Moving Shadow Detection Algorithm Using Multiple Features
Wen-Hua Xie, Ji-Yong Zhao, Yanming Hu · 2015
A shadow detection method based on color and regional texture is proposed in this paper to solve the problem of that single color or texture feature can't be used to detect moving shadow effectively. Firstly, a new texture-based method of shadow detection is proposed based on the conclusion of that regional gradient direction histogram keeps constant when the illumination changes. By this way, the noise effect when using one pixel as the smallest detecting unit in traditional methods can be significantly reduced. Secondly, in order to detect shadows more effectively, color feature and regional gradient direction feature are used respectively in different image areas according to the complexity of texture. Experimental results show that the proposed algorithm can detect shadows under different scenes effectively and stably.