Journals Articles Conferences News About CLP
Submit a paper Email Alert
Search
Search
Articles
Journals
News
Advanced Search
Top Searches
2D MaterialsTransformation opticsQuantum PhotonicsSmall lasersSemiconductor UV PhotonicsSupersymmetric microring laser arrays

Acta Optica Sinica, Volume. 39, Issue 9, 0915005(2019)

Salient Object Detection Algorithm Based on Dual-Attention Recurrent Convolution

Xueli Xie, Chuanxiang Li, Xiaogang Yang, and Jianxiang Xi*
Author Affiliations
  • College of Missile Engineering, Rocket Force University of Engineering, Xi'an, Shaanxi 710025, China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (12)
    Equations (0)
    References (29)
    Cited By (3)
    Tools
    Paper Information
    Topics
    Figures & Tables(12)
    Overall structure of algorithm

    Fig. 1. Overall structure of algorithm

    Download full size
    Internal structure of DAM

    Fig. 2. Internal structure of DAM

    Download full size
    Schematic of multi-scale receptive field

    Fig. 3. Schematic of multi-scale receptive field

    Download full size
    Internal structure of MFRCM

    Fig. 4. Internal structure of MFRCM

    Download full size
    Connectionschematic of MFRCM

    Fig. 5. Connectionschematic of MFRCM

    Download full size
    Comparison of subjective visual effects

    Fig. 6. Comparison of subjective visual effects

    Download full size
    Characteristic curves of seven algorithms on ECSSD-250 testset. (a) PR curves; (b) ROC curves

    Fig. 7. Characteristic curves of seven algorithms on ECSSD-250 testset. (a) PR curves; (b) ROC curves

    Download full size
    Characteristic curves of seven algorithms on HKU-IS-1447 testset. (a) PR curves; (b) ROC curves

    Fig. 8. Characteristic curves of seven algorithms on HKU-IS-1447 testset. (a) PR curves; (b) ROC curves

    Download full size
    Characteristic curves of seven algorithms on DUT-OMRON-1500 testset. (a) PR curves; (b) ROC curves

    Fig. 9. Characteristic curves of seven algorithms on DUT-OMRON-1500 testset. (a) PR curves; (b) ROC curves

    Download full size

    • Table 1. Comparison of objective quantitative indicators

      View table

      Table 1. Comparison of objective quantitative indicators

      DatasetEvaluating indicatorValue
      OursAMUUCFSRMKSRMDFDSS
      ECSSD-250F-measure0.9280.9180.9120.9160.8270.8340.917
      MAE0.0430.0540.0710.0540.1170.1040.056
      HKU-IS-1447F-measure0.9230.9070.8930.9120.7840.8500.910
      MAE0.0380.0490.0580.0420.1270.1180.040
      DUT-OMRON-1500F-measure0.7820.7480.7370.7630.6340.6470.758
      MAE0.0560.0710.1310.0680.1310.0870.074

    • Table 2. Running time of different algorithms

      View table

      Table 2. Running time of different algorithms

      AlgorithmAMUUCFKSRSRMMDFDSSOurs
      Time /s0.080.0847.200.107.640.480.11

    • Table 3. Parameters statistics of different algorithms

      View table

      Table 3. Parameters statistics of different algorithms

      AlgorithmU-NetOursAMUUCFSRMDSS
      Number of parameters133953291906225617740720304064004548460819130619
    Tools

    Get Citation

    Copy Citation Text

    Xueli Xie, Chuanxiang Li, Xiaogang Yang, Jianxiang Xi. Salient Object Detection Algorithm Based on Dual-Attention Recurrent Convolution[J]. Acta Optica Sinica, 2019, 39(9): 0915005

    Download Citation

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

    Category: Machine Vision

    Received: Mar. 11, 2019

    Accepted: May. 23, 2019

    Published Online: Sep. 9, 2019

    The Author Email: Xi Jianxiang (xijx07@mails.tsinghua.edu.cn)

    DOI:10.3788/AOS201939.0915005

    Topics

    laser devices and laser physics
    Lasers and Laser Optics
    Laser physics
    laser manufacturing
    Instrumentation, Measurement and Metrology