Panoramic Crack Detection for Steel Beam Based on Structured Random Forests
Author(s) -
Sen Wang,
Xiaoqin Liu,
Tangfeng Yang,
Xing Wu
Publication year - 2018
Publication title -
ieee access
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.587
H-Index - 127
ISSN - 2169-3536
DOI - 10.1109/access.2018.2812141
Subject(s) - aerospace , bioengineering , communication, networking and broadcast technologies , components, circuits, devices and systems , computing and processing , engineered materials, dielectrics and plasmas , engineering profession , fields, waves and electromagnetics , general topics for engineers , geoscience , nuclear engineering , photonics and electrooptics , power, energy and industry applications , robotics and control systems , signal processing and analysis , transportation
Condition monitoring and fault diagnosis are the most important process in manufacturing industries. In this paper, a steel beam panoramic crack detection method based on structured random forests has been proposed to obtain more efficient maintenance of manufacturing equipment. The structured random forests method and semi-reconstruction method of anti-symmetrical bi-orthogonal wavelets are combined to detect the edges of the cracks. Candidate features of the crack images are randomly chosen to train the crack classifier. Besides, the fast-multi-image stitching method is applied to stitch the entire image. The generated crack detection classifier is also used to determine the classification by voting the feature vector of each image. The prescribed characteristics, i.e., area, height, and weight, are introduced to select those cracks that satisfy the prescribed conditions. The experimental results show that the approach is effective and efficient in recognizing the surface cracks of the panoramic steel beam.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom