An algorithm to compare two-dimensional footwear outsole images using maximum cliques and speeded-up robust feature

FILNet: Fast Image-Based Indoor Localization Using an Anchor Control Network

This paper designs a fast image-based indoor localization method based on an anchor control network (FILNet) to improve localization accuracy and shorten the duration of feature matching. Particularly, two stages are developed for the proposed algorithm. The offline stage is to construct an anchor feature fingerprint database based on the concept of an anchor control network. This introduces detailed surveys to infer anchor features according to the information of control anchors using the visual-inertial odometry (VIO) based on Google ARcore. In addition, an affine invariance enhancement algorithm based on feature multi-angle screening and supplementation is developed to solve the image perspective transformation problem and complete the feature fingerprint database construction. In the online stage, a fast spatial indexing approach is adopted to improve the feature matching speed by searching for active anchors and matching only anchor features around the active anchors. Further, to improve the correct matching rate, a homography matrix filter model is used to verify the correctness of feature matching, and the correct matching points are selected. Extensive experiments in real-world scenarios are performed to evaluate the proposed FILNet. The experimental results show that in terms of affine invariance, compared with the initial local features, FILNet significantly improves the recall of feature matching from 26% to 57% when the angular deviation is less than 60 degrees. In the image feature matching stage, compared with the initial K-D tree algorithm, FILNet significantly improves the efficiency of feature matching, and the average time of the test image dataset is reduced from 30.3 ms to 12.7 ms. In terms of localization accuracy, compared with the benchmark method based on image localization, FILNet significantly improves the localization accuracy, and the percentage of images with a localization error of less than 0.1m increases from 31.61% to 55.89%.

Toward Developing Techniques─Agnostic Machine Learning Classification Models for Forensically Relevant Glass Fragments

Glass fragments found in crime scenes may constitute important forensic evidence when properly analyzed, for example, to determine their origin. This analysis could be greatly helped by having a large and diverse database of glass fragments and by using it for constructing reliable machine learning (ML)-based glass classification models. Ideally, the samples that make up this database should be analyzed by a single accurate and standardized analytical technique. However, due to differences in equipment across laboratories, this is not feasible. With this in mind, in this work, we investigated if and how measurement performed at different laboratories on the same set of glass fragments could be combined in the context of ML. First, we demonstrated that elemental analysis methods such as particle-induced X-ray emission (PIXE), laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS), scanning electron microscopy with energy-dispersive X-ray spectrometry (SEM-EDS), particle-induced Gamma-ray emission (PIGE), instrumental neutron activation analysis (INAA), and prompt Gamma-ray neutron activation analysis (PGAA) could each produce lab-specific ML-based classification models. Next, we determined rules for the successful combinations of data from different laboratories and techniques and demonstrated that when followed, they give rise to improved models, and conversely, poor combinations will lead to poor-performing models. Thus, the combination of PIXE and LA-ICP-MS improves the performances by ∼10-15%, while combining PGAA with other techniques provides poorer performances in comparison with the lab-specific models. Finally, we demonstrated that the poor performances of the SEM-EDS technique, still in use by law enforcement agencies, could be greatly improved by replacing SEM-EDS measurements for Fe and Ca by PIXE measurements for these elements. These findings suggest a process whereby forensic laboratories using different elemental analysis techniques could upload their data into a unified database and get reliable classification based on lab-agnostic models. This in turn brings us closer to a more exhaustive extraction of information from glass fragment evidence and furthermore may form the basis for international-wide collaboration between law enforcement agencies.

Examination of the possibility to use Siamese networks for the comparison of firing pin marks

One of the most discussed issues in forensic firearms identification is the subjectivity of conclusions. The main part of firearms examiners' work is to make a microscopic comparison of the marks on cartridge cases and bullets. In this process, examiners have to decide if the quantity and the quality of the observed characteristics are sufficient for identification. This decision is based on the personal experience of an examiner, so examiners with different backgrounds can come to different conclusions, and this fact presents a problem. Besides, the calculation of the error rate for this type of examination is a debatable issue. Different mathematical and statistical models were proposed, and computer-based algorithms were developed in order to avoid subjectivity and to determine error rates. This article investigates the possibility to use methods of machine learning for the comparison of marks of the firing pin impressions on cartridge cases. In the research, the Siamese network model, which included two similar Convolutional Neural Networks, was prepared and trained. For the training and validation of the model, the database of firing pin impressions was prepared. This database included images of cartridge cases discharged from 300 firearms that came from regular casework and clone images used for data augmentation. The model was trained and examined using the validation part of the database. The metrics, such as accuracy, sensitivity, and specificity were calculated. The results of the research show the possibility of using the Siamese network for building an objective forensic firearms examination system with a known error rate.

