基于改进Harris算法的汉字图像角点检测

根据汉字图像的特点，借鉴加速分割检测特征算法的思想，提出一种改进的Harris算法对汉字图像进行角点检测。首先，计算像素值初步判断出非角点并排除；然后，通过计算传统Harris算法中的角点响应函数对剩余的像素进行角点检测；最后，借鉴加速分割检测特征算法的思想对伪角点进行删除。最终检测出的角点是汉字笔画的起点和末端的角点，为下一步特征提取中确定线段的位置和计算线段的长度提供有利的技术基础。通过对一定数量的汉字图像的实验仿真，将本文方法与几种常用的角点检测方法进行比较，本文方法在检测正确率方面有所提高，但在运行时间上没有达到最短，综合考虑正确率和运行时间，本文方法较其他几种方法有所提高。

Improved Harris Corner Detection for Chinese Characters Image

According to characteristics of Chinese characters image, we proposed an improved corner detection method based on Harris algorithm, and the ideal of FAST to improve detection rate for next feature extraction. Firstly, the pixels were computed to remove non-corners, and then the image of Chinese characters was detected for comers using Harris algorithm. Secondly, FAST for reference was used and the false corners were removed. The corners found lastly were endpoints of line segments, providing the length of line segments for shape feature extraction. The proposed method was compared with several corner detection methods over a number of images. Experimental results show that the proposed method is better in terms of detection rate, but not best for running time.