CN-121985140-A - Image decoding and encoding method, device, equipment and storage medium

Abstract

The application belongs to the technical field of image processing, and discloses an image decoding and encoding method, an image decoding and encoding device, image decoding and encoding equipment and a storage medium. The application decodes the image code stream, determines the characteristic reconstruction value corresponding to the current image characteristic obtained by decoding, enhances the characteristic reconstruction value to obtain an enhanced characteristic value, and synthesizes and transforms the enhanced characteristic value to obtain a reconstructed image block. Before image reconstruction, the feature reconstruction value is subjected to feature enhancement, and then the image reconstruction is performed according to the enhanced feature enhancement value, so that the distortion of the image features in the processes of quantization and the like is reduced, and the image quality of the reconstructed image is improved.

Inventors

• CHEN FANGDONG
• WU XIAOYANG

Assignees

• 杭州海康威视数字技术股份有限公司

Dates

Publication Date

20260505

Application Date

20230113

Claims (20)

1. 1. An image decoding method, characterized in that the image decoding method comprises the steps of: Decoding the image code stream, and determining a feature reconstruction value corresponding to the current image feature obtained by decoding; Performing feature enhancement on the feature reconstruction value to obtain an enhanced feature value; and carrying out synthesis transformation on the enhanced characteristic values through a synthesis transformation network to obtain a reconstructed image block.
2. 2. The image decoding method of claim 1, wherein the current image feature is a three-dimensional feature matrix; the step of performing feature enhancement on the feature reconstruction value to obtain an enhanced feature value comprises the following steps: Acquiring characteristic standard deviation and standard deviation characterization values corresponding to matrix elements in the current image characteristics; determining a feature mask corresponding to each matrix element in the current image feature according to the feature standard deviation, the standard deviation representation value and a preset definition threshold; and carrying out feature enhancement on the feature reconstruction value based on the feature mask to obtain an enhanced feature value.
3. 3. The image decoding method according to claim 2, wherein the determining the feature mask corresponding to each matrix element in the current image feature according to the feature standard deviation, the standard deviation characterization value, and a preset definition threshold value includes: If the characteristic standard deviation and the standard deviation representation value corresponding to the matrix element meet the preset enhancement condition, setting the characteristic mask corresponding to the matrix element as a first type value; If the characteristic standard deviation and the standard deviation representation value corresponding to the matrix element do not meet the preset enhancement condition, setting the characteristic mask corresponding to the matrix element as a second type value.
4. 4. The image decoding method as claimed in claim 3, wherein before setting the feature mask corresponding to the matrix element to the first type value if the feature standard deviation and the standard deviation representation value corresponding to the matrix element satisfy the preset enhancement condition, the method further comprises: If the characteristic standard deviation corresponding to the matrix element is larger than a preset defining threshold value and the standard deviation representation value is a first type representation value, judging that the characteristic standard deviation and the standard deviation representation value corresponding to the matrix element meet a preset enhancement condition; Or alternatively, the first and second heat exchangers may be, If the characteristic standard deviation corresponding to the matrix element is smaller than a preset defined threshold value and the standard deviation representation value is a second type representation value, judging that the characteristic standard deviation and the standard deviation representation value corresponding to the matrix element meet a preset enhancement condition.
5. 5. The image decoding method according to claim 2, wherein the feature enhancing the feature reconstruction value based on the feature mask to obtain an enhanced feature value includes: taking the matrix element with the corresponding feature mask as a first type value as a target matrix element; and enhancing the characteristic reconstruction value corresponding to the target matrix element to obtain an enhanced characteristic value.
6. 6. The image decoding method as claimed in claim 5, wherein the enhancing the feature reconstruction value corresponding to the target matrix element to obtain an enhanced feature value includes: acquiring a characteristic reconstruction value, a residual reconstruction value and a predicted characteristic value corresponding to the target matrix element; determining a first enhancement value according to the first scaling factor and the residual error reconstruction value, and determining a second enhancement value according to the second scaling factor and the prediction characteristic value; And determining an enhancement characteristic value according to the characteristic reconstruction value, the first enhancement value and the second enhancement value.
7. 7. The method of image decoding according to claim 6, wherein before determining a first enhancement value according to a first scaling factor and the prediction feature value and determining a second enhancement value according to a second scaling factor and the residual reconstruction value, further comprising: Acquiring a characteristic channel corresponding to the target matrix element; and determining a first scaling factor and a second scaling factor according to the characteristic channels, wherein different characteristic channels correspond to different first scaling factors and different second scaling factors.
8. 8. The image decoding method as claimed in claim 5, wherein the enhancing the feature reconstruction value corresponding to the target matrix element to obtain an enhanced feature value includes: Acquiring a characteristic reconstruction value and a predicted characteristic value corresponding to the target matrix element; Determining a first enhancement value according to the first scaling factor and the characteristic reconstruction value, and determining a second enhancement value according to the second scaling factor and the predicted characteristic value; And determining an enhancement characteristic value according to the first enhancement value and the second enhancement value.
