Face recognition through Hough transform for irises extraction and projection procedures for parts localization: For facial caricaturing system PICASSO

Abstract

In order to generate facial caricatures automatically, it is necessary to recognize facial parts in advance. Since it is difficult to realize the face recognition, the primal description of face has been provided manually so far also in our project PICASSO for facial caricaturing. This paper presents a robust method to recognize human faces by detecting irises via Hough transform with a preprocessing based on a hierarchical binarization process and localizing parts-regions via projection process. We experimented on this method with 30 face images and irises were successfully recognized for 29 images among them. Enforcing these results by employing a localizing procedure by means of projection process, a failure was quickly recovered and therefore the recognition rate of irises become 100%. In this paper, a procedure to extract other facial parts based on the information of irises were also proposed.