Face image captcha generation using particle swarm optimization approach

CAPTCHA is a software programming which is introduced to differentiate the human from the robots. CATCHA intends to generate a code which can only be identified by the human and machines cannot. In the real world, due to the massive increase in the usage of smart phones, tablets and other devices with the touch screen functionality poses a many online security threats. The traditional CAPTCHA requires a help of keyboard input and does dependant of language which will not be efficient in the smart phone devices. The face CAPTCHA is the one which intends to generate a CAPTCHA by using a combination of facial images and the fake images. It is based on generating a CAPTCHA with noised real face images and the fake images which cannot be identified by the machines but humans do. In the existing work, genetic algorithm is used to select the optimized face images by using which the better optimized fpso CAPTCHA can be created. However this work lacks from the local convergence problem where it can only select the best images within the local region. To overcome this problem in this work, the particle swarm optimization method is propose which can generate the globalize solution. Particle Swarm Optimization (PSO) is a popular and bionic algorithm based on the social behavior associated with bird flocking for optimization problems. The experimental tests that were conducted were proved that the proposed methodology improves in accuracy and generates an optimized solution than the existing methodologies.