Study Of Multibiometrics System Based On Fingerprint And Finger Vein

21st century has become a sophisticated biological-based identification of the times. Biometric technology has been resolved from a sophisticated disciplines subject into a production technique, and the formation of the products are considerable servicing for our daily life. With the needing of accuracy of the personal identification system and the increasing security requirements in most countries in the world, only relying on a single method or single biometric sometimes difficult to meet the actual needs. Multibiometrics is the automated recognition of individuals based on their biological or behavioral characteristics, and involves different characteristics or different identify method to constitute the fusion system. With the appropriate fusion algorithm, multibiometrics normally are better than the performance of unimodal biometric systems, and it has become an important biological identification of the contents and direction of development. As a biology feature, fingerprint recognition is earlier than others. The technology of fingerprint recognition manufacture is mature and the price is inexpensive. Fingerprint manufactures have advantages in market research. Fingerprint sensor is prone to tumble in other systems, and the fingerprint recognition is prone to handle by users, so most of the multibiometrics systems researches are based on fingerprint. But fingerprint is not enough secure in biometrics. Finger vein is a good choice. As a biologic feature in human body, finger vein is provided with highly secure than fingerprint.This dissertation studies key techniques and application problems in designing the method of finger vein image synthesis technology, finger vein pattern extraction and matching algorithm design and the key technologies of information fusion in a mutltibiometrics system based on finger vein and fingerprints. For designing synthesis finger vein image method, a binary like-vein image and ellipse function are used to generate the vein pattern with various width and brightness. Design feature extract and match method of finger vein recognition technology where wavelet analysis is studied. A new method to hold vein pattern in finger vein image is proposed, and Position-Gray Profile Image is defined. Framework of multibiometrics based on fingerprint and finger vein is designed, and the fusion level method is studied at matching score level.This dissertation includes following aspects:1. Firstly, the aim and the meaning of this paper are given. The conception of biometircs is proposed. Then introduce the typical biology features, and recite the advantage and disadvantage of these typical features. Review the history of biometrics technology and summarize the research present situation. Finally, introduce the experiment toolboxes and databases that are used in this work.2. This dissertation studies finger vein image synthesis method, a similar like-vein binary image and ellipse function are used to generate finger vein pattern, then combines the vein pattern and background which is created with real finger vein image to get the synthesis finger vein image. In this work, the characteristic of width and brightness in infrared finger vein image is studied, and a distributing modal of infrared penetrate through a finger is assumed. Using this method establish synthesis finger vein image database for algorithm evaluation and system testing. The experiment data shows that the performance parameter of synthesis image is similar to the real infrared finger vein image. The result indicates the algorithm in this paper for synthesizing image can well simulate the finger vein in human body.3. This dissertation studies feature extraction and matching problem in the wavelet domain for finger vein recognition. Using dyadic wavelet transform and texture describe the characteristics of the finger vein. And induct the information fusion method for determining. In Matlab environment, the algorithm is realized and validated by both real finger vein image database and the synthesis image database. And the recognition result shows that wavelet multiresolution analysis is suitable to this work.4. This dissertation studies another finger vein recognition method, which is based on the Position-Gray Profile Image. This profile image reflects the veins have various widths and brightnesses. To observe the width and brightness of veins, we check the Position-Gray Profile of a finger-vein image. Curvatures of the profile are calculated to detect the vein patterns. To realize a robust feature extraction method in order to against a fluctuation of the width of veins, the positions of the center of veins are only extracted. The testing result indicates that the method of calculation curvature in multi-direction using the Position-Gray Profile Image of the finger vein image to recognition finger vein is robust.5. This dissertation studies multibiometrics technology based on fingerprint and finger vein. And launch the integration at matching score level. This fusion method using the matching scores of fingerprint and finger vein, then adopting the suitable fusion algorithm integrates the scores. The multibiometrics system can consumedly improve the veracity and the security of the personal identification.The multibiometrics system, which integrates with fingerprint and finger vein, is better than others fusion systems. Capturing the fingerprint and finger vein at the same time in equipment is more predominant than multi-equipment. So the cost plays down. This multibiometrics system based on fingerprint and finger vein is convenient for users.