恶劣气象条件下机场跑道侵入的预警研究

Research on the Early Warning of Airport Runway Invasion Under Severe Weather Conditions

飞行安全是民航发展的基础,经统计,有40%~50%的飞行事故发生在起飞降落的“黑色十分钟”;为保证恶劣气象条件下飞机的安全率,文章使用Retinex方法和形态学的权重自适应算法对机场恶劣天气下的监视视频进行去雾降噪处理,之后采用光流法与混合高斯-卡尔曼滤波等机器视觉算法实现同时对多目标的检测、跟踪和预测;最后建立摄像机标定算法完成摄像机视角与机场视角坐标的转换,构建一套完整的机场跑道侵入的预警系统。

Flight safety is the basis for the development of civil aviation.According to statistics,40%~50%of flight accidents occur in take-off and landing"black ten minutes".In order to ensure the safety rate of aircraft under severe weather conditions,the histogram optimization technique and the weight adaptive algorithm of morphology are used to de-fog and reduce noise in the surveillance video airports.After that,the optical flow method and the hybrid Gauss-Kalman filter are used to detect,track and predict multiple targets simultaneously.Finally,the camera calibration algorithm is established to transform the camera angle of view from the airport angle of view,and a complete early warning system for airport runway intrusion is constructed.