9. 9. The image decoding method as claimed in claim 5, wherein the enhancing the feature reconstruction value corresponding to the target matrix element to obtain an enhanced feature value includes: acquiring a characteristic reconstruction value and a residual reconstruction value corresponding to the target matrix element; determining a first enhancement value according to the first scaling factor and the characteristic reconstruction value, and determining a second enhancement value according to the second scaling factor and the residual reconstruction value; And determining an enhancement characteristic value according to the first enhancement value and the second enhancement value.
10. 10. The image decoding method as claimed in claim 5, wherein the enhancing the feature reconstruction value corresponding to the target matrix element to obtain an enhanced feature value includes: acquiring a characteristic reconstruction value and a characteristic standard deviation corresponding to the target matrix element; determining a first enhancement value according to the first scaling factor and the characteristic reconstruction value, and determining a second enhancement value according to the second scaling factor and the characteristic standard deviation; And determining an enhancement characteristic value according to the first enhancement value and the second enhancement value.
11. 11. The image decoding method of claim 5, wherein the feature reconstruction values include a luminance reconstruction value and a chrominance reconstruction value, and the enhancement feature values include a luminance enhancement feature value and a chrominance enhancement feature value; The enhancing the feature reconstruction value corresponding to the target matrix element to obtain an enhanced feature value comprises the following steps: Enhancing the chromaticity reconstruction value corresponding to the target matrix element to obtain a chromaticity enhancement characteristic value; and enhancing the brightness reconstruction value corresponding to the target matrix element to obtain a brightness enhancement characteristic value.
12. 12. The image decoding method as claimed in claim 11, wherein the enhancing the chroma reconstruction value corresponding to the target matrix element, after obtaining the chroma enhancement feature value, further comprises: determining a first enhancement value according to the brightness reconstruction value and a first scaling factor; And secondarily enhancing the chroma enhancement characteristic value according to the first enhancement value.
13. 13. The method of image decoding according to claim 12, wherein said determining a first enhancement value from said luminance reconstruction value and a first scaling factor further comprises: extracting component indication parameters from the image code stream; And if the component indicating parameter is a chroma enhancement type parameter, determining a first enhancement value according to the brightness reconstruction value and a first scaling factor.
14. 14. The image decoding method according to any one of claims 1 to 13, wherein decoding the image code stream and determining a feature reconstruction value corresponding to a current image feature obtained by decoding, includes: decoding the image code stream, and determining a residual error reconstruction value corresponding to the current image characteristic obtained by decoding; Predicting according to the characteristic reconstruction value of the reconstructed characteristic to obtain a predicted characteristic value; and determining a characteristic reconstruction value corresponding to the current image characteristic according to the residual error reconstruction value and the predicted characteristic value.
15. 15. The image decoding method as claimed in claim 14, wherein after decoding the image code stream and determining a residual reconstruction value corresponding to a current image feature obtained by decoding, the method further comprises: predicting according to the enhanced feature value of the reconstructed feature to obtain a predicted feature value; and determining a characteristic reconstruction value corresponding to the current image characteristic according to the residual error reconstruction value and the predicted characteristic value.
16. 16. The image decoding method of claim 14, wherein the enhancement feature values of the reconstructed feature are the same or different in feature enhancement from enhancement feature values corresponding to the current image feature.
17. 17. The image decoding method of claim 16, wherein the enhancement feature value of the reconstructed feature is determined to be the same as or different from the feature enhancement mode of the enhancement feature value corresponding to the current image feature by a syntax flag bit read from an image code stream.
18. 18. The image decoding method as claimed in claim 14, wherein the feature enhancing the feature reconstruction value to obtain an enhanced feature value comprises: extracting grammar application interval parameters from an image code stream, and acquiring feature position information corresponding to the current image features; Determining an enhancement grammar parameter according to the feature position information and the grammar application interval parameter; And carrying out characteristic enhancement on the characteristic reconstruction value according to the enhancement grammar parameter to obtain an enhancement characteristic value.
19. 19. An image decoding device, characterized in that it comprises the following modules: The code stream decoding module is used for decoding the image code stream and determining a feature reconstruction value corresponding to the current image feature obtained by decoding; The characteristic enhancement module is used for carrying out characteristic enhancement on the characteristic reconstruction value to obtain an enhanced characteristic value; and the image reconstruction module is used for carrying out synthesis transformation on the enhanced characteristic values through a synthesis transformation network to obtain a reconstructed image block.
20. 20. An image encoding method, characterized in that the image encoding method comprises: extracting features of an image block to be encoded, and taking the extracted features as current image features; predicting according to the feature reconstruction value corresponding to the reconstructed feature to obtain a predicted feature value; determining a coding residual coefficient corresponding to the current image characteristic according to the prediction characteristic value; And writing the coding residual error coefficients into an image code stream corresponding to the image block to be coded.

