A new Gabor wavelet transform feature extraction technique for ear biometric recognition

Ear Biometrics Recognition is an important research area in pattern recognition for identifying the person using ear digital images. In this research area, issues are to increases accuracy of correct identification by optimizing the feature extraction step in the methodology of Ear Biometrics Recognition System. Feature Extraction is the key step of ear biometrics recognition on which recognition rate depends. In the Gabor filter feature selection technique, Gabor equation is projected on digital image using different angles and scale but it generates a high dimension Gabor coefficient matrix of redundant features. The redundancy increases the confusion and reduces the correct recognition rate (CRR) of system. For increasing the CCR, redundancy of features should be reduced using filtering process. The proposed Gabor DWT filters technique for ear biometrics recognition system reduces redundant features using new filtering technique and optimized features are selected.