Wear debris shape classification

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)


Wear debris is produced in all machines containing moving parts. Wear debris or particles separate from these moving parts because of close contacts and friction and are contained in oil in an oil-wetted system. Analysis of wear debris provides important information about the condition of a machine. The produced particles come in different shapes, sizes, colors, and surface texture. This paper describes the morphological analysis of wear particles by using computer vision and image processing techniques. The aim is to classify these particles according to their shape attributes. Four particle shapes are classified by using Histogram of Oriented Gradients (HOG) and shape attributes including eccentricity, extent, major and minor axis length, equiv-diameter, and centroid distance. The shape classification can be used to identify origin of particle generation and thus predict wear failure modes in engines and other machinery. The objective of particle classification obviates reliance on visual inspection techniques and the need for specialists in the field.

Original languageEnglish
Title of host publicationModelling, Simulation and Intelligent Computing - Proceedings of MoSICom 2020
EditorsNilesh Goel, Shazia Hasan, V. Kalaichelvi
Number of pages9
ISBN (Print)9789811547744
Publication statusPublished - 2020
EventInternational Conference on Modelling, Simulation and Intelligent Computing, MoSICom 2020 - Dubai, United Arab Emirates
Duration: Jan 29 2020Jan 31 2020

Publication series

NameLecture Notes in Electrical Engineering
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119


ConferenceInternational Conference on Modelling, Simulation and Intelligent Computing, MoSICom 2020
Country/TerritoryUnited Arab Emirates


  • Computer vision
  • Histogram of oriented gradients
  • Image processing
  • Particle shape classification
  • Wear debris

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering


Dive into the research topics of 'Wear debris shape classification'. Together they form a unique fingerprint.

Cite this