Class Of Defects Research Articles

Enhanced detection of unknown defect patterns on wafer bin maps based on an open-set recognition approach

It is crucial to detect and classify defect patterns on wafers in semiconductor-manufacturing processes for wafer-quality management and prompt analysis of defect causes. In recent years, continuous technological innovation and advancements in semiconductor-industry processes have led to an increase in unknown defect patterns, which must be detected and classified. However, detection of unknown defect patterns is difficult due to complex reasons, such as training on non-existent defect classes, closed datasets owing to industrial security, and labeling large volumes of manufacturing data. Owing to these challenges, methods for detecting unknown defect patterns in an actual semiconductor-manufacturing environment primarily rely on qualitative indicators, such as intuition and experience of engineers. To overcome these problems, this study proposes a methodology based on open-set recognition to accurately detect unknown defect patterns. This methodology begins with two preprocessing steps: constrained mean filtering (C-mean filtering); and Radon transform to diminish noise and efficiently extract features from wafer-bin maps. This study then develops an entropy-estimation one-class support vector machine (EEOC-SVM), which accounts for the uncertainty in the one-class SVM classification results. EEOC-SVM computes entropy-uncertainty scores based on the distance between decision boundaries and samples and then reclassifies uncertain samples using a weighted sum of uncertainties for each class. This method can effectively detect unknown defect patterns. The proposed method achieves a detection performance of over 98 % for various defect classes based on experiments conducted with new defect patterns occurring in actual semiconductor-manufacturing environments. These results confirm that the proposed method is an effective tool for detecting and addressing unknown defect patterns.

Perforated gastroduodenal ulcers: perioperative prognosis and prevention of complications

Objective. To improve the effectiveness of treatment for patients with perforated gastroduodenal ulcers (PGDU) by developing criteria for predicting postoperative complications and mortality. Materials and methods. The treatment outcomes of 127 patients with PGDU were analyzed. Prognostic scales ASA, SOFA, Peptic Ulcer Perforation Score (PULP), and the Mannheim Peritonitis Index (MPI) were used for the assessment in all the patients. Specialized classifications (DEP and ulcer defect classes) developed at Sklifosovsky Research Institute For Emergency Medicine were employed to determine the surgical approach and the extent of surgical intervention. Results. Duodenal ulcers were observed in 97(76.4 %) patients, gastric ulcers in 28(22.0 %), and combined ulcers in 2(1.6 %). According to the clinical form, chronic ulcers were identified in 81 (63.8 %) patients, while acute ones were detected in 46(36.2 %) patients. Helicobacter pylori infection was revealed in all patients with chronic ulcers. The majority of patients sought medical care within the first 6 hours after perforation (53 patients, 41.7 %); from 6 to 12 hours – 24 patients (18.9 %), 12–24 hours – 17 patients (13.4 %), and in more than 24 hours – 33 patients (26.0 %). Fatal outcome occurred in 45 (35.4 %) patients, while 82 (64.4 %) patients were discharged after the recovery had been observed. The mortality rate was significantly higher in patients with MPI grades 2 and 3, in those with an initially high risk according to the PULP scale, and those who were hospitalized more than 24 hours after perforation. Indications for laparoscopic ulcer suturing were proposed based on the scales used. Conclusions. The risk factors for adverse outcomes in PGDU include the following criteria: surgery performed more than 24 hours after perforation, MPI 2-3 grades, and a PULP scale score more than 8. In patients with PGDU who are at high surgical-anesthetic risk and have a complicated comorbid background, the risk of sepsis development is more than 30 %. Laparoscopic techniques for suturing PGDU are preferable when performed within 6 hours of perforation, with a PULP score not more than 7, MPI grade 1, and DEP classification scores less than 9, specifically in IIC, IIIC, IVA, IVB, and IVC ulcer defect classes.

Automated tablet defect detection and the prediction of disintegration time and crushing strength with deep learning based on tablet surface images

This paper presents novel measurement methods, where deep learning was used to detect tableting defects and determine the crushing strength and disintegration time of tablets on images captured by machine vision. Five different classes of defects were used and the accuracy of the real-time defect recognition performed with the deep learning algorithm YOLOv5 was 99.2 %. The system can already match the production capability of tablet presses, with still further room left for improvement. The YOLOv5 algorithm was also used to determine the disintegration time and crushing strength of tablets produced at different compression force settings based on their surface texture. With these accurate, low-cost methods, the 100 % screening of the produced tablets could be carried out, resulting in the improvement of quality control and effectiveness of pharmaceutical production.

