A method for detecting wear and damage on railcar wheel tread surface using a combination of laser measurement and machine vision

• DOI: 10.1299/mej.23-00352
• Publication Date: 2024-04-04T22:13:01Z
• Authors: Takeshi EMOTO; Takanori EMARU; Ankit RAVANKAR; Abhijeet RAVANKAR; Yukinori KOBAYASHI
• Keywords: automatic inspection system; railcar; wheel tread surface condition; TJ1-1570; wheel tread dimension; machine vision; Mechanical engineering and machinery; wheel tread surface defect; 01 natural sciences; laser sensor; 0105 earth and related environmental sciences

Skilled technicians primarily use manual tools to maintain the geometry and surface conditions of railcar wheels. To ensure efficient wheel maintenance, development an automated inspection system is required. Herein, method for wheels proposed based on assessment tread dimensions investigation results are presented. The parameters considered because they crucial railway operation safety manually inspecting them requires a large workforce. improve efficiency system, combined through laser measurement machine vision used, both techniques compensate each other's shortcomings. experiments conducted using sensors techniques, including defect detection AI model YOLOv5, as well image data captured via digital video camera obtain high-accuracy automatic measurements determine feasibility such measurements. Additionally, accuracy verified sample having defects. In this paper, various defects used.