Defect Detection Approach Research Articles

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.

Automatic Detection of Concrete Surface Defects Using PRE-TRAINED CNN and Laser Ultrasonic Visualization Testing

In recent years, nondestructive testing for civil engineering structures has become increasingly important. Ultrasonic testing is one of nondestructive inspection methods for civil structures. However, the inspection of civil engineering structures takes much time because of the extensive scope of the inspection. Moreover, in the field of nondestructive testing, there are also concerns about a future shortage of inspectors, so that an innovative effective nondestructive method needs to be developed. This study proposes an automatic defect detection approach using pre-trained convolutional neural network for laser ultrasonic visualization testing. The effectiveness of the proposed method is confirmed by applying it to a concrete structure with a surface defect. Grad-CAM demonstrates that the created CNN model in this study accurately predicts the position of a surface defect of concrete specimens.

VQGNet: An Unsupervised Defect Detection Approach for Complex Textured Steel Surfaces

Defect detection on steel surfaces with complex textures is a critical and challenging task in the industry. The limited number of defect samples and the complexity of the annotation process pose significant challenges. Moreover, performing defect segmentation based on accurate identification further increases the task’s difficulty. To address this issue, we propose VQGNet, an unsupervised algorithm that can precisely recognize and segment defects simultaneously. A feature fusion method based on aggregated attention and a classification-aided module is proposed to segment defects by integrating different features in the original images and the anomaly maps, which direct the attention to the anomalous information instead of the irregular complex texture. The anomaly maps are generated more confidently using strategies for multi-scale feature fusion and neighbor feature aggregation. Moreover, an anomaly generation method suitable for grayscale images is introduced to facilitate the model’s learning on the anomalous samples. The refined anomaly maps and fused features are both input into the classification-aided module for the final classification and segmentation. VQGNet achieves state-of-the-art (SOTA) performance on the industrial steel dataset, with an I-AUROC of 99.6%, I-F1 of 98.8%, P-AUROC of 97.0%, and P-F1 of 80.3%. Additionally, ViT-Query demonstrates robust generalization capabilities in generating anomaly maps based on the Kolektor Surface-Defect Dataset 2.

Application of the Semi-Supervised Learning Approach for Pavement Defect Detection.

Road surface quality is essential for driver comfort and safety, making it crucial to monitor pavement conditions and detect defects in real time. However, the diversity of defects and the complexity of ambient conditions make it challenging to develop an effective and robust classification and detection algorithm. In this study, we adopted a semi-supervised learning approach to train ResNet-18 for image feature retrieval and then classification and detection of pavement defects. The resulting feature embedding vectors from image patches were retrieved, concatenated, and randomly sampled to model a multivariate normal distribution based on the only one-class training pavement image dataset. The calibration pavement image dataset was used to determine the defect score threshold based on the receiver operating characteristic curve, with the Mahalanobis distance employed as a metric to evaluate differences between normal and defect pavement images. Finally, a heatmap derived from the defect score map for the testing dataset was overlaid on the original pavement images to provide insight into the network's decisions and guide measures to improve its performance. The results demonstrate that the model's classification accuracy improved from 0.868 to 0.887 using the expanded and augmented pavement image data based on the analysis of heatmaps.

Rotating machinery early fault detection integrating variational mode decomposition and multiscale singular value decomposition

Abstract Security and reliability are important issues that must be paid attention to during the operation of rotating machinery. If defects can be found in the early stage, there will be enough time to take maintenance measures and realize the stable operation of equipment. However, the presence of noise, shaft rotation signals, gear meshing signals, and other interfering factors often obfuscate fault signals, rendering the early detection of defects an arduous undertaking. Against this backdrop, this study presents an advanced approach for early defect detection, integrating the virtues of variational mode decomposition (VMD) and multiscale singular value decomposition (MSVD). Initially, a novel evaluation index is constructed by combining envelope entropy and envelope spectrum sparsity. Based on this a method is proposed to adaptively determine the critical parameters of VMD, enabling the adaptive decomposition of vibration signals into a series of modal components. The optimal sensitive components are then discerned utilizing the characteristic frequency intensity coefficient index. Subsequently, to address the limitations of single VMD methods in effectively suppressing low-frequency noise, the MSVD method is proposed for effective noise reduction, which reconstructs the signal after SVD of the signal within each segment through the operation of successive signal segmentation. Ultimately, envelope spectrum analysis is conducted on the reconstructed signal, facilitating the precise extraction of fault characteristic frequency information and enabling early fault identification. The efficacy of this novel methodology is evaluated through simulations and actual vibration signals, successfully discerning early faults afflicting rotating machinery.