High-Precision Defect Detection in Solar Cells Using YOLOv10 Deep Learning Model

This study presents an advanced defect detection approach for solar cells using the YOLOv10 deep learning model. Leveraging a comprehensive dataset of 10,500 solar cell images annotated with 12 distinct defect types, our model integrates Compact Inverted Blocks (CIBs) and Partial Self-Attention (PSA) modules to enhance feature extraction and classification accuracy. Training on the Viking cluster with state-of-the-art GPUs, our model achieved remarkable results, including a mean Average Precision (mAP@0.5) of 98.5%. Detailed analysis of the model’s performance revealed exceptional precision and recall rates for most defect classes, notably achieving 100% accuracy in detecting black core, corner, fragment, scratch, and short circuit defects. Even for challenging defect types such as a thick line and star crack, the model maintained high performance, with accuracies of 94% and 96%, respectively. The Recall–Confidence and Precision–Recall curves further demonstrate the model’s robustness and reliability across varying confidence thresholds. This research not only advances the state of automated defect detection in photovoltaic manufacturing but also underscores the potential of YOLOv10 for real-time applications. Our findings suggest significant implications for improving the quality control process in solar cell production. Although the model demonstrates high accuracy across most defect types, certain subtle defects, such as thick lines and star cracks, remain challenging, indicating potential areas for further optimization in future work.

High-performance surface defect detection of aluminum substrate based on event camera

Abstract Traditional industrial surface defect detection often employs CCD/CMOS cameras, but they are difficult to detect the minute defects on aluminum substrates in highly dynamic industrial scenes due to their nature. Event camera is a novel high-resolution vision sensor which measures per-pixel brightness changes in an asynchronous manner and outputs as event information flow (EIF). Small and weak defects on aluminum substrate can be captured by event camera effectively, but the EIF contain a large amount of noise, making it difficult to perform accurate and high-precision defect detection. To address this problem, we propose a frame aggregation method to realize good event information flow processing (EIFP), and then use an improved circle detection method to locate the aluminum substrate in each frame, removing abundant event information outside the aluminum substrate. Subsequently, we enhance the event signals under different frames based on optical flow tracking using multiple features, and construct a semi-supervised detector based on pseudo-labels to achieve high-precision defect localization. Finally, considering the small inter-class differences in defects on the surface of aluminum substrates, we construct a defect class corrector based on ensemble learning to enhance the ability to determine defect classes, achieving high-precision automatic quality inspection of aluminum substrate surfaces. The performance of our method is compared with other advanced methods based on event camera data of aluminum substrates in real industrial scenarios. The experimental results show that our method has improved the detection accuracy by about 10% and the classification accuracy by about 25% compared to the original state-of-the-art methods.

Polycrystalline silicon photovoltaic cell defects detection based on global context information and multi-scale feature fusion in electroluminescence images

In photovoltaic (PV) cell inspection, electroluminescence (EL) imaging provides high spatial resolution for detecting various types of defects. The recent integration of EL imaging with deep learning models has enhanced the recognition of defects in PV cells. However, the high surface impurity content in polycrystalline silicon PV cells presents a challenge. Defect features can be similar to complex background textures, making highlighting features for small target defects difficult and leading to potential misclassification as background or other classes. To address these challenges, we propose a novel deep convolutional neural network (CNN) model for effectively identifying small target defects in polycrystalline PV cells. We first utilize a global context information (GCI) block to improve CNN’s modeling of global information, aiding in distinguishing PV cell defects with similar local details. Moreover, we design a channel weight feature pyramid (CWFP) that dynamically adjusts the channel weights of extracted features. We further achieve multi-scale feature fusion through the pyramid structure to enhance the resolution capability for small targets. The proposed model was evaluated on a publicly available dataset of 8 defect classes in polycrystalline PV cells, achieving an accuracy of 96.36 %. The experimental results show that the proposed model outperforms existing deep learning models in detecting small target defects with higher precision.

