Defect Density and Atomic Defect Recognition in the Middle Layer of a Trilayer MoS2 Stack.

Molybdenum disulfide (MoS2) is one of the most intriguing two-dimensional materials, and moreover, its single atomic defects can significantly alter the properties. These defects can be both imaged and engineered using spherical and chromatic aberration-corrected high-resolution transmission electron microscopy (CC/CS-corrected HRTEM). In a few-layer stack, several atoms are vertically aligned in one atomic column. Therefore, it is challenging to determine the positions of missing atoms and the damage cross-section, particularly in the not directly accessible middle layers. In this study, we introduce a technique for extracting subtle intensity differences in CC/CS-corrected HRTEM images. By exploiting the crystal structure of the material, our method discerns chalcogen vacancies even in the middle layer of trilayer MoS2. We found that in trilayer MoS2 the middle layer's damage cross-section is about ten times lower than that in the monolayer. Our findings could be essential for the application of few-layer MoS2 in nanodevices.

Automated Defect Recognition for Additive Manufactured Parts Using Machine Perception and Visual Saliency.

Metal additive manufacturing (AM) is known to produce internal defects that can impact performance. As the technology becomes more mainstream, there is a growing need to establish nondestructive inspection technologies that can assess and quantify build quality with high confidence. This article presents a complete, three-dimensional (3D) solution for automated defect recognition in AM parts using X-ray computed tomography (CT) scans. The algorithm uses a machine perception framework to automatically separate visually salient regions, that is, anomalous voxels, from the CT background. Compared with supervised approaches, the proposed concept relies solely on visual cues in 3D similar to those used by human operators in two-dimensional (2D) assuming no a priori information about defect appearance, size, and/or shape. To ingest any arbitrary part geometry, a binary mask is generated using statistical measures that separate lighter, material voxels from darker, background voxels. Therefore, no additional part or scan information, such as CAD files, STL models, or laser scan vector data, is needed. Visual saliency is established using multiscale, symmetric, and separable 3D convolution kernels. Separability of the convolution kernels is paramount when processing CT scans with potentially billions of voxels because it allows for parallel processing and thus faster execution of the convolution operation in single dimensions. Based on the CT scan resolution, kernel sizes may be adjusted to identify defects of different sizes. All adjacent anomalous voxels are subsequently merged to form defect clusters, which in turn reveals additional information regarding defect size, morphology, and orientation to the user, information that may be linked to mechanical properties, such as fatigue response. The algorithm was implemented in MATLAB™ using hardware acceleration, that is, graphics processing unit support, and tested on CT scans of AM components available at the Center for Innovative Materials Processing through Direct Digital Deposition (CIMP-3D) at Penn State's Applied Research Laboratory. Initial results show adequate processing times of just a few minutes and very low false-positive rates, especially when addressing highly salient and larger defects. All developed analytic tools can be simplified to accommodate 2D images.

Deep Learning Neural Network Performance on NDT Digital X-ray Radiography Images: Analyzing the Impact of Image Quality Parameters-An Experimental Study.

In response to the growing inspection demand exerted by process automation in component manufacturing, non-destructive testing (NDT) continues to explore automated approaches that utilize deep-learning algorithms for defect identification, including within digital X-ray radiography images. This necessitates a thorough understanding of the implication of image quality parameters on the performance of these deep-learning models. This study investigated the influence of two image-quality parameters, namely signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), on the performance of a U-net deep-learning semantic segmentation model. Input images were acquired with varying combinations of exposure factors, such as kilovoltage, milli-ampere, and exposure time, which altered the resultant radiographic image quality. The data were sorted into five different datasets according to their measured SNR and CNR values. The deep-learning model was trained five distinct times, utilizing a unique dataset for each training session. Training the model with high CNR values yielded an intersection-over-union (IoU) metric of 0.9594 on test data of the same category but dropped to 0.5875 when tested on lower CNR test data. The result of this study emphasizes the importance of achieving a balance in training dataset according to the investigated quality parameters in order to enhance the performance of deep-learning segmentation models for NDT digital X-ray radiography applications.

3D Scanner-Based Identification of Welding Defects-Clustering the Results of Point Cloud Alignment.

This paper describes a framework for detecting welding errors using 3D scanner data. The proposed approach employs density-based clustering to compare point clouds and identify deviations. The discovered clusters are then classified according to standard welding fault classes. Six welding deviations defined in the ISO 5817:2014 standard were evaluated. All defects were represented through CAD models, and the method was able to detect five of these deviations. The results demonstrate that the errors can be effectively identified and grouped according to the location of the different points in the error clusters. However, the method cannot separate crack-related defects as a distinct cluster.

A Robust Deep Learning Ensemble-Driven Model for Defect and Non-Defect Recognition and Classification Using a Weighted Averaging Sequence-Based Meta-Learning Ensembler.

