Optics and Precision Engineering, Volume. 32, Issue 8, 1186(2024)

Gas leakage detection based on spatiotemporal information of low contrast infrared images

Jinhui ZUO1...2, Wenbin XU1,3, Shijie ZHOU4, Daobin SHENG5, Xiangdong XU7, Zhengqiang LI1,*, Yinghui HAN6, Chunjiang WU4 and Lei ZHANG5 |Show fewer author(s)
Author Affiliations
  • 1State Environmental Protection Key Laboratory of Satellite Remote Sensing & State Key Laboratory of Remote Sensing Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing000, China
  • 2University of Chinese Academy of Sciences, Beijing100049, China
  • 3Science and Technology on Optical Radiation Laboratory, Beijing Institute of Environmental Characteristics, Beijing100854, China
  • 4School of Information and Software Engineering, University of Electronic Science and Technology , Chengdu610000, China
  • 5Jiangsu Ancline Technology Co, Nantong226000, China
  • 6College of Resources and Environment, University of Chinese Academy of Sciences, Beijing100049, China
  • 7Sichuan Yifang Intelligent Technology Co, Chengdu610054, China
  • show less
    Figures & Tables(10)
    Uncooled long-wave infrared imaging system
    Infrared absorption spectra of Butane
    Experimental diagram (Case2)
    Filter results for different methods
    Results of different gas leak detection algorithms
    Comparison of Invalid probability and Error under different algorithms
    • Table 1. Basic process of the proposed algorithm

      View table
      View in Article

      Table 1. Basic process of the proposed algorithm

      算法1:基于低对比度红外图像时空信息的气体泄漏检测

      Input:ft(x,y)

      f¯t(x,y)←Time-domain adaptive inter frame filtering ft(x,y)

      if t<N

      spatiotemporal GMM update (ft_new,α=0.01*e1/N

      else:

      spatiotemporal GMM update (ft_new,αi,t=0.01*match(fi,t)

      end if

      Ft(x,y)←spatiotemporal GMM (f¯t(x,y)

      FFRFCM&GCt(x,y)←FRFCM&GC (Ft(x,y)

      Output:FFRFCM&GCt(x,y)

    • Table 2. Comparison of filtering effects of different methods on HMDB51 dataset

      View table
      View in Article

      Table 2. Comparison of filtering effects of different methods on HMDB51 dataset

      MethodGolf (0,0.1)Walk (0,0.1)Golf (0,0.05)Walk (0,0.05)
      PSNRSSIMPSNRSSIMPSNRSSIMPSNRSSIM
      Anisotropic filtering22.3010.47821.0800.47926.8290.529725.3150.5487
      Bilateral filtering21.9210.48820.4800.52326.1670.59424.7770.6194
      Inter-frame filtering21.9890.60721.6560.73226.4020.66427.0210.825
      Proposed filtering23.5800.69623.4200.74528.6120.784629.0560.834
    • Table 3. Comparison of filtering effects of different methods on measured data

      View table
      View in Article

      Table 3. Comparison of filtering effects of different methods on measured data

      MethodCase1Case2Case3Case4
      PSNRSSIMPSNRSSIMPSNRSSIMPSNRSSIM
      Anisotropic filtering44.2190.96849.7770.91746.3680.90948.4060.919
      Bilateral filtering39.0830.95841.8550.92542.4740.92940.1370.919
      Inter-frame filtering48.8750.98955.6350.99344.4560.99857.49710.981
      Proposed filtering51.4280.99157.6240.99749.7410.99659.0290.993
    • Table 4. Comparison of detection performance of different algorithms

      View table
      View in Article

      Table 4. Comparison of detection performance of different algorithms

      CaseCase 1Case 2Case 3Case 4
      AccIpErrAccIpErrAccIpErrAccIpErr
      FD81.910.37.886.88.74.584.68.37.181.47.810.8
      GMM85.76.87.591.15.63.389.65.25.288.25.66.2
      ViBe84.88.46.890.26.92.9895.65.485.66.87.6
      LBAdaptive90.24.75.194.82.62.693.63.33.190.64.64.8
      Proposed93.13.13.896.32.31.495.22.22.692.32.55.2
      CaseCase 5Case 6Case 7Case 8
      AccIpErrAccIpErrAccIpErrAccIpErr
      FD84.47.18.581.78.69.783.79.17.291.83.74.5
      GMM88.15.36.687.66.95.590.24.75.194.22.63.2
      ViBe89.85.64.688.56.35.287.66.85.694.63.12.3
      LBAdaptive92.62.25.290.64.25.293.42.6495.41.92.7
      Proposed93.22.64.292.32.94.893.63.23.294.81.63.6
    Tools

    Get Citation

    Copy Citation Text

    Jinhui ZUO, Wenbin XU, Shijie ZHOU, Daobin SHENG, Xiangdong XU, Zhengqiang LI, Yinghui HAN, Chunjiang WU, Lei ZHANG. Gas leakage detection based on spatiotemporal information of low contrast infrared images[J]. Optics and Precision Engineering, 2024, 32(8): 1186

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category:

    Received: Jul. 23, 2023

    Accepted: --

    Published Online: May. 29, 2024

    The Author Email:

    DOI:10.37188/OPE.20243208.1186

    Topics