A deep learning approach for automated PCB defect detection: A comprehensive review

The ever-expanding market for electronic devices has significantly heightened the demand for high-quality printed circuit boards (PCBs). Even minor defects in PCBs can pose substantial safety risks for end-users. This article provides a comprehensive review on deep learning-based approaches for PCB defect detection. Our exploration covers various critical aspects, including the classification of PCB defects, automated vision inspection (AVI) techniques, object detection methodologies, and the widespread adoption of deep learning models. Specifically, we focus on the state-of-the-art approach known as region-based Fully Convolutional Network with Feature Pyramid Networks (FPN-RFCN). Additionally, we discuss effective data augmentation techniques and commonly used evaluation metrics in this domain. This review provides valuable insights for researchers, practitioners, and industry professionals engaged in PCB quality assurance.

A two-stage defect detection method for unevenly illuminated self-adhesive printed materials

The process of printing defect detection usually suffers from challenges such as inaccurate defect extraction and localization, caused by uneven illumination and complex textures. Moreover, image difference-based defect detection methods often result in numerous small-scale pseudo defects. To address these challenges, this paper proposes a comprehensive defect detection approach that integrates brightness correction and a two-stage defect detection strategy for self-adhesive printed materials. Concretely, a joint bilateral filter coupled with brightness correction corrects uneven brightness properly, meanwhile smoothing the grid-like texture in complex printed material images. Then, in the first detection stage, an image difference method based on a bright–dark difference template group is designed to effectively locate printing defects despite slight brightness fluctuations. Afterward, a discriminative method based on feature similarity is employed to filter out small-scale pseudo-defects in the second detection stage. The experimental results show that the improved difference method achieves an average precision of 99.1% in defect localization on five different printing pattern samples. Furthermore, the second stage reduces the false detection rate to under 0.5% while maintaining the low missed rate.

Defect detection with ego-noise reduction based on multimodal information in UAV hammering inspection

In this paper, we introduce a novel approach for defect detection in hammering inspections using Unmanned Aerial Vehicles (UAVs). Despite the promising application of UAVs for inspecting hard-to-reach structures, like bridges, their efficiency is often compromised by the significant ego noise produced by their motor-propeller systems. This noise complicates the discrimination between healthy and defective hammering sounds. In previous research, methods to improve robustness through supervised learning have been proposed; however, these methods require the labeling of hammering sounds by skilled inspectors to train the discrimination model. To overcome this problem, we propose an ego-noise reduction method based on propeller vibrations. By reducing ego noise and thereby making the characteristics of hammering sound more dramatically clear, we enable unsupervised defect detection amidst ego noise. Our experiments with concrete specimens demonstrate that our technique achieves defect detection with an accuracy on par with the supervised method. The proposed method proves especially beneficial for hammering inspections, in which the domain gap–the variability in acoustic signatures of hammering sounds caused by differences in concrete mix ratios and curing conditions from one site to another–presents a significant challenge. Our approach effectively adapts to these variations, ensuring reliable defect detection across diverse construction environments.

Local and Global Context-Enhanced Lightweight CenterNet for PCB Surface Defect Detection.

Printed circuit board (PCB) surface defect detection is an essential part of the PCB manufacturing process. Currently, advanced CCD or CMOS sensors can capture high-resolution PCB images. However, the existing computer vision approaches for PCB surface defect detection require high computing effort, leading to insufficient efficiency. To this end, this article proposes a local and global context-enhanced lightweight CenterNet (LGCL-CenterNet) to detect PCB surface defects in real time. Specifically, we propose a two-branch lightweight vision transformer module with local and global attention, named LGT, as a complement to extract high-dimension features and leverage context-aware local enhancement after the backbone network. In the local branch, we utilize coordinate attention to aggregate more powerful features of PCB defects with different shapes. In the global branch, Bi-Level Routing Attention with pooling is used to capture long-distance pixel interactions with limited computational cost. Furthermore, a Path Aggregation Network (PANet) feature fusion structure is incorporated to mitigate the loss of shallow features caused by the increase in model depth. Then, we design a lightweight prediction head by using depthwise separable convolutions, which further compresses the computational complexity and parameters while maintaining the detection capability of the model. In the experiment, the LGCL-CenterNet increased the mAP@0.5 by 2% and 1.4%, respectively, in comparison to CenterNet-ResNet18 and YOLOv8s. Meanwhile, our approach requires fewer model parameters (0.542M) than existing techniques. The results show that the proposed method improves both detection accuracy and inference speed and indicate that the LGCL-CenterNet has better real-time performance and robustness.

