Finger Knuckle Based Biometric Identifier Using Principal Component Analysis, Feature Extraction and K-NN Classifier

Ms. Sampada H. Raut, Mr. V. B. Raskar

finger knuckle print, biometric identifier PCA, feature extraction, K-nnclasssifier.

Amidst several biometric measures, the figure knuckle surface is becoming a preferred choice of researchers due to its natural ease of reproducibility and verification. For any purpose of personal identification or crime analysis, figure knuckles surface do not need to be a voluntarily presented, they get exposed naturally. Specific line pattern on the figure knuckle surfaces can be used as effective biometric measure on their own or in combination with other biometrics. Present paper demonstrates the development of a figure knuckle based biometric identification system. The system incorporates principal component analysis (PCA) for feature extraction out of pre-processed and enhanced input image as extracted from knuckle surface video capture. Secondly the system employs k-nn classifier as personal identification algorithm. The system has been tested, verified and validated with many sample test experiments. The paper illustrates the working of the system with detailed intermittent snapshots.