3D Footprint Shape Characteristic Pick-up And Biological Characteristic Analysis

Recently, with the rapid development of information technology, the identity identification based on the human body's biological characteristic has been the research focus in the field of information technology. Under the conditions of standing or walking, the sole surface of the human body marks a footprint on the ground. Footprint is a type of important trace information belonging to biometrics. Although the long-term experiences in the criminal detection domain have shown the special relationship between footprint and people's gender, height,physique and age, the domestic and foreign research on footprint biometrics is far from sufficiency for the reason that the footprint's forming condition is variable, and lot of difficulty should be faced in data acquisition and characteristic recognition.This paper studies the pick-up of the three-dimensional footprint characteristic and biological characteristic analysis.Aimed at three-dimensional footprint characteristic, the uniform coordinate system which is based on the footprint trace expert knowledge has been established for footprint image. And on this foundation, the paper has picked-up the footprint position characteristics, measure characteristics such as distance and shape characteristics which can describle each heavy pressed face. There are 227 parameters in all. Because the correlation between some characteristics is serious, and some characteristics may not include useful information about recognition, we conduct feature selection by correlation analysis. After this operation, 118 parameters are retained.The footprint expert knowledge shows that the footprint has relationship with each part of organism physiological condition of human body nearly. So the paper analyzes footprint characteristics based on 514 footprint range images to conclude the gender, physique, height, weight and age of the human body. Firstly, the internal relationship is researched between footprint characteristics and the biological features by canonical correlation analysis. Secondly, the paper studies stratified random training sample to acquire the relation between footprint characteristics and the gender by Bayes discriminant analysis and Logistic regression analysis separately, the gender verification result of testing sample achieves a 0.953125 correct classification rate. Thirdly, the paper researches the relation between footprint characteristics and the physique by Bayes discriminant analysis, the physique verification result of testing sample achieves a 0.765625 correct classification rate. Fourthly, we analyze the relation between footprint characteristics and the height by multivariate linear regression analysis, the...