基于背景像素被阴影覆盖后RGB值的变化,提出了一种智能交通系统(ITS)中运动阴影的检测算法。首先定义了归一化亮度比例坐标系并将阴影检测视为该坐标系中的分类问题,然后采用阴影像素在该坐标系中的分布信息构建一椭球,并根据像素的坐标值是否处在椭球内部以及RGB3种颜色归一化亮度比例的相对大小确定阴影像素。对比实验表明,该方法能够有效地检测运动阴影,性能优于SNP算法和DNM算法。 Based on the change of RGB values ​​after the background pixels are covered by shadows, a detection algorithm of motion shadows in Intelligent Transportation Systems (ITS) is proposed. Firstly, the normalized luminance scale coordinate system is defined and the shadow detection is regarded as the classification problem in the coordinate system. Then, an ellipsoid is constructed by using the distribution information of shadow pixels in the coordinate system. According to whether the pixel coordinate value is in The shadow pixels are determined by the relative sizes of normalized brightness ratios within the ellipsoid and RGB3 colors. Contrast experiments show that this method can effectively detect motion shadows and has better performance than SNP and DNM algorithms.