Ear Localization and Validation Using Ear Candidate Set

Ear biometrics has been found to be a good and reliable technique for human recognition. Initially ear biometrics could not gain popularity because there were doubts about its uniqueness. But, it started to gain momentum after a theory which came into existence and which said that it was very unlikely for any two years to be completely identical in all respects. The implemented methodology consists of steps such as pre-processing, feature extraction and matching based on the selected features. Our technique determines the extent to which these features support matching. The proposed work has been carried out on on a dataset containing 60 images for analyzing their features and matching of the source image with the dataset images. The results have been obtained on the basis of images correctly classified. The system accuracy telling us the extent to which matching could be performed on the basis of selected features.