
The image block effect detection method aims at calculating the block effect intensity by combining pixel discrete information and initial block effect information of the coding block so as to consider the sensitivity degree of human eyes to the block effect, accurately detect the block effect intensity of the image coding block and reduce the detection error of the block effect.
Compression of video data is extremely important because the vast amount of information carried by video in network transmission scenarios can make transmission of video data difficult. According to modern coding theory and method, video data contains a large amount of redundant information. The redundant information is removed by the video encoding and decoding technology, so that the video data volume can be reduced, and the video information can be effectively acquired and utilized.
Among them, block transform-based codecs are widely used in compression techniques for images and video. When compression coding video images, the pixels of the images are divided into coding units which are not overlapped with each other, and a plurality of transformation units can be possibly divided inside the coding units. The coded prediction residual is transformed from the spatial domain to the frequency domain using a discrete DCT transform for each transform unit, and the resulting DCT coefficients are quantized and entropy coded. Block-based DCT transforms exploit the spatial correlation of images, and since each transform unit is encoded separately as a single entity, the correlation between neighboring blocks is split during encoding. The quantization process of the DCT coefficients after the DCT transform is a lossy compression, and there may be errors between the reconstructed pixel values after the inverse quantization and inverse transform and the pixel values of the original image. When the error is continuous and large enough at the edge, obvious jump occurs at the edge of a pixel block of the reconstructed image in the time domain, so that the pixel value of the image at the edge is discontinuous, namely, a square phenomenon of a 'pseudo boundary' occurs, and subjective visual perception of people is influenced. Especially in the real-time low-code rate coding scene, the error of the pixel value at the edge is easier to enlarge due to the larger quantization step length. Causing adjacent pixel values that were originally continuously changing to make transitions at the block boundary after encoding, thereby forming a "pseudo-boundary" causing strong blocking effects.
It can be seen that the essential cause of the generation of the blocking effect is the discontinuity that the block-level DCT transform coding generates in the quantization process pixel errors at the block edges during video coding. Video blocking is one of the most common coding losses, especially in low rate coded scenes, where strong blocking is often generated at the smooth background and strong edges of the video. This situation not only affects the subsequent optimization of the image in the video link, but also directly affects the viewing experience of the user.
And because of lack of detection of the video image blocking strength, most of the deblocking algorithms are not strong enough in adaptability, and cannot adjust the algorithm strength according to the blocking strengths of different positions of the video image. The equal degree of deblocking is carried out on all images, so that the quality loss of the image without the blocking effect can be caused, the problems of insufficient removal strength of a strong blocking effect area and the like can be caused, and meanwhile, the certain degree of calculation power waste can be caused. Based on the above, an image blocking effect detection method of the embodiment of the application is provided to solve the technical problem that errors exist in blocking effect intensity detection of different positions of a video image.
Examples:
fig. 1 is a flowchart of an image blocking effect detection method provided in the embodiment of the present application, where the image blocking effect detection method provided in the embodiment may be performed by an image blocking effect detection device, and the image blocking effect detection device may be implemented by software and/or hardware, and the image blocking effect detection device may be configured by two or more physical entities or may be configured by one physical entity. In general, the image blocking detection device may be a processing device such as a video server, a codec, a computer, or the like.
The following description will take the image blocking detecting apparatus as an example of a main body that performs the image blocking detecting method. Referring to fig. 1, the image blocking effect detection method specifically includes:
s110, reading a coding block of a video image, determining an internal gradient and an edge gradient of the coding block, and determining initial block effect information of the coding block based on the internal gradient and the edge gradient;
s120, determining pixel discrete information of the coding block according to pixel values of all pixel points in the coding block, and determining a picture content weight coefficient of the coding block based on the pixel discrete information;
s130, calculating the blockiness intensity of the coding block based on the initial blockiness information and the picture content weight coefficient.
The image blocking effect detection method of the embodiment of the application aims to accurately detect the blocking effect by combining subjective feeling of human eyes, so that the intensity detection accuracy of the blocking effect is improved. The blocking effect removing operation of the image is carried out, so that the understanding of the service side on the subjective image quality of the video can be improved, and the viewing experience of a user is optimized.
