一种融合图像区域属性特征的运动阴影消除算法
Moving cast shadow removing algorithm based on blocked texture and color feature fusion

针对传统C1C2C3彩色不变性特征模型中分母只考虑值最大的颜色信息导致阴影消除结果不准确及局部二值模式(local binary pattern，LBP)易受噪声扰动的缺陷，在C1C2C3模型中采用两个颜色通道的均值作为分母以及在LBP模型中加入抗噪声因子的方法分别对以上两个模型进行改进，提高了两个模型的阴影检测准确率. 为了有效融合各种基于属性阴影消除方法的优势，引入LBP纹理复杂度测量函数，根据LBP纹理复杂度分区域融合文中改进的两个模型，实验结果表明，改进的算法提高了运动阴影消除准确率.
For the traditional C1C2C3 invariant color features，the largest color value in the denominator was selected to remove the cast shadow. Thus，the result is not accurate. And local binary pattern (LBP) is apt to be confused by noises. In order to solve the problem effectively，we improved the C1C2C3 model by using the average of the two color value in the denominator and added a divisor for resisting noises in LBP. Furthermore，we proposed the LBP texture measure function to calculate the coefficient of texture complexity in the neighbor field of center pixel. Then whether the improved LBP model is replaced by the improved C1C2C3 model or not is determined by the coefficient of texture complexity. Experimental results show that the improved shadow removing algorithm performs better