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

Laser & Optoelectronics Progress, Volume. 60, Issue 2, 0228004(2023)

Lightweight Remote Sensing Object Detector based on YOLOX-Tiny

Lei Lang1, Kuan Liu2, and Dong Wang1、*
Author Affiliations
  • 1School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China
  • 2Computer and Communication Engineering, Zhengzhou University of Light Industry, Zhengzhou 450001, Henan , China
    • show less
    AbstractGet PDF(in Chinese)
    Figures&Tables (16)
    Equations (3)
    References (38)
    Cited By (2)
    Tools
    Paper Information
    Topics
    Figures & Tables(16)
    YOLOX-Tiny network structure

    Fig. 1. YOLOX-Tiny network structure

    Download full size
    Mapping of anchor in original image at different scales

    Fig. 2. Mapping of anchor in original image at different scales

    Download full size
    Improved YOLOX-Tiny

    Fig. 3. Improved YOLOX-Tiny

    Download full size
    Coordinate attention module

    Fig. 4. Coordinate attention module

    Download full size
    Deformable convolution schematic diagram

    Fig. 5. Deformable convolution schematic diagram

    Download full size
    Images and objects in DIOR dataset

    Fig. 6. Images and objects in DIOR dataset

    Download full size
    Size distribution of objects in DIOR dataset by category

    Fig. 7. Size distribution of objects in DIOR dataset by category

    Download full size
    AP curve during training

    Fig. 8. AP curve during training

    Download full size
    Loss curve during training

    Fig. 9. Loss curve during training

    Download full size
    Thermal map visualization results

    Fig. 10. Thermal map visualization results

    Download full size
    Comparison of detection results between YOLOX-Tiny and optimized model. (a) YOLOX-Tiny;(b) proposed algorithm

    Fig. 11. Comparison of detection results between YOLOX-Tiny and optimized model. (a) YOLOX-Tiny;(b) proposed algorithm

    Download full size

    • Table 1. Training parameters

      View table

      Table 1. Training parameters

      NameValue
      OptimizerSGD
      Momentum0.9
      Weight decay5×10-4
      NesterovTrue
      Learning rate scheduler

      Type is CosineAnealing，

      Learning rate is 0.0025，

      Min_lr_ratio is 0.05

      Batch16
      Epoch100
      MosaicImg_scale is （640，640）
      Random affineScaling_ratio_range is （0.5，1.5）
      Photometric distortion

      Brightness_delta is 32，

      Contrast_range is （0.5， 1.5），

      Saturation_range is （0.5， 1.5），

      Hue_delta is 18

    • Table 2. Comparison of progressively improved algorithms

      View table

      Table 2. Comparison of progressively improved algorithms

      MethodParameters /106AP /%AP50 /%
      YOLOX-Tiny5.0446.071.66
      + multi-scale prediction method5.4047.4（+1.4）75.10（+3.44）
      + coordinate attention module5.4147.7（+0.3）75.30（+0.20）
      + deformable convolution5.649.8（+2.1）75.60（+0.30）
      +loss function（proposed optimized model）5.650.1（+0.3）76.08（+0.48）

    • Table 3. Improved algorithm detection results under different scales

      View table

      Table 3. Improved algorithm detection results under different scales

      MethodAPAPSAPMAPLAP50AP50SAP50MAP50L
      YOLOX-Tiny46.09.535.766.371.6624.158.590.4
      Proposed optimized model50.112.838.670.276.0831.563.191.8

    • Table 4. AP50 results under different categories

      View table

      Table 4. AP50 results under different categories

      MethodC1C2C3C4C5C6C7C8C9C10C11C12C13C14C15C16C17C18C19C20
      YOLOX-Tiny6884.679.986.642.678.371.386.566.180.878.861.559.879.769.861.587.567.440.182.3
      Proposed optimized model74.489.183.888.347.278.677.588.876.182.481.264.562.287.774.370.988.870.349.186.4

    • Table 5. Comparison of results in DIOR dataset

      View table

      Table 5. Comparison of results in DIOR dataset

      MethodYearBackboneParameters /106FLOPs /109AP50 /%FPSDevice
      CF2PN［34］2021VGG1691.6>3167.2519.7RTX 2080
      ASSD［35］2021VGG16>40>3171.821RTX TITAN
      LO-Det［36］2020MobileNetv26.936.42465.8560.03RTX 3090
      Proposed optimized model2021Modified CSPNet5.67.69576.0846.0RTX 2080Ti
    Tools

    Get Citation

    Copy Citation Text

    Lei Lang, Kuan Liu, Dong Wang. Lightweight Remote Sensing Object Detector based on YOLOX-Tiny[J]. Laser & Optoelectronics Progress, 2023, 60(2): 0228004

    Download Citation

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

    Category: Remote Sensing and Sensors

    Received: Oct. 11, 2021

    Accepted: Nov. 29, 2021

    Published Online: Jan. 6, 2023

    The Author Email: Dong Wang (wangdong@bjtu.edu.cn)

    DOI:10.3788/LOP212699

    Topics

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