基于图像信息检测的船-桥智能避碰系统研究

将避碰决策系统分为船舶运动检测、视觉几何模型和船舶运动态势计算、碰撞预报及预控3部分.针对图像质量问题采取图像预处理增强、中值滤波、直方图均衡处理方法.采用相位相关算法基于背景模态模式静态空间信息记录图像信息,补偿图像抖动.基于视觉几何模型计算船舶运动态势计算船舶目标空间坐标位置、船舶型尺度、船速、船首向等.基于实时运动状态信息和桥区船舶运动态势预报碰撞危险以及避碰决策模式.引入避碰等级术语通过计算桥区船舶操纵避碰相关参数,进行避碰早期预警并得出避碰操纵决策方案.形成合理化的船-桥避碰决策系统,以达到船-桥避碰预警预控的智能效果.

Intelligent Ship-bridge Collision Avoidance Decision-making System Research Based on Image Information Detection

The Intelligent decision system for ship-bridge collision avoidance presented in this paper is composed by 3 parts： ship motion detection, visual geometric model and ship＇s motion situation calcu- lation, and collision forecast and control. The image quality is improved by image enhancement pre- treatment, median filtering and histogram equalization. The image offset is compensated by using phase correlation algorithm based on the statistical information-airspace of the background modeling method. Ship motion situation is calculated based on visual geometry model, including： the movement state information of target ship＇s position in space coordinate, ship size, speed, heading etc. The col- lision risk prediction and collision avoidance decision are implemented based on the information of ship real-time motion state information, and bridge area shifts motion. The collision avoidance early warn- ing and maneuvering decision are made through the introduction of the collision risk to rank terms, and the calculation of ship maneuvering related parameter in bridge area. So as to form the rationaliza- tion collision avoidance decision system, and realize the ship-bridge collision avoidance warning and pre-controlling inlelligent interfacial effect.