• Infrared and Laser Engineering
  • Vol. 51, Issue 3, 20210468 (2022)
Bing Xie1, Shuhui Wan1, and Yunhua Yin2、3
Author Affiliations
  • 1Department of Network Security, Henan Police College, Zhengzhou 450046, China
  • 2School of Electronics and Control Engineering, North University of China, Taiyuan 030051, China
  • 3Science and Technology on Transient Impact Laboratory, Beijing 102202, China
  • show less
    DOI: 10.3788/IRLA20210468 Cite this Article
    Bing Xie, Shuhui Wan, Yunhua Yin. SR reconstruction algorithm of regularization based on improve of sparse representation[J]. Infrared and Laser Engineering, 2022, 51(3): 20210468 Copy Citation Text show less
    Sparse representation model in the transform domain ψ变换域下的稀疏表示模型
    Fig. 1. Sparse representation model in the transform domain ψ变换域 下的稀疏表示模型
    Sparse representation model under the dictionary
    Fig. 2. Sparse representation model under the dictionary
    Algorithm design framework
    Fig. 3. Algorithm design framework
    Schematic diagram of quadratic function is used to gradually approximate the optimized problem
    Fig. 4. Schematic diagram of quadratic function is used to gradually approximate the optimized problem
    Four HR test images
    Fig. 5. Four HR test images
    Results of each SR reconstruction algorithm in the simulation experiment
    Fig. 6. Results of each SR reconstruction algorithm in the simulation experiment
    LR waypoint image
    Fig. 7. LR waypoint image
    The results of each SR reconstruction algorithm in the actual test experiment
    Fig. 8. The results of each SR reconstruction algorithm in the actual test experiment
    Test imagesTowerBoatHouseAircraftAverage
    Sparse represent-ation regulariza-tion 34.5922.4527.2330.2128.62
    Autoregressive regulariza-tion 34.6522.5327.5130.3228.74
    Non-local similarity regulariza-tion 34.7922.3627.6030.6728.85
    Proposed algorithm34.9222.7128.6931.3329.41
    Table 1. PSNR results of each SR reconstruction algorithm (Unit: dB)
    Evaluation indexSparse representation regularizationAutoregressive regularizationNon-local similarity regularizationThe proposed algorithm
    SNR2.123.714.686.12
    IE2.312.863.234.23
    Table 2. SNR and IE comparison results of each SR reconstruction algorithm
    Bing Xie, Shuhui Wan, Yunhua Yin. SR reconstruction algorithm of regularization based on improve of sparse representation[J]. Infrared and Laser Engineering, 2022, 51(3): 20210468
    Download Citation