A defect classification algorithm for gas tungsten arc welding process based on unsupervised learning and few-shot learning strategy

Welding defect prediction is the foundation for ensuring welding quality in gas tungsten arc welding (GTAW). In the prediction process, method based on molten pool vision is the most effective. Since the classification of molten pool defects relies on a substantial volume of labeled data, it is challenging for the models to be applied industrially. This paper presents an algorithm, FS-Classifier, that can achieve high prediction accuracy based on a limited amount of labeled data. The FS-Classifier comprises two stages: Firstly, an unsupervised training approach named RaP is designed to pre-train the feature extractor using extensive unlabeled daily datasets. The RaP consists of a rotation angle prediction task and a position prediction task, which ensure that the network focuses on salient features and precise elements, respectively. Secondly, the support vectors constructed from limited labeled data are used for the feature classifier. The input data is classified to certain class by computing its distances to support vector. The model achieves an accuracy of 94.5 % on the private dataset and 92.8 % on the public dataset for the six classes of defects using 5 % of labeled data volume. In addition, comparative experiments show that our method only requires 5 % of labeled data to achieve accuracy comparable to traditional supervised learning methods. The proposed algorithm addresses the issue of relying on a substantial amount of labeled data in welding process defect classification.

Hybrid inspection of printed board defects

Objectives. A hybrid approach to the problem of searching and classifying defects in printed circuit boards (PCB) is proposed. Key factors and trends in the design and production of PCBs are considered. The relevance of the study is determined by the use of new materials and production technologies.Methods. A hybrid approach based on the algorithm of comparison with reference and the use of the YOLO family of neural network models for detecting objects is used to solve the problem.Results. Models were trained on public sets of PCB images with six classes of defects, and their accuracy was assessed using generally accepted metrics.Conclusion. Experiments have shown that the YOLOv8 neural network architecture has high accuracy of defect detection, low sensitivity to image quality, presence of text and graphic objects on the PCB, but the low quality of training datasets imposes restrictions on the use of only neural networks for defect detection. It is proposed to use a hybrid approach to improve the quality of defect inspection by applying different methods depending on the quality assessment of the analyzed images.

Automatic Classification of Defective Photovoltaic Module Cells Based on a Novel CNN-PCA-SVM Deep Hybrid Model in Electroluminescence Images

In today’s world, the rapid development of photovoltaic (PV) power plants has facilitated sustainable energy production. Maintenance and fault detection play a crucial role in ensuring the continuity of energy production. Manual inspection of electroluminescence (EL) images of PV modules requires significant human resources and time investment. This study presents a method for automatic fault detection of PV cells in EL images using hybrid deep features optimized with the principal component analysis (PCA) feature selection algorithm. A lightweight and high-performance model that combines the strengths of convolutional neural network (CNN) architectures is proposed. Firstly, data augmentation techniques are employed due to the imbalance between defective and functional classes in the dataset containing EL images. In experimental studies conducted by integrating the PCA algorithm into MobileNetV2, DenseNet201, and InceptionV3 CNN models, accuracy, precision, recall, and F1-score values of 92.19%, 92%, 90% and 91%, respectively, were achieved. When the results were analyzed, it was observed that the proposed method had an effective performance in detecting faults in PV panel cells.

First-principles investigation of near-field energy transfer between localized quantum emitters in solids

We present a predictive and general approach to investigate near-field energy transfer processes between localized defects in semiconductors, which couples first-principles electronic structure calculations and a nonrelativistic quantum electrodynamics description of photons in the weak-coupling regime. The approach is general and can be readily applied to investigate broad classes of defects in solids. We apply our approach to investigate an exemplar point defect in an oxide, the F center in MgO, and we show that the energy transfer from a magnetic source, e.g., a rare-earth impurity, to the vacancy can lead to spin nonconserving long-lived excitations that are dominant processes in the near field, at distances relevant to the design of photonic devices and ultrahigh dense memories. We also define a descriptor for coherent energy transfer to predict geometrical configurations of emitters to enable long-lived excitations, that are useful to design optical memories in semiconductor and insulators. Published by the American Physical Society 2024

