Binocular Matching Method for Detecting and Locating Marine Resources Using Binocular Cameras and Deep Learning
Xiang Li, Yejun KOU, Yanchun Liang · 2023
It is well known that Artificial Intelligence and underwater robots play a significant role in energy exploration. This paper proposes a novel method using binocular cameras to detect, locate, and track marine resources. The method combines the YOLOv5 object detection method with the BM algorithm binocular ranging method. The method then uses a neural network to identify and detect the position of the target in the image, matches the detection framework with the depth chart, and employs a weighted average model to calculate the relative distance between the target and the object detected by the camera. Finally, underwater robots can use the relative distance to locate and track marine resources.