2例足迹检验出现差错的原因分析

足迹检验鉴定理论的发展已经突破了根据鞋印形象特征检验的范畴,发展到根据鞋(或鞋印)上所反映的磨损特征、动力形态特征、结合步法检验技术对同一人穿用不同鞋进行人身同一认定,这是足迹检验鉴定理论一个质的飞跃。这一理论在实际办案中对于串并案件、揭露证实犯罪提供了有力的支持。但是,不可否认的是近几年在实际案例操作时,足迹检验中根据同一人穿用不同鞋对穿鞋人进行同一认定有时会出现差错。我们在实际工作中,就曾经遇到两例,报道如下:1案例简介案例12002年某日,在一起盗窃案件的外围现场,提取到一枚平面灰尘加层足迹,该足迹边缘完整…     

模压鞋底细小纹理的应用

一个重大案件发生后，如果有现场鞋印，紧接着一项重要任务就是寻找现场鞋印的样本鞋，通过样本鞋的销售范围来确定犯罪嫌疑人的活动范围；通过样本鞋的价格档次来确定犯罪嫌疑人的生活水平；通过样本鞋来还原现场的残缺鞋印，并在周围地区通过鞋印进行案件串并；通过样本鞋的销售人员来回忆近期购鞋人员的特征，以期从中发现犯罪嫌疑人。     

身高与足迹长度相关关系综合研究

目的全面分析研究足迹长度与身高之间的相关关系及其变化规律。方法在全国范围内大量抽样收集不同地区、不同年龄、不同性别人群的足迹样本,逐一测量其足迹长度,建立数据库,利用计算机进行综合统计、回归分析、比较研究。结果获得大量的分析数据及回归方程。结论足迹长度与身高之间的相关关系明显,可以根据足迹长度分析留痕人的身高。     

利用残缺赤足足迹推算身高

目的由于主客观条件的影响,犯罪现场中常见的足迹多为不完整的残缺足迹,如何能够有效地利用残缺足迹,对案件破获往往具有至关重要的作用。方法从残缺赤足足迹的反映形态出发,选择足迹中明显、稳定的特征点,确定特征量,通过数理统计分析,建立残缺赤足足迹推算身高的经验公式。结果在4个特征量中,跖前缘突点到跟后缘突点的距离与身高的相关性最好,跖内、外缘突点到跟后缘突点的距离与身高具有一定的相关性。结论根据现场残缺赤足足迹的反映情况,也可以近似推断嫌疑人的身高特点,为侦查提供方向。     

基于多尺度特征融合的穿鞋足迹识别网络

近年来，随着人工智能的快速发展，赤足足迹和穿袜足迹的自动识别均取得了不错的效果，但因为需要脱鞋，应用场景受到了很大的限制。单枚穿鞋足迹识别由于鞋底花纹的多样性、随机性给智能识别造成了很大的阻碍，识别准确率普遍较低，成为了一个具有挑战性的任务。本文聚焦于不同人穿同一种鞋的足迹识别，提出一种基于多尺度特征融合的穿鞋足迹识别网络。采集大量的足迹样本，通过旋转、平移、加噪声等方式模拟犯罪现场足迹图像可能产生的变化，制作了足迹数据集，然后以ResNet为骨干网络，利用双向金字塔特征融合模块充分融合足迹图像的深层特征和浅层特征，最后针对鞋底花纹变化造成网络识别准确率降低的问题，提出了用迁移学习解决的思路，让网络学习未知花纹与现有花纹之间的联系，快速拟合出模型。经过训练测试，本文提出的方法在自制足迹数据集上识别准确率达到了93.1%,CMC评价指标也明显优于其他网络。在新的足底花纹上，迁移学习对比从头开始训练，速度更快、准确率更高。大量的实验证明，本文提出的穿鞋足迹识别网络识别准确率更高，提出的迁移学习的方法在面对新的鞋底花纹时，能够实现更好的效果。     

对足迹检验鉴定理论的思考

我国的足迹检验技术在80年代和90年代得到了较大的发展,足迹检验技术从步法检验、形象检验,发展到鞋底磨损检验、足迹动力形态检验、足迹定量化检验等。特别是近几年,足迹检验中各学派、各门类的各种检验技术内容丰富,方法多样,正在逐步形成具有中国特色的足迹学体系,在侦查破案中发挥了重要作用。目前,常用的足迹鉴定包括:1赤足足迹的鉴定主要以足迹的大小、形象、轮廓、重压部位的密度、乳突花纹、残缺部位、是否变形等特征为赤足足迹进行同一认定,赤足足迹的鉴定又分为平面赤足足迹的鉴定和立体赤足足迹的鉴定。2鞋印足迹的形象鉴定主要…     

