光视觉在水下管线巡检中的应用

Application of Vision-based System in Underwater Pipeline Active Inspection

水下观察一般多由位于水面支持母船上的操作员遥控ROV完成。然而,水下管线巡检任务冗长而又艰巨,尤其在长距离、低照明度或者水流环境下。较为详细地介绍了一套自主研制的基于光视觉的水下管线自动巡检系统。该系统可以搭载在AUV上进行长距离的海底管道、光缆、电缆等水下管线智能巡检。在该系统的设计中,试图克服深海图像照明微弱、含有大量悬浮杂质颗粒等困难,并考虑管线植物附着、周围干扰物等较不利于辨识的恶劣状态。进一步的研究还包括三维环境建模、自适应扩展kalman滤波以及统计信号检测技术等。

Usually,underwater inspections are performed by remote operated vehicles（ROVs） driven by human operators on water surface.Howerver,this task is often challenging,especially in conditions of poor visibility or in presence of strong currents.A vison-based system for underwater pipeline active inspection by an autonomous underwater vehicle（AUV） is recommended in this paper,including 3-D scene model,extended Kalman fiter,statistical signal detection and so on.This system allows the guidance and follow of AUV in real time.