NCJ Number
305931
Date Published
2020
Length
2 pages
Annotation
This article presents a workflow for unsupervised clustering a large collection of forensic images.
Abstract
Large collections of images, if curated, drastically contribute to the quality of research in many domains. Unsupervised clustering is an intuitive, yet effective step towards curating such datasets. The workflow in the current project utilizes classic clustering on deep feature representation of the images in addition to domain-related data to group them together. The manual evaluation shows a purity of 89% for the resulted clusters. (Published abstract provided)
Date Published: January 1, 2020
Similar Publications
- A Systematic Study of Liquid Chromatography to Separate Eighteen Natural Cannabinoids for Potency Testing of Hemp-Based Products Using Diode Array Detector and Electrospray Ionization Mass Spectrometry
- Clinical characteristics of patients exposed to medetomidine in the illicit opioid drug supply in Philadelphia – a case series
- Identifying Thermal Decomposition Products of Nitrate Ester Explosives Using Gas Chromatography–Vacuum Ultraviolet Spectroscopy: An Experimental and Computational Study