A Machine Vision Attack Model on Image Based CAPTCHAs Challenge: Large Scale Evaluation

Over the past decade, several public web services made an attempt to prevent automated scripts and exploitation by bots by interrogating a user to solve a Turing-test challenge (commonly known as a CAPTCHA) before using the service. A CAPTCHA is a cryptographic protocol whose underlying hardness assumption is based on an artificial intelligence problem. CAPTCHAs challenges rely on the problem of distinguishing images of living or non-living objects (a task that is easy for humans). User studies proves, it can be solved by humans 99.7% of the time in under 30 s while this task is difficult for machines. The security of image based CAPTCHAs challenge is based on the presumed difficulty of classifying CAPTCHAs database images automatically. In this paper, we proposed a classification model which is 95.2% accurate in telling apart the images used in the CAPTCHA database. Our method utilizes layered features optimal tuning with an improved VGG16 architecture of Convolutional Neural Networks. Experimental simulation is performed using Caffe deep learning framework. Later, we compared our experimental results with significant state-of-the-art approaches in this domain.