This study investigates the application of Learnable Memory Vision Transformers(LMViT)for detecting metal surface flaws,comparing their performance with traditional CNNs,specifically ResNet18 and ResNet50,as well as o...This study investigates the application of Learnable Memory Vision Transformers(LMViT)for detecting metal surface flaws,comparing their performance with traditional CNNs,specifically ResNet18 and ResNet50,as well as other transformer-based models including Token to Token ViT,ViT withoutmemory,and Parallel ViT.Leveraging awidely-used steel surface defect dataset,the research applies data augmentation and t-distributed stochastic neighbor embedding(t-SNE)to enhance feature extraction and understanding.These techniques mitigated overfitting,stabilized training,and improved generalization capabilities.The LMViT model achieved a test accuracy of 97.22%,significantly outperforming ResNet18(88.89%)and ResNet50(88.90%),aswell as the Token to TokenViT(88.46%),ViT without memory(87.18),and Parallel ViT(91.03%).Furthermore,LMViT exhibited superior training and validation performance,attaining a validation accuracy of 98.2%compared to 91.0%for ResNet 18,96.0%for ResNet50,and 89.12%,87.51%,and 91.21%for Token to Token ViT,ViT without memory,and Parallel ViT,respectively.The findings highlight the LMViT’s ability to capture long-range dependencies in images,an areawhere CNNs struggle due to their reliance on local receptive fields and hierarchical feature extraction.The additional transformer-based models also demonstrate improved performance in capturing complex features over CNNs,with LMViT excelling particularly at detecting subtle and complex defects,which is critical for maintaining product quality and operational efficiency in industrial applications.For instance,the LMViT model successfully identified fine scratches and minor surface irregularities that CNNs often misclassify.This study not only demonstrates LMViT’s potential for real-world defect detection but also underscores the promise of other transformer-based architectures like Token to Token ViT,ViT without memory,and Parallel ViT in industrial scenarios where complex spatial relationships are key.Future research may focus on enhancing LMViT’s computational efficiency for deployment in real-time quality control systems.展开更多
Efficient detection of surface defects is primary for ensuring product quality during manufacturing processes.To enhance the performance of deep learning-based methods in practical applications,the authors propose Den...Efficient detection of surface defects is primary for ensuring product quality during manufacturing processes.To enhance the performance of deep learning-based methods in practical applications,the authors propose Dense-YOLO,a fast surface defect detection network that combines the strengths of DenseNet and you only look once version 3(YOLOv3).The authors design a lightweight backbone network with improved densely connected blocks,optimising the utilisation of shallow features while maintaining high detection speeds.Additionally,the authors refine the feature pyramid network of YOLOv3 to increase the recall of tiny defects and overall positioning accuracy.Furthermore,an online multi-angle template matching technique is introduced based on normalised cross-correlation to precisely locate the detection area.This refined template matching method not only accelerates detection speed but also mitigates the influence of the background.To validate the effectiveness of our enhancements,the authors conduct comparative experiments across two private datasets and one public dataset.Results show that Dense-YOLO outperforms existing methods,such as faster R-CNN,YOLOv3,YOLOv5s,YOLOv7,and SSD,in terms of mean average precision(mAP)and detection speed.Moreover,Dense-YOLO outperforms networks inherited from VGG and ResNet,including improved faster R-CNN,FCOS,M2Det-320 and FRCN,in mAP.展开更多
In response to themissed and false detections that are easily caused by the large variety of and significant differences among aluminum surface defects,a detection algorithm based on an improved You Only Look Once(YOL...In response to themissed and false detections that are easily caused by the large variety of and significant differences among aluminum surface defects,a detection algorithm based on an improved You Only Look Once(YOLO)v8n network is proposed.First,a C2f_DWR_DRB module is constructed by introducing a dilation-wise residual(DWR)module and a dilated reparameterization block(DRB)to replace the C2f module at the high level of the backbone network,enriching the gradient flow information and increasing the effective receptive field(ERF).Second,an efficient local attention(ELA)mechanism is fused with the high-level screening-feature pyramid networks(HS-FPN)module,and an ELA_HSFPN is designed to replace the original feature fusion module,enhancing the ability of the network to cope with multiscale detection tasks.Moreover,a lightweight shared convolutional detection head(SCDH)is introduced to reduce the number of parameters and the computational complexity of the module while enhancing the performance and generalizability of the model.Finally,the soft intersection over union(SIoU)replaces the original loss function to improve the convergence speed and prediction accuracy of the model.Experimental results show that compared with that of the original YOLOv8n model,the mAP@0.5 of the improved algorithm is increased by 5.1%,the number of parameters and computational complexity are reduced by 33.3%and 32.1%,respectively,and the FPS is increased by 4.9%.Compared with other mainstream object detection algorithms,the improved algorithm still leads in terms of core indicators and has good generalizability for surface defects encountered in other industrial scenarios.展开更多
