| Title | CAPTCHA Recognition Algorithm Based on the Relative Shape Context and Point Pattern Matching |
| Publication Type | Conference Paper |
| Year of Publication | 2017 |
| Authors | An, G., Yu, W. |
| Conference Name | 2017 9th International Conference on Measuring Technology and Mechatronics Automation (ICMTMA) |
| Date Published | jan |
| Keywords | authorisation, CAPTCHA, CAPTCHA recognition, CAPTCHA recognition algorithm, captchas, character recognition, composability, Context, CSDN site verification code, distance uneven grouping, Human Behavior, human beings, human factors, image matching, image recognition, Pattern matching, point pattern matching, pubcrawl, recognition rate, relative shape context, response time, Shape, shape context, shape context descriptors, shape context invariance, shape feature description, target contour point set deformation invariance, twisted adhesion verification code |
| Abstract | Using shape context descriptors in the distance uneven grouping and its more extensive description of the shape feature, so this descriptor has the target contour point set deformation invariance. However, the twisted adhesions verification code have more outliers and more serious noise, the above-mentioned invariance of the shape context will become very bad, in order to solve the above descriptors' limitations, this article raise a new algorithm based on the relative shape context and point pattern matching to identify codes. And also experimented on the CSDN site's verification code, the result is that the recognition rate is higher than the traditional shape context and the response time is shorter. |
| DOI | 10.1109/ICMTMA.2017.0049 |
| Citation Key | an_captcha_2017 |
CAPTCHA Recognition Algorithm Based on the Relative Shape Context and Point Pattern Matching