Laser & Optoelectronics Progress, Volume. 62, Issue 6, 0628001(2025)

Hyperspectral Target Detection Based on Spatial-Spectral Reconstruction and Operator Weighting

Tie Li, Hongfeng Jin*, and Zhiqiu Li
Author Affiliations
  • School of Electronic and Information Engineering, Liaoning Technical University, Huludao 125105, Liaoning , China
  • show less
    Figures & Tables(10)
    Schematic of proposed SSROW target detection method
    Experimental diagram of spatial domain model
    Experimental diagram of spectral domain model
    Impact of regularization parameters on various datasets
    Comparison of results of each method
    ROC curves of each method
    • Table 1. Information of hyperspectral datasets

      View table

      Table 1. Information of hyperspectral datasets

      DatasetSpace size /(pixel×pixel)Bands numberTotal number of target pixels /pixelSpatial resolution /m
      Segundo250×30022420487.1
      Abu-airport-4100×100191603.4
      Sandiego100×100189573.5
      Abu-urban-1100×100204677.1
      Abu-urban-2100×1002071557.1
      HYDICE80×100162211.0
    • Table 2. AUC values of each method

      View table

      Table 2. AUC values of each method

      HSIsCEMACEHTD-IRNBLTSCCEM-VAESSROW
      Segundo0.93541550.62730810.99380020.93605420.99435840.9969429
      Abu-airport-40.87020960.83630280.93980050.99815730.97302150.9981992
      Sandiego0.95124580.89725730.99065020.98795500.99680640.9976445
      Abu-urban-10.99496100.78062200.98697090.97339800.98633080.9992848
      Abu-urban-20.98678650.73081240.96214220.94838220.99923720.9991940
      Hydice0.97543510.94426900.99752700.88676400.99957790.9998154
      Average0.95234200.80276200.97848200.95511850.99155540.9983468
    • Table 3. Total running time of each method

      View table

      Table 3. Total running time of each method

      DatasetCEMACEHTD-IRNBLTSCCEM-VAESSROW
      Segundo0.10911.144370.1141876.330449.862559.777
      Abu-airport-40.0371.402332.704271.52779.85994.043
      Sandiego0.0451.476258.530219.65680.83794.692
      Abu-urban-10.0351.564299.372253.17547.64462.493
      Abu-urban-20.0371.507373.954233.53228.40764.383
      Hydice0.0201.134184.783105.84359.42975.092
    • Table 4. Ablation experiment

      View table

      Table 4. Ablation experiment

      Experimental conditionSegundoAbu-airport-4SandiegoAbu-urban-1Abu-urban-2HYDICE
      Without feature enhancement0.99462640.79280350.32961740.98211750.99929360.8580642
      Without P-operator weighted0.98694290.92652410.95848090.99816980.98678620.9935175
      Without feature enhancement and P-operator0.99451400.72379280.43454180.98607090.99934210.8424614
      Without spatial-spectral fusion0.99619730.99573270.99771590.99923890.99892690.9997824
      SSROW0.99694290.99819920.99764450.99928480.99919400.9998154
    Tools

    Get Citation

    Copy Citation Text

    Tie Li, Hongfeng Jin, Zhiqiu Li. Hyperspectral Target Detection Based on Spatial-Spectral Reconstruction and Operator Weighting[J]. Laser & Optoelectronics Progress, 2025, 62(6): 0628001

    Download Citation

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

    Category: Remote Sensing and Sensors

    Received: Jun. 26, 2024

    Accepted: Jul. 29, 2024

    Published Online: Mar. 6, 2025

    The Author Email: Hongfeng Jin (jinhongfeng1109@163.com)

    DOI:10.3788/LOP241556

    CSTR:32186.14.LOP241556

    Topics