Fault Diagnosis Method for Tractor Transmission System Based on Improved Convolutional Neural Network–Bidirectional Long Short-Term Memory

In response to the problems of limited algorithms and low diagnostic accuracy for fault diagnosis in large tractor transmission systems, as well as the high noise levels in tractor working environments, a defect detection approach for tractor transmission systems is proposed using an enhanced convolutional neural network (CNN) and a bidirectional long short-term memory neural network (BILSTM). This approach uses a one-dimensional convolutional neural network (1DCNN) to create three feature extractors of varying scales, directly extracting feature information from different levels of the raw vibration signals. Simultaneously, in order to enhance the model’s predicted accuracy and learn the data features more effectively, it presents the multi-head attention mechanism (MHA). To overcome the issue of high noise levels in tractor working environments and enhance the model’s robustness, an adaptive soft threshold is introduced. Finally, to recognize and classify faults, the fused feature data are fed into a classifier made up of bidirectional long short-term memory (BILSTM) and fully linked layers. The analytical findings demonstrate that the fault recognition accuracy of the method described in this article is over 98%, and it also has better performance in noisy environments.

Robotic-based terahertz spectroscopy imaging for non-destructive monitoring of water loss-induced damage in timber materials under low temperature stress

Timber materials are widely used in various industries, from construction to furniture manufacturing. One of the critical factors affecting the structural integrity and longevity of timber is water loss-induced damage, which often occurs under low-temperature stress conditions. Monitoring and assessing this damage traditionally involve invasive and time-consuming methods. This study explores the application of robotic-based terahertz spectroscopy 3D imaging as a non-destructive and efficient system for assessing the impact of water loss on timber materials under low-temperature stress. The detection and localization of damage in a various species of timber-based structures were experimentally investigated by performing pulsed terahertz time-domain spectroscopy and using a six- DOF (degree-of-freedom) robotic arm. The thickness and damage length were monitored by designing the reflection geometry of the terahertz scanner. Additionally, the terahertz source was controlled with a six-DOF robot arm for path planning. The corresponding measured dielectric property reference value for the damage identification was computed using a captured terahertz wave at 12% of moisture content. The reflected signals were analyzed to determine the thickness, defectiveness, and locations of the defects through signal processing in the time and frequency domains. Finally, the validity of the structural health monitoring approach for 3D detection and visualization of water loss-induced defects in timber structures using terahertz waves, was discussed.

FP‐Flow: Feature pyramid flow model for fabric defect detection

AbstractFabric defect detection is a crucial aspect of fabric production. At present, deep learning detection methods mostly rely on supervised learning. To tackle this issue, this study proposes an unsupervised fabric defect detection approach based‐on normalising flow. The method only needs to train the mapping of the feature probability distribution of defect‐free samples to a Gaussian distribution. In the inference process, the location of defects can be determined by testing the distance between the probability distribution of image features and the estimated distribution. To adapt to the complex background and various minor defects of the fabric, a feature pyramid structure is adopted. Moreover, considering the gradient vanishing and network degradation caused by deep layers during training, a residual structure is incorporated into the model. Experimental results demonstrate that the feature pyramid flow model outperforms other methods in defect detection across multiple datasets, with an average score rate of 98.7% and 100% for pixel‐level area under the curve (AUC) receiver operating characteristic and image‐level AUC, respectively, compared to an average score rate of 91.1% and 85.4% for other methods.

Automated detection of underwater cracks based on fusion of optical and texture information

This paper presents a novel method for the underwater crack detection by fuse the optical and texture information. Underwater crack images were influenced by the harsh underwater environment, lead to making crack detection challenging, especially difficult to capturing the details of cracks accurately. To improve detection accuracy, it was necessary to obtain more feature information related to cracks. Therefore, this paper proposes a dual-input branch semantic segmentation model to achieve the fusion of optical and texture information, and introduces the Convolutional Block Attention Module (CBAM) module to enhance the performance of the semantic segmentation model. The optimal network architecture was determined by selecting the backbone network and optimizer, and a custom Tversky loss function was introduced to make the semantic segmentation model pay more attention to the crack area. The results show that the detection accuracy, IoU, and F1-Score can reach 96.07 %, 0.95, and 0.96 respectively. Through multiple comparative experiments, the effectiveness of the proposed method was validated, particularly compared to the non-fused texture information method, where the accuracy, IoU, and F1-Score were increased by 3.30 %, 6.74 %, and 7.88 % respectively. Finally, by visualizing the variation pattern of cracks in the detection model, the operational mechanism of the proposed method was explained. This confirms that the proposed method significantly improves the accuracy of underwater crack detection, and provides a novel approach for the underwater defect detection.

