摘要
在局部二值模式(LBP)基础上,运用一种改进的局部三值模式(LTP)纹理特征提取方法。并把这种提取方法运用到运动阴影去除中。该方法首先利用自适应高斯混合模型进行背景建模,得到背景和含有阴影的前景目标;并用亮度属性得到疑似阴影分块,然后再把疑似阴影区域和已获取背景相应位置的LTP纹理相似性进行判断;最后得到准确的阴影区域并实现阴影去除。实验结果表明,该算法能够很好地抑制分割噪声,准确地去除出运动阴影,具有较好的实验效果。
Based on the local binary patterns, a new improved method for extract local texture is introduced, namely local ternary patterns. This extraction method can be used to remove the shadow of movement. The background is modeled with adaptive gaussian mixture models, to get the background and foreground object. Then the intensity property is used to obtain the probable-shadow blocks. The shadow detected is improved based on the similarity of texture represented by LTP between shadow region and corresponding region in the background. Finally, the shadow can be removed accurately. The experimental results show that the algorithm can inhibit partition noise, and accurately remove the shadow of the movement.
出处
《计算机系统应用》
2010年第5期80-83,共4页
Computer Systems & Applications
基金
国家自然科学基金(60772071)
浙江省科技计划(2008C14063)
