Underwater Object Detection based on YOLO-v3 network

Recently, side scan sonar (SSS) is increasingly applied to underwater search, which can display the microgeomorphic morphology and distribution, and obtain a continuous two-dimensional submarine acoustic map with a certain width. Automatic underwater object detection methods can help a lot in case of long searches, where sonar operators may feel exhausted and therefore miss the possible object. This paper proposes an underwater object detection method based on YOLO-v3 network. We first establish a real side scan sonar image data-set, which includes 7000 sonar images with four types of objects. Secondly, we propose an underwater object detection system based on side scan sonar images and YOLO-v3 network. Finally, we carried out extensive experiments in the real underwater environment to prove the effectiveness of our algorithm. Our work indicates that the YOLO-v3 network is an effective way to improve the accuracy of underwater object detection.