基于6种不同行走状态下步幅特征对3种体态特征推定的研究  

Study on the Estimation of Three Kinds of Individual's Stature Based on Stride Characteristics under Six Different Walking Conditions

作　　者：史洪飞[1,2] 魏晓辉 蒋敬 徐伯芃[1] SHI Hongfei;WEI Xiaohui;JIANG Jing;XU Bopeng(Nanjing Forest Police College,Nanjing 210023,China;Forest Police Judicial Expertise Center of National Forest Bureau,Nanjing 210023,China;Criminal Investigation Detachment of Anqing Subbureau of Yangtze River Shipping Public Security Bureau,Anqing,246000,China)

机构地区：[1]南京森林警察学院,江苏南京210023 [2]国家林业局森林公安司法鉴定中心,江苏南京210023 [3]长江航运公安局安庆分局刑侦支队,安徽安庆246000

出　　处：《中国人民公安大学学报（自然科学版）》2021年第4期5-13,共9页Journal of People’s Public Security University of China(Science and Technology)

基　　金：中央高校专项资金项目(LGYB201812)。

摘　　要：足迹是犯罪现场最常见的痕迹之一,通过现场遗留足迹推断嫌疑人身高体重,对案件侦破意义重大。首先,获取138个研究对象在正常行走、快速大步、快速小步、慢速正常、快速正常、倒退行走6种行走状态下赤足成趟的足迹样本,利用SPSS软件,对这6种行走状态下的赤足成趟足迹步幅特征和研究样本的3种体态特征等参数进行分析,得出在6种行走状态下步幅特征与体态特征的综合相关性,再通过主成分分析和线性回归得出相关推断公式。经研究得出身高平均综合相关系数为R=0.623,具有强相关性;体重平均综合相关系数为R=0.448,具有中度相关性;肥胖指数平均综合相关系数为R=0.508,具有中度相关性。主成分分析和线性回归分别得出身高、体重及肥胖指数对应6种行走状态下的具体公式,通过化简得到3个总公式。带入数据验证,分别得出身高总公式偏差率为0.64%,体重总公式的偏差率为1.26%,肥胖指数总公式的偏差率为1.50%,公式具有较强准确性。本文为今后公安工作开展提供新的有效的方法和思路,为案件侦破提供相关依据。Footprint is one of the most common traces at the crime scene.The suspects'individual stature could be inferred from the footprints left at the scene,which is of great significance in the investigation of certain cases.In this paper,the footprints of 138 subjects in six walking states were collected,including normal walking,fast big step,fast small step,slow normal walking,fast normal walking and backward walking.The SPSS software was used to analyze the relevance between stride characteristics of bare footprints and individual stature under six walking conditions.Furthermore,the correlation inference formula was obtained by principal component analysis and linear regression.The results showed that the average comprehensive correlation coefficient of height was 0.623,indicating high correlation.The average comprehensive correlation coefficient of weight and obesity index were 0.448 and 0.508,which were moderately correlated.Principal component analysis and linear regression were used to obtain the specific formulas of height,weight and obesity index corresponding to six walking states,and three general formulas were obtained by simplification.According to the data verification,the total formula deviation rate of height,weight and obesity index were 0.64%,1.26%and 1.50%,respectively,indicating high accuracy of these formulas.This paper demonstrated a new and effective method for the development of public security,and provided relevant basis for the case investigation.

关 键 词：步幅特征 体态特征 相关性 线性回归 

分 类 号：D918.91[政治法律—法学]