Finger- image Segmentation using K-means clustering and a comparative analysis with L*A*B* method and detection of the segment using Correlation and Fourier-Series
Dept. of EXTC, Pillai Institute of Information Technology, Engineering, Media Studies & Research, New Panvel, India, India
ABSTRACT
Image segmentation is an important component in many image analysis and computer vision tasks. Finger-print recognition security systems deploy dormant, contact based scanners that rob the system of its viable flexibility. In this paper, foundation for a contactless finger-image recognition system is laid. Using K-means clustering method, finger images, captured using smart-phones, are segmented to extract the finger region. The same is accomplished using L*A*B* image segmentation algorithm, and a comparative analysis is framed. Basically a computer based application, Contactless security system demands an automated tool dedicated for identifying the desired finger-segment from the complete image. For this purpose, two independent algorithms, namely- correlation and FFT are practiced. The earlier segmentation results, obtained from both the methods, are employed in these algorithms and the results are scrutinized.