It will be appreciated that the degree of sensitivity of the human eye to blocking will vary from image content to image content. The perception of blocking by the human eye is more pronounced in flat areas and less sensitive in areas of rich detail. Therefore, the method starts from subjective perception of human eyes, and divides the image into different areas to perform block effect estimation. And when the block effect estimation is carried out, the sensitivity degree of human eyes to the block effect of different areas is considered, and the block effect intensity of the different areas is accurately determined.
After decoding a video image, each encoding block in the video image is subjected to block effect detection respectively, so that the block effect detection is carried out on different positions of the video image respectively, and the effect of precisely and accurately detecting the block effect of the video image is achieved. When detecting the blockiness intensity of a coding block, firstly, determining the internal gradient of the coding block according to the difference of each pixel point in the coding block. And determining an edge gradient according to pixel point differences of four edges of the coding block and the adjacent coding block. The edge gradient and the internal gradient are combined to calculate the initial blockiness intensity of the current coding block, and the initial blockiness intensity is defined as initial blockiness information.
In order to combine the subjective feeling of human eyes to accurately detect the blockiness, the application also introduces a picture content weight coefficient so as to characterize the influence of the subjective feeling of human eyes on the blockiness intensity through the picture content weight coefficient, thereby realizing more accurate blockiness detection. Wherein the perception of blocking by the human eye is more pronounced in flat areas and less sensitive in areas with rich details. Therefore, the embodiment of the application defines the discrete condition of each pixel in the coding block as the pixel discrete information by acquiring the pixel values of each pixel point in the coding block and determining the discrete condition of each pixel in the coding block according to the pixel values. It can be appreciated that for a flat region of the encoded block, the difference between the pixels is small and the pixel values are relatively uniform. For the region with rich details, the pixel points of the pixels have larger difference, and the pixel values are more scattered. Therefore, the subjective feeling of human eyes on the block effect intensity of the current coding block can be accurately reflected by determining the pixel discrete information to calculate the picture content weight coefficient of the coding block. Finally, the block effect intensity of the coding block is calculated by combining the picture content weight coefficient and the initial block effect information, so that a block effect detection result fused with subjective feeling of human eyes can be obtained, and the sensitivity degree of human eyes to the block effect of the current coding block is accurately reflected. The block effect removing operation can be performed, so that the reasonable removal of the block effect can be ensured, and the video image obtained through optimization is better in image quality, and meets the requirement of subjective feeling of human eyes on the image quality.
Specifically, the present application detects the presence of blocking artifacts based on discontinuous edges of an image generated at block boundaries, and evaluates the extent of blocking artifacts based on the intensity of such discontinuous edges in combination with the image details to match the subjective perception of the human eye. In a video transmission scene, when the blocking effect detection is required to be performed on a video image, a compressed code stream of the video is often obtained from a transmission link, the compressed code stream is decoded, and the blocking effect detection is performed in an auxiliary manner according to related information obtained by decoding the compressed code stream.
Wherein, by decoding the compressed code stream, the relevant decoding information of each frame of video, such as the frame type of each frame of video image, the coding block boundary, the block type of each coding block, the quantization step qp of each coding block, is stored in the decoding process. And then reading in each frame of video image obtained by decoding in turn, and detecting the blocking effect intensity of each frame of image.
Wherein, whether to perform the block effect detection is decided according to the frame type of the current video image. Optionally, in order to improve the detection efficiency, only the video image with the frame type of P frame may be detected, and if the frame type is a non-P frame, the block effect detection of the current frame is ended, and the next frame video image is skipped. Because the P frame video image is easy to generate the blocking effect and is a part of the video image with relatively large blocking effect intensity, the pertinence of the blocking effect detection can be improved and the blocking effect detection efficiency can be improved by adaptively selecting the part of the video image for detection.
Further, a blocking detection of a frame of video image is determined. For each frame of video image, the block effect intensity detection is carried out on a coding block-by-coding block basis. Since the blocking effect only appears at the coding boundary, starting from the information of the video coding and decoding, the block boundary can be accurately positioned and the blocking effect detection can be carried out only at the block boundary, thereby avoiding the erroneous judgment of the edge of the real image to a great extent. And the detection precision is improved.
