水下视觉图像处理和识别技术研究

Research on the Underwater Optical Imaging Processing and Identification

自主式水下机器人(AUV,简称水下机器人)在商业与军事上具有重要意义,受到越来越多科学家和技术人员的重视。视觉系统是水下机器人最重要的探测手段之一,而水下环境的复杂性以及光在水下传播的规律不同,导致水下视觉处理及识别面临新的难题。论文研究了几种常规的光学处理技术在水下视觉中的问题和分析,并使用采集的实际水下图像验证了算法的可靠性,为水下视觉应用提供了可靠手段。

With the rapid increase and requirement of the Automatic Underwater Vehicle(AUV)in business and military application,the underwater optical imaging technology,one of the important underwater detection ways,plays the most important role. The complicated underwater environment will bring new challenge to optical processing due to lighting and decay. This paper discusses some algorithms,which can be applied on underwater optical processing,and some collected data is used to verify the good performance of these algorithms.