Cellular neural networks for welding arc thermograms segmentation

•2D discrete time CNN was studied for welding arc thermogram segmentation.•For welding arc segmentation CNN provide good arc-joint region separability.•Batter diagnostic signals as in the case of region growing can be obtained with CNN.•Emissivity setting error has marginal influence on CNN segmenta...

Full description

Saved in:
Bibliographic Details
Published in:Infrared physics & technology 2014-09, Vol.66, p.18-28
Main Author: Jamrozik, Wojciech
Format: Article
Language:English
Subjects:
Algorithms
Arc welding
Cellular
Cellular neural networks
Computer simulation
Image segmentation
Neural networks
Segmentation
Thermovision
Welding
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•2D discrete time CNN was studied for welding arc thermogram segmentation.•For welding arc segmentation CNN provide good arc-joint region separability.•Batter diagnostic signals as in the case of region growing can be obtained with CNN.•Emissivity setting error has marginal influence on CNN segmentation. Machine vision systems are used in many areas for monitoring of technological processes. Among this processes welding takes important place, where often infrared cameras are used. Besides reliable hardware, successful application of vision systems requires suitable software based on proper algorithms. One of most important group of image processing algorithms is connected to image segmentation. Obtainment of exact boundary of an object that changes shape in time, such as the welding arc, represented on a thermogram is not a trivial task. In the paper a segmentation method using supervised approach based on a cellular neural networks is presented. Simulated annealing and genetic algorithm were used for training of the network (template optimization). Comparison of proposed method to a well elaborated segmentation method based on region growing approach was made. Obtained results prove that the cellular neural network can be a valuable tool for infrared welding pool images segmentation.
ISSN:1350-4495
1879-0275
DOI:10.1016/j.infrared.2014.05.005