Face detection is a technique of detecting faces from pictures, video footages, etc. There are various face detection algorithms; one of the widely used algorithm is the Viola Jones algorithm for object detection. The success rate of this algorithm for detecting faces is about 78.4%. In this paper, we present a technique, which is an improvisation on the existing Viola Jones algorithm. We have improvised the algorithm to clearly detect the eyes in a face of both people wearing glasses or not. The detection of glasses on the face is done by training a neural network. This algorithm primarily identifies a face with the presence of eyes, which has improved the detection rate and today our observations have yielded 90% success.