SVM-KNN: Discriminative Nearest Neighbor Classification for Visual Category Recognition
CVPR '06 Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 2
Asirra: a CAPTCHA that exploits interest-aligned manual image categorization
Proceedings of the 14th ACM conference on Computer and communications security
Machine learning attacks against the Asirra CAPTCHA
Proceedings of the 15th ACM conference on Computer and communications security
| Hi-index | 0.00 |
CAPTCHA (Completely Automatic Public Turing Test to Tell Computer and Human Apart) systems are used to distinguish human users from computer programs automatically. The goal of them is to ask questions which human users can easily answer, but current computers cannot. Most current CAPTCHA methods are based on the weak points of OCR (Optical Character Recognition) systems. In this paper, a new CAPTCHA method is presented on the basis of object categorization. In this method, a number of objects are chosen randomly and the pictures of these objects are searched in the Internet and downloaded. The pictures are then shown to the user and the user is asked to mark the objects which belong to a specific category. If the user marks the right objects, it can be assumed that the user is a human being and not a computer program. The main advantage of this method is that it enables the human user pass even if makes a few mistakes, without compromising the security for that.