• Laser & Optoelectronics Progress
  • Vol. 56, Issue 8, 080001 (2019)
Haoxian Wang*, Heng Dong**, and Zhiquan Zhou
Author Affiliations
  • School of Information and Electrical Engineering, Harbin Institute of Technology, Weihai, Shandong 264209, China
  • show less
    DOI: 10.3788/LOP56.080001 Cite this Article Set citation alerts
    Haoxian Wang, Heng Dong, Zhiquan Zhou. Review on Dim Small Target Detection Technologies in Infrared Single Frame Images[J]. Laser & Optoelectronics Progress, 2019, 56(8): 080001 Copy Citation Text show less
    Infrared images and three-dimensional (3D) intensity distributions. (a) Actual infrared image 1 containing small target; (b) 3D intensity distribution in small target area of image 1; (c) actual infrared image 2 containing small target; (d) 3D intensity distribution in small target area of image 2
    Fig. 1. Infrared images and three-dimensional (3D) intensity distributions. (a) Actual infrared image 1 containing small target; (b) 3D intensity distribution in small target area of image 1; (c) actual infrared image 2 containing small target; (d) 3D intensity distribution in small target area of image 2
    Infrared dim small targets in complex background. (a) Dim small target in sea-sky background; (b) bright target in sea-sky background; (c) dim small target in sky cloud background; (d) bright target in sky-ground background[12]
    Fig. 2. Infrared dim small targets in complex background. (a) Dim small target in sea-sky background; (b) bright target in sea-sky background; (c) dim small target in sky cloud background; (d) bright target in sky-ground background[12]
    Filter template structure of Robinson Guard with window size of 5×5
    Fig. 3. Filter template structure of Robinson Guard with window size of 5×5
    Nine parts of sliding window
    Fig. 4. Nine parts of sliding window
    Patch-image construction from original image by using image patches
    Fig. 5. Patch-image construction from original image by using image patches
    Steps of algorithm
    Fig. 6. Steps of algorithm
    Representative images in four datasets. (a) Dataset 1; (b) dataset 2; (c) dataset 3; (d) dataset 4
    Fig. 7. Representative images in four datasets. (a) Dataset 1; (b) dataset 2; (c) dataset 3; (d) dataset 4
    Four representative images in dataset 5. (a) Image 1; (b) image 2; (c) image 3; (d) image 4
    Fig. 8. Four representative images in dataset 5. (a) Image 1; (b) image 2; (c) image 3; (d) image 4
    MethodMax-MedianTop-HatLCMIPI
    Dataset 11.141.641.5140.27
    Dataset 21.472.171.7868.65
    Dataset 30.901.251.3357.55
    Dataset 41.001.451.5245.69
    Dataset 51.164.051.28123.86
    Table 1. GSCR of datasets 1-5 processed by different algorithms
    MethodMax-MedianTop-HatLCMIPI
    Dataset 17.517.250.9921.55
    Dataset 213.1612.671.0021.65
    Dataset 35.875.610.997.69
    Dataset 411.5211.021.0017.69
    Dataset 515.0713.241.0382.25
    Table 2. Background suppress factor (BSF) values of datasets 1-5 processed by different algorithms
    Haoxian Wang, Heng Dong, Zhiquan Zhou. Review on Dim Small Target Detection Technologies in Infrared Single Frame Images[J]. Laser & Optoelectronics Progress, 2019, 56(8): 080001
    Download Citation