Fusion of Geometric and Texture Features For Ear Recognition

The field of ear recognition has been getting a lot of attention in recent years in biometric recognition systems field, which have seen a rise in interest in the present day, not only because it provides a better substitute for conventional identification and authentication systems, but also because of the high rise in global insecurities which makes the biometric research field a very active research field. The aim of our study is to contribute to the field of ear recognition by applying both texture and geometric features. This research in respect to its aims and objectives carried out experiments on AMI ear database by extracting the Local Binary Patterns features, and also applying Laplacian filter on the raw images separately to extract the geometric features. The ear database was processed by dividing ear images into four quarters and experimented on them individually in a pursuit of identifying the region of high interest in human ear images. Texture and geometric features are then fused and experiments were carried out on to verify the contribution of the fused features.