提出了一种新的二维形状描述方法,该方法利用对目标图像边界搜索的结果,将目标边界曲线像素的坐标(xi,Yi)分别用两个数组x=X[i];y=Y[i]描述。将一个二维图像的问题转换成两个一维数组的分析问题,简化了分析的复杂度。通过分别对两个数组的处理可以对原目标边界进行一系列处理,如边界圆滑、锐化、拐点检测等。介绍了该描述方法及其基本性质,并利用该描述方法通过小波分解对工件目标边界的平滑进行了具体分析。
There is put forward a new methods of 2D(two-dimensional) shape representation which stores the object borderline pixels coordinate value(xi ,yi ) into two arrays x = X[ i ] ;y = Y[ i ] by the sequence of object borderline tracking result. It changes a 2D image problem into two single dimension arrays'problem. It lowers down the complexity of analysis. A series of processes about object borderline can be implemented such as smoothing, enhancement, comer detection etc. By operating on the two arrays. The representation of this method and its basic property are Introduced. And a concrete analysis has been done for the course of workpiece boundary smoothing based on this 2D shape representation by wavelet.