NOVEL APPROACH FOR FACE SPOOF DETECTION IN IMAGE PROCESSING

Abstract

The face spoof technique was proposed to identify and detect the spoofed and non-spoofed images. The DWT technique is used to analyze the textual features present within the test images. There is a possibility of presence of exceptional disturbances like geometric disturbances and the artificial texture disturbances. The time which is required for the classification is very long and it is difficult to evaluate optimal value of k . According to the data, the value of k is evaluated. By increasing the value of k, noise is reduced. The distinct boundary value is very less as compared to the other generated classes. By using several k-values of different classes KNN methodology is increased in order to overcome such limitations. Comparisons are being made to analyze the proposed approach performance according to the accuracy of the proposed algorithm and the existed algorithm in terms of accuracy and time of execution.