模式识别与人工智能
Thursday, Apr. 3, 2025 Home      About Journal      Editorial Board      Instructions      Ethics Statement      Contact Us                   中文
Pattern Recognition and Artificial Intelligence  2023, Vol. 36 Issue (5): 419-432    DOI: 10.16451/j.cnki.issn1003-6059.202305003
Papers and Reports Current Issue| Next Issue| Archive| Adv Search |
Lightweight Inverse Separable Residual Information Distillation Network for Image Super-Resolution Reconstruction
ZHAO Xiangqiang1,2,3, LI Xiyao1, SONG Zhaoyang1
1. College of Electrical and Information Engineering, Lanzhou University of Technology, Lanzhou 730050;
2. Key Laboratory of Gansu Advanced Control for Industrial Processes, Lanzhou University of Technology, Lanzhou 730050;
3. National Experimental Teaching Center of Electrical and Control Engineering, Lanzhou University of Technology, Lanzhou 730050

Download: PDF (1549 KB)   HTML (1 KB) 
Export: BibTeX | EndNote (RIS)      
Abstract  The application of the deep learning-based image super-resolution reconstruction algorithm on mobile devices is limited, due to the sharp increase of parameters and high computational cost caused by performance requirement. To solve this problem, a lightweight inverse separable residual information distillation network for image super-resolution reconstruction is proposed in this paper. Firstly, a progressive separable distillation shuffle module is designed to extract multi-level features and in the meantime keep the model lightweight. Multiple feature extraction connections are employed to learn a more distinguishing feature representation, and thus the network acquires more useful information from distillation. Then, a contrast perception coordinate attention module is designed to fully leverage channel-aware and position-sensitive information, enhancing the feature selection capability. Finally, a progressive compensation residual connection is put forward to improve the utilization of shallow features and compensate for the texture detail features of the network. Experiments show that the proposed algorithm achieves a good balance between model complexity and reconstruction performance, yielding excellent subjective and objective quality in the reconstructed high-resolution images.
Key wordsConvolutional Neural Network      Super-Resolution      Residual Network      Attention Mechanism      Information Distillation     
Received: 16 March 2023     
ZTFLH: TP391  
Fund:National Key Research and Development Program(No.2020YFB1713600), National Natural Science Foundation of China(No.61763029)
Corresponding Authors: ZHAO Xiaoqiang , Ph.D., professor. His research interests include image processing, fault diagnosis and data mining.   
About author:: LI Xiyao, master student. His research interests include image processing and deep learning.SONG Zhaoyang, Ph.D., lecturer. His research interests include image processing and deep learning
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
ZHAO Xiangqiang
LI Xiyao
SONG Zhaoyang
Cite this article:   
ZHAO Xiangqiang,LI Xiyao,SONG Zhaoyang. Lightweight Inverse Separable Residual Information Distillation Network for Image Super-Resolution Reconstruction[J]. Pattern Recognition and Artificial Intelligence, 2023, 36(5): 419-432.
URL:  
http://manu46.magtech.com.cn/Jweb_prai/EN/10.16451/j.cnki.issn1003-6059.202305003      OR     http://manu46.magtech.com.cn/Jweb_prai/EN/Y2023/V36/I5/419
Copyright © 2010 Editorial Office of Pattern Recognition and Artificial Intelligence
Address: No.350 Shushanhu Road, Hefei, Anhui Province, P.R. China Tel: 0551-65591176 Fax:0551-65591176 Email: bjb@iim.ac.cn
Supported by Beijing Magtech  Email:support@magtech.com.cn