COTS recognition and detection based on Improved YOLO v5 model

Yufeng Jiang

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

9 Citations (Scopus)

Abstract

At present, the ecological environment of the Great Barrier Reef is becoming more and more fragile. Stopping the propagation and spread of COTS is an important part of protecting the environment of the Great Barrier Reef. It is becoming more and more important to identify and detect the distribution of COTS. With the development of computer science, deep learning technology has been widely used in the field of image recognition. Based on YOLOv5 algorithm and WBF model, this paper constructs a more accurate and efficient detection model to frame the distribution position of COTS. Our algorithm has been verified on the KAGGLE platform. The results show that our algorithm has great advantages in detection performance compared with other detection models. Quantitatively, the F2 value of our model is 39.241% and 5.263% higher than that of Faster R-CNN and YOLO v5 algorithms, respectively, which provides a certain reference for the protection of the Great Barrier Reef.

Original languageEnglish
Title of host publication2022 7th International Conference on Intelligent Computing and Signal Processing, ICSP 2022
Pages830-833
Number of pages4
ISBN (Electronic)9781665478571
DOIs
Publication statusPublished - 2022
Event7th International Conference on Intelligent Computing and Signal Processing, ICSP 2022 - Xi'an, China
Duration: 15 Apr 202217 Apr 2022

Publication series

Name2022 7th International Conference on Intelligent Computing and Signal Processing, ICSP 2022

Conference

Conference7th International Conference on Intelligent Computing and Signal Processing, ICSP 2022
Country/TerritoryChina
CityXi'an
Period15/04/2217/04/22

Keywords

  • COTS
  • Image detection
  • WBF
  • YOLO v5

Fingerprint

Dive into the research topics of 'COTS recognition and detection based on Improved YOLO v5 model'. Together they form a unique fingerprint.

Cite this