Open AccessAdaptive matching algorithm for laser detection signals of linear cutting tool marks
Author(s) -
Pan Nan,
Liu Yi,
Wu Sixing
Publication year - 2017
Publication title -
iet signal processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.384
H-Index - 42
ISSN - 1751-9683
DOI - 10.1049/iet-spr.2015.0372
Subject(s) - smoothing , similarity (geometry) , algorithm , computer science , feature (linguistics) , matching (statistics) , pattern recognition (psychology) , displacement (psychology) , artificial intelligence , noise (video) , mathematics , computer vision , image (mathematics) , psychology , linguistics , philosophy , statistics , psychotherapist
Since it is very difficult to compare linear cutting tool marks quickly and quantitatively using existing image‐processing and three‐dimensional scanning methods, an adaptive matching algorithm for laser detection signals of linear cutting marks is proposed. Using locally weighted scatterplot smoothing regression, the proposed algorithm first performs noise reduction on the surface signals of linear cutting tool marks that are detected by a laser displacement sensor. Trends of the thick and consistent features of the signal data are then identified and the feature vectors are quantised via cosine vector curve fitting to individually calculate the spatial distances of the samples. Finally, the most similar samples are matched via batch similarity comparison using a dynamic programming (DP) algorithm. The accuracy and validity of the proposed algorithm are verified by similarity comparison tests of actual cutting marks from a variety of samples.