Fingerprint Verification Based on Fixed Length Square Finger Code Pradeep M. Patil Shekhar R Suralkar Hemant K Abhyankar VIT, Pune. SSBT’s COET, Jalgaon. VIT, Pune. patil_pm@rediffmail.com Abstract In this paper a novel method is presented for generation of a fixed length square finger code of size 4 16 16 × × . It uses a set of Gabor filters for extracting fingerprint features from gray scale image cropped in the size of 128 128 × pixels using its core point as the center. Experimental results show that the recognition rate based on the Euclidean distance between the two corresponding Gabor filter finger codes with verification accuracy of 93%. Since, the fingerprint matching is based on the Euclidean distance between two corresponding finger codes, it is extremely fast. This reveals that by setting the parameters to appropriate values, the finger code generated is more efficient and suitable than conventional methods for a small-scale fingerprint recognition system. Key words: Gabor features, Poincare index, fingerprint verification, fixed length square finger code, cropping of fingerprint image. 1. Introduction A fingerprint is the pattern of ridges and valleys on the surface of the finger [1] . Although the fingerprints possess the discriminatory information, designing a reliable automatic fingerprint-matching algorithm is very challenging as the images of two different fingers may have the same global configuration (Fig 1). (a) (b) Fig 1 fingerprints having same global configuration. The uniqueness of a fingerprint can be determined by the overall pattern of ridges and valleys as well as the local ridge anomalies called minutiae points. The smooth flow pattern of ridges and valleys in a fingerprint can be viewed as an oriented texture field. The image intensity surface in an ideal fingerprint image is comprised of ridges whose direction and height vary continuously, which constitute an oriented texture. Most textured images contain limited range of spatial frequencies, and mutually distinct textures differ significantly in their dominant frequency. Textured regions possessing different spatial frequency, orientation, or phase can be easily discriminated by decomposing the texture in several spatial frequency and orientation channels. Fingerprints can be identified by using quantitative measures associated with the flow of pattern (oriented texture) as features. In literature schemes using local landmarks i.e. minutiae based fingerprint matching systems [2-5] or exclusively global information [6-8] are available. The minutiae based automatic identification techniques first locate the minutiae points and then match their relative position in a given finger and the stored template [2]. A good quality fingerprint contains 60 to 80 minutiae, but different fingerprints have different number of minutiae. The main steps for minutiae extraction are smoothing, local ridge orientation estimation, ridge extraction, thinning, and minutiae detection [9]. For a small-scale fingerprint recognition system, however, it would not be efficient to process all the steps. Also the recognition result will heavily dependent on the accuracy of each step. The characteristics of the Gabor filter, especially the frequency and orientation representations, are similar to those of human visual system. Hence, Gabor filter based features have been successfully and widely applied to texture segmentation [10], face recognition [11], handwriting recognition [12], and fingerprint enhancement [13]. Fingerprint images present a strong orientation tendency and have a well-defined spatial frequency f for each local neighborhood that does not contains singular points (Fig 2). Proceedings of the 17th IEEE International Conference on Tools with Artificial Intelligence (ICTAI’05) 1082-3409/05 $20.00 © 2005 IEEE