中文摘要：

对二值化的图像进行细化处理在目标识别中有着重要的应用.Zhang快速并行细化算法是一种常用的细化算法,其细化结果保持了原图像的连通性,且形态结构保持较好,无毛刺现象.但是细化后结果不能保证为单一像素,这为图像的后期处理带来了困难.本文对Zhang并行细化算法进行了改进,实验结果表明改进后的细化结果保证为单一像素,且又保持原有算法的优点.

英文摘要：

The thinning of the binary image has important application in target recognition.Zhang＇s fast parallel thinning algorithm is a kind of common thinning algorithm.Its thinning result maintains the connection of the original image,has a good structural form and has no burs.However it can＇t ensure the single pixel width,which brings difficulties for post-processing.In this paper,we make some improvement about Zhang＇s fast parallel thinning algorithm.Experimental results show that the results of the improved algorithm can not only ensure a single pixel width,but maintain the advantages of the original algorithm.