This paper presents a text line segmentation method for printed or handwritten historical Arabic documents. Documents are first classified into 2 classes using a K-means scheme. These classes correspond to document complexity (easy or not eas

VC++ 手写文字识别程序源码。实现手写数字给出不同的分类器识别结果，可心采用模板匹配分类器（最邻近模板匹配法）识别。手写数字或者打开已有的手写数字图像后，在右视图空白处，单击鼠标左键，激活右视图，选择菜单中的各种分类算法，可以对手写数字进行分类。-VC++ handwriting recognition program source code. Handwritten numbers give different classification recognition result satisf