Quantifying the similarity of 2D images using edge pixels: an application to the forensic comparison of footwear impressions

dc.contributor.author Park, Soyoung
dc.contributor.author Carriquiry, Alicia
dc.contributor.department Center for Statistics and Applications in Forensic Evidence
dc.contributor.department Statistics
dc.date 2020-08-07T15:19:54.000
dc.date.accessioned 2021-02-25T00:41:34Z
dc.date.available 2021-02-25T00:41:34Z
dc.date.copyright Wed Jan 01 00:00:00 UTC 2020
dc.date.issued 2020-06-11
dc.description.abstract <p>We propose a novel method to quantify the similarity between an impression (<em>Q</em>) from an unknown source and a test impression (<em>K</em>) from a known source. Using the property of geometrical congruence in the impressions, the degree of correspondence is quantified using ideas from graph theory and maximum clique (MC). The algorithm uses the <em>x</em> and <em>y</em> coordinates of the edges in the images as the data. We focus on local areas in <em>Q</em> and the corresponding regions in <em>K</em> and extract features for comparison. Using pairs of images with known origin, we train a random forest to classify pairs into mates and non-mates. We collected impressions from 60 pairs of shoes of the same brand and model, worn over six months. Using a different set of very similar shoes, we evaluated the performance of the algorithm in terms of the accuracy with which it correctly classified images into source classes. Using classification error rates and ROC curves, we compare the proposed method to other algorithms in the literature and show that for these data, our method shows good classification performance relative to other methods. The algorithm can be implemented with the R package shoeprintr.</p>
dc.description.comments <p>This is a manuscript of an article published as Park, Soyoung, and Alicia Carriquiry. "Quantifying the similarity of 2D images using edge pixels: an application to the forensic comparison of footwear impressions." <em>Journal of Applied Statistics</em> (2020): 1-28. Posted with permission of CSAFE.</p>
dc.format.mimetype application/pdf
dc.identifier archive/lib.dr.iastate.edu/csafe_pubs/53/
dc.identifier.articleid 1052
dc.identifier.contextkey 18818807
dc.identifier.s3bucket isulib-bepress-aws-west
dc.identifier.submissionpath csafe_pubs/53
dc.identifier.uri https://dr.lib.iastate.edu/handle/20.500.12876/93862
dc.language.iso en
dc.source.bitstream archive/lib.dr.iastate.edu/csafe_pubs/53/_JoAS_Spark_ACarriquiry_2020May.pdf|||Sat Jan 15 00:50:48 UTC 2022
dc.source.uri 10.1080/02664763.2020.1779194
dc.subject.disciplines Forensic Science and Technology
dc.subject.keywords Maximum clique
dc.subject.keywords learning algorithms
dc.subject.keywords shoe outsole comparison
dc.subject.keywords pattern matching
dc.subject.keywords image analysis
dc.title Quantifying the similarity of 2D images using edge pixels: an application to the forensic comparison of footwear impressions
dc.type article
dc.type.genre article
dspace.entity.type Publication
relation.isOrgUnitOfPublication d8a3c72b-850f-40f6-87c4-8812547080c7
relation.isOrgUnitOfPublication 264904d9-9e66-4169-8e11-034e537ddbca
File
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
_JoAS_Spark_ACarriquiry_2020May.pdf
Size:
3.29 MB
Format:
Adobe Portable Document Format
Description:
Collections