Research On Fingerprint Segmentation And Referece Point Detection Algorithms

Fingerprint identification is one of the most successful applications of biology identification technology and has developed quickly in recent years. Fingerprint pre-processing technology is foundation of feature extraction.At the same time it's a key for fast and efficient fingerprint identification. The fingerprint segmentation and the reference point detection are two important aspects of the fingerprint pre-processing and can decide the performance of the whole fingerprint identification system.To improve the performance of automatic fingerprint recognition and promoting its application, the dissertation discusses the problems of fingerprint segmentation and the reference point detection. The main work is as follows:1 This dissertation describes the overall process of Fingerprint Identification System.2 The fingerprint image segmentation algorithms are studied deeply in this dissertation. It described several major fingerprint segmentation algorithms. A novel segmentation algorithm for low quality image is presented.3 This dissertation researches the genetic algorithm theory and discusses the fingerprint segmentation technique using genetic algorithm. The Otsu value is choosed as fitness function and a complete scheme for fingerprint segmentation based on the annealing genetic algorithm is proposed. The dissertation does simulation using MATLAB.4 The fingerprint reference point detection algorithms are discussed in the dissertation. And a novel method for the reference point detection using the Poincare Index, which is combined with the adaptive smoothing and the directional consistency factor is proposed. Experimental results show it is effective and practical.5 The dissertation summarizes the work, analyzes the improvements which need to be done later and give the directions of future work.