OpenCV based computer vision of deep-sea visual exploration system

Abstract

Deep-sea visual exploration system plays an important role in the area of deep-sea expedition. Because of the harsh ocean environment, the deep-sea visual exploration system may face many problems such as high press of deep sea, instantaneous impact, deep-sea surge, local high temperature etc. Therefore, most current deep-sea visual exploration systems are not practicable. In this paper, considering the strongpoint of MPEG-4 coding and decoding technology with G.SHDSL modulation and demodulation technology, we designed a deep-sea visual exploration device based on 10000 m coaxial cable and MPEG-4 video server. Moreover, as a step up, computer vision based on OpenCV development kit has been introduced in this deep-sea visual exploration device. As a scientific discipline, computer vision is concerned with the theory for building artificial systems that obtain information from images, on the other hand, computer vision technology studies and describes artificial vision system that is implemented in software and/or hardware. With computer vision technology, deep-sea visual exploration system can obtain better performance and is intelligent enough to deal with some simply instance. As a consequence, the proposed deep-sea visual exploration device has been working on the ship named as ocean-one for many years, and has obtained many wonderful deep-sea video material, as a result, has made important contributions to ocean investigation enterprise of China.