Machine vision significantly improves the efficiency, quality, and reliability of defect detection. In visual inspection, excellent optical illumination platforms and suitable image acquisition hardware are the prerequisites for obtaining high-quality images. Image processing and analysis are key technologies in obtaining defect information, while deep learning is significantly impacting the field of image analysis. In this study, a brief history and the state of the art in optical illumination, image acquisition, image processing, and image analysis in the field of visual inspection are systematically discussed. The latest developments in industrial defect detection based on machine vision are introduced. In the further development of the field of visual inspection, the application of deep learning will play an increasingly important role. Thus, a detailed description of the application of deep learning in defect classification, localization and segmentation follows the discussion of traditional defect detection algorithms. Finally, future prospects for the development of visual inspection technology are explored.

/pdf/state-of-the-art-in-defect-detection-based-on-machine-vision-2rtzvgsnqy.pdf

State of the Art in Defect Detection Based on Machine Vision

Active infrared thermography is a safe, fast, and low-cost solution for subsurface defects inspection, providing quality control in many industrial production tasks. In this paper, we explore deep learning-based approaches to analyze lock-in thermography image sequences for non-destructive testing and evaluation (NDT&E) of subsurface defects. Different from most existing Convolutional Neural Network (CNN) models that directly classify individual regions/pixels as defective and non-defective ones, we present a novel two-stream CNN architecture to extract/compare features in a pair of 1D thermal signal sequences for accurate classification/differentiation of defective and non-defective regions. In this manner, we can significantly increase the size of the training data by pairing two individually captured 1D thermal signals, thereby greatly easing the requirement for collecting a large number of thermal sequences of specimens with defects to train deep CNN models. Moreover, we experimentally investigate a number of network alternatives, identifying the optimal fusion scheme/stage for differentiating the thermal behaviors of defective and non-defective regions. Experimental results demonstrate that our proposed method, directly learning how to construct feature representations from a large number of real-captured thermal signal pairs, outperforms the well-established lock-in thermography data processing techniques on specimens made of different materials and at various excitation frequencies.

https://ris.utwente.nl/ws/files/254342322/yang_two.pdf

Two-stream convolutional neural network for non-destructive subsurface defect detection via similarity comparison of lock-in thermography signals

Progress and trends in fault diagnosis for renewable and sustainable energy system based on infrared thermography: A review

A study on depth classification of defects by machine learning based on hyper-parameter search

This paper proposes a methodology for managing complex sewerage networks based on the concomitant use of two performance evaluation methods, namely, the failure modes, effects, and criticality anal...

Managing Sewerage Networks Using both Failure Modes, Effects and Criticality Analysis (FMECA) and Analytic Hierarchy Process (AHP) methods

Carbon fiber reinforced polymer (CFRP) has been increasingly used in aviation industry since it significantly enhances the performance of aircraft. However, imperfections inside the CFRP structures pose a threat to aviation safety. Apart from the defect shape and size, flaw depth is crucial to assess the defect severity. In this work, we utilize a laser infrared thermography (LIT) system to inspect an aviation CFRP sheet and adopt a long-short term memory recurrent neural network (LSTM-RNN) to determine the defect depth. Thermographic sequences obtained by LIT are processed using thermographic signal reconstructions (TSR) method. Raw data and TSR processed data are separately used to train and test the LSTM-RNN. Results show that background noises in the original thermal signals can be effectively reduced by the TSR method, which is helpful for the models to learn the signal characteristics. Compared with two traditional methods, recurrent neural network (RNN) and convolutional neural network (CNN), we find the LSTM-RNN outperforms.

/pdf/defect-depth-determination-in-laser-infrared-thermography-4tyggd2ik6.pdf

Defect Depth Determination in Laser Infrared Thermography Based on LSTM-RNN

Automatic defect prediction in glass fiber reinforced polymer based on THz-TDS signal analysis with neural networks

A new infrared preprocessing modality called sequence differential processing is proposed. Different from the cold image subtraction which only decreases the background noise, the proposed modality takes the parallel test into account to preprocess the raw thermal data sequence prior to applying the advanced postprocessing techniques such as principal component thermography and pulsed phase thermography. The results show that the preprocessing method not only effectively reduces the influence of uneven laser energy distribution on detection efficiency but also enhances internal defect information. Moreover, the combination of the proposed preprocessing method with the pulsed phase thermography and principal component analysis algorithms improves the ability of laser infrared thermography to detect defects inside aviation carbon-fiber-reinforced plastics.

Image enhancement method for laser infrared thermography defect detection in aviation composites

Recently, terahertz (THz) nondestructive testing technology has prevailed in the defect detection of composite materials yet is limited by identifying deeper flaws. To solve this problem, a time-domain piecewise imaging method based on terahertz signal extraction is proposed in this article. First, the surface reflection echo signal is removed by applying a time correction, and the time-domain signal is then segmented according to the thickness of the sample. Finally, high-quality terahertz images are obtained by image fusion technology. All imaging methods, including traditional ones, are utilized to detect defects in glass fiber reinforced plastic (GFRP) sample then the signal-to-noise ratio (SNR) and defect edge contrast of each method are calculated. The experimental results show that the proposed method can effectively improve the terahertz imaging of deeper defects in GFRP sample especially for multi-depth defects and possesses the highest SNR and defect edge contrast.

/pdf/time-segmented-image-fusion-based-multi-depth-defects-3yhnhk5xr6.pdf

Ruicong Xia

Papers

Defect Depth Determination in Laser Infrared Thermography Based on LSTM-RNN

Automatic defect prediction in glass fiber reinforced polymer based on THz-TDS signal analysis with neural networks

Image enhancement method for laser infrared thermography defect detection in aviation composites

Time Segmented Image Fusion Based Multi- Depth Defects Imaging Method in Composites With Pulsed Terahertz

Pixel-based thermal sequence processing algorithm based on R2 fractile threshold of non-linear fitting in active infrared thermography