足迹个体识别技术自动化——足迹计算机自动鉴定系统

现场遗留的足迹,是直接认定罪犯的法庭物证,有着重要作用.近年来罪犯作案手段狡诈,换穿鞋作案或作案后把鞋仍、毁掉,致使现场足迹失去了检验特征及认定鞋子的条件.1975～1986年笔者成功研究出足迹鉴定进行个体识别的科学理论——“足迹动力学”和足迹动力形态检验新技术.1991年利用该技术开始研制“足迹计算机自动鉴定系统”,用于检验、鉴定一个人换穿鞋足迹的同一认定,及不同人穿相同鞋足迹的个体识别,实现了反映人体运动规律、个体特点的足迹动力形态特征数据的输入——检验、比对、鉴定——自动生成成果——输出打印鉴定报告、图形自动化.     

软胶底平面足迹推算身高方法的探讨

利用足迹推断身高,确定重点嫌疑人,方法简单,结果准确,是缩小侦察范围的有力武器.常规计算公式是:足迹全长X 7(或6.876)-3,在这里,足迹全长,即是足迹的前端最突出点到后跟最突出的距离.它不同于鞋长,在立体足迹中,它要长于鞋长;在平面足迹中,它要短于鞋长.它也不同于脚长,在立体足迹中,它要长于脚长;在平面足迹中,既可能长于脚长,也可能短于脚长.即鞋印长.然而,笔者在利用这个公式进行深人研究的时候,却发现,对于软胶底平面足迹利用这个公式来进行推算留足迹人的身高其结果可能与实际情况相差很远.为此,笔者在这方面进行了一些探讨、研究,报告如下.     

手按压伪装足迹的检验1例

罪犯作案,为了逃避打击或嫁祸于人,往往伪装足迹.常见的伪装方法有:改变步法、换穿鞋、倒穿鞋等.在实际工作中,我们发现并鉴别了1例手按压伪装足迹.案例简介:1992年9月4日晚,萍乡市某供销社值班营业员戴某(男)被杀死在营业间内.现场勘查时,发现尸体周围的血泊水泥地面上,有不成行的两种女式中跟血迹鞋印9个,鞋尖方向均以血泊为起点,向四周扩展,尸体短裤脱至臀部,生殖器外露,尸体旁还发现沾有血迹的卫生纸、绣花线等女性用品.其中一较完整的右脚鞋印全长22.5cm,前掌宽8.4cm,后跟宽3.7cm,前     

利用擦划痕迹辅助认定鞋子1例

鞋印形象特征检验是指对现场鞋印和犯罪嫌疑人的鞋印样本所反映的鞋底外表结构特征进行检验、鉴别,确定现场鞋印是否为犯罪嫌疑人的鞋所留。在实践中,鞋底外太结构特征一般反映为鞋底面上的形态特征.在有些犯罪现场，由于受地面条件等因素的影响,鞋印并不能很好地反映出鞋底面上的特征，有时会留下以鞋底边缘擦划为主的痕迹。     

The Use of Scale‐Invariance Feature Transform Approach to Recognize and Retrieve Incomplete Shoeprints

Shoeprints left at the crime scene provide valuable information in criminal investigation due to the distinctive patterns in the sole. Those shoeprints are often incomplete and noisy. In this study, scale‐invariance feature transform is proposed and evaluated for recognition and retrieval of partial and noisy shoeprint images. The proposed method first constructs different scale spaces to detect local extrema in the underlying shoeprint images. Those local extrema are considered as useful key points in the image. Next, the features of those key points are extracted to represent their local patterns around key points. Then, the system computes the cross‐correlation between the query image and each shoeprint image in the database. Experimental results show that full‐size prints and prints from the toe area perform best among all shoeprints. Furthermore, this system also demonstrates its robustness against noise because there is a very slight difference in comparison between original shoeprints and noisy shoeprints.     

A comparative research of two lifting methods: electrostatic lifter and gelatin lifter

Two-dimensional dust shoeprints are often of very high resolution and contain unique features. Lifting these prints in the most effective method may contribute much to preserving these fine details. A research was conducted by experts from Israel and Switzerland to compare gelatin lifters and electrostatic lifters for lifting shoeprints. Several substrates were chosen, and on each material a set of dry dust shoeprints was made. A set of wet prints was made on paper as well. The shoeprints were approximately of the same quality, and the only variable was the nature of the material. On substrates indifferent to the method used, the preferable sequence was tested. Gelatin lifter was superior on most substrates and for wet prints. The superior sequence for using both methods is electrostatic lifting followed by gelatin lifter.     

