Automatic Defect Recognition Research Articles

A Systematic Review of Machine-Vision-Based Leather Surface Defect Inspection

Machine-vision-based surface defect inspection is one of the key technologies to realize intelligent manufacturing. This paper provides a systematic review on leather surface defect inspections based on machine vision. Leather products are regarded as the most traded products all over the world. Automatic detection, location, and recognition of leather surface defects are very important for the intelligent manufacturing of leather products, and are challenging but noteworthy tasks. This work investigates a large amount of literature related to leather surface defect inspection. In addition, we also investigate and evaluate the performance of some edge detectors and threshold detectors for leather defect detection, and the identification accuracy of the classical machine learning method SVM for leather surface defect identification. A detailed and methodical review of leather surface defect inspection with image analysis and machine learning is presented. Main challenges and future development trends are discussed for leather surface defect inspection, which can be used as a source of guidelines for designing and developing new solutions in this field.

Defect Recognition and Morphology Operation in Binary Images Using Line-Scanning-Based Induction Thermography

Active infrared thermography is an attractive and highly reliable technique used for the non-destructive evaluation of test objects. In this paper, defect detection on the subsurface of the STS304 metal specimen was performed by applying the line-scanning method to induction thermography. In general, the infrared camera and the specimen are fixed in induction thermography, but the line-scanning method can excite a uniform heat source because relative movement occurs. After that, the local heating area due to Joule’s heating effect was removed, and filtering was applied for the 1st de-noising. Threshold-value-based binarization processing using the Otsu algorithm was performed for clear defect object recognition. After performing the 2nd de-noising, automatic defect recognition was performed using a boundary tracking algorithm. As a result, the conditions due to the parameters of the scanning line for the thermal image were determined.

Automatic Recognition Method of Broken Transmission Line Defect Image Based on Deep Transfer Learning

The material of ACSR in transmission line is prone to local damage, which leads to broken strand defect and reducing power consumption safety. Therefore, an automatic recognition method of broken strand defect image of transmission line based on deep transfer learning is designed to improve the automatic recognition effect of broken strand defect image. The multi-scale algorithm is used to enhance the image. In the feature extraction part of the depth transfer learning framework in the confusion domain, the multi-source domain transfer and dual flow fusion algorithm are used to extract the features of the enhanced image, and the Euclidean distance between the feature vector and the template image feature vector is used to correct the image features; using the corrected image feature training network propagated to the automatic defect recognition part and the domain classification part, the loss function and back propagation algorithm are used to reduce the loss of feature extraction and automatic defect recognition part, and the optimal results of automatic defect recognition and domain classification are obtained. The experimental results show that the method can enhance the image effectively with high definition. At different image angles, the recognition accuracy of this method is as high as 0.96, which has better automatic recognition effect of defect image.

Automatic Recognition of Welding Seam Defects in TOFD Images Based on TensorFlow

Automatic Recognition of Welding Seam Defects in TOFD Images Based on TensorFlow

The Technique of Automatic Fault Recognition of the RDPU

The Technique of Automatic Fault Recognition of the RDPU

Enhancement of Multi-Class Structural Defect Recognition Using Generative Adversarial Network

Recently, in the building and infrastructure fields, studies on defect detection methods using deep learning have been widely implemented. For robust automatic recognition of defects in buildings, a sufficiently large training dataset is required for the target defects. However, it is challenging to collect sufficient data from degrading building structures. To address the data shortage and imbalance problem, in this study, a data augmentation method was developed using a generative adversarial network (GAN). To confirm the effect of data augmentation in the defect dataset of old structures, two scenarios were compared and experiments were conducted. As a result, in the models that applied the GAN-based data augmentation experimentally, the average performance increased by approximately 0.16 compared to the model trained using a small dataset. Based on the results of the experiments, the GAN-based data augmentation strategy is expected to be a reliable alternative to complement defect datasets with an unbalanced number of objects.

Instance contrastive learning with dynamic weighted variance for small sample steel defect recognition

As an essential element in industrial steel, automatic defect recognition can guarantee the surface quality through focused supervised learning with ample labelled samples. However, defect recognition inevitably features with data-limiting characteristic under the influence of costly expert labelling. To address this problem, a novel framework, Instance Contrast (InCo), is proposed with the inspiration of contrastive learning. This framework consists of two streams. One with instance labels attributed to the unlabelled data in each batch for classification, which is called Batch Instance Discrimination (BID). The other with different enhanced samples embedding of the same image aggregated by a new function named dynamic weighted variance loss (DWV loss). Therefore, better semantic features can be learned by model due to the moderation of embedding distance between similar steel defect images. Experimental results on the NEU-CLS database validate that the proposed method achieves 89.86% classification accuracy with only fine-tuning on the 1:32 training data, outperforming other general contrastive learning methods.

