Fingerprint Recognition using Orthogonal Polynomial Based Transform

Biometric features such as face, iris, voice, retina, handwriting and fingerprint plays an important role in security for authentication. Among biometrics, fingerprint systems have been one of most widely researched and deployed because of their easy access and low price of fingerprint sensors. Nowadays, for personal identification automatic fingerprint recognition is preferred due to their reliability and uniqueness features. Traditional minutia based and pattern based techniques face difficulty in extraction of minutia points and patterns in fingerprints. This paper proposed an efficient feature extraction technique for fingerprint recognition based on Orthogonal Polynomial based Transform. Statistical analysis on the fingerprint image is done and a candidate region suitable for feature extraction is selected. This region is subjected to the Orthogonal Polynomial Transform and transform coefficients obtained forms the feature vector. These feature vectors act as representatives for a fingerprint image and recognition is done using Support Vector Machine. Index Terms — Biometrics, fingerprint recognition, Orthogonal Polynomial Transform, Support Vector Machine.