The contact area of static and dynamic footprints: Forensic implications

When found at crime scenes, footprints may be evidentially valuable and can assist with the identity of a perpetrator based on their features and/or measurements. Footprints can be either static (made while standing) or dynamic (made while walking). While extensive research has been performed on the linear measurements obtained from static and dynamic footprints, research on the comparisons between the contact area of static and dynamic footprints in the forensic context are limited. The present study compares the contact area of static and dynamic bare footprints to determine if statistically significant differences exist between the two. Static and dynamic footprints were obtained from a sample of randomly-selected 461 Jatt Sikh adults (230 males and 231 females) of Indian origin between the ages of 19 and 32 years. The footprint contact area was calculated from each footprint (excluding the toes) using a PedoGRID® sheet. No statistically significant differences were observed between the contact area of static and dynamic footprints for each foot among males and females. However, statistically significant differences between both the sexes were found in the footprint contact areas of both footprint types. The right dynamic footprint contact area was found to be the most predictive measurement for classifying and estimating sex from a footprint’s contact area. The study has implications in the analysis of footprints recovered from crime scenes.     

Shoeprint image retrieval and crime scene shoeprint image linking by using convolutional neural network and normalized cross correlation

A shoeprint image retrieval process aims to identify and match images of shoeprints found at crime scenes with shoeprint images from a known reference database. It is a challenging problem in the forensic discipline of footwear analysis because a shoeprint found at the crime scene is often imperfect. Recovered shoeprints may be partial, distorted, left on surfaces that do not mark easily, or perhaps come from shoes that do not transfer marks easily. In this study, we present a shoeprint retrieval method by using a convolutional neural network (CNN) and normalized cross-correlation (NCC). A pre-trained CNN was used to extract features from the pre-processed shoeprint images. We then employed NCC to compute a similarity score based on the extracted image features. We achieved a retrieval accuracy of 82% in our experiments, where a “successful” retrieval means that the ground truth image was returned in the top 1% of returned images. We also extend our shoeprint retrieval method to the problem of linking shoeprints recovered from crime scenes. This new method can provide a linkage between two crime scenes if the two recovered shoeprints originated from the same shoe. This new method achieved a retrieval accuracy of 88.99% in the top 20% of returned images.     

Computerized classification of the shoeprints of burglars' soles

In the matter of shoe soles, there are many different types of designs which make the creation of an effective group classification difficult. The system created at the forensic technical laboratory in Neuchâtel, Switzerland, offers a new means of classification concerning the standards of the shoeprints of burglars' soles. This classification system has been created using the data processing programm FileMaker Pro 2.1 of Macintosh. It has been established in Neuchâtel for 4 years and contains over 1500 reference files. During this period of time it has been possible to identify four times as many burglars through shoeprints than with fingerprints.     

Two-dimensional linear analysis of dynamic bare footprints: A comparison of measurement techniques

In forensic intelligence-gathering, footprints have been shown to be valued evidence found at crime scenes. Forensic podiatrists and footprint examiners use a variety of techniques for measuring footprints for comparison of the crime scene evidence with the exemplar footprints. This study examines three different techniques of obtaining two-dimensional linear measurement data of dynamic bare footprints. Dynamic bare footprints were gathered from 50 students from a podiatric medical school using the Identicator® Inkless Shoe Print Model LE 25P system. After obtaining 100 bilateral footprints from the participants, the quantitative measurement data were collected by using three different measurement techniques: (i) a manual technique using a ruler (direct technique); (ii) an Adobe® Photoshop® technique; and (iii) a GIMP (GNU Image Manipulation Program) technique. The seven Reel linear measurement methodology was used for producing measurements using these three techniques.This study showed that all the mean bare footprint measurements on the right and left feet obtained using the direct technique were larger than those obtained using GIMP and Adobe® Photoshop® images. Differences were also observed in measurements produced using GIMP software and Photoshop images. However, the differences observed in the three techniques used for bare footprint measurements were not found to be statistically significant. The study concludes that there are no significant differences between the three measurement techniques when applied to two-dimensional bare footprints using the Reel method. It further concluded that any of these measurement techniques can be used when employing the Reel methodology for footprint analysis without significant difference.