摘要
随着数字图像处理和模式识别技术的发展,字符识别技术得到了广泛的应用。二值化技术作为字符预处理的一个重要环节,对后续的字符分割和特征提取有着很大的影响。本文针对字符图像的特点,提出了一种改进的基于直方图特征和形态学处理的Otsu二值化算法。仿真表明,该算法在减少运算量的同时,可以有效处理字符笔划残缺造成的干扰,并增强字符的连通性。
With the development of Digital Image Processing and Pattern Recognition technology, Character recognition technology has been widely used. As an important part of image preprocesSing,the binaryzation technology has a great impact to the following algo- rithm, such as character segmentation and feature extraction. According to the characteristics of the character image,an improved Otso binarization algorithm based on histogram features and morpho!ogical processing has been proposed in this paper. Simulation shows that the improved algorithm can lighten the computation and effectively deal with the interference caused by incomplete character strokes and can improve the connectivity of the characters.
出处
《微计算机信息》
2012年第3期140-141,180,共3页
Control & Automation
关键词
二值化
灰度直方图
阈值分割
Binaryzation
Gray-level Histograms
Threshold Segmentation
