1.College of Artificial Intelligence,Nankai University;2.Naval Research Academy
National Natural Science Foundation of China (No. 62103204)
Underwater target recognition is a core technology of underwater unmanned detection, which has important applications in military and civil areas. According to the current research, the principles and methods of underwater target recognition based on sonar images were comprehensively reviewed in this paper, which is of positive significance to summarize the research achievements, find the existing problems and tap the potential research directions. Aiming at the problems of sonar image based underwater target recognition, the main development and their extensions in underwater target detection, image denoising and image segmentation were discussed. The feature extraction, feature classification methods and main technologies in underwater target sonar image recognition were also analyzed in the following sections. Furthermore, the latest development of sonar image target recognition based on deep learning was presented. Through the discussion and analysis of underwater target processing development, the key scientific problems and possible solutions for underwater target recognition were pointed out, and the future development direction of this area was further prospected.