• Laser & Optoelectronics Progress
  • Vol. 59, Issue 12, 1210003 (2022)
Wenkai Zhang1、2, Feng Xu1、2、*, and Jin Li1、2
Author Affiliations
  • 1School of Information Engineering, Southwest University of Science and Technology, Mianyang 621010, Sichuan , China
  • 2Robot Technology Used for Special Environment Key Laboratory of Sichuan Province, Mianyang 621010, Sichuan , China
  • show less
    DOI: 10.3788/LOP202259.1210003 Cite this Article Set citation alerts
    Wenkai Zhang, Feng Xu, Jin Li. Image Feature Extraction in Nuclear Radiation Environment[J]. Laser & Optoelectronics Progress, 2022, 59(12): 1210003 Copy Citation Text show less
    Image in nuclear environment
    Fig. 1. Image in nuclear environment
    Distribution of noise block
    Fig. 2. Distribution of noise block
    FAST features in image patch block
    Fig. 3. FAST features in image patch block
    Noise location detection in real-world environment. (a) Threshold is 0.4; (b) threshold is 0.5; (c) threshold is 0.6;(d) threshold is 0.7
    Fig. 4. Noise location detection in real-world environment. (a) Threshold is 0.4; (b) threshold is 0.5; (c) threshold is 0.6;(d) threshold is 0.7
    Screening results of features in real-world environment. (a) Image before screening; (b) set 50 features; (c) set 100 features; (d) set 150 features
    Fig. 5. Screening results of features in real-world environment. (a) Image before screening; (b) set 50 features; (c) set 100 features; (d) set 150 features
    Extraction results in real-world environment. (a) Image 1; (b) image 2; (c) image 3
    Fig. 6. Extraction results in real-world environment. (a) Image 1; (b) image 2; (c) image 3
    Results of ANF algorithm compared with FAST algorithm
    Fig. 7. Results of ANF algorithm compared with FAST algorithm
    Number of featuresBRISKKAZEANF
    505.1286.9665.077
    1005.8287.0735.454
    1506.3387.6276.111
    Table 1. Comparison of relative time of feature extraction
    Is /%ANFFASTBRISK
    0.010.99110.98540.9823
    0.030.97540.93770.9429
    0.050.92790.88930.8988
    0.070.92650.86220.8611
    0.090.89920.83710.8258
    Table 2. Rejection rate of feature extraction
    Is /%ANFFASTBRISK
    0.01112325
    0.0324113128
    0.0554166186
    0.0770214232
    0.09125326333
    Table 3. Number of noise points contained in features
    Wenkai Zhang, Feng Xu, Jin Li. Image Feature Extraction in Nuclear Radiation Environment[J]. Laser & Optoelectronics Progress, 2022, 59(12): 1210003
    Download Citation