The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLVI-3/W1-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLVI-3/W1-2022, 301–306, 2022
https://doi.org/10.5194/isprs-archives-XLVI-3-W1-2022-301-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLVI-3/W1-2022, 301–306, 2022
https://doi.org/10.5194/isprs-archives-XLVI-3-W1-2022-301-2022
 
22 Apr 2022
22 Apr 2022

REAL-TIME MARINE ANIMAL DETECTION USING YOLO-BASED DEEP LEARNING NETWORKS IN THE CORAL REEF ECOSYSTEM

J. Zhong1, M. Li1, J. Qin1, Y. Cui2, K. Yang3, and H. Zhang1 J. Zhong et al.
  • 1State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Luoyu Road NO.129, Wuhan, China
  • 2Hongyi Honor College of Wuhan University, Bayi Road NO. 299, Wuhan, China
  • 3School of Water Resources and Hydropower Engineering, Wuhan University, Wuhan, China

Keywords: underwater images, object detection, deep learning, neural networks, coral reef, YOLO

Abstract. In recent years, with the advancement of marine resources and environment research, the ecological functions of reef-building coral reef ecosystems distributed in warm shallow waters of the ocean are being continuously discovered and valued by people. It is important for ecosystem protection to monitor the population of marine animals. Besides, many projects of Autonomous Underwater Vehicle (AUV) also need technology to perceive and understand environment information in real-time for better decision-making. Therefore, marine animal detection has become a challenge for researchers to study nowadays. Deep neural network models have been used to solve fish-related tasks and gained encouraging achievements, but there are still many problems in this field. In this paper, several YOLO-based methods are chosen for comparison. Experiment results indicate that these methods can recognize the marine animals in coral reef quickly and accurately. Finally, several recommendations for model improvement according to assessment results are presented.