为了解决智能交通系统中运动车辆检测鲁棒性问题,分析了阴影产生的物理机制和运动投射阴影的视觉特性,提出一种新的阴影抑制方法。该方法首先采用帧间差分法获得运动车辆轮廓和运动阴影轮廓,然后应用霍特林变换消除RGB颜色分量的相关性,最后构造了一种阴影测度来进行运动车辆和阴影的检测。实验表明该方法能有效地抑制阴影并完整地分割出运动车辆目标。

A shadow detection and remove method is proposed based on analyzing visual properties concerning shadows for intelligent transport system. The method firstly uses inter-frame difference to remove the static backgrounds and obtains the vehicle contour and moving shadow contour. And then removes the correlation between RGB components by hotelling transform. Finally it defines a shadow measure to detect the shadows and extract the vehicle contour. Experimental results demonstrate the effectiveness and practicality of the method.