When detecting the blocking effect intensity of a coding block, firstly determining the internal gradient and the edge gradient of the coding block to calculate the initial blocking effect information. When determining the internal gradient and the edge gradient of the coding block, obtaining the internal gradient by calculating the first-order difference of the pixels in the coding block and superposing; and calculating first-order differences between pixel points of four edges of the coding block and pixel points of adjacent coding blocks, and overlapping to obtain edge gradients.
And calculating the first-order difference of the pixels in the current coding block by acquiring the pixel value of each pixel point of the current coding block. The first-order difference comprises two directions of horizontal and vertical, and then the difference in the horizontal direction and the difference in the vertical direction are accumulated to obtain the internal gradient of the current coding block, which is recorded as inner_sum.
On the other hand, the first-order difference between the pixel points of the four edges of the current coding block and the pixel points of the corresponding adjacent coding blocks is calculated. And accumulating the differences obtained by the four edge calculation to obtain the edge gradient of the current coding block, and marking the edge gradient as a coder_sum.
Based on the internal gradient and the edge gradient, initial blockiness information for the encoded block may be determined. Referring to fig. 2, the initial blockiness information calculation flow includes:
s1101, acquiring the number of internal differential pixel pairs and the number of edge differential pixel pairs of a coding block, determining a first ratio of an internal gradient to the number of internal differential pixel pairs, and determining a second ratio of the edge gradient to the number of edge differential pixel pairs;
s1102, under the condition that the internal gradient is not zero, dividing the second ratio value by the first ratio value to obtain initial blockiness information; in the case where the internal gradient is zero, the second ratio is taken as the initial blockiness information.
When calculating the internal gradient and the edge gradient, each pair of pixels participating in the differential calculation is counted, so that the number of the internal differential pixel pairs is recorded as inner_num, and the number of the edge differential pixel pairs is recorded as boredr_num. According to the inner gradient inner_sum, the edge gradient boundary_sum, the inner differential pixel pair number inner_num and the edge differential pixel pair number boredr_num of the current coding block, a set calculation formula is adopted to calculate initial block effect information, and the initial block effect information is recorded as a blockiness_simple. The initial blockiness information calculation formula is as follows:
based on the initial blockiness information calculation formula, a first ratio of the internal gradient to the number of internal differential pixel pairs and a second ratio of the edge gradient to the number of edge differential pixel pairs are determined. And then according to different value results of the internal gradient, when the internal gradient is 0, directly taking a second ratio of the edge gradient to the number of the edge difference pixels as initial blockiness information, and obtaining corresponding initial blockiness information by adaptive accurate calculation. Because in an actual coding scene, the discontinuous boundary caused by coding errors can bring about the difference between the edge gradient and the internal gradient of the coding block, based on the difference, the embodiment of the application defines an initial blockiness information calculation formula according to the difference between the internal gradient and the edge gradient, and combines different values of the internal gradient to realize more accurate initial blockiness information calculation and provide accurate basic data for subsequent blockiness intensity calculation.
Based on the determined initial blockiness information, the blockiness intensity is calculated as a function of the influence that the human eye is required to experience in combination with the dominant perception. The embodiment of the application determines the picture content weight coefficient of the coding block based on the pixel discrete information by determining the pixel discrete information of the coding block according to the pixel value of each pixel point in the coding block. By introducing a picture content weight coefficient, the influence of subjective feeling of human eyes on the blocking strength is represented by the picture content weight coefficient, and more accurate blocking detection is realized.
Referring to fig. 3, determining pixel discrete information of an encoding block according to pixel values of each pixel point in the encoding block according to an embodiment of the present application includes:
s1201, calculating an average pixel value of the coding block according to the pixel values of all pixel points in the coding block;
s1202, calculating pixel standard deviation of the coding block based on the average pixel value, and taking the average pixel value and the pixel standard deviation as pixel discrete information of the coding block.
The block size of the current coding block is presumed through the block boundary, and then the sum of pixel values of all pixel points in the current coding block is calculated. Dividing the sum of the pixel values by the number of the pixel points of the coding block to obtain an average pixel value mu of the current coding block. And then according to the average pixel value and the pixel value of each pixel point, the pixel standard deviation sigma of the current coding block can be calculated, so as to obtain the pixel discrete information of the current coding block.