Defect Detection in Synthetic Fibre Ropes using Detectron2 Framework

Fibre ropes with the latest technology have emerged as an appealing alternative to steel ropes for offshore industries due to their lightweight and high tensile strength. At the same time, frequent inspection of these ropes is essential to ensure the proper functioning and safety of the entire system. The development of deep learning (DL) models in condition monitoring (CM) applications offers a simpler and more effective approach for defect detection in synthetic fibre ropes (SFRs). The present paper investigates the performance of Detectron2, a state-of-the-art library for defect detection andinstance segmentation. Detectron2 with Mask R-CNN architecture is used for segmenting defects in SFRs. Mask R-CNN with various backbone configurations has been trained and tested on an experimentally obtained dataset comprising 1,803 high-dimensional images containing seven damage classes (placking high, placking medium, placking low, compression, core out, chafing, and normal respectively) for SFRs. By leveraging the capabilities of Detectron2, this study aims to develop an automated and efficient method for detecting defects in SFRs, enhancing the inspection process, and ensuring the safety of the fibre ropes.

Enhancing the reliability of power grids: A YOLO based approach for insulator defect detection

Insulators are crucial components of power grid systems, safeguarding against electrical conductor breaks. However, their prolonged exposure to complex outdoor environments renders them susceptible to defects. In this study, we address the importance of accurate insulator defect detection and propose an approach using the You Only Look Once (YOLO) object detection framework. In particular, we compare the performance of YOLO v8 against YOLO v7 in detecting two specific types of insulator defects—broken insulators and flashover damaged insulators. Leveraging the Insulator Defect Image Dataset, our results demonstrate that YOLO v8 achieves superior accuracy with a rate of 98.99 percent along with a mean average precision (mAP) of 99.10 percent. The findings underscore the efficacy of YOLO v8 in improving the reliability and resilience of power grid systems by allowing timely and accurate detection of insulator defects in complex outdoor environments. This research contributes to advancing the field of power grid infrastructure monitoring and maintenance, ultimately facilitating more effective strategies for mitigating the consequence of insulator defects on power grid system performance and reliability.

Investigation on a lightweight defect detection model for photovoltaic panel

The detection of defect types of photovoltaic (PV) panel is a crucial task in PV system. Existing detection models face challenges in effectively balancing the trade-off between detection accuracy and resource consumption. To address this issue, this paper proposes a new defect detection method for PV panel based on the improved YOLOv8 model, which realizes both the high detection accuracy and the lightweight. Firstly, Reversible Column Networks (RevCol) is used as the Backbone of YOLOv8, which makes sure to preserve the feature information in the process of network transmission and also reduces the number of parameters and Giga floating-point operations per second (GFLOPs). Subsequently, a new lightweight Bottleneck fused with Efficient Multi-Scale Attention (EMA) is designed to optimize the CSPDarknet53 to 2-Stage FPN (C2f) module of Neck in YOLOv8 to enhance the robustness and further decrease network parameters. Finally, Squeeze-and-Excitation (SE) Attention is integrated into the Head of YOLOv8 to prioritize the important channel features and thus enhance the detection performance. The experimental results on the PVEL-AD dataset demonstrate that parameters and GFLOPs of the proposed model are declined by 38.46% and 34.39% respectively, and mAP0.5:0.95 is increased by 2.6% compared with the baseline model. The lightweight improved YOLOv8 model facilitates the deployment of deep learning model on edge devices and provides a novel approach for the online detection of PV panel defects.

Process defect analysis and visual detection of aluminum/copper cable joints with magnetic pulse crimping

Magnetic pulse crimping (MPC) has been demonstrated to enhance the electrical and mechanical performance of the aluminum/copper cable joints compared to traditional mechanical crimping. However, little attention has been given to the impact of defects during the manufacture on the performance. This paper uses MPC technology to connect joints and investigates the influence of various defects (including scratches, skew, length inadequate, wire broken, and depth inadequate), on the electrical and mechanical performance of the joints. The results indicated that the defects lead to significant changes performance of joints. Compared to the normal joint, skew, length inadequate, wire broken, and depth inadequate contributed to decreased tightness and the strength of the joints, resulting in an increase in the average resistance by 518.8 %, 1133.2 %, 1533.8 %, and 1377.9 %. Simultaneously, average loads of joints decrease by 16.4 %, 23.5 %, 28.8 %, and 91.3 %, respectively. Scratches have no discernible impact on performance. To address these defects systematically, a hierarchical defect detection approach was proposed. Faster R-CNN was employed to search three distinct regions: a large region (L), a middle region (M), and a small region (S), each associated with different defects that proved to impact performance via previous experiments. A COCO metric of 84.85 % was achieved. Four ResNet models were then utilized to classify defects within these regions. ResNet50 exhibited the best overall performance: 95 % accuracy for L, 100 % for M, and 87 % for S.

