• Laser & Optoelectronics Progress
  • Vol. 56, Issue 22, 221004 (2019)
Yakang Duan, Lin Luo, Jinlong Li*, and Xiaorong Gao
Author Affiliations
  • School of Physical Science and Technology, Southwest Jiaotong University, Chengdu, Sichuan 610031, China
  • show less
    DOI: 10.3788/LOP56.221004 Cite this Article Set citation alerts
    Yakang Duan, Lin Luo, Jinlong Li, Xiaorong Gao. Super-Resolution Reconstruction of Astronomical Images Based on Centralized Sparse Representation[J]. Laser & Optoelectronics Progress, 2019, 56(22): 221004 Copy Citation Text show less
    Flowchart of agglomerative hierarchical clustering
    Fig. 1. Flowchart of agglomerative hierarchical clustering
    Hierarchical clustering tree
    Fig. 2. Hierarchical clustering tree
    Test image set. (a) Cluster; (b) Galaxy; (c) Jupiter; (d) Satellite; (e) Saturn
    Fig. 3. Test image set. (a) Cluster; (b) Galaxy; (c) Jupiter; (d) Satellite; (e) Saturn
    Effects of different experimental parameters on reconstruction results. (a) Clustering number; (b) γ
    Fig. 4. Effects of different experimental parameters on reconstruction results. (a) Clustering number; (b) γ
    Super-resolution reconstruction results of Satellite with scale factor of 3. (a) Original image; (b) bicubic interpolation algorithm; (c) ScSR algorithm; (d) Zeyde algorithm; (e) ANR algorithm; (f) ASDS algorithm; (g) NCSR algorithm; (h) proposed algorithm
    Fig. 5. Super-resolution reconstruction results of Satellite with scale factor of 3. (a) Original image; (b) bicubic interpolation algorithm; (c) ScSR algorithm; (d) Zeyde algorithm; (e) ANR algorithm; (f) ASDS algorithm; (g) NCSR algorithm; (h) proposed algorithm
    Super-resolution reconstruction results of Saturn with scale factor of 3. (a) Original image; (b) bicubic interpolation algorithm; (c) ScSR algorithm; (d) Zeyde algorithm; (e) ANR algorithm; (f) ASDS algorithm; (g) NCSR algorithm; (h) proposed algorithm
    Fig. 6. Super-resolution reconstruction results of Saturn with scale factor of 3. (a) Original image; (b) bicubic interpolation algorithm; (c) ScSR algorithm; (d) Zeyde algorithm; (e) ANR algorithm; (f) ASDS algorithm; (g) NCSR algorithm; (h) proposed algorithm
    ImageBicubicScSRZeydeANRASDSNCSRProposed method
    Cluster25.6426.8128.3628.2728.4528.9729.12
    Galaxy33.5735.3737.3837.2437.3338.1538.35
    Jupiter31.7434.2535.2035.3635.7236.0536.06
    Satellite28.1733.1934.0734.5037.3137.4637.69
    Saturn27.2431.3032.9932.3534.0734.2334.51
    Average29.2832.1833.6033.5434.5834.9735.15
    Table 1. Comparison of PSNR of reconstructed images obtained by different methodsdB
    ImageBicubicScSRZeydeANRASDSNCSRProposed method
    Cluster0.65780.66180.76330.76240.77270.79130.7999
    Galaxy0.90620.90270.94120.94010.93940.94710.9505
    Jupiter0.88240.90580.92460.92740.92980.93570.9362
    Satellite0.94280.94700.97380.97420.98570.98770.9885
    Saturn0.94030.92450.97030.96140.97970.97870.9807
    Average0.86590.86840.91460.91310.92150.92810.9312
    Table 2. Comparison of SSIM of reconstructed images obtained by different methods
    ImageANRASDSNCSRProposed method
    Cluster0.3190.2355.5362.5
    Galaxy0.3157.2337.3348.3
    Jupiter0.6331.7702.4680.3
    Satellite0.7374.8681.4665.8
    Saturn0.3183.1348.8335.6
    Average0.4247.4485.1478.5
    Table 3. Comparison of reconstruction time of different methodss
    Yakang Duan, Lin Luo, Jinlong Li, Xiaorong Gao. Super-Resolution Reconstruction of Astronomical Images Based on Centralized Sparse Representation[J]. Laser & Optoelectronics Progress, 2019, 56(22): 221004
    Download Citation