Deep convolutional neural network for automatic fault recognition from 3D seismic datasets

With the explosive growth in seismic data acquisition and the successful application of deep convolutional neural networks (DCNN) to various image processing tasks within multidisciplinary fields, many researchers have begun to research DCNN based automatic seismic interpretation techniques. Due to the vast number of parameters considered in deep neural networks, deep learning methods usually require a large amount of data for training. However, collecting a large number of expert interpretations is very time consuming, so related research usually uses synthetic datasets and ignores the practical problems of field datasets. In this paper, we open-source a multi-gigabyte expert-labelled field dataset in response to the challenge of accessing large-scale expert-labelled field datasets. We show that 2D fault recognition within this dataset is an image segmentation or edge detection problem in the computer vision field, that can be expressed as a pixel-level fault/non-fault binary classification. Both types of DCNNs are compared, and we propose a novel fault recognition workflow, which involves processing and screening of seismic images and labels, training DCNNs and automatic numerical evaluation. We have also demonstrated for three case study datasets that effective image augmentation methods can reduce the required labelled crosslines while maintaining satisfactory performance. Our experimental results show that our workflow not only outperforms two state-of-the-art DCNN solutions but also achieves performance comparable to humans on an expert labelled image dataset, even predicting subtle faults that an expert interpreter did not annotate. We suggest that the proposed workflow could reduce the fault interpretation life cycle from months to hours and improve the quality, and define the confidence, of fault interpretation results.

A sliding window method to identify defects in 3D printing lattice structure based on the difference principle

Structural defects are inevitably produced by residual stress in the 3D printing process, which reduce the structural-functional properties. For a typical 3D printing lattice structure, a new sliding window approach to intelligently identify defects, based on the difference principle, in the inspection process was proposed. Industrial computed tomography (CT) was used for image information acquisition. The grayscale standard deviation in the local range of the CT slice is calculated to characterize the grayscale variation of the defect location. On this basis, a sliding window method is proposed for traversing CT slices. Three different sizes of window and the corresponding calculation methods were established, and then the appropriate window was selected through comparison. After differential processing is performed on the traversed image to extract the defect feature, automatic defect recognition rules are created. The experimental results show that the recognition rate of the proposed method is 98.5% for typical internal defects of metal 3D lattice structures and the validity of the method is verified. Compared with a manual marking method, this method can effectively improve the efficiency of defect detection.

Automatic recognition of surface defects for hot-rolled steel strip based on deep attention residual convolutional neural network

Generally, the existence of surface defects in hot-rolled steel strip can lead to adverse influences on the appearance and quality of industrial products. Therefore, it is significant to timely recognize the surface defects for hot-rolled steel strip. In order to improve the efficiency and accuracy of surface defects, a deep neural network, namely, deep attention residual convolutional neural network (DARCNN), is proposed to automatically distinguish 6 kinds of hot-rolled steep strip surface defects. In this network, a channel attention mechanism is combined with residual blocks so that the network can focus on the significant feature channels without information loss. The experimental results show that the accuracy, precision and area under curve (AUC) of DARCNN reach 99.5%, 99.51% and 99.98%, respectively, and the application of DARCNN can improve the accuracy, precision and AUC for surface defect recognition tasks by 1.17%, 1.03% and 0.58%, respectively, which verifies the applicability of deep learning technologies to materials.

Automated Defect Recognition on X-ray Radiographs of Solid Propellant Using Deep Learning Based on Convolutional Neural Networks

For defense applications, rapid X-ray inspection of propellant samples is essential for the identification and assessment of defects. Automation of this process using artificial intelligence is possible by properly training a neural network model. Convolution Neural Networks (CNNs) have recently demonstrated excellent success in both the tasks of image recognition and localisation using an adequate amount of data. In real-world, it’s not an easy task to produce the correct amount of experimental data required for the deep neural network to operate. In this work, we propose a method for producing synthetic radiographic data that is supported by ray tracing based radiographic simulations for the deep learning algorithms to automatically detect anomaly in X-ray images. The simulation results, which are then supplemented by noise extracted from the experimental data, show a good comparison with the measurements. This Simulation assisted Automatic Defect Recognition (Sim-ADR) system simultaneously perform defect detection and defect instance segmentation. The accuracy of the defect detection system is more than 87% on a testing set included 416 images.

