Fine-Grained Visual Categorization of Fasteners in Overhaul Processes

Sajjad Taheritanjani, Juan Haladjian, Bernd Bruegge

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

8 Scopus citations

Abstract

Commercial aircraft engines must be overhauled approximately every six years, during which hundreds of different parts must be disassembled, checked, and then reassembled. This includes undoing up to thousands of fasteners, cleaning, checking, refitting, and tightening them. Prior to refitting the fasteners, they must be classified and packaged. In this paper, we describe a system for classifying fasteners automatically, by use of computer vision and machine learning. Using the proposed system, we created sample datasets and performed a fine-grained visual categorization of the fasteners. Our trained model classifies 20 bolts and 14 washers with an accuracy of 99.4%. Our work is the first step towards an automated fastener classification system in overhaul processes.

Original languageEnglish
Title of host publication2019 5th International Conference on Control, Automation and Robotics, ICCAR 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages241-248
Number of pages8
ISBN (Electronic)9781728133263
DOIs
StatePublished - Apr 2019
Event5th International Conference on Control, Automation and Robotics, ICCAR 2019 - Beijing, China
Duration: 19 Apr 201922 Apr 2019

Publication series

Name2019 5th International Conference on Control, Automation and Robotics, ICCAR 2019

Conference

Conference5th International Conference on Control, Automation and Robotics, ICCAR 2019
Country/TerritoryChina
CityBeijing
Period19/04/1922/04/19

Keywords

  • FGVC
  • Fine-grained Visual Categorization of Fasteners
  • Industrial Automation
  • Overhaul Processes
  • Small Parts Categorization

Fingerprint

Dive into the research topics of 'Fine-Grained Visual Categorization of Fasteners in Overhaul Processes'. Together they form a unique fingerprint.

Cite this