Abstract

Recent advancements of nondestructive testing and evaluation (NDT&E) with machine learning, artificial intelligence, and the internet of things as key enablers in parallel with industry 4.0 reached the fourth industrial revolution. Nevertheless, active thermography (AT) is a noncontact, whole field, safe, remote, cost-efficient, and widely used NDT technique for subsurface anomaly detection. In AT, the automatic defect detection is modelled as object localization and semantic segmentation in thermograms. This paper presents a feature fusion network that fuses the global features extracted using a deep neural network (DNN) with the deep features extracted using a convolutional neural network (CNN). A set of handcrafted timedomain statistical and frequency domain features of thermal profiles are given to the DNN subnetwork whereas, the CNN subnetwork is fed with the thermal profiles in the feature fusion network. Experimentation is carried out over carbon fiber reinforced polymer (CFRP) sample with artificially drilled flat bottom holes excited by quadratic frequency-modulated optical stimulus. Experimental results showed that the feature fusion enhanced the defect detection capability compared to the local networks with a significant increment in signal-to-noise ratio, accuracy, and F-score.

中文翻译：

---

用于二次频率调制热波成像中自动缺陷检测的深度和手工特征融合

摘要

无损检测和评估 (NDT&E) 的最新进展以机器学习、人工智能和物联网作为与工业 4.0 并行的关键推动因素，达到了第四次工业革命。尽管如此，主动热成像 (AT) 是一种非接触式、全场、安全、远程、经济高效且广泛使用的无损检测技术，用于地下异常检测。在 AT 中，自动缺陷检测被建模为热图中的对象定位和语义分割。本文提出了一种特征融合网络，该网络将使用深度神经网络 (DNN) 提取的全局特征与使用卷积神经网络 (CNN) 提取的深度特征进行融合。一组手工制作的热分布的时域统计和频域特征被提供给 DNN 子网，而，CNN 子网络被输入特征融合网络中的热分布。实验是在碳纤维增强聚合物 (CFRP) 样品上进行的，该样品具有由二次频率调制光刺激激发的人工钻孔平底孔。实验结果表明，与局部网络相比，特征融合增强了缺陷检测能力，信噪比、准确度和 F 值都有显着增加。