The detection of surface defects in concrete bridges using deep learning is of significant importance for reducing operational risks,saving maintenance costs,and driving the intelligent transformation of bridge defect...The detection of surface defects in concrete bridges using deep learning is of significant importance for reducing operational risks,saving maintenance costs,and driving the intelligent transformation of bridge defect detection.In contrast to the subjective and inefficient manual visual inspection,deep learning-based algorithms for concrete defect detection exhibit remarkable advantages,emerging as a focal point in recent research.This paper comprehensively analyzes the research progress of deep learning algorithms in the field of surface defect detection in concrete bridges in recent years.It introduces the early detection methods for surface defects in concrete bridges and the development of deep learning.Subsequently,it provides an overview of deep learning-based concrete bridge surface defect detection research from three aspects:image classification,object detection,and semantic segmentation.The paper summarizes the strengths and weaknesses of existing methods and the challenges they face.Additionally,it analyzes and prospects the development trends of surface defect detection in concrete bridges.展开更多
A novel dual-branch decoding fusion convolutional neural network model(DDFNet)specifically designed for real-time salient object detection(SOD)on steel surfaces is proposed.DDFNet is based on a standard encoder–decod...A novel dual-branch decoding fusion convolutional neural network model(DDFNet)specifically designed for real-time salient object detection(SOD)on steel surfaces is proposed.DDFNet is based on a standard encoder–decoder architecture.DDFNet integrates three key innovations:first,we introduce a novel,lightweight multi-scale progressive aggregation residual network that effectively suppresses background interference and refines defect details,enabling efficient salient feature extraction.Then,we propose an innovative dual-branch decoding fusion structure,comprising the refined defect representation branch and the enhanced defect representation branch,which enhance accuracy in defect region identification and feature representation.Additionally,to further improve the detection of small and complex defects,we incorporate a multi-scale attention fusion module.Experimental results on the public ESDIs-SOD dataset show that DDFNet,with only 3.69 million parameters,achieves detection performance comparable to current state-of-the-art models,demonstrating its potential for real-time industrial applications.Furthermore,our DDFNet-L variant consistently outperforms leading methods in detection performance.The code is available at https://github.com/13140W/DDFNet.展开更多
Detecting surface defects on unused rails is crucial for evaluating rail quality and durability to ensure the safety of rail transportation.However,existing detection methods often struggle with challenges such as com...Detecting surface defects on unused rails is crucial for evaluating rail quality and durability to ensure the safety of rail transportation.However,existing detection methods often struggle with challenges such as complex defect morphology,texture similarity,and fuzzy edges,leading to poor accuracy and missed detections.In order to resolve these problems,we propose MSCM-Net(Multi-Scale Cross-Modal Network),a multiscale cross-modal framework focused on detecting rail surface defects.MSCM-Net introduces an attention mechanism to dynamically weight the fusion of RGB and depth maps,effectively capturing and enhancing features at different scales for each modality.To further enrich feature representation and improve edge detection in blurred areas,we propose a multi-scale void fusion module that integrates multi-scale feature information.To improve cross-modal feature fusion,we develop a cross-enhanced fusion module that transfers fused features between layers to incorporate interlayer information.We also introduce a multimodal feature integration module,which merges modality-specific features from separate decoders into a shared decoder,enhancing detection by leveraging richer complementary information.Finally,we validate MSCM-Net on the NEU RSDDS-AUG RGB-depth dataset,comparing it against 12 leading methods,and the results show that MSCM-Net achieves superior performance on all metrics.展开更多
The service life of internal combustion engines is significantly influenced by surface defects in cylinder liners.To address the limitations of traditional detection methods,we propose an enhanced YOLOv8 model with Sw...The service life of internal combustion engines is significantly influenced by surface defects in cylinder liners.To address the limitations of traditional detection methods,we propose an enhanced YOLOv8 model with Swin Transformer as the backbone network.This approach leverages Swin Transformer's multi-head self-attention mechanism for improved feature extraction of defects spanning various scales.Integrated with the YOLOv8 detection head,our model achieves a mean average precision of 85.1%on our dataset,outperforming baseline methods by 1.4%.The model's effectiveness is further demonstrated on a steel-surface defect dataset,indicating its broad applicability in industrial surface defect detection.Our work highlights the potential of combining Swin Transformer and YOLOv8 for accurate and efficient defect detection.展开更多
