International Journal of Computer Applications (0975 – 8887) Volume 40– No.11, February 2012 8 The Study of Adoption of Neural Network Approach in Fingerprint Recognition Divyakant T. Meva Marwadi Education Foundation Rajkot – Morbi Highway Rajkot, Gujarat, India C. K. Kumbharana Phd, Dept. of Comp. Science Saurashtra University Rajkot, Gujarat, India Amit D. Kothari Phd, Marwadi Education Foundation Rajkot – Morbi Highway Rajkot, Gujarat, India ABSRACT Fingerprint identification and verification are one of the well established methods for implementing security aspects. This technique has become mature due to lots of research in this area. One more feature of artificial intelligence is clubbed with image processing for fingerprint that is artificial neural network. ANN is used at different levels in fingerprint recognition. In this paper we will study the use of neural network approach in fingerprint recognition at different stages. General Terms Multimodal Biometrics, Neural Network, Pattern Recognition, Keywords Fingerprint recognition, artificial neural network, backpropagation network, fuzzy logic, LCNN, Biometrics 1. HISTORY AND INTRODUCTION TO FINGERPRINT RECOGNITION Biometrics, the word is well known in the area of authentication and authorization now a day. The technology uses different human traits for identification and verification. We can consider two types of biometrics technologies based on physical and behavioral traits. Biometrics technologies based on physical traits include fingerprint, face, iris, retina, ear shape, hand geometry, palmprint etc.. Biometric technologies based on behavioral traits include voice, keystroke, gait recognition etc.. Among all these technologies, fingerprint recognition is the oldest biometric recognition technology. It is in the use since ancient times. It was not until the late sixteenth century that the modern scientific fingerprint tech- technique was first initiated. In 1684, the English plant morphologist, Nehemiah Grew, published the first scientific paper reporting his systematic study on the ridge, furrow, and pore structure in fingerprints. In the late nineteenth century, Sir Francis Galton conducted an extensive study on fingerprints. He introduced the minutiae features for fingerprint matching in 1888. In the early twentieth century, fingerprint recognition was formally accepted as a valid per- personal identification method and became a standard routine in forensics. Fingerprint identification agencies were set up worldwide and criminal fingerprint databases were established. Various fingerprint recognition techniques, including latent fingerprint acquisition, fingerprint classification, and fingerprint matching were developed. For example, the FBI fingerprint identification division was set up in 1924 with a database of 810,000 fingerprint cards [1]. Now a day, two widely used approaches for fingerprint recognition are minutiae based and correlation based. As each fingerprint is unique in reference to various available on it, so we can adopt any one of the two methods. To reduce the complexity, fingerprints are classified into five classes based on curve pattern in the fingerprint. These classes are 1. Right loop 2. Left loop 3. Arc 4. Tented Arc 5. Whorl. It reduces total number of comparisons during the matching phase and time is also reduced. 2. NEURAL NETWORK Neurons are biological elements present in the human brain. They perform information processing in the brain. The network of interconnected neurons is known as neural network. A neural network is composed of a number of nodes, or units, connected by links. Each link has a numeric weight associated with it. The architecture can be made up of multiple layers. It is also called multilayer feed forward network. The fig. of multilayer feed forward network is shown below: Fig 1: Multilayer Feed Forward Network The most popular method for learning in multilayer networks is called back-propagation. You can perform learning in back propagation network.