The need to overcome the challenges of visual inspections conducted by domain experts drives the recent surge in visual inspection research. Typical manual industrial data analysis and inspection for defects conducted by trained personnel are expensive, time-consuming, and characterized by mistakes. Thus, an efficient intelligent-driven model is needed to eliminate or minimize the challenges of defect identification and elimination in processes to the barest minimum. This paper presents a robust method for recognizing and classifying defects in industrial products using a deep-learning architectural ensemble approach integrated with a weighted sequence meta-learning unification framework. In the proposed method, a unique base model is constructed and fused together with other co-learning pretrained models using a sequence-driven meta-learning ensembler that aggregates the best features learned from the various contributing models for better and superior performance. During experimentation in the study, different publicly available industrial product datasets consisting of the defect and non-defect samples were used to train, validate, and test the introduced model, with remarkable results obtained that demonstrate the viability of the proposed method in tackling the challenges of the manual visual inspection approach.

A Hard Voting Policy-Driven Deep Learning Architectural Ensemble Strategy for Industrial Products Defect Recognition and Classification.

Manual or traditional industrial product inspection and defect-recognition models have some limitations, including process complexity, time-consuming, error-prone, and expensiveness. These issues negatively impact the quality control processes. Therefore, an efficient, rapid, and intelligent model is required to improve industrial products' production fault recognition and classification for optimal visual inspections and quality control. However, intelligent models obtained with a tradeoff of high accuracy for high latency are tedious for real-time implementation and inferencing. This work proposes an ensemble deep-leaning architectural framework based on a deep learning model architectural voting policy to compute and learn the hierarchical and high-level features in industrial artefacts. The voting policy is formulated with respect to three crucial viable model characteristics: model optimality, efficiency, and performance accuracy. In the study, three publicly available industrial produce datasets were used for the proposed model's various experiments and validation process, with remarkable results recorded, demonstrating a significant increase in fault recognition and classification performance in industrial products. In the study, three publicly available industrial produce datasets were used for the proposed model's various experiments and validation process, with remarkable results recorded, demonstrating a significant increase in fault recognition and classification performance in industrial products.

A New Bolt Defect Identification Method Incorporating Attention Mechanism and Wide Residual Networks.

Bolts are important components on transmission lines, and the timely detection and exclusion of their abnormal conditions are imperative to ensure the stable operation of transmission lines. To accurately identify bolt defects, we propose a bolt defect identification method incorporating an attention mechanism and wide residual networks. Firstly, the spatial dimension of the feature map is compressed by the spatial compression network to obtain the global features of the channel dimension and enhance the attention of the network to the vital information in a weighted way. After that, the enhanced feature map is decomposed into two one-dimensional feature vectors by embedding a cooperative attention mechanism to establish long-term dependencies in one spatial direction and preserve precise location information in the other direction. During this process, the prior knowledge of the bolts is utilized to help the network extract critical feature information more accurately, thus improving the accuracy of recognition. The test results show that the bolt recognition accuracy of this method is improved to 94.57% compared with that before embedding the attention mechanism, which verifies the validity of the proposed method.

Defects Recognition Algorithm Development from Visual UAV Inspections.

Aircraft maintenance plays a key role in the safety of air transport. One of its most significant procedures is the visual inspection of the aircraft skin for defects. This is mainly carried out manually and involves a high skilled human walking around the aircraft. It is very time consuming, costly, stressful and the outcome heavily depends on the skills of the inspector. In this paper, we propose a two-step process for automating the defect recognition and classification from visual images. The visual inspection can be carried out with the use of an unmanned aerial vehicle (UAV) carrying an image sensor to fully automate the procedure and eliminate any human error. With our proposed method in the first step, we perform the crucial part of recognizing the defect. If a defect is found, the image is fed to an ensemble of classifiers for identifying the type. The classifiers are a combination of different pretrained convolution neural network (CNN) models, which we retrained to fit our problem. For achieving our goal, we created our own dataset with defect images captured from aircrafts during inspection in TUI's maintenance hangar. The images were preprocessed and used to train different pretrained CNNs with the use of transfer learning. We performed an initial training of 40 different CNN architectures to choose the ones that best fitted our dataset. Then, we chose the best four for fine tuning and further testing. For the first step of defect recognition, the DenseNet201 CNN architecture performed better, with an overall accuracy of 81.82%. For the second step for the defect classification, an ensemble of different CNN models was used. The results show that even with a very small dataset, we can reach an accuracy of around 82% in the defect recognition and even 100% for the classification of the categories of missing or damaged exterior paint and primer and dents.

Investigation on SMT Product Defect Recognition Based on Multi-Source and Multi-Dimensional Data Reconstruction.