Deep Learning Based Defect Detection Research on Printed Circuit Boards

Abstract As computer vision and deep learning detection techniques advance rapidly, their use in identifying defects has become more common across various industries. The significance of Printed Circuit Boards (PCBs) in contemporary electronic devices is undeniable, as they substantially influence the functionality and durability of these products. Thus, utilizing deep learning models for identifying flaws in Printed Circuit Boards (PCBs) is of particular importance. The focus of this study is primarily on examining PCB defect identification utilizing deep learning techniques. Firstly, it introduces the importance and development history of PCBs in the electronics and information industry. It then offers a comprehensive review of the existing research on conventional PCB defect detection approaches alongside methodologies grounded in deep learning. Following that, the structure of the YOLOv8 object detection model and its key technologies are elaborated. Lastly, the superior performance of YOLOv8 in PCB defect detection tasks is verified through comparative experiments. According to the evaluation metrics of the algorithm, the average detection accuracy reaches 92.3%, and the Frames Per Second (FPS) value reaches 157.2, meeting the accuracy requirements for PCB defect detection in the industrial domain.

An improved hybrid solar cell defect detection approach using Generative Adversarial Networks and weighted classification

Quality control has a vital role in manufacturing processes. Electroluminescence (EL) imaging is one of the main non-destructive inspection methods for quality assessment in the Photovoltaic (PV) module production industry. EL test reveals PV cell defects such as micro cracks, broken cells, finger interruptions and provides detailed information about production quality. In recent years, automated detection and classification systems using deep neural networks for PV module inspection have gained increasing attention. However, deep learning-based systems usually require large amount of labeled data and high computational power. In this work, we proposed a compact classification framework based on hybrid data augmentation and deep learning models for detection of the defective solar cells. In the proposed method, the limited and imbalanced EL datasets were augmented through various Generative Adversarial Networks (GAN), and defect detection was achieved by customized pre-trained Convolutional Neural Networks (CNN). A novel hybrid EL dataset for training was formed by combining public ELPV dataset, our custom real-world dataset and synthetic images created with different GAN architectures such as GAN, cGAN and WGAN-GP. The datasets were classified through proposed customized VGG-16 and other state-of-the-art CNN models. The best results obtained by the proposed CNN model with WGAN-GP augmented dataset are 94.11% mean accuracy in the ELPV test dataset and 93.08% mean accuracy in the custom real-world test dataset. Therefore, the proposed detection system achieved superior performance with lesser resource usage and limited data.

Optimizing Insulator Defect Detection with Improved DETR Models

With the increasing demand for electricity, the power grid is undergoing significant advancements. Insulators, which serve as protective devices for transmission lines in outdoor high-altitude power systems, are widely employed. However, the detection of defects in insulators captured under challenging conditions, such as rain, snow, fog, sunlight, and fast-moving drones during long-distance photography, remains a major challenge. To address this issue and improve the accuracy of defect detection, this paper presents a novel approach: the Multi-Scale Insulator Defect Detection Approach using Detection Transformer (DETR). In this study, we propose a multi-scale backbone network that effectively captures the features of small objects, enhancing the detection performance. Additionally, we introduce a self-attention upsampling (SAU) module to replace the conventional attention module, enhancing contextual information extraction and facilitating the detection of small objects. Furthermore, we introduce the insulator defect (IDIoU) loss, which mitigates the instability in the matching process caused by small defects. Extensive experiments were conducted on an insulator defect dataset to evaluate the performance of our proposed method. The results demonstrate that our approach achieves outstanding performance, particularly in detecting small defects. Compared to existing methods, our approach exhibits a remarkable 7.47% increase in the average precision, emphasizing its efficacy in insulator defect detection. The proposed method not only enhances the accuracy of defect detection, which is crucial for maintaining the reliability and safety of power transmission systems but also has broader implications for the maintenance and inspection of high-voltage power infrastructure.