Real-Time Visualization System for Deep-Sea Surveying

摘要: Remote robotic exploration holds vast potential for gaining knowledge about extreme environments, which is difficult to be accessed by humans. In the last two decades, various underwater devices were developed detecting mines and mine-like objects in deep-sea environment. However, there are some problems recent equipment, like poor accuracy of mineral detection, without real-time processing, low resolution video frames. Consequently, recognition a task, because physical properties medium, captured frames, distorted seriously. this paper, we considering use modern image processing methods determine location recognize actually within little computation complex. We firstly analyze imaging models propose novel optical model, much closer light propagation model our system, remove electrical noise dual-tree complex wavelet transform. And then solve nonuniform illumination artificial lights fast guided trilateral bilateral filter recover color through automatic equalization. Finally, shape-based algorithm proposed detection. These designed execution on limited-memory platforms. This pipeline suitable practice experiences. The initial results presented experiments demonstrate effectiveness visualization system.