基于主动形状模型算法的局部灰度模型的加权改进方法

摘要

In the searching of Active Shape Model (ASM) algorithm for target point, only surrounding local gray-level information of fixed point is adopted. Thus, the two points that have both similar gray-level and widely different texture details are often be confused. Consequently, the accuracy of positioning cannot be ensured. Thus, a weighted improved algorithm based on ASM was suggested in the paper. In the improved algorithm, some coefficients obey discrete Gaussian distribution were adopted. The coefficients represented the real feature point probability of bilateral candidate points in the normal direction in turn. Then, a weighted local gray-level model was constructed. The experimental results show that the new algorithm improved the accuracy of positioning by six percent than traditional algorithm.%主动形状模型( ASM)算法在对目标点搜索的过程中,只采用了标定点周围的局部灰度信息,这样往往会使灰度相似但其实微观纹理细节差别很大的两个点混为一谈,最终使算法的定位不够精确.为此,提出一种基于主动形状模型算法的局部灰度模型的加权改进方法,该算法采用一系列服从离散高斯分布的加权系数依次表示标定点法线方向上两侧的候选点是真实特征点的可能性,从而建立局部加权灰度模型.通过实验结果表明,改进算法比传统算法精度提高了6％.