And comparing the pixel standard deviation with the set pixel weight coefficient, calculating the picture content weight coefficient based on the pixel discrete information and the set frame coefficient under the condition that the pixel standard deviation is smaller than or equal to the pixel weight coefficient, and calculating the picture content weight coefficient based on the pixel discrete information under the condition that the pixel standard deviation is larger than the pixel weight coefficient, wherein the set frame coefficient is used for representing the sensitivity degree of human eyes to the video image.
It will be appreciated that the degree of sensitivity of the human eye to blocking will vary from image content to image content, taking into account the effect of masking of the human eye on subjective perception of the intensity of blocking. Wherein the perception of blocking by the human eye is more pronounced in flat areas and less sensitive in areas of rich detail. Detecting image textures according to the pixel standard deviation sigma of the current coding block, wherein the smaller the pixel standard deviation sigma is, the flatter the current region is, the more sensitive the human eyes are to block effect perception, and the picture content weight coefficient weight_texture value of the human eyes is relatively larger; the larger the pixel standard deviation sigma is, the more abundant the detail of the current region is, the less sensitive the human eyes are to the blocking effect perception, and the weight coefficient weight_texture of the picture content should take on a relatively smaller value.
Based on this, a picture content weight coefficient weight_texture calculation formula is provided as follows:
wherein ζ represents the pixel weight coefficient, and the default value is 81. The pixel weight coefficient is expressed as a peak value of the sensitivity of the human eye to the pixels of the image, and the closer the pixel value is to the pixel weight coefficient, the more sensitive the human eye is. λ represents the set frame coefficient, and the set frame coefficient calculation formula is as follows:
and according to the pixel weight coefficient zeta, the average pixel value mu of the current coding block and the standard deviation sigma of the current coding block, inputting the parameters into the picture content weight coefficient weight_texture calculation formula, and calculating the picture content weight coefficient weight_texture of the current coding block.
And multiplying the surface content weight coefficient weight_texture and the initial blockiness information based on the calculated surface content weight coefficient weight_texture and the initial blockiness information to obtain the blockiness strength of the comprehensive human eye subjective feeling influence.
Alternatively, the essential cause of the generation due to the blocking effect is an error generated when the DCT transformed coefficients are quantized. The larger the quantization step qp, the larger such error will be. Likewise, the larger the quantization step qp difference between adjacent blocks, the larger such error. Therefore, when the block effect intensity is calculated, the block effect intensity is weighted and calculated according to the quantization step qp of the current coding block and the adjacent block, so that the block effect intensity detection precision is further improved.
Also, according to actual observations of a video bitstream, when a block type is intra-coded block intra, coding of intra blocks is more code-rate consuming and thus more prone to blocking artifacts. Based on this, the embodiment of the application further performs a weighted calculation on the blockiness intensity according to the block type of the encoding block, so as to further improve the blockiness intensity detection precision.
Finally, by determining the block type weight coefficient and the quantization step size weight coefficient of the coding block, taking the product of the block type weight coefficient, the quantization step size weight coefficient, the picture content weight coefficient and the initial blockiness information as the blockiness intensity of the coding block.
The blockiness calculation formula is expressed as
blockiness＝blockiness simple *weight texture *weight qp *weight type
Wherein weight is qp Representing quantization step weight coefficients, weight type Representing block type weighting coefficients.
When determining the block type weight coefficient and the quantization step weight coefficient of the coding block, determining the block type weight coefficient and the quantization step weight coefficient according to the block type of the coding block, acquiring a first quantization step length of the coding block and a second quantization step length of an adjacent coding block with the largest first-order difference with the coding block under the condition that the block type of the coding block is of a specified type, and determining the quantization step length weight coefficient based on the first quantization step length and the second quantization step length.
And selecting the edge with the largest first-order difference of the current coding block according to the calculated edge gradient, and recording the direction of the edge. And further acquiring quantization step length and block type information of the adjacent coding blocks of the maximum edge. And then determining a block type weight coefficient and a quantization step weight coefficient according to the block type of the current coding block.
