Search for a command to run...
Palm print recognition is widely studied in the past few years and many efforts are done to use it as a biometric modality for various applications. Existing research on palm prints is based on low resolution images [1]and hence matching is based on the creases present on the palm prints[2]. Recently it was analyzed that ridges on palm prints can be used for matching since it is unique and persistent for humans and can be used for large forensic applications. Since ridges are insensitive to distortion and discrimination power, they are very reliable and hence used for palm print matching. The issues in existing systems are that those algorithms for palm print matching followed the fingerprint algorithms and hence not much speed and matching accuracy and was inefficient. The palm print databases are not in same coordinate system and hence the computational complexity in matching is more and is not able to handle noise and distortion. The aim of this project is to design an efficient authentication system using ridge features for palm print recognition with reduced computational complexity and hence to increase the matching speed and accuracy. Here, the proposed system is extracting the features such as orientation field and region mask[3]. For matching minutiae extraction and cascade filtering is done to increase the speed of matching. Hence palm print matching is being done efficiently with an accurate speed and good performance.