| Fingerprint identification technology contains four parts, fingerprint classification, fingerprint enhancement, minutiae acquisition, fingerprint matching. Embedded applications of fingerprint matching technology have become a new research direction. uClinux operation system is the best choose, because it is a OS designed for embedded system applications. In this thesis, the design of the fingerprint matching system based on uClinux OS is mainly discussed.Fingerprint matching mainly compare the similarity of input fingerprint pattern and standard one in the conditions of the known standard fingerprint pattern. Fingerprint similarity higher, the greater probability of matching. According to the difference of the fingerprint pattern, the fingerprint matching algorithms can be divided into two kinds, the first one is based on the minutiae, the second one is based on architecture. The coordinates and the orientations of the minutia are utilized for identification in matching algorithm based on minutiae. The relations between the minutiae and the lines are utilized by matching algorithm based on structure.Fingerprint matching algorithms, the hardware system design based on uClinux and the transplantation of the uClinux OS are the main contents of this thesis. In the aspects of the fingerprint matching algorithms aimed at the application of embedded system, the fingerprint matching algorithms based on minutia matching theory were studied. The algorithm was divided into three implementation steps, the computation of the supporting number, the parameters confirmation of the calibrated functions, the calibrated minutia matching were discussed in detail. In the aspects of the system hardware platform design, system hardware platform based on S3C44B0X including the memory module of FLASH, SDRAM and the functions of serials communication and network communication have been designed in detail according to the requirement of the system. In the aspects of the transplantation of the uClinux OS, the structural features of the uClinux OS and the JFFS2 file systems were studied, the measures of kernel reduction and compiling and the workflow of Bootloader program are discussed, the implementation steps were presented follow.Staggered achievements were obtained from the project. Fingerprint matching algorithms were test under the Matlab. As is shown from the experiments, the ordinary applications can be satisfied. The system hardware platform was implemented and the uClinux OS was transplanted successfully. Those achievements are the preparations for the future application of the embedded fingerprint matching. |
PDF Full Text Request