If the block type of the current coding block is skip, the current coding block does not have a quantization step qp, at this time, a quantization step weight coefficient weight_qp is directly set to be 1, and a block type weight coefficient weight_type is set to be 1, and the current coding block is directly substituted into the block effect intensity calculating formula to calculate the block effect intensity.
If the block type of the current coding block is not skip, a quantization step weight coefficient weight_qp is calculated according to the first quantization step qp of the current coding block and the second quantization step qp_ nei of the neighboring block. The quantization step weight coefficient weight_qp is calculated as follows:
further, if the block type of the current coding block is intra coding block intra and the standard deviation sigma of the current coding block is greater than the set threshold Th (Th default to 5), weight_type is directly set to 1.5.
And determining block type weight coefficients and quantization step length weight coefficients according to different block type adaptability, substituting the block type weight coefficients and the quantization step length weight coefficients into a quantization step length weight coefficient calculation formula, and calculating the final blockiness.
To this end, the subjective feeling of the human eye is matched by detecting the exact position where the video image appears to be blocky and evaluating the intensity of the blocky effect. The method can solve the problems of inaccurate positioning of the block boundary and mismatching of the human eye perception intensity in the block effect detection process, greatly avoid misjudgment of the real edge, obtain accurate block effect position and block effect intensity evaluation results, provide accurate and proper detection data for subsequent block effect removal operation, and optimize video image quality.
The method comprises the steps of determining an internal gradient and an edge gradient of a coding block by reading the coding block of a video image, and determining initial blockiness information of the coding block based on the internal gradient and the edge gradient; determining pixel discrete information of the coding block according to pixel values of all pixel points in the coding block, and determining a picture content weight coefficient of the coding block based on the pixel discrete information; the blockiness intensity of the encoded block is calculated based on the initial blockiness information and the picture content weight coefficient. By adopting the technical means, the block effect sensitivity degree of human eyes to the current coding block is determined through the pixel discrete information of the coding block, and then the block effect intensity is calculated by combining the pixel discrete information and the initial block effect information of the coding block, so that the block effect intensity of the image coding block can be accurately detected by considering the sensitivity degree of human eyes to the block effect, and the detection error of the block effect is reduced. Therefore, the blocking effect is removed, the blocking effect removing effect can be improved, and the image quality of the video image is improved.
Fig. 4 is a schematic structural diagram of an image blocking effect detection system according to the present application based on the above embodiment. Referring to fig. 4, the image blocking effect detection system provided in this embodiment specifically includes: an information determination module 21, a coefficient determination module 22 and a calculation module 23.
Wherein the information determination module 21 is configured to read an encoded block of the video image, determine an internal gradient and an edge gradient of the encoded block, and determine initial blockiness information of the encoded block based on the internal gradient and the edge gradient;
the coefficient determining module 22 is configured to determine pixel discrete information of the encoding block according to pixel values of each pixel point in the encoding block, and determine a picture content weight coefficient of the encoding block based on the pixel discrete information;
the calculation module 23 is configured to calculate the blockiness intensity of the encoded block based on the initial blockiness information and the picture content weight coefficient.
Specifically, determining the internal gradient and the edge gradient of the encoded block includes:
calculating the first-order difference of pixels in the coding block, and superposing to obtain an internal gradient;
and calculating first-order differences between pixel points of four edges of the coding block and pixel points of adjacent coding blocks, and overlapping to obtain edge gradients.
Specifically, determining initial blockiness information of the encoded block based on the internal gradient and the edge gradient includes:
acquiring the number of internal differential pixel pairs and the number of edge differential pixel pairs of a coding block, determining a first ratio of an internal gradient to the number of internal differential pixel pairs, and determining a second ratio of the edge gradient to the number of edge differential pixel pairs;
dividing the second ratio by the first ratio under the condition that the internal gradient is not zero to obtain initial blockiness information; in the case where the internal gradient is zero, the second ratio is taken as the initial blockiness information.
Specifically, determining the pixel discrete information of the coding block according to the pixel value of each pixel point in the coding block includes:
calculating an average pixel value of the coding block according to the pixel values of all the pixel points in the coding block;
the pixel standard deviation of the coding block is calculated based on the average pixel value, and the average pixel value and the pixel standard deviation are used as the pixel discrete information of the coding block.
