A fusion adaptive recognition network based on intensity and polarization imaging

Ning Ma, Yunan Wu, Jun Chang, Wancheng Liu*, Yining Yang, Jinjin Wang, Xin Liu

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

As a new photoelectric detection method, polarization imaging can effectively improve the detection range and recognition accuracy of key targets under harsh environmental conditions by its excellent imaging effect of de-fog reconstruction. This paper proposes an image detection and recognition network based on polarization information and intensity information. The network is based on the yolov5 network integrated with DYHEAD block for recognition, realizing scale perception, space perception and task perception in a unified manner. Meanwhile, Res2Net is integrated for multi-scale characterization at the granularity level. Finally, the attention mechanism is introduced to realize the adaptive extraction and fusion of multi-scale features at the granularity level. The results show that the proposed method can effectively improve the recognition accuracy in low visibility environment.

Original languageEnglish
Title of host publicationSixth Conference on Frontiers in Optical Imaging and Technology
Subtitle of host publicationImaging Detection and Target Recognition
EditorsChao Zuo, Jiangtao Xu
PublisherSPIE
ISBN (Electronic)9781510679726
DOIs
Publication statusPublished - 2024
Event6th Conference on Frontiers in Optical Imaging and Technology: Imaging Detection and Target Recognition - Nanjing, China
Duration: 22 Oct 202324 Oct 2023

Publication series

NameProceedings of SPIE - The International Society for Optical Engineering
Volume13156
ISSN (Print)0277-786X
ISSN (Electronic)1996-756X

Conference

Conference6th Conference on Frontiers in Optical Imaging and Technology: Imaging Detection and Target Recognition
Country/TerritoryChina
CityNanjing
Period22/10/2324/10/23

Keywords

  • Convolutional neural network
  • Multidimensional information fusion
  • Polarization imaging
  • Target recognition

Fingerprint

Dive into the research topics of 'A fusion adaptive recognition network based on intensity and polarization imaging'. Together they form a unique fingerprint.

Cite this