Vibration Response-Based Intelligent Non-Contact Fault Diagnosis of Bearings

Abstract Accelerometers, used as vibration pickups in machine health monitoring systems, need physical connection to the machine tool through cables, complicating physical systems. A non-contact laser based vibration sensor has been developed and used for bearing health monitoring in this article. The vibration data have been acquired under speed and load variation. Hilbert transform (HT) has been applied for denoising the vibration signal. An extraction of condition monitoring indicators from both raw and envelope signals has been made, and the dimensionality of these extracted indicators was deducted with principal component analysis (PCA). Sequential floating forward selection (SFFS) method has been implemented for ranking the selected indicators in order of significance for reduction in the input vector size and for finalizing the most optimal indicator set. Finally, the selected indicators are passed to k-nearest neighbor (kNN) and weighted kNN (WkNN) for diagnosing the bearing defects. The comparative analysis of the effectiveness of kNN and WkNN has been executed. It is evident from the experimental results that the vibration signals obtained from developed non-contact sensor compare adequately with the accelerometer data obtained under similar conditions. The performance of WkNN has been found to be slower compared to kNN. The proposed fault detection methodology compares very well with the other reported methods in the literature. The non-contact fault detection methodology has an enormous potential for automatic recognition of defects in the machine, which can provide early signals to avoid catastrophic failure and unplanned equipment shutdowns.

Fast automatic multi-defects recognition based on the spatial characteristics of Freeman chain code

Shearography is a very effective non-destructive testing method for detecting the internal defects of composite materials. How to improve the detection efficiency of this technique to help achieve the rapid and automatic recognition of multiple defects has always been a hot research topic for researchers. Based on the results of the shearography, we propose a method to identify multiple defects in the phase map using the spatial characteristics of Freeman chain code. First of all, we use Canny operator to binarize the phase map, then refine the fringe by morphological means, and finally identify the defects by the spatial characteristics chain code. In addition, Otsu is used to assist in judging the fracture or incomplete defects. For the fringes under vacuum loading or thermal loading, this method can detect the defects quickly and successfully. The theoretical description is given and the effectiveness and practicability of the method are verified by experiments.

Automatic Defect Recognition

Automatic defect recognition in industrial x-ray applications allows the ability to automate the pass / fail decisions on individual parts based on system settings programmed by a quality engineer

Adaptive depth and receptive field selection network for defect semantic segmentation on castings X-rays

In order to ensure the safety of important castings, the ADR (Automatic-Defect-Recognition) system should recognize, locate, and count the area of internal defects that are undetectable to the naked eye. However, small differences between inter-classes, large defect scale change, and uncertainly annotation limit the achievement for ADR system. To solve these challenges, this paper presents an adaptive depth and receptive field selection network for defect semantic segmentation on castings X-rays. Firstly, the Resnet18 with ADSM (adaptive depth selection mechanism) is elaborately designed to extract and adaptively aggregate the different depth features, which is beneficial to distinguish the similar defect. Then the ARFB (adaptive receptive field block) is proposed to select the optimum receptive field in a data-driven manner to adapt to the scale change of defects. To overcome the problem of inaccurate labeling caused by the ambiguity of defect edges and the subjectivity of manual annotation, we propose a data augmentation method called “lazy-label”. Finally, we set up a castings defect segmentation dataset, called SRIF-CDS, to train and evaluate our method. Experiments on this dataset indicate that our method achieves 0.86 mIoU (mean intersection-over-union) and 0.92 mAcc (mean accuracy), which has better performance than the state-of-the-art semantic segmentation baseline.

SurfNetv2: An Improved Real-Time SurfNet and Its Applications to Defect Recognition of Calcium Silicate Boards.

This paper presents an improved Convolutional Neural Network (CNN) architecture to recognize surface defects of the Calcium Silicate Board (CSB) using visual image information based on a deep learning approach. The proposed CNN architecture is inspired by the existing SurfNet architecture and is named SurfNetv2, which comprises a feature extraction module and a surface defect recognition module. The output of the system is the recognized defect category on the surface of the CSB. In the collection of the training dataset, we manually captured the defect images presented on the surface of the CSB samples. Then, we divided these defect images into four categories, which are crash, dirty, uneven, and normal. In the training stage, the proposed SurfNetv2 is trained through an end-to-end supervised learning method, so that the CNN model learns how to recognize surface defects of the CSB only through the RGB image information. Experimental results show that the proposed SurfNetv2 outperforms five state-of-the-art methods and achieves a high recognition accuracy of 99.90% and 99.75% in our private CSB dataset and the public Northeastern University (NEU) dataset, respectively. Moreover, the proposed SurfNetv2 model achieves a real-time computing speed of about 199.38 fps when processing images with a resolution of 128 × 128 pixels. Therefore, the proposed CNN model has great potential for real-time automatic surface defect recognition applications.