Fabric Defect Detection in Real World Manufacturing Using Deep Learning

Defect detection is very important for guaranteeing the quality and pricing of fabric. A considerable amount of fabric is discarded as waste because of defects, leading to substantial annual losses. While manual inspection has traditionally been the norm for detection, adopting an automatic defect detection scheme based on a deep learning model offers a timely and efficient solution for assessing fabric quality. In real-time manufacturing scenarios, datasets lack high-quality, precisely positioned images. Moreover, both plain and printed fabrics are being manufactured in industries simultaneously; therefore, a single model should be capable of detecting defects in all kinds of fabric. So training a robust deep learning model that detects defects in fabric datasets generated during production with high accuracy and lower computational costs is required. This study uses an indigenous dataset directly sourced from Chenab Textiles, providing authentic and diverse images representative of actual manufacturing conditions. The dataset is used to train a computationally faster but lighter state-of-the-art network, i.e., YOLOv8. For comparison, YOLOv5 and MobileNetV2-SSD FPN-Lite models are also trained on the same dataset. YOLOv8n achieved the highest performance, with a mAP of 84.8%, precision of 0.818, and recall of 0.839 across seven different defect classes.

A multi-scale attention mechanism for detecting defects in leather fabrics

Defect detection is critical to industrial quality control in leather production engineering. The various sizes and locations of defects in leather, as well as significant differences within the same class and indistinctive variations between different classes of defects, contribute to the complexity of the problem. To address this challenge, we propose a Multi-Layer Residual Convolutional Attention (MLRCA) approach. MLRCA enhances its ability to capture both intra-class and inter-class differences by enhancing the semantic feature representation in the backbone network. To improve multiscale fusion effects, we also incorporate the MLRCA module into the feature pyramid network (FPN) and propose a new multi-layer residual convolution attention feature pyramid network (ML-FPN). This approach enables more accurate identification of leather defects at a more detailed level by selectively capturing contextual information from different domains. We then implement the Side-Aware Boundary Localization (SABL) detection head, which accurately locates defects and helps the network distinguish between similar defect categories for more precise positioning. To validate the effectiveness of our approach, we conducted ablation experiments on the created leather dataset. Comparative experiments demonstrate the excellent capability of our model to detect minor defects. The model achieved 83.4, 89.7, and 85.6 for the AP, AP50, and AP75 evaluation metrics. In addition, the model achieves 71.3, 89.9, and 88.9 for APS, APM, and APL. Our approach has been confirmed feasible through experimentation and provides new insights for automated leather defect detection methods.

Data Complexity: A New Perspective for Analyzing the Difficulty of Defect Prediction Tasks

Defect prediction is crucial for software quality assurance and has been extensively researched over recent decades. However, prior studies rarely focus on data complexity in defect prediction tasks, and even less on understanding the difficulties of these tasks from the perspective of data complexity. In this article, we conduct an empirical study to estimate the hardness of over 33,000 instances, employing a set of measures to characterize the inherent difficulty of instances and the characteristics of defect datasets. Our findings indicate that: (1) instance hardness in both classes displays a right-skewed distribution, with the defective class exhibiting a more scattered distribution; (2) class overlap is the primary factor influencing instance hardness and can be characterized through feature, structural, and instance-level overlap; (3) no universal preprocessing technique is applicable to all datasets, and it may not consistently reduce data complexity, fortunately, dataset complexity measures can help identify suitable techniques for specific datasets; (4) integrating data complexity information into the learning process can enhance an algorithm’s learning capacity. In summary, this empirical study highlights the crucial role of data complexity in defect prediction tasks, and provides a novel perspective for advancing research in defect prediction techniques.

