基于双背景建模与差分图像的轨道异物识别

Recognition of Foreign Object Intrusion for Railway Track on Double Background Modeling and Difference Image

针对铁路轨道异物入侵的识别精度,提出了一种基于双背景建模和差分图像的识别方法.基于多帧平均背景和当前帧的前一帧背景建立模型,采用背景逐帧更新,使用平均背景差分和帧间差分的异物检测方法,界定基于边缘提取的铁轨危险区域,再通过黑白像素方法对轨道异物进行识别.实验表明,该方法具有较好的抑噪性和环境适应性,可有效提高异物识别精度.

In order to improve the accuracy of foreign object recognition on railway track,the foreign object recognition method on railway track is put forward based on double background modeling and difference image.In this method,the double background modeling method that includes the multi-frame average background modeling and the previous frame of current frame background modeling method are used to model the background model,and two backgrounds are updated respectively to achieve background updating by using the newest image.The foreign object detection method is done by fusing the background difference and interframe difference,and the hazardous area of railway track is divided based on edge extraction,then the track foreign object has been discerned through the recognition method of the black and white pixels.The simulink results show that this method has better noise immunity performance and environmental adaptability,and the precision of foreign object recognition is improved effectively.