Combination of Convolutional and Generative Adversarial Networks for Defect Image Demoiréing of Thin-Film Transistor Liquid-Crystal Display Image

In thin-film transistor liquid crystal display (TFT-LCD) manufacturing, the automatic recognition and classification of defects can help manufacturers monitor abnormalities, identify potential process problems, and swiftly respond to these process problems. Thus, yield loss can be reduced. However, capturing the content displayed on screen using cameras is challenging because it is often contaminated with moiré patterns. Moiré patterns originate from the interference between the pixel grids of inspection camera sensors and panel screens and adversely affect the visual quality of images, causing difficulty in determining the defect type, especially mura defects. Therefore, eliminating moiré patterns in defect images without impairing image quality is critical. However, moiré patterns are often dynamic and vary with sensor resolution, distance, and direction. Moreover, the frequency distribution of moiré patterns is broad, encompassing both low- and high-frequency components. Therefore, demoiréing is more challenging than other image restoration tasks. We investigated this problem and proposed an approach to eliminate moiré patterns from defect images based on a generative adversarial network architecture by using the U-net network as a generator and adding a discriminator. Moreover, we added an attention mechanism to focus on the local consistency of the restored moiré regions. The peak-signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM) obtained using the proposed model were compared with those obtained using the U-net and pixel-to-pixel models. The experimental results of this study demonstrated that the proposed method can efficiently remove moiré patterns from defect images, and the proposed method quantitatively and qualitatively outperforms other methods.

Automatic Recognition of Weld Defects in TOFD D-Scan Images Based on Faster R-CNN

Abstract The type recognition of a weld defect in an ultrasonic Time of Flight Diffraction (TOFD) image is lack of efficiency, stability, and reliability that is due to the limitation of experience and professional knowledge of the inspector. In this study, characteristics of a weld defect of TOFD D-scan images were analyzed, and the Faster Region-based Convolutional Neural Networks (Faster R-CNN) was adopted for autorecognition of the defect’s type. In the course of the training process, the proposed box configuration was optimized to improve training and recognition efficiency, and the imaging samples were expanded for training, adjusting, and verifying the Faster R-CNN. Eventually, the recognition effect and influence factors of misrecognition were analyzed. Research results show that a D-scan image of the weld defect is closely related to the defect’s shape, which can be used to decide the type a weld defect’s type. Automatic recognition of the defect’s type based on the Faster R-CNN possesses the advantages of high recognition rate, robustness, and antijamming ability, which can achieve a recognition accuracy of 80–97 % for the weld defect’s type. In addition, the analysis of misrecognition shows that it is necessary to denoise the D-scan images before autorecognition because, in the light of the noise, stripes are liable to be misrecognized as porosity and cracks in the recognition experiments.

Metal Additive Manufacturing Parts Inspection Using Convolutional Neural Network

Metal additive manufacturing (AM) is gaining increasing attention from academia and industry due to its unique advantages compared to the traditional manufacturing process. Parts quality inspection is playing a crucial role in the AM industry, which can be adopted for product improvement. However, the traditional inspection process has relied on manual recognition, which could suffer from low efficiency and potential bias. This study presented a convolutional neural network (CNN) approach toward robust AM quality inspection, such as good quality, crack, gas porosity, and lack of fusion. To obtain the appropriate model, experiments were performed on a series of architectures. Moreover, data augmentation was adopted to deal with data scarcity. L2 regularization (weight decay) and dropout were applied to avoid overfitting. The impact of each strategy was evaluated. The final CNN model achieved an accuracy of 92.1%, and it took 8.01 milliseconds to recognize one image. The CNN model presented here can help in automatic defect recognition in the AM industry.

Automatic Recognition and Positioning of Wheel Defects in Ultrasonic B-Scan Image Using Artificial Neural Network and Image Processing

Abstract Wheels are one of the most important testing components in rail transport that play a significant role in the safety of train, and thence, research on wheel defect detection is of great significance. In this article, a method using image processing techniques and artificial neural network techniques is proposed for the purpose of recognizing defects in ultrasound B-scan image. A noise reduction and filtering algorithm and a feature extraction algorithm are proposed to simplify identification steps and improve the accuracy of the later recognition. Then, two back propagation neural networks with two hidden layers are built respectively for two classification steps. One is to identify noise, while the other is to identify the echo of real defects. Finally, a 93 % recognition rate is achieved by using the algorithm proposed in this article. The result shows that appropriate feature extraction algorithms and artificial neural network techniques are efficient and reliable in defects recognition of ultrasound B-scan images.