查看文章

CAPTCHA is a security technique to allow a computer application to distinguish between computer and human access. Most of now a day CAPTCHA words have their characters connected together which makes them very difficult to be segmented and recognized. In this paper we present an efficient technique that first benefits from the intersections between characters in a word and then segment it based on the recognition of each segmented character. Edge corners (ECs) are used in both segmentation and recognition phases. A novel fuzzy logic‐based scheme is proposed to match characters using their ECs. Experimental results show the efficiency of the proposed scheme in terms of success rate using a large set of CAPTCHA's words. Copyright © 2015 John Wiley & Sons, Ltd.