A Survey On Occluded Image Based Captcha For Online Authentication

Abstract

CAPTCHA is a security mechanism that distinguishes human from malicious computer. Text based Captcha ask the user to identify distorted text. It brings complexity for the user to recognize the text. In order to overcome such difficulty Image based Captcha is developed. Image based Captcha has another type of Captcha such as Confident Captcha that asks the users to click on all images that gives a specific type of symbols such as cats or birds. Such kind of image based Captcha faces difficulty, since it could be easily identified by the unauthenticated user. The proposed system focus on providing authentication by using Occluded Image based Captcha. Occlusion is applied on images. Occlusion used is affine transformation. Affine transformation is used for providing complexity to user. Transformations such as Rotation, Translation, Scaling and Shearing will be applied to the images. Rotation is used for rotating the image to certain angle. Translation will translate the shape of an image. Shearing slants the shape of an image. Scaling will changes the size of an image. The user has to select the image during enrollment and the details will be stored into database. During login the user has to select the shuffled image that is selected in the enrollment. The selected image will be compared with database. If it matches, the user is authenticated. If the match is not detected, the user will not be authenticated. To bring complexity in authentication OTP is generated and sent to registered ID.