Specifically, determining the picture content weighting factor of the encoded block based on the pixel discrete information includes:
comparing the pixel standard deviation with the set pixel weight coefficient, calculating the picture content weight coefficient based on the pixel discrete information and the set frame coefficient under the condition that the pixel standard deviation is smaller than or equal to the pixel weight coefficient, and calculating the picture content weight coefficient based on the pixel discrete information under the condition that the pixel standard deviation is larger than the pixel weight coefficient, wherein the set frame coefficient is used for representing the sensitivity degree of human eyes to the video image.
Specifically, calculating the blockiness intensity of the encoded block based on the initial blockiness information and the picture content weight coefficient includes:
and determining a block type weight coefficient and a quantization step size weight coefficient of the coding block, wherein the product of the block type weight coefficient, the quantization step size weight coefficient, the picture content weight coefficient and the initial block effect information is used as the block effect intensity of the coding block.
Specifically, determining a block type weight coefficient and a quantization step weight coefficient of an encoded block includes:
and determining a block type weight coefficient and a quantization step size weight coefficient according to the block type of the coding block, acquiring a first quantization step size of the coding block and a second quantization step size of an adjacent coding block with the largest first-order difference with the coding block under the condition that the block type of the coding block is of a specified type, and determining the quantization step size weight coefficient based on the first quantization step size and the second quantization step size.
The method comprises the steps of determining an internal gradient and an edge gradient of a coding block by reading the coding block of a video image, and determining initial blockiness information of the coding block based on the internal gradient and the edge gradient; determining pixel discrete information of the coding block according to pixel values of all pixel points in the coding block, and determining a picture content weight coefficient of the coding block based on the pixel discrete information; the blockiness intensity of the encoded block is calculated based on the initial blockiness information and the picture content weight coefficient. By adopting the technical means, the block effect sensitivity degree of human eyes to the current coding block is determined through the pixel discrete information of the coding block, and then the block effect intensity is calculated by combining the pixel discrete information and the initial block effect information of the coding block, so that the block effect intensity of the image coding block can be accurately detected by considering the sensitivity degree of human eyes to the block effect, and the detection error of the block effect is reduced. Therefore, the blocking effect is removed, the blocking effect removing effect can be improved, and the image quality of the video image is improved.
The image blocking effect detection system provided by the embodiment of the application can be configured to execute the image blocking effect detection method provided by the embodiment, and has corresponding functions and beneficial effects.
On the basis of the above practical example, the embodiment of the present application further provides an image blocking effect detection apparatus, referring to fig. 5, including: processor 31, memory 32, communication module 33, input device 34 and output device 35. The memory 32 is configured as a computer-readable storage medium configured to store a software program, a computer-executable program, and modules, such as program instructions/modules (e.g., an information determination module, a coefficient determination module, and a calculation module in an image blocking effect detection system) corresponding to the image blocking effect detection method according to any embodiment of the present application. The communication module 33 is configured to perform data transmission. The processor 31 executes various functional applications of the apparatus and data processing by executing software programs, instructions and modules stored in the memory, i.e., implements the image blocking detection method described above. The input device 34 may be configured to receive input numeric or character information and to generate key signal inputs related to user settings and function control of the apparatus. The output means 35 may comprise a display device such as a display screen. The image blocking effect detection device provided by the embodiment can be configured to execute the image blocking effect detection method provided by the embodiment, and has corresponding functions and beneficial effects.
On the basis of the above embodiments, the present application further provides a computer-readable storage medium storing computer-executable instructions that, when executed by a computer processor, are configured to perform an image blocking detection method, and the storage medium may be any of various types of memory devices or storage devices. Of course, the computer-readable storage medium provided in the embodiments of the present application, whose computer-executable instructions are not limited to the image blocking effect detection method described above, may also perform the related operations in the image blocking effect detection method provided in any embodiment of the present application.
On the basis of the above embodiments, the embodiments of the present application further provide a computer program product, where the technical solution of the present application is essentially or a part contributing to the prior art or all or part of the technical solution may be embodied in the form of a software product, and the computer program product is stored in a storage medium, and includes several instructions to cause a computer device, a mobile terminal or a processor therein to execute all or part of the steps of the image blocking effect detection method according to the embodiments of the present application.