Location distribution of randomly acquired characteristics on a shoe sole

Footwear comparison is used to link between a suspect's shoe and a shoeprint found at a crime scene. Forensic examiners compare the two items, and the conclusion reached is based on class characteristics and randomly acquired characteristics (RACs), such as scratches or holes. An important question concerns the distribution of the location of RACs on shoe soles, which can serve as a benchmark for comparison. This study examines the probability of observing RACs in different areas of a shoe sole using a database of approximately 13,000 RACs observed on 386 outsoles. The analysis is somewhat complicated as the shoes are differentiated by shape and contact surface, and the RACs' locations are subject to measurement errors. A method that takes into account these challenges is presented. All impressions are normalized to a standardized axis to allow for inter‐comparison of RACs on outsoles of different sizes and contact areas, and RACs are localized to one of 14 subareas of the shoe sole. Expected frequencies in each region are assumed to be Poisson distributed with rate parameters that depend on the subarea and the contact surface. Three different estimation approaches are studied: a naive crude approach, a shoe‐specific random effects model, and an estimate that is based on conditional maximum likelihood. It is shown that the rate is not uniform across the shoe sole and that RACs are approximately twice as likely to appear at certain locations, corresponding to the foot's morphology. The results can guide investigators in determining a shoeprint's evidential value.

The effect of image descriptors on the performance of classifiers of footwear outsole image pairs

Shoe prints are commonly found at the scene of a crime and can sometimes help link a suspect to the scene. Because prints tend to be partially observed or smudgy, comparing crime scene prints with reference images from a putative shoe can be challenging. Footwear examiners rely on guidelines such as those published by SWGTREAD [1] to visually assess the similarity between two or more footwear impressions, one reason being that reliable, quantitative methods have yet to be validated for use in real cases. To help in the development of such methods, we created a study dataset of images of outsole impressions that shared class characteristics and degree of wear and that were subject to a specific type of degradation. We also propose a method to quantify the similarity between two outsole images that extends the capabilities of MC-COMP [2]. The proposed method is composed of three steps; (1) extracting image descriptors, (2) aligning images using the maximum clique, (3) calculating similarity values using two different classifiers; (a) degree of overlap between the two images, and (b) a score produced by a random forest. To explore the performance of the algorithm we propose, we compared degraded, crime scene-like images to high-quality reference images produced by the same or by different shoes. Even though comparisons involved matches or very close non-matches, and one of the images was blurry, the algorithm shows good source classification performance.

Quantitative evaluation of footwear evidence: Initial workflow for an end-to-end system

In the United States, footwear examiners make decisions about the sources of crime scene shoe impressions using subjective criteria. This has raised questions about the accuracy, repeatability, reproducibility, and scientific validity of footwear examinations. Currently, most footwear examiners follow a workflow that compares a questioned and test impression with regard to outsole design, size, wear, and randomly acquired characteristics (RACs). We augment this workflow with computer algorithms and statistical analysis so as to improve in the following areas: (1) quantifying the degree of correspondence between the questioned and test impressions with respect to design, size, wear, and RACs, (2) reducing the potential for cognitive bias, and (3) providing an empirical basis for examiner conclusions by developing a reference database of case-relevant pairs of impressions containing known mated and known nonmated impressions. Our end-to-end workflow facilitates all three of these points and is directly relatable to current practice. We demonstrate the workflow, which includes obtaining and interpreting outsole pattern scores, RAC comparison scores, and final scores, on two scenarios-a pristine example (involving very high quality Everspry EverOS scanner impressions) and a mock crime scene example that more closely resembles real casework. These examples not only demonstrate the workflow but also help identify the algorithmic, computational, and statistical challenges involved in improving the system for eventual deployment in casework.

Comparing footwear impressions that are close non-matches using correlation-based approaches

Forensic activities related to footwear evidence may be broadly classified into the following two categories: (1) intelligence gathering and (2) evidential value assessment. Intelligence gathering provides additional leads for investigators. Assessment of evidential value, as practiced in the United States, involves a trained footwear examiner evaluating the degree of similarity between a known shoe of interest (together with its test impressions) and footwear impressions obtained from a crime scene, by performing side-by-side visual comparisons. However, the need for developing quantitative approaches for expressing similarities during such comparisons is being increasingly recognized by the forensic science community. In this paper, we explore the ability of similarity metrics to discriminate between impressions made by a shoe of interest and impressions made by close non-matching shoes. Close non-matching shoes largely share the same design and size. Therefore, the ability to effectively discriminate between them requires considering, either explicitly or implicitly, not only design and size, but also wear patterns and, to some extent, individual characteristics. This type of discrimination is necessary for assessment of evidential value. The similarity metrics examined in this paper are correlation-based metrics, including normalized cross-correlation, phase-only correlation, AvNCC, and AvPOC. The latter two metrics are based on features obtained from a convolutional neural network. Experiments are performed using Everspry impressions, FBI boot impressions, and the West Virginia University footwear impression collection. The results show that phase-only correlation performs as well as or better than the other metrics in all cases for the datasets we considered.