The recognition of defects in the solder paste printing process significantly influences the surface-mounted technology (SMT) production quality. However, defect recognition via inspection by a machine has poor accuracy, resulting in a need for the manual rechecking of many defects and a high production cost. In this study, we investigated SMT product defect recognition based on multi-source and multi-dimensional data reconstruction for the SMT production quality control process in order to address this issue. Firstly, the correlation between features and defects was enhanced by feature interaction, selection, and conversion. Then, a defect recognition model for the solder paste printing process was constructed based on feature reconstruction. Finally, the proposed model was validated on a SMT production dataset and compared with other methods. The results show that the accuracy of the proposed defect recognition model is 96.97%. Compared with four other methods, the proposed defect recognition model has higher accuracy and provides a new approach to improving the defect recognition rate in the SMT production quality control process.

Identifying defective solar cells in electroluminescence images using deep feature representations.

Electroluminescence (EL) imaging is a technique for acquiring images of photovoltaic (PV) modules and examining them for surface defects. Analysis of EL images has been manually performed by visual inspection of images by experts. This manual procedure is tedious, time-consuming, subjective, and requires deep expert knowledge. In this work, a hybrid and fully-automated classification system is developed for detecting different types of defects in EL images. The system fuses the deep feature representations extracted from two different deep learning models (Inception-V3 and ResNet50) to form more discriminative feature vectors. These feature vectors are then fed into the classifier layer to assign them into one of different types of defects. A large-scale, challenging solar cells dataset composed of 2,624 EL images was used to assess the performance of the proposed system in both the binary classification (functional vs defective) task and multi-class classification (functional, mild, moderate, and severe) task. The proposed system has managed to detect the correct defect type with less than 1 s per image with an accuracy rate of 98.15% and 95.35% in the binary classification and multi-classification task, respectively.

An Intelligence Method for Recognizing Multiple Defects in Rail.

Ultrasonic guided waves are sensitive to many different types of defects and have been studied for defect recognition in rail. However, most fault recognition algorithms need to extract features from the time domain, frequency domain, or time-frequency domain based on experience or professional knowledge. This paper proposes a new method for identifying many different types of rail defects. The segment principal components analysis (S-PCA) is developed to extract characteristics from signals collected by sensors located at different positions. Then, the Support Vector Machine (SVM) model is used to identify different defects depending on the features extracted. Combining simulations and experiments of the rails with different kinds of defects are established to verify the effectiveness of the proposed defect identification techniques, such as crack, corrosion, and transverse crack under the shelling. There are nine channels of the excitation-reception to acquire guided wave detection signals. The results show that the defect classification accuracy rates are 96.29% and 96.15% for combining multiple signals, such as the method of single-point excitation and multi-point reception, or the method of multi-point excitation and reception at a single point.

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.

The Combination of Adaptive Convolutional Neural Network and Bag of Visual Words in Automatic Diagnosis of Third Molar Complications on Dental X-Ray Images.

In dental diagnosis, recognizing tooth complications quickly from radiology (e.g., X-rays) takes highly experienced medical professionals. By using object detection models and algorithms, this work is much easier and needs less experienced medical practitioners to clear their doubts while diagnosing a medical case. In this paper, we propose a dental defect recognition model by the integration of Adaptive Convolution Neural Network and Bag of Visual Word (BoVW). In this model, BoVW is used to save the features extracted from images. After that, a designed Convolutional Neural Network (CNN) model is used to make quality prediction. To evaluate the proposed model, we collected a dataset of radiography images of 447 patients in Hanoi Medical Hospital, Vietnam, with third molar complications. The results of the model suggest accuracy of 84% ± 4%. This accuracy is comparable to that of experienced dentists and radiologists.

Highly sensitive photoelectrochemical biosensor for kinase activity detection and inhibition based on the surface defect recognition and multiple signal amplification of metal-organic frameworks.

A turn-on photoelectrochemical (PEC) biosensor based on the surface defect recognition and multiple signal amplification of metal-organic frameworks (MOFs) was proposed for highly sensitive protein kinase activity analysis and inhibitor evaluation. In this strategy, based on the phosphorylation reaction in the presence of protein kinase A (PKA), the Zr-based metal-organic frameworks (UiO-66) accommodated with [Ru(bpy)3]2+ photoactive dyes in the pores were linked to the phosphorylated kemptide modified TiO2/ITO electrode through the chelation between the Zr4+ defects on the surface of UiO-66 and the phosphate groups in kemptide. Under visible light irradiation, the excited electrons from [Ru(bpy)3]2+ adsorbed in the pores of UiO-66 injected into the TiO2 conduction band to generate photocurrent, which could be utilized for protein kinase activities detection. The large surface area and high porosities of UiO-66 facilitated a large number of [Ru(bpy)3]2+ that increased the photocurrent significantly, and afforded a highly sensitive PEC analysis of kinase activity. The detection limit of the as-proposed PEC biosensor was 0.0049UmL-1 (S/N!=!3). The biosensor was also applied for quantitative kinase inhibitor evaluation and PKA activities detection in MCF-7 cell lysates. The developed visible-light PEC biosensor provides a simple detection procedure and a cost-effective manner for PKA activity assays, and shows great potential in clinical diagnosis and drug discoveries.