Skip to content

Archived

All archived opportunities can be found on Youtube.

Visual Diagnostics for Algorithmic Cartridge Case Comparisons

Automatic cartridge case comparison algorithms can be used to measure the similarity between two fired cartridge cases. However, many of these algorithms are difficult to interpret, understand, and audit. We introduce a suite of visual diagnostic tools that are useful for understanding the behavior of automatic comparison algorithms and introduce a novel algorithm that returns a probability that two cartridge cases originated from the same firearm.

This webinar is available on YouTube.

Algorithmic Assessment of Striation Similarity between Wire Cuts

There is no systematic algorithm to inspect striations on wires without manual extraction. A new, reproducible, automatic algorithm is proposed to analyze the similarity between wires. Interpreting the result correctly promises a consistent way of using wires as forensic evidence in the future.

This webinar is available on YouTube.

Multi-Camera Smartphone Image Identification

Source camera identification uses small imperfections in a camera’s sensor array, called a camera fingerprint. We introduce the new CSAFE multi-camera smartphone image database and discuss how we adapt traditional camera identification methods to work on multi-camera smartphones.

This webinar is available on YouTube.

A panel discussion with NIST researchers and the CSAFE team on the likelihood ratio

This CSAFE Center Wide Webinar was presented on March 1, 2018 by a panel of speakers who examine the impact of the likelihood ratio in a legal context, how to communicate the likelihood ratio to lay audiences and more, followed by a time for questions from the audience.

This webinar is available on YouTube.

Covering the Basic Concepts Surrounding the Weight and Strength of Evidence

In this two-part presentation, we will focus on aspects of the forensic identification of source problems where the main question of interest is determining the origin of evidence with unknown source.

This webinar is on-demand and available on YouTube.

Footwear Impression Evidence Examination: Overview, challenges, research and the future

This presentation will provide an overview of footwear impression examination, explore directions that research in this discipline.

This webinar is available on YouTube.

Statistical and Computational Tools for Automated Matching of Footwear Class Characteristics

We investigate the problem of automatically determining shoe outsole class characteristics from crime scene impression evidence using computer vision and machine learning techniques.

This webinar is available on YouTube.

Learning Algorithms for Evaluating Forensic Glass Evidence

Glass evidence may arise when a glass object is broken during the commission of a crime. Small fragments can transfer to the perpetrator.

This webinar is available on YouTube.

Probabilistic Reporting in American Criminal Cases: A Baseline Study

The reporting of forensic results is a topic of crucial importance and increasing interest. Existing scholarship primarily addresses how forensic results should or could be reported. Our purpose is to understand empirically how forensic results are actually reported in American trials today. Given that many forensic statisticians are advocating for the greater use of probabilistic reporting, this research may allow us to establish a baseline so we can measure progress toward that goal.

This webinar is available on YouTube.

Similarity of Two-Dimensional Images: An application to the forensic comparison of shoe outsole impressions

Shoe outsole prints are often found in crime scenes. If a suspect is apprehended and her shoes are potential sources of the prints in the crime scene, how might a forensic scientist go about quantifying the degree of similarity between the two?

This webinar is available on YouTube.