Exploring the Potential of Facial Skin Regions for the Provision of Identity Information

This work presents a novel framework to investigate the possibility of using texture information from facial skin regions for biometric person recognition. Such information will be practically useful when the entire facial image is not available for identifying the individuals. Four facial regions have been investigated (i.e. forehead, right cheek, left cheek, and chin) since they are relatively easy to distinguish in frontal images. Facial landmarks are automatically detected to facilitate the extraction of these facial regions of interest. A new skin detection technique is applied to identify regions with significant skin content. Each such skin regions are then processed independently using features based on Local Binary Patterns and Gabor wavelet filters. Feature fusion is then used prior to classification of the images. Experiments were carried out using the publicly available Skin Segmentation database and the XM2VTS databases to evaluate the skin detection technique and the biometric recognition performances respectively. The results indicate that the skin detection algorithm provided an acceptable results when compared with other state-of-the-art skin detection algorithms. In addition, the forehead and the chin regions where found to provide a rich source of biometric information.