Abstract: The segmentation of off-line handwritten Chinese characters adhesion is one of the issues to be resolved.Therefore，a new segmentation algorithm is proposed in this paper for the adhesion of handwritten Chinese characters.The algorithm first looking for the dividing line to extract adhesion stroke.And then extracting the adhesion stroke to analyze and determine its type（straight line or curve），determine adhesion points and cut direction.Finally，using background thinning algorithm to find out the segmentation curve.This algorithm is not only applicable to the conditions of two Chinese characters width varies with adhesive，contains many adhesion points but also has a good anti-noise effect.

摘要： 粘连手写汉字的切分是手写汉字切分中亟待解决的问题之一。因此，针对粘连手写汉字提出一种新的切分算法。该算法首先通过寻找分界线的方法来提取粘连笔段，分界线的位置是通过粘连汉字骨架图像的聚类和笔段端点类属可信度的信息来确定的。然后提取粘连笔段并对其进行分析和类型（直线或曲线）判定，从而确定切分点及切分方向。最后利用背景细化算法找到分割曲线。该算法不仅能够很好地适用于两个粘连汉字宽窄不一、含有多个粘连点等粘连情况，而且具有良好的抗噪声效果。