Single Image Super-Resolution Reconstruction with Preservation of Structure and Texture Details
In recent years, deep-learning-based single image super-resolution reconstruction has achieved good performance. However, most existing methods pursue a high peak signal-to-noise ratio (PSNR), while ignoring the quality of the structure and texture details, resulting in unsatisfactory performance of...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/11/1/216 |
| _version_ | 1827624462206369792 |
|---|---|
| author | Yafei Zhang Yuqing Huang Kaizheng Wang Guanqiu Qi Jinting Zhu |
| author_facet | Yafei Zhang Yuqing Huang Kaizheng Wang Guanqiu Qi Jinting Zhu |
| author_sort | Yafei Zhang |
| collection | DOAJ |
| description | In recent years, deep-learning-based single image super-resolution reconstruction has achieved good performance. However, most existing methods pursue a high peak signal-to-noise ratio (PSNR), while ignoring the quality of the structure and texture details, resulting in unsatisfactory performance of the reconstruction results in terms of human subjective perception. To solve this issue, this paper proposes a structure- and texture-preserving image super-resolution reconstruction method. Specifically, two different network branches are used to extract features for image structure and texture details. A dual-coordinate direction perception attention (DCDPA) mechanism is designed to highlight structure and texture features. The attention mechanism fully considers the complementarity and directionality of multi-scale image features and effectively avoids information loss and possible distortion of image structure and texture details during image reconstruction. Additionally, a cross-fusion mechanism is designed to comprehensively utilize structure and texture information for super-resolution image reconstruction, which effectively integrates the structure and texture details extracted by the two branch networks. Extensive experiments verify the effectiveness of the proposed method and its superiority over existing methods. |
| first_indexed | 2024-03-09T12:07:56Z |
| format | Article |
| id | doaj.art-b6f3a5c78c4c4f5dace65afb90b7b23c |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-09T12:07:56Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-b6f3a5c78c4c4f5dace65afb90b7b23c2023-11-30T22:55:42ZengMDPI AGMathematics2227-73902023-01-0111121610.3390/math11010216Single Image Super-Resolution Reconstruction with Preservation of Structure and Texture DetailsYafei Zhang0Yuqing Huang1Kaizheng Wang2Guanqiu Qi3Jinting Zhu4Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, ChinaFaculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, ChinaFaculty of Electrical Engineering, Kunming University of Science and Technology, Kunming 650500, ChinaDepartment of Computer Information Systems, State University of New York at Buffalo State, Buffalo, NY 14222, USASchool of Natural and Computational Sciences, Massey University at Auckland, Auckland 0632, New ZealandIn recent years, deep-learning-based single image super-resolution reconstruction has achieved good performance. However, most existing methods pursue a high peak signal-to-noise ratio (PSNR), while ignoring the quality of the structure and texture details, resulting in unsatisfactory performance of the reconstruction results in terms of human subjective perception. To solve this issue, this paper proposes a structure- and texture-preserving image super-resolution reconstruction method. Specifically, two different network branches are used to extract features for image structure and texture details. A dual-coordinate direction perception attention (DCDPA) mechanism is designed to highlight structure and texture features. The attention mechanism fully considers the complementarity and directionality of multi-scale image features and effectively avoids information loss and possible distortion of image structure and texture details during image reconstruction. Additionally, a cross-fusion mechanism is designed to comprehensively utilize structure and texture information for super-resolution image reconstruction, which effectively integrates the structure and texture details extracted by the two branch networks. Extensive experiments verify the effectiveness of the proposed method and its superiority over existing methods.https://www.mdpi.com/2227-7390/11/1/216deep neural networkssuper-resolution image reconstructionstructure and texture detailsattention mechanism |
| spellingShingle | Yafei Zhang Yuqing Huang Kaizheng Wang Guanqiu Qi Jinting Zhu Single Image Super-Resolution Reconstruction with Preservation of Structure and Texture Details Mathematics deep neural networks super-resolution image reconstruction structure and texture details attention mechanism |
| title | Single Image Super-Resolution Reconstruction with Preservation of Structure and Texture Details |
| title_full | Single Image Super-Resolution Reconstruction with Preservation of Structure and Texture Details |
| title_fullStr | Single Image Super-Resolution Reconstruction with Preservation of Structure and Texture Details |
| title_full_unstemmed | Single Image Super-Resolution Reconstruction with Preservation of Structure and Texture Details |
| title_short | Single Image Super-Resolution Reconstruction with Preservation of Structure and Texture Details |
| title_sort | single image super resolution reconstruction with preservation of structure and texture details |
| topic | deep neural networks super-resolution image reconstruction structure and texture details attention mechanism |
| url | https://www.mdpi.com/2227-7390/11/1/216 |
| work_keys_str_mv | AT yafeizhang singleimagesuperresolutionreconstructionwithpreservationofstructureandtexturedetails AT yuqinghuang singleimagesuperresolutionreconstructionwithpreservationofstructureandtexturedetails AT kaizhengwang singleimagesuperresolutionreconstructionwithpreservationofstructureandtexturedetails AT guanqiuqi singleimagesuperresolutionreconstructionwithpreservationofstructureandtexturedetails AT jintingzhu singleimagesuperresolutionreconstructionwithpreservationofstructureandtexturedetails |