Optics and Precision Engineering, Volume. 31, Issue 6, 905(2023)

Vehicle detection method based on remote sensing image fusion of superpixel and multi-modal sensing network

Yuanfeng LIAN1...2,*, Guangyang LI1 and Shaochen SHEN1 |Show fewer author(s)
Author Affiliations
  • 1China University of Petroleum, Beijing02249, China
  • 2Beijing Key Laboratory of Petroleum Data Mining, China University of Petroleum., Beijing1049, China
  • show less
    Figures & Tables(13)
    Overall network structure diagram
    Region adjacent graph
    Superpixel segmentation remote sensing image results under differentλp
    Superpixel segmentation of remote sensing image results
    Precision and recall are compared between the proposed model and other method
    Comparison of segmentation results of this method and other methods on ISPRS Potsdam dataset
    Comparison of segmentation results of this method and other methods on ISPRS Vaihingen dataset
    Influence of OPT-FPN module on local details of vehicle detection
    Influence of multi-mode fusion module on local details of vehicle detection
    • Table 1. Proposed model was compared with other models on the ISPRS Potsdam dataset

      View table
      View in Article

      Table 1. Proposed model was compared with other models on the ISPRS Potsdam dataset

      模型SurfaceBuildingLow-vegTreeCarMIoUmF1OA
      FCN-32s76.1489.8172.0168.9866.8474.7684.4887.15
      FCN-8s77.5290.3472.6265.9877.3276.7685.5487.44
      U-net77.2091.1673.2369.0079.2977.9886.6787.84
      SegNet72.8487.9967.6860.3375.0072.7782.7684.93
      PSPNet79.1691.7373.3769.7675.6177.9286.6188.26
      DeepLabv3+79.6091.2175.4870.3378.1978.9687.2888.92
      G-FRNet85.1695.5580.5676.7781.5383.9190.6591.94
      our85.7296.0081.9176.8282.5984.6191.0592.39
    • Table 1. [in Chinese]

      View table
      View in Article

      Table 1. [in Chinese]

      输入:遥感光学图像和高程数据
      输出:光学图像区域合并结果

      1:分别读入光学图像和高程图像,得到的预分割结果AB

      2:使用定义9对遥感影像预分割结果进行分裂操作;

      3:构建二分图;

      3.1:设置V=IPPS

      3.2:W=

      3.3:for i=1,,m+n do

      3.4:  for j=1,,n do

      3.5:    计算Wij

      3.6:   W=Wwij

      3.7:  end for

      3.8:end for

      4:使用Transfer Cuts算法计算分割向量V

      5:基于向量VI进行区域融合。

    • Table 2. Proposed model was compared with other models on the ISPRS Vaihingen dataset

      View table
      View in Article

      Table 2. Proposed model was compared with other models on the ISPRS Vaihingen dataset

      模型SurfaceBuildingLow-vegTreeCarbackgroundMIoUmF1OA
      FCN-32s73.1283.4759.9973.5141.2010.5056.9768.7684.10
      FCN-8s78.3085.5359.8774.6061.3618.2862.9974.5685.91
      U-net79.0985.8363.9576.0664.6117.9264.5875.6786.76
      SegNet73.4882.4957.2472.8143.751.2355.1765.9183.43
      PSPNet78.2187.1562.5875.5656.1319.3463.1674.7186.44
      DeepLabv3+79.4887.9062.9075.9464.1219.6264.9976.1086.98
      G-FRNet80.1389.3368.1579.4152.6842.5068.7080.2688.48
      our86.1393.5064.9475.3471.3858.1274.9085.1189.66
    • Table 3. Results of ablation experiment

      View table
      View in Article

      Table 3. Results of ablation experiment

      ModelsSurfaceBuildingLow-vegTreeCarmIoU
      N185.7296.0081.9176.8282.5984.61
      N285.1095.7680.5575.9481.1683.70
      M185.7296.0081.9176.8282.5984.61
      M283.2694.0577.4273.6681.3581.95
    Tools

    Get Citation

    Copy Citation Text

    Yuanfeng LIAN, Guangyang LI, Shaochen SHEN. Vehicle detection method based on remote sensing image fusion of superpixel and multi-modal sensing network[J]. Optics and Precision Engineering, 2023, 31(6): 905

    Download Citation

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

    Category: Information Sciences

    Received: Nov. 14, 2022

    Accepted: --

    Published Online: Apr. 4, 2023

    The Author Email: LIAN Yuanfeng (lianyuanfeng@cup.edu.cn)

    DOI:10.37188/OPE.20233106.0905

    Topics