Human Ear Recognition Using an Integrated Method of ICP and SCM Techniques

Abstract

The ear recognition techniques in image processing become a key issue in ear identification and analysis for many geometric applications. Some current specialized feature extraction methods attempted to examine the effects of pose variation and lighting changes that potentially alter the visual characteristics of the structure of the ear. In addition, one of the main issues to be addressed is the need for larger datasets of ear images. Where in a more accurate estimate of the recognition performance can be obtained, and the potential variations in the performance can be analyzed. The classifier combination problem can be defined as a problem of finding the combination function accepting dimensional score vectors from classifiers and outputting final classification scores.. The main objectives of this research are: To enhance the pose variations including differing angulations and distances by combining the Iterative Closest Point (ICP) algorithm matching with the Stochastic Clustering Method (SCM) and to propose an effective surface matching scheme based on the modified ICP algorithm combined SCM method. ICP is widely used for 3D shape When the input is a 2D image, the result is usually affected by the two limiting factors; lighting and angle of image, on the other hand, when the input is a 3D image the weakness is that the time for processing usually takes longer compared to processing 2D images and that the method is usually hard to apply in real life situation. An efficient ear recognition system therefore is one that integrates different methods such as the ICP and SCM into a neural network. The neural network should be able to perform logical and real-time assessment from minimal information inputs and it should be affected minimally by different factors that influence image quality. The most important steps in the research methodology are :- Integration of ICP and SCM Algorithms, Ear Feature Extraction, a local feature extraction technique (SURF) used to enhance the images to minimize the effect of pose variations and reduced image registration, the SURF feature carried out on enhanced images to gain the sets of local features for each enhanced image, Ear Learning, Ear Classification and Ear Matching.