小波仿射不变函数与凸壳相结合的目标识别

Object Recognition Based on Wavelet Affine Invariant Function and Convex Hull

提出一种将小波仿射不变函数与凸壳相结合进行目标识别的方法。基于小波仿射不变函数构造了绝对小波仿射不变函数,作为待识别目标的特征向量,与凸壳相结合进行目标识别。该方法首先提取待识别目标的轮廓线,然后以轮廓线的一个凸壳顶点为起始点重新构造目标轮廓线点列,计算轮廓线点列的绝对小波仿射不变函数,与模板库中的模板进行匹配,根据最大相关原则识别目标。实验结果表明了该方法的有效性。

A method which combined Wavelet Affine Invariant Function （WAIF） with Convex Hull for object recognition is proposed. The Absolute Wavelet Affine Invariant Function （AWAIF） is constructed based on WAIF,and is used as the eigenvector of the object to be recognized,and is combined with Convex Hull for object recognition. First,contour line of the object to be recognized is extracted. Then,a new contour line whose start point is one of vertex of the convex hull of the object, and AWAIF of the contour line is calculated and used as the eigenvector. The object is then recognized according to the maximum correlation rule. The experiment results show efficiency of the proposed method.