Underwater Image Target Detection with Cascade Classifier and Image Preprocessing Method

摘要

Underwater image target detection is an important part of exploring the ocean. This paper adopts cascade classifier and image preprocessing method. Firstly, it selects candidate regions on a given picture, then extracts feature from them and finally uses the trained classifier to detect. It focuses on the self-defined training of the cascade classifier, and trains the cascade classifier by collecting a large number of underwater target images. Secondly, it uses some of image preprocessing to make the detection effect more accurate. Finally, the simulation results show that it can achieve the target detection of underwater image by using the method of self-defined cascade classifier and image preprocessing.