Applications of Biometrics by ear print recognition

Abstract<br /> Biometrics has received considerable attention in recent years to its importance in the life process that starts from security systems at airports and criminal investigations to electronic banking. The use of (biometrics) is the most important parameters used in the disclosure of the identity of persons. The fingerprint, ear print recognition and automatically signature are important in the area of biometrics technology. So ear research recognition is improved broadly during the past decades and current years of other biometric research because their features are fixed and their change is expected. Biometrics are defined in many definitions as one of them "as any part of the body properties and that can be used to identify persons". And also know as "as a measure of physical properties or character traits".<br /> The aim of the proposed work is to build an efficient system of person identification based on ear biometrics through the use of (Geometrical Feature Extraction of Ear). Images are collected from a database of ear color images from (Zsged) database. The completed initial processing of these images, by changing the size of these images using the method (Bilinear). Then, the image has shifted to the grayscale and define the edges of the gray image using filter (Canny). An Eleven features are extracted from the right ear of a person. The right ear has been chosen after reviewing the researches and analysis the area that related to the ear that the right ear is more efficient to distinguish sounds instead of the left one, which distinguish the music. These attributes are stored in the feature database to be beyond the testing phase and compared ear to be identified after extraction characteristics in the same proposed algorithm with recipes pictures prayer in the database. Correlation coefficient is dependent as a similarity measure between the input images entered to the system with image database. The results of the proposed work represented by FAR (False Accepted Rate) and FRR (False Rejected Rate) a great performance compared with other works in the same field. The system programs are applied by the use of (MATLAB Version R2009a) system.