Occluded Ear Recognition Using Block-Based PCA
In the real world, two non-intrusive biometric methods, namely face recognition and ear recognition, are interesting as they have flexibility in scanning the subject being tested. In particular, ear recognition is quite interesting as the ear pattern is stable in all emotions. This paper presents a novel way using block-based PCA to recognize ear even in case of partial occlusion. The other significance of the proposed method is that it can decide whether an input image is occluded or not. Conducted experiments, used a standard data set and shown that min–min fusion technique with city block distance metric is the apt ear recognition method when block-based PCA is used. The recognition rate with the proposed method is greater than 94%, and the equal error rate (EER) is less than 15% for a 25% occluded ear image.
KeywordsEar recognition Principal component analysis Euclidean distance City block distance Recognition rate Equal error rate
This work is supported by University Grants Commission of India under the scheme—Minor Research Project No. MRP6023 Dated: 31. 10. 2016.
- 3.Iannarelli, A.: Ear Identification, Forensic Identification Series. Paramount Publishing Company, Fremont, CA (1989)Google Scholar
- 4.Nejati, H., Zhang, L., Sim, T., Martinez-Marroquin, E., Dong, G.: Wonder ears: identification of identical twins from ear images. In: 21st International Conference on Pattern Recognition, pp. 1201–1204 (2012)Google Scholar
- 6.Turk, M., Pentland, A.: Face recognition using eigenfaces. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Maui, Hawaii, USA, pp. 586–591 (1991)Google Scholar
- 7.Fan, Z., Ni, M., Sheng, M., Wu, Z., Xu, B.: Principal component analysis integrating Mahalanobis distance for face recognition. In: Second International Conference on Robot, Vision and Signal Processing, pp. 89–92 (2013)Google Scholar
- 8.Cha, S.-H.: Comprehensive survey on distance/similarity measures between probability density functions. Int. J. Math. Models Methods Appl. Sci. 1(4), 300–307 (2007)Google Scholar
- 10.Liu, L., Wang, Y., Wang, Q., Tan, T.: Fast principal component analysis using eigen space merging. IEEE Int. Conf. Image Process. (ICIP) 6, 457–460 (2007)Google Scholar
- 11.Tharwat, A., Ibrahim, A., Ali, H.A.: Personal identification using ear images based on fast and accurate principal component analysis. In: INFOS2012, pp. 56–59, May 2012Google Scholar
- 12.Aggarwal, C.C., Hinneburg, A., Keim, D.A.: On the surprising behavior of distance metrics in high dimensional space, pp. 420–434 (2001)Google Scholar
- 13.Querencias-uceta, D., Carmen, S.: Principal component analysis for ear-based biometric verification. IEEE conference, pp. 1–6 (2017)Google Scholar