An efficient system for recognition of human face in different expressions by some measured features of the face using laplacian operator

In this paper we present an efficient system for face recognition with high recognition rate. In our proposed method, at first we detect the face from an image, then two main significant edge lines - chin-line and nose-line are determined and next we apply third order polynomial regression on these two lines to get a third order polynomial equation with four coefficients for each line. Here we use Laplacian operator to determine the curve of chin line and nose line. Then we measure distances from the middle point of the nose curve to the chin line horizontally and vertically, the width of the two eyebrows, the width of forehead and the distance from pupil to eyebrow. We also determine two regions - eye-region and nose-region. For each of these regions, we determine the average value of each three basic colors: red, green and blue. We then store the coefficients of the detected edge lines, the average values of the three basic colors and other distances and perform the task of recognition process. The existing face is recognized for which the weighted error is minimum and higher than a predefined threshold value. Experimental results show that our proposed method successfully recognizes face at a very high rate.