基于广义霍夫变换的粘连字符验证码的识别  

Recognition of Character Merged CAPTCHA Based on Generalized Hough Transform

作　　者：汪志华[1] WANG Zhihua(School of Computer Engineering, Jimei University, Xiamen 361021, China)

机构地区：[1]集美大学计算机工程学院,福建厦门361021

出　　处：《集美大学学报（自然科学版）》2018年第1期75-80,共6页Journal of Jimei University：Natural Science

基　　金：福建省中青年教师教育科研项目(JA14184)

摘　　要：针对具有粘连字符的验证码,提出了一种基于广义霍夫变换的识别算法。首先,对验证码图片和字符模板进行预处理和骨架化,提取字符模板中目标点的局部特征保存到参考表中;然后,对验证码图片进行逐像素比较,并进行投票,通过投票累加器的极大值定位出字符模板的参考点;最后,对相邻字符之间的干扰进行分析处理,从而提高了字符识别的正确率。提取的局部特征具有平移和旋转的不变性,因而可以对验证码图片中出现的字符部分形变、字符粘连、字符旋转进行有效处理和识别。An approach based on generalized Hough transform algorithm is presented for the recognition of character merged CAPTCHA.Firstly,pre-processing and skeletonization are applied to the CAPTCHA picture and character model images,with the aim that the local features of object pixels in model images are computed and saved to the reference table.Then,each pixel of the CAPTCHA picture is compared with the reference table in the voting phase.The coordinate of the reference point is located by finding the maxima of the voting accumulator.Finally,the interference of neighboring characters is analyzed,with the correct recognition rate heightened by removing the interference characters.Given the shift and rotational invariance of the local features in this algorithm,the CAPTCHA picture with local shape variation,merged characters or rotational characters can be processed and recognized effectively.

关 键 词：验证码识别 粘连字符 广义霍夫变换 骨架化 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]