• Laser & Optoelectronics Progress
  • Vol. 57, Issue 22, 222801 (2020)
Xinlei Ren1、* and Yangping Wang1、2、3、4
Author Affiliations
  • 1School of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou, Gansu 730070, China
  • 2National Experimental Teaching Demonstration Center of Computer Science and Technology, Lanzhou Jitotong University, Lanzhou, Gansu 730070, China
  • 3Gansu Provincial Engineering Research Center for Artificial Intelligence and Graphics & Image Processing, Lanzhou, Gansu 730070, China;
  • 4Gansu Provincial Key Laboratory of System Dynamics and Reliability of Rail Transport Equipment, Lanzhou, Gansu 730070, China
  • show less
    DOI: 10.3788/LOP57.222801 Cite this Article Set citation alerts
    Xinlei Ren, Yangping Wang. Super-Pixel Segmentation of Remote Sensing Image Based on Improved Simple Linear Iterative Clustering Algorithm[J]. Laser & Optoelectronics Progress, 2020, 57(22): 222801 Copy Citation Text show less
    Search space of K-means and SLIC algorithm. (a) K-means algorithm; (b) SLIC algorithm
    Fig. 1. Search space of K-means and SLIC algorithm. (a) K-means algorithm; (b) SLIC algorithm
    Super-pixel segmentation results of SLIC algorithm. (a) Original image; (b) K=100; (c) K=300; (d) K=500
    Fig. 2. Super-pixel segmentation results of SLIC algorithm. (a) Original image; (b) K=100; (c) K=300; (d) K=500
    Initialization process of clustering center. (a) Random selection; (b) computational gradient; (c) clustering center
    Fig. 3. Initialization process of clustering center. (a) Random selection; (b) computational gradient; (c) clustering center
    Super-pixel segmentation result obtained by SLIC algorithm. (a) Input image; (b) super-pixel segmentation result; (c) partial enlarged drawing
    Fig. 4. Super-pixel segmentation result obtained by SLIC algorithm. (a) Input image; (b) super-pixel segmentation result; (c) partial enlarged drawing
    Comparison of segmentation results. (a) SLIC algorithm; (b) improved algorithm
    Fig. 5. Comparison of segmentation results. (a) SLIC algorithm; (b) improved algorithm
    Super-pixel segmentation results of different algorithms (K=100). (a) SLIC0 algorithm; (b) Ref.[23]; (c) TurboPixel algorithm; (d) SLIC algorithm; (e) our algorithm
    Fig. 6. Super-pixel segmentation results of different algorithms (K=100). (a) SLIC0 algorithm; (b) Ref.[23]; (c) TurboPixel algorithm; (d) SLIC algorithm; (e) our algorithm
    Average running time of 5 algorithms
    Fig. 7. Average running time of 5 algorithms
    Segmentation accuracy of different algorithms. (a) UE; (b) ASA
    Fig. 8. Segmentation accuracy of different algorithms. (a) UE; (b) ASA
    AlgorithmSLIC0Ref.[23]TurboPixelSLICOurs
    Running time2.531.5518.053.4251.25
    Table 1. Running time of 5 algorithms (K=100) unit: s
    AlgorithmSLIC0Ref.[23]TurboPixelSLICOurs
    UE0.2660.2330.3580.2960.222
    ASA0.9590.9450.9350.9500.964
    Table 2. Edge fitting capabilities of different algorithms (K=100)
    Xinlei Ren, Yangping Wang. Super-Pixel Segmentation of Remote Sensing Image Based on Improved Simple Linear Iterative Clustering Algorithm[J]. Laser & Optoelectronics Progress, 2020, 57(22): 222801
    Download Citation