An adaptive detection approach for multi-scale defects on wind turbine blade surface

Multi-scale defects will inevitably appear in wind turbine blade defect detection within practical detection applications. An adaptive multi-scale detection approach is proposed to accurately classify and locate defects on the wind turbine blade surface. The proposed approach includes two main detection procedures: rough detection and precise detection. In the rough detection, the multi-level features extraction module with an adaptive bounding box proposal module is used to depict multi-scale defect regions and train a binary classifier to distinguish defects from non-defects. At the precise detection stage, three defect categories and four coordinates representing defect locations are obtained based on a multi-class defect classifier and regression of rough location boxes. The proposed method is evaluated on a real wind turbine blade surface defect dataset collected in a commercial wind farm and annotated manually. Results show that (1) the proposed model can detect the class of the blade multi-scale defects and outperforms other schemes with 96.89% mAP in the same model training epochs, (2) the positioning performance analysis of the model for multi-scale defects is conducted to validate the accuracy of the proposed model for multi-scale defect location.

DeepSEM-Net: Enhancing SEM defect analysis in semiconductor manufacturing with a dual-branch CNN-Transformer architecture

Due to the increasing complexity of modern semiconductor integrated circuit fabrication processes, various defects may occur at each process step, leading to a semiconductor integrated circuit yield loss as well as reliability problems. This study introduces DeepSEM-Net, a novel dual-branch architecture that harmoniously integrates Convolutional Neural Networks (CNN) with Transformers. This innovative approach significantly enhances defect classification and precise segmentation in Scanning Electron Microscope (SEM) defect images, crucial for semiconductor manufacturing processes. DeepSEM-Net consists of a classification branch, which synergizes CNN and Transformers for robust global feature extraction, and a segmentation branch, which incorporates a Pyramid Attention module to meticulously recalibrate and exploit the feature map, thereby enriching the segmentation quality. A semi-supervised defect analysis system based on this model markedly reduces manual inspection efforts and is adept at operating in both simultaneous and sequential classification-segmentation modes. Rigorously tested on a diverse dataset from a real 12-inch wafer fab, DeepSEM-Net not only demonstrated a commendable classification accuracy of 97.25% on a 5-class dataset and a segmentation IoU of 84.40% but also exhibited remarkable resilience against the introduction of new defect classes in an unbalanced dataset. The dual-branch strategy of DeepSEM-Net, thus, significantly elevates the efficiency and reliability of SEM defect image analysis, offering a substantial leap forward in addressing yield loss and ensuring the reliability of semiconductor products.

Topological disclination mode in photonic Chern insulators

Topological defects in topological materials offer novel routes for creating topological modes and probing bulk topology. Disclination, a class of topological defects, has been recently shown to host fractional charges in topological crystalline insulators with well-defined Wannier centers. Here, we study the effects of disclinations in gyromagnetic photonic crystals with non-zero Chern numbers that prohibit the Wannier center picture. We find the emergence of topological disclination modes carrying orbital angular momentum from the interplay between the Chern-type topology and the effective flux induced by the disclination. When the Chern number changes its sign, the chirality of the disclination mode also flips, revealing the bulk-disclination correspondence. Furthermore, we perform numerical experiments to probe the disclination mode. Our results expand the study of disclination physics in photonic crystals to time-reversal-broken systems.

The Rac pathway prevents cell fragmentation in a nonprotrusively migrating leader cell during C. elegans gonad organogenesis

The C.elegans hermaphrodite distal tip cell (DTC) leads gonadogenesis. Loss-of-function mutations in a C.elegans ortholog of the Rac1 GTPase (ced-10) and its GEF complex (ced-5/DOCK180, ced-2/CrkII, ced-12/ELMO) cause gonad migration defects related to directional sensing; we discovered an additional defect class of gonad bifurcation in these mutants. Using genetic approaches, tissue-specific and whole-body RNAi, and invivo imaging of endogenously tagged proteins and marked cells, we find that loss of Rac1 or its regulators causes the DTC to fragment as it migrates. Both products of fragmentation-the now-smaller DTC and the membranous patch of cellular material-localize important stem cell niche signaling (LAG-2 ligand) and migration (INA-1/integrin subunit alpha) factors to their membranes, but only one retains the DTC nucleus and therefore the ability to maintain gene expression over time. The enucleate patch can lead a bifurcating branch off the gonad arm that grows through germ cell proliferation. Germ cells in this branch differentiate as the patch loses LAG-2 expression. While the nucleus is surprisingly dispensable for aspects of leader cell function, it is required for stem cell niche activity long term. Prior work found that Rac1-/-;Rac2-/- mouse erythrocytes fragment; in this context, our new findings support the conclusion that maintaining a cohesive but deformable cell is a conserved function of this important cytoskeletal regulator.

