Cited time in webofscience Cited time in scopus

Automatic Brittle Fracture Ratio Estimation Using Convolutional Neural Network Regression Based on Classmap Regulation

Title
Automatic Brittle Fracture Ratio Estimation Using Convolutional Neural Network Regression Based on Classmap Regulation
Author(s)
Jeong, Seung HyunWoo, Min WooKoo, GyogwonLee, Jong-HakYun, Jong Pil
Issued Date
2021-10
Citation
IEEE Access, v.9, pp.136552 - 136560
Type
Article
Author Keywords
Surface cracksDiscrete wavelet transformsConvolutional neural networksFeature extractionEstimationTrainingInspectionBrittle fracture rate estimatorconvolutional neural network regressiondrop-weight tear testheatmap regulation
Keywords
WEIGHT-TEAR TESTLOW-TEMPERATURE TOUGHNESSX70MICROSTRUCTURESTEEL
ISSN
2169-3536
Abstract
A convolutional neural network (CNN) based regression is proposed for estimating the brittle fracture ratio (BFR) in a fracture image of a drop weight tear test (DWTT) specimen. Different with the previous complex semantic segmentation-based estimator, the method extracts the feature vector through global average pooling of feature map and calculates the BFR directly through the fully connected layer. By removing decoder network, the number of weights, training time, and required GPU memory dramatically reduced. To train the proposed CNN, a new loss function, which is the sum of L1-norm between class activation map and ground truth inspection image and L1-norm of BFR error, is also designed. To validate the present method, fracture images of 1532, 79, and 158 DWTT specimens obtained from real industrial site were used for training, validation, and test, respectively. The accuracy of the proposed method was evaluated based on the number of test samples with an error of 5% or less divided by the total number of test samples, which is the measure used in real industrial application. Despite having dramatically reduced the number of weights and inference time by 85.8% and 64.8%, respectively, the proposed method has a higher accuracy (96.2%) compared to that of the existing segmentation based BFR estimation method (94.9%).
URI
http://hdl.handle.net/20.500.11750/15662
DOI
10.1109/ACCESS.2021.3117579
Publisher
Institute of Electrical and Electronics Engineers Inc.
Related Researcher
  • 구교권 Koo, Gyogwon
  • Research Interests Computer vision; 컴퓨터 비전; Deep learning; 딥러닝; Defect detection; 결함 검사; Product number recognition; 제품번호인식; Anomaly detection; 이상 탐지; Smart factory; 스마트 공장
Files in This Item:

There are no files associated with this item.

Appears in Collections:
Division of Intelligent Robot 1. Journal Articles

qrcode

  • twitter
  • facebook
  • mendeley

Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE