Deep Boltzmann machine for corrosion classification using eddy current pulsed thermography
Deep Boltzmann machine for corrosion classification using eddy current pulsed thermography
复制标题
DOI：
10.1016/j.ijleo.2020.164828
复制
发表时间：
2020-10
期刊：
影响因子：
3.1
通讯作者：
Yuming Chen;F. Sohel;S. A. A. Shah-S.-A.-A.-Shah-2112405071;S. Ding
中科院分区：
文献类型：
--
作者：
Yuming Chen;F. Sohel;S. A. A. Shah-S.-A.-A.-Shah-2112405071;S. Ding
本文的目的是用涡流脉冲热成像技术对导电材料的腐蚀进行分类。热瞬态图像产生的数据量大，难以对不同腐蚀材料进行准确检测和分类，特别是具有隐蔽性的腐蚀。我们应用深度玻尔兹曼机（DBM）网络从整个测量区域自动提取和分类特征。腐蚀分类测试了几种不同的机器学习算法，包括：聚类，PCA，多层DBM分类器。所提出的框架的性能是衡量的准确性，灵敏度，特异性和精度。在四种不同腐蚀程度的涡流信号样本数据集上进行了几个实验。实验结果表明，该方法在分类准确率（97.9%）、灵敏度（96.1%）、精确度（97.1%）和特异度（98.4%）方面均优于现有算法。
The aim of this paper is to classify conductive material corrosion by eddy current pulsed thermography. Thermal transient images generate a large of amount of data which is difficult for accurate detection and classification of the different corrosion materials, especially with the hidden corrosion. We apply deep Boltzmann machines (DBM) network to automatically extract and classify features from the whole measured area. Corrosion classification is tested with several different machine learning based algorithms including: clustering, PCA, multi-layer DBM classifier. The performance of the proposed framework is measured in terms of accuracy, sensitivity, specificity and precision. Several experiments are performed on a dataset of eddy current signal samples for four different corrosion degrees. The results show that our method outperforms the existing algorithms in classification accuracy (97.9%), sensitivity (96.1%), precision (97.1%), and especially specificity (98.4%).