Personal identification using lip print furrows

Abstract

There are common techniques for personal identification such as fingerprint recognition, face recognition, and iris recognition. In this paper, we suggest a new method for personal identification using lip print furrows. This method will be useful in forensics services. This study aims to identify people by their lip print patterns that are unique for each person. Lip print features are extracted from the lip print images by using the vertical Sobel operator, diagonal ±45 lines, and hough transform. A lip print pattern was formed for each person containing only the furrows (the extracted features) that were extracted from 3 lip print images for that person. The correlation coefficients and the mean squared error (MSE) are provided and passed to the support vector machine (SVM) for the classification process. The suggested method gives good results. A comparison between the results of the proposed method and other methods was presented.