Windows Device Driver Design Of AFIS And Optimization Of Character Recognition Algorithms On DSP

Fingerprint is the inherent physical characteristics of the human body with uniqueness and invariance. Therefore, it becomes an important means of identification, with more applications in various areas of life. Automatic Fingerprint Identification System (AFIS)is an identification system,which includes fingerprint capture, fingerprint pre-processing, feature extraction, fingerprint matching, and output results.So,it is desirable to have an efficient fingerprint identification system.Fingerprint capture is the major front-end part in the acquisition quality, and its computational speed has a direct impact on the overall performance of the system. For MBF200 fingerprint collection instrument, we developed a new USB driver, which is based on WDF. Comparing with traditional driver, image quality of our proposed driver is better, and it.also meets the transfer speed requirements based on USB2.0.Fingerprint matching part in the fingerprint identification system need higher speed in order to match fingerprint with large database. In many practical applications,in order to resolve the contradiction between the complex algorithm and the speed of the system based on large fingerprint database, we must use hardware-accelerated devices to improve the fingerprint comparison accuracy and speed.A modified algorithm for PCI based WDM driver for the coarse fingerprint comparison accelerator is developed.License Plate Recognition System (LPRS)is an important part of Intelligent Transportation System.License plate recognition process is divided into vehicle license plate location, license plate image segmentation,and character recognition process.Finally, we also propose an optimization of the character recognition algorithms,and successfully integrate into TI5509A platform.