Research On Fingerprint Ridge Distance Estimation And Fingerprint Matching Under Embedded Environment

Fingerprint has become a mainstream means of biometrics due to its uniqueness, reliability and convenience. Fingerprint recognition has been widely used in daily attendance, identity authentication, data encryption, e-commerce, e-government systems and has supplied great convenience for daily life and greatly meets the requirements of modern society. Traditional automatic fingerprint identification system is online system based on personal computer, which can provide convenient and efficient identity authentication service; however, it tends not to meet the requirements of miniaturization and portability of AFIS because of its huge size and is not easy to carry as well as power consumption. With the continuous innovation of high-performance digital signal processing chip, embedded fingerprint identification system has been more and more widely used in portable devices.Automatic fingerprint identification system(AFIS) often contains fingerprint acquisition, fingerprint preprocessing, fingerprint matching, where fingerprint preprocessing includes normalization, fingerprint segmentation, orientation computation, ridge distance estimation, fingerprint enhancement, fingerprint binarization, fingerprint thinning, feature extraction. Feature information after feature extraction is used for matching, which is used to get the final matching result. AFIS under embedded environment has a steep demand for space complexity, time complexity and real-timing, thus, ordinary fingerprint recognition algorithm can run under the embedded environment only if it is greatly improved or even be rewritten. The thesis aims to do some research on fingerprint ridge distance estimation and fingerprint matching under embedded environment, the main research contents are as follows:An efficient fingerprint ridge distance estimation is proposed, firstly, we obtain some typical candidate image blocks through the initial selection according to the orientation curvature of each block; Then better quality blocks from these candidate blocks are selected by taking quality strategy into account, while the remaining blocks are used to compute ridge distance using statistical window method; Finally, the average ridge distance is estimated through averaging several ridge distances in the remaining blocks. The experimental results show that our method is fast enough for online and embedded applications and is robust and reliable in estimating average fingerprint ridge distance.How to find the corresponding minutiae pairs and thereafter accurately match minutiae in fingerprints has always been a hot topic in minutiae-based fingerprint matching algorithm. To achieve this goal, we propose a multilevel-similarity-based fingerprint matching algorithm. Firstly, a discriminative sub-structure is defined for each minutia, based on which we calculate their similarities and obtain some potential reference pairs. Then a similarity scoring system is employed to give awards or penalties to such candidates according to their similarities of structural relationship concerning edges and angles. Based on a predefined threshold, those candidate reference pairs with higher scores are kept as the final reliable ones and applied in a global matching respectively. The maximal matching score is then chosen as the final matching score of two fingerprints. The experimental results show that our matching algorithm can greatly reduce the time consumption with ignorable performance degeneration.