AdvancingTire Safety: Explainable Artificial Intelligence-Powered Foreign Object Defect Detection with Xception Networks and Grad-CAM Interpretation

Automatic detection of tire defects has become an important issue for tire production companies since these defects cause road accidents and loss of human lives. Defects in the inner structure of the tire cannot be detected with the naked eye; thus, a radiographic image of the tire is gathered using X-ray cameras. This image is then examined by a quality control operator, and a decision is made on whether it is a defective tire or not. Among all defect types, the foreign object type is the most common and may occur anywhere in the tire. This study proposes an explainable deep learning model based on Xception and Grad-CAM approaches. This model was fine-tuned and trained on a novel real tire dataset consisting of 2303 defective tires and 49,198 non-defective. The defective tire class was augmented using a custom augmentation technique to solve the imbalance problem of the dataset. Experimental results show that the proposed model detects foreign objects with an accuracy of 99.19%, recall of 98.75%, precision of 99.34%, and f-score of 99.05%. This study provided a clear advantage over similar literature studies.

A new ViT-Based augmentation framework for wafer map defect classification to enhance the resilience of semiconductor supply chains

Wafer map defect classification plays a crucial role in sustaining the semiconductor supply chain during industrial disruptions by ensuring continuity, resilience, and efficiency while aligning with sustainability principles. Up to 30 percent of production costs is lost to chip testing and yield losses for semiconductor manufacturing supply chain. In the semiconductor practice, wafer map defects recognition plays a linchpin role in the front-end-of-line stage. Defect pattern recognition can directly pinpoint the assignable causes and provide the domain experts with actionable insights. However, various wafer defect types occur differently from each other, which makes the collected wafer map dataset to be highly imbalanced in defect classes. In the face of data imbalance, the classification model usually cannot provide satisfactory classification performance. In this aspect, this paper intends to investigate the wafer defect map classification problem by using Vision Transformer (ViT) as an alternative data augmentation approach. The primary purpose is to alleviate the class-imbalance issue and then the performance of a deep learning-based convolutional neural network for wafer defect classification can be effectively improved. The experimental results demonstrate that the proposed data augmentation method by using ViT proves to be a potential generative model for improving wafer map defect classification, particularly robust on the individual minority class. In a word, the proposed augmentation framework for wafer map defect classification facilitates a more targeted and efficient approach to quality control, resource utilization, and production optimization in maintaining a sustainable semiconductor supply chain, particularly in times of industrial disruption.

Acoustic emission signatures for quantifying damage patterns in half grouted sleeve connection under tensile load

The half-grouted sleeve connection (HGSC) have been widely used as connection tools for prefabricated concrete (PC) buildings. It is crucial to maintain the safety and stability of the nodes and detecting internal defects in the grouting sleeve for the safety of the entire structure. Therefore, understanding the mechanical behavior and damage mechanism of HGSC is essential for the design and construction of PC. Currently, there are few on-site techniques applicable for monitoring the damage progress. The acoustic emission (AE) as a passive method shows great potential to characterize the damage in the HGSC. Therefore, in this paper, we propose to use AE to quantify the damage of the HGSC with defects under the tensile load. The artificial defects were introduced into HGSC with different defective rates. The monotonic tensile load was applied to each HGSC sample with a loading rate of 0.33 mm/s and tested to the failure. In this process, the AE signals were collected by AE transducers. For the classification and prediction of damage signals, we utilized a machine learning framework tailored for HGSC. This framework integrates unsupervised k-means++ cluster analysis and supervised K-Nearest Neighbors (K-NN) for damage classification and prediction. The experimental results showed that four AE signal clusters can be defined during the progress of tensile failure:(i) microcrack expansion;(ii) the friction signals of rebar and high-strength grouts; (iii)the generation of microcracks; (iv) plastic deformation of rebar. The corresponding AE signatures were identified. The K-NN model trained on the training data provides a high accuracy in predicting different class of defects. It was concluded that the proposed machine learning framework is a promising and powerful tool for realizing the on-site damage detection and prediction for HGSC.