Claims (11)
1. An image blocking detection method, comprising:
reading a coding block of a video image, determining an internal gradient and an edge gradient of the coding block, and determining initial blockiness information of the coding block based on the internal gradient and the edge gradient;
determining pixel discrete information of the coding block according to pixel values of all pixel points in the coding block, and determining a picture content weight coefficient of the coding block based on the pixel discrete information;
and calculating the blockiness intensity of the coding block based on the initial blockiness information and the picture content weight coefficient.
2. The image blocking effect detection method according to claim 1, wherein the determining the pixel discrete information of the encoded block according to the pixel values of the respective pixel points in the encoded block includes:
calculating an average pixel value of the coding block according to the pixel values of all pixel points in the coding block;
and calculating the pixel standard deviation of the coding block based on the average pixel value, wherein the average pixel value and the pixel standard deviation are used as pixel discrete information of the coding block.
3. The image blocking effect detection method according to claim 2, wherein the determining the picture content weight coefficient of the encoded block based on the pixel discrete information includes:
and comparing the pixel standard deviation with a set pixel weight coefficient, calculating the picture content weight coefficient based on the pixel discrete information and a set frame coefficient under the condition that the pixel standard deviation is smaller than or equal to the pixel weight coefficient, and calculating the picture content weight coefficient based on the pixel discrete information under the condition that the pixel standard deviation is larger than the pixel weight coefficient, wherein the set frame coefficient is used for representing the sensitivity degree of human eyes to the video image.
4. The image blocking artifact detection method according to claim 1, wherein the determining the internal gradient and the edge gradient of the encoded block comprises:
calculating a first-order difference of pixels in the coding block, and superposing to obtain the internal gradient;
and calculating first-order differences between pixel points of four edges of the coding block and pixel points of adjacent coding blocks, and superposing to obtain the edge gradient.
5. The image blocking effect detection method according to claim 1, wherein the determining initial blocking effect information of the encoded block based on the internal gradient and the edge gradient includes:
acquiring the number of internal differential pixel pairs and the number of edge differential pixel pairs of the coding block, determining a first ratio of the internal gradient to the number of internal differential pixel pairs, and determining a second ratio of the edge gradient to the number of edge differential pixel pairs;
dividing the second ratio by the first ratio to obtain the initial blockiness information under the condition that the internal gradient is not zero;
and taking the second ratio as the initial blockiness information in the case that the internal gradient is zero.
6. The image blocking effect detection method according to claim 1, wherein the calculating the blocking effect intensity of the encoded block based on the initial blocking effect information and the picture content weight coefficient includes:
and determining a block type weight coefficient and a quantization step size weight coefficient of the coding block, wherein the product of the block type weight coefficient, the quantization step size weight coefficient, the picture content weight coefficient and the initial blockiness information is used as the blockiness intensity of the coding block.
7. The image blocking artifact detection method according to claim 6, wherein the determining the block type weight coefficient and the quantization step weight coefficient of the encoded block comprises:
and determining the block type weight coefficient and the quantization step size weight coefficient according to the block type of the coding block, acquiring a first quantization step size of the coding block and a second quantization step size of an adjacent coding block with the largest first-order difference with the coding block under the condition that the block type of the coding block is a specified type, and determining the quantization step size weight coefficient based on the first quantization step size and the second quantization step size.
8. An image blocking detection system, comprising:
an information determination module configured to read a coding block of a video image, determine an internal gradient and an edge gradient of the coding block, and determine initial blockiness information of the coding block based on the internal gradient and the edge gradient;
a coefficient determining module configured to determine pixel discrete information of the encoding block according to pixel values of each pixel point in the encoding block, and determine a picture content weight coefficient of the encoding block based on the pixel discrete information;
and a calculation module configured to calculate a blockiness intensity of the encoded block based on the initial blockiness information and the picture content weight coefficient.
9. An image blocking effect detection apparatus, characterized by comprising:
a memory and one or more processors;
the memory is configured to store one or more programs;
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the image blocking detection method of any of claims 1-7.
10. A computer readable storage medium storing computer executable instructions which, when executed by a computer processor, are configured to perform the image blocking detection method according to any of claims 1-7.
