An image-processing methodology for extracting bloodstain pattern features

Highlights • An image-processing methodology for analysing bloodstain patterns is proposed. • Local and global features based on traditional spatter properties were extracted. • Quantitative pattern data will enhance current bloodstain pattern taxonomies. • Pattern data will enable analysts to reach...

Full description

Saved in:
Bibliographic Details
Published in:Forensic science international 2017-08, Vol.277, p.122-132
Main Authors: Arthur, Ravishka M, Humburg, Philomena J, Hoogenboom, Jerry, Baiker, Martin, Taylor, Michael C, Bruin, Karla G.de
Format: Article
Language:English
Subjects:
Animals
Blood
Blood Stains
Bloodstain Pattern Analysis
Classification
Data processing
Feature extraction
Forensic engineering
Forensic science
Forensic sciences
Forensic Sciences - methods
Global features
Humans
Identification methods
Image processing
Image Processing, Computer-Assisted - methods
Local features
Methods
Neural networks
Numerical methods
Pathology
Pattern recognition
Stains & staining
Statistics as Topic
Taxonomy
Vocabularies & taxonomies
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Highlights • An image-processing methodology for analysing bloodstain patterns is proposed. • Local and global features based on traditional spatter properties were extracted. • Quantitative pattern data will enhance current bloodstain pattern taxonomies. • Pattern data will enable analysts to reach objective pattern conclusions. • This study lays the groundwork for an automated pattern classification system.
ISSN:0379-0738
1872-6283
DOI:10.1016/j.forsciint.2017.05.022