Personal Recognition with Ear Biometrics

Research of ear recognition technology, as well as its application, is a new subject in the field of biometrics recognition. The earlier research has shown that human ear is one of the representative human biometrics with uniqueness and stability. The paper discusses the feasibility and characteristics of ear recognition firstly, and presents recent research outcomes on 2D and 3D based ear image recognition and ear print recognition. The 2D ear recognitoin has the following two approaches: feature--based methods such as pure geometry methods, neural networks and force field transformation etc. and holistic methods such as PCA and moment invariants. The 3 D ear detection and ear recognition are carried out with range images. The ear print recogniton is mostly used in forensic research. Furthermore, the authors bring forward some essential aspects of this technology including description and extraction of era features, processing of partially occluded ear images and deformed ear images, the methods for recognition and ear detection, and the construction of ear database. The ear recognition is believed to be a promising branch in personal identification either by itself or combining with other biometrics recognition.