Description

Image decoding and encoding method, device, equipment and storage medium The present invention is a divisional application of Chinese patent application with application date 2023, 01, 13 and application number 202310055970.0, named "image decoding and encoding method, device, equipment and storage medium". Technical Field The present invention relates to the field of image processing technologies, and in particular, to an image decoding and encoding method, apparatus, device, and storage medium. Background Nowadays, deep learning and a neural network are continuously broken through in the field of video image compression, an image coding technology based on the deep learning has greatly led the traditional coding standard in coding performance, and how to improve the image quality of an image obtained by reconstruction after compression by an image coding technology based on the deep learning is a difficult problem. The foregoing is provided merely for the purpose of facilitating understanding of the technical solutions of the present invention and is not intended to represent an admission that the foregoing is prior art. Disclosure of Invention The invention mainly aims to provide an image decoding and encoding method, an image decoding and encoding device and a storage medium, and aims to improve the technical problem of image quality of an image obtained by reconstruction after compression by an image encoding technology based on depth learning. To achieve the above object, the present invention provides an image decoding method comprising the steps of: Decoding the image code stream, and determining a feature reconstruction value corresponding to the current image feature obtained by decoding; Performing feature enhancement on the feature reconstruction value to obtain an enhanced feature value; And carrying out synthesis transformation on the enhanced characteristic values to obtain a reconstructed image block. In a possible embodiment of the present application, the current image feature is a three-dimensional feature matrix; the step of performing feature enhancement on the feature reconstruction value to obtain an enhanced feature value comprises the following steps: Acquiring characteristic standard deviation and standard deviation characterization values corresponding to matrix elements in the current image characteristics; determining a feature mask corresponding to each matrix element in the current image feature according to the feature standard deviation, the standard deviation representation value and a preset definition threshold; and carrying out feature enhancement on the feature reconstruction value based on the feature mask to obtain an enhanced feature value. In one possible implementation manner of the present application, the determining the feature mask corresponding to each matrix element in the current image feature according to the feature standard deviation, the standard deviation representation value and a preset defining threshold value includes: If the characteristic standard deviation and the standard deviation representation value corresponding to the matrix element meet the preset enhancement condition, setting the characteristic mask corresponding to the matrix element as a first type value; If the characteristic standard deviation and the standard deviation representation value corresponding to the matrix element do not meet the preset enhancement condition, setting the characteristic mask corresponding to the matrix element as a second type value. In one possible implementation manner of the present application, before setting the feature mask corresponding to the matrix element to the first type value if the feature standard deviation and the standard deviation representation value corresponding to the matrix element satisfy the preset enhancement condition, the method further includes: If the characteristic standard deviation corresponding to the matrix element is larger than a preset defining threshold value and the standard deviation representation value is a first type representation value, judging that the characteristic standard deviation and the standard deviation representation value corresponding to the matrix element meet a preset enhancement condition; Or alternatively, the first and second heat exchangers may be, If the characteristic standard deviation corresponding to the matrix element is smaller than a preset defined threshold value and the standard deviation representation value is a second type representation value, judging that the characteristic standard deviation and the standard deviation representation value corresponding to the matrix element meet a preset enhancement condition. In a possible implementation manner of the present application, the feature enhancement is performed on the feature reconstruction value based on the feature mask, to obtain an enhanced feature value, which includes: taking the matrix element with the corresponding featu