Design Of FPGA-Based Finger Multi-Modal Personal Identification System

With the development of information technology,personal identification system is widely used in national security and consumption.Nowadays,there are two big challenges: low recognition rate for large scale population,and low ability of anti-counterfeiting.Compared to the single-mode biological recognition system,multi-modal biological recognition has fused variety of biological information,it has better security features and distinction,and also can improve the recognition rate.After analyzing the research of the finger image acquisition and the preprocessing algorithms of image,this thesis proposed a multi-modal identification of finger based on the Xilinx's Zynq series SOC processor.This thesis designed the preprocessing algorithms of image,image quality evaluation methods and multi-modal images fusion method.The main work and achievements of this thesis are summarized as follows:(1)The hardware design of finger multi-modal identification system: After analyzing the requirement of the hardware system,the prototype is designed,which includes the processing system and the light source control system.(2)The design of multi-modal finger image acquisition and regions of interest image acquisition: By analyzing the imaging modality of finger image,an image acquisition mode was designed to acquire multi-modal finger image.For the characteristics of the acquired images,a method for extracting an ROI based on the largest cut in the matrix was presented.By analyzing the factors that affecting the quality of the finger vein image,a method based on the image quality of two-dimensional entropy was presented to capture high-quality ROI images.(3)The software design of finger multi-modal system: Designed the normalization,denoising enhancement,feature extraction and matching method for the captured ROI images.By combining with finger multi-modal system features,a fusion architecture mode was presented: firstly,use the finger shape features for the preliminary matches,then use the knuckle and finger vein features for the second matches in the decision-making aspect.