11. A computer program product comprising instructions which, when executed on a computer or processor, cause the computer or processor to perform the image blocking detection method according to any of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310019966.9A CN116132697A (en) | 2023-01-06 | 2023-01-06 | Image blocking effect detection method, system, equipment and storage medium |
| PCT/CN2023/137328 WO2024146329A1 (en) | 2023-01-06 | 2023-12-08 | Image blocking artifact detection method, system and device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310019966.9A CN116132697A (en) | 2023-01-06 | 2023-01-06 | Image blocking effect detection method, system, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116132697A true CN116132697A (en) | 2023-05-16 |
Family
ID=86307574
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310019966.9A Pending CN116132697A (en) | 2023-01-06 | 2023-01-06 | Image blocking effect detection method, system, equipment and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN116132697A (en) |
| WO (1) | WO2024146329A1 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024146329A1 (en) * | 2023-01-06 | 2024-07-11 | 百果园技术(新加坡)有限公司 | Image blocking artifact detection method, system and device, and storage medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100345450C (en) * | 2005-01-31 | 2007-10-24 | 浙江大学 | Deblocking filtering method and apparatus of video frequency or image |
| JP4945533B2 (en) * | 2008-09-09 | 2012-06-06 | 株式会社東芝 | Image processing apparatus and image processing method |
| CN111711825B (en) * | 2020-06-23 | 2024-04-30 | 腾讯科技（深圳）有限公司 | Deblocking filtering method, device, equipment and medium in video coding and decoding |
| CN114125472B (en) * | 2021-11-26 | 2023-10-24 | 湖南中科助英智能科技研究院有限公司 | Blocking effect detection method, device, equipment and storage medium based on airspace perception |
| CN116132697A (en) * | 2023-01-06 | 2023-05-16 | 百果园技术(新加坡)有限公司 | Image blocking effect detection method, system, equipment and storage medium |
-
2023
- 2023-01-06 CN CN202310019966.9A patent/CN116132697A/en active Pending
- 2023-12-08 WO PCT/CN2023/137328 patent/WO2024146329A1/en unknown
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024146329A1 (en) * | 2023-01-06 | 2024-07-11 | 百果园技术(新加坡)有限公司 | Image blocking artifact detection method, system and device, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2024146329A1 (en) | 2024-07-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20240155147A1 (en) | Content adaptive deblocking during video encoding and decoding | |
| CN101175209B (en) | Adaptive de-blocking filtering method for mpeg video decoder | |
| CN101390401B (en) | Enhanced image/video quality through artifact evaluation | |
| US20110255589A1 (en) | Methods of compressing data and methods of assessing the same | |
| WO2023274074A1 (en) | Systems and methods for image filtering | |
| US20140321552A1 (en) | Optimization of Deblocking Filter Parameters | |
| JP2022522217A (en) | Bit width control methods and devices for bidirectional optical flow | |
| WO2013075318A1 (en) | Video quality measurement | |
| CN109803146B (en) | Method, apparatus, medium, and device for secondary compression of video | |
| Kim et al. | Compression artifacts removal by signal adaptive weighted sum technique | |
| WO2024146329A1 (en) | Image blocking artifact detection method, system and device, and storage medium | |
| US20110317912A1 (en) | Method, apparatus and computer-readable medium coding and decoding depth image using color image | |
| US20130235931A1 (en) | Masking video artifacts with comfort noise | |
| CN112001854A (en) | Method for repairing coded image and related system and device | |
| KR20040106416A (en) | Method of processing digital images for low-rate applications | |
| CN104113761A (en) | Code rate control method for video encoding and encoder | |
| CN101729883A (en) | Method for adaptively adjusting video decoding complexity | |
| US20140334555A1 (en) | Method and apparatus for video quality measurement | |
| JP4266227B2 (en) | Video signal processing device | |
| US20120133836A1 (en) | Frame level quantization estimation | |
| CN109587487B (en) | Method and system for evaluating structural distortion degree of RDO strategy | |
| US20020076119A1 (en) | Quality rating function for a discrete decoded picture | |
| US7706440B2 (en) | Method for reducing bit rate requirements for encoding multimedia data | |
| JP2001346208A (en) | Image signal decoder and method | |
| CN117115433B (en) | Display abnormality detection method, device, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |