基于自组织映射的验证码识别分析-信息安全专业论文.docxVIP

华 中科 技大学 华 中 科 技 大 学 硕 士 学 位 论 文 II II Abstract CAPTCHA is now almost a standard security technology. The most widely deployed CAPTCHAs are text-based schemes, which typically require users to solve a text recognition task. The state of the art of CAPTCHA design suggests that such text-based schemes should rely on segmentation resistance to provide security guarantee, as individual character recognition after segmentation can be solved with a high success rate by standard methods such as neural networks. In this issue, firstly, we use self-organizing map (SOM) neural network to recognize a non-adhesive CAPTCHA, it is able to achieve satisfactory learning outcomes with less training samples and time, the recognition rate of the CAPTCHA by SOM is almost 100%. Compared with other methods ,SOM could achieve high success rate with less training samples and less time; Furthermore, we present new character segmentation techniques to attack a text CAPTCHAs of Web QQ, in which characters touch or overlap with each other; we present a novel method to remove the arc or weaken the arc’s effect, which is based on image skeleton extraction algorithm, and two algorithm to character image segmentation; then input the individual character to the self-organizing map neural network for training or recognition, the recognition rate is approximately 80%. Secondly, we give the CAPTCHA recognition system framework and two sub- module design, introduce the related third-party libraries. Then we present the corresponding experimental results of these two kinds of CAPTCHAs validation experiments and their analysis. The future work and the development trend of CAPTCHA is at the end of the issue. Keywords: CAPTCHA, image segmentation, self-organizing map, pattern recognition III III 目 录 摘要 I HYPERLINK \l _bookmark0 Abstract II HYPERLINK \l _bookmark1 1 绪论 HYPERLINK \l _bookmark2 1.1 研究背景 (1) HYPERLINK \l _bookmark3 1.2 国内外研究现状 (2) HYPERLINK \l _bookmark4 1.3 研究的内容及论文的结构 (6)