In order to meet the requirements of accurate identification of surface defects on copper strip in industrial production,a detection model of surface defects based on machine vision,CSC-YOLO,is proposed.The model uses...In order to meet the requirements of accurate identification of surface defects on copper strip in industrial production,a detection model of surface defects based on machine vision,CSC-YOLO,is proposed.The model uses YOLOv4-tiny as the benchmark network.First,K-means clustering is introduced into the benchmark network to obtain anchor frames that match the self-built dataset.Second,a cross-region fusion module is introduced in the backbone network to solve the difficult target recognition problem by fusing contextual semantic information.Third,the spatial pyramid pooling-efficient channel attention network(SPP-E)module is introduced in the path aggregation network(PANet)to enhance the extraction of features.Fourth,to prevent the loss of channel information,a lightweight attention mechanism is introduced to improve the performance of the network.Finally,the performance of the model is improved by adding adjustment factors to correct the loss function for the dimensional characteristics of the surface defects.CSC-YOLO was tested on the self-built dataset of surface defects in copper strip,and the experimental results showed that the mAP of the model can reach 93.58%,which is a 3.37% improvement compared with the benchmark network,and FPS,although decreasing compared with the benchmark network,reached 104.CSC-YOLO takes into account the real-time requirements of copper strip production.The comparison experiments with Faster RCNN,SSD300,YOLOv3,YOLOv4,Resnet50-YOLOv4,YOLOv5s,YOLOv7,and other algorithms show that the algorithm obtains a faster computation speed while maintaining a higher detection accuracy.展开更多
Automatic surface defect detection is a critical technique for ensuring product quality in industrial casting production.While general object detection techniques have made remarkable progress over the past decade,cas...Automatic surface defect detection is a critical technique for ensuring product quality in industrial casting production.While general object detection techniques have made remarkable progress over the past decade,casting surface defect detection still has considerable room for improvement.Lack of sufficient and high-quality data has become one of the most challenging problems for casting surface defect detection.In this paper,we construct a new casting surface defect dataset(CSDD)containing 2100 high-resolution images of casting surface defects and 56356 defects in total.The class and defect region for each defect are manually labeled.We conduct a series of experiments on this dataset using multiple state-of-the-art object detection methods,establishing a comprehensive set of baselines.We also propose a defect detection method based on YOLOv5 with the global attention mechanism and partial convolution.Our proposed method achieves superior performance compared to other object detection methods.Additionally,we also conduct a series of experiments with multiple state-of-the-art semantic segmentation methods,providing extensive baselines for defect segmentation.To the best of our knowledge,the CSDD has the largest number of defects for casting surface defect detection and segmentation.It would benefit both the industrial vision research and manufacturing applications.Dataset and code are available at https://github.com/Kerio99/CSDD.展开更多
Aiming at the problems of low detection efficiency and difficult positioning of traditional steel surface defect detection methods,a lightweight steel surface defect detection model based on you only look once version...Aiming at the problems of low detection efficiency and difficult positioning of traditional steel surface defect detection methods,a lightweight steel surface defect detection model based on you only look once version 7(YOLOv7)is proposed.First,a cascading style sheets(CSS)block module is proposed,which uses more lightweight operations to obtain redundant information in the feature map,reduces the amount of computation,and effectively improves the detection speed.Secondly,the improved spatial pyramid pooling with cross stage partial convolutions(SPPCSPC)structure is adopted to ensure that the model can also pay attention to the defect location information while predicting the defect category information,obtain richer defect features.In addition,the convolution operation in the original model is simplified,which significantly reduces the size of the model and helps to improve the detection speed.Finally,using efficient intersection over union(EIOU)loss to focus on high-quality anchors,speed up convergence and improve positioning accuracy.Experiments were carried out on the Northeastern University-defect(NEU-DET)steel surface defect dataset.Compared with the original YOLOv7 model,the number of parameters of this model was reduced by 40%,the frames per second(FPS)reached 112,and the average accuracy reached 79.1%,the detection accuracy and speed have been improved,which can meet the needs of steel surface defect detection.展开更多
T he residual stray magnetic fields present in ferromagnetic casting slabs were investigated in this work,which result from the magnetic fields generated during the steel casting process.Existing optical detection met...T he residual stray magnetic fields present in ferromagnetic casting slabs were investigated in this work,which result from the magnetic fields generated during the steel casting process.Existing optical detection methods face challenges owing to surface oxide scales,and conventional high-precision magnetic sensors are ineffective at high temperatures.To overcome these limitations,a small coil sensor was employed to measure the residual magnetism strength in oscillation traces,using metal magnetic memory and electromagnetic induction methods,which can carry out detection without an external excitation source.Using this technology,the proposed scheme successfully detects defects at high tempe-ratures(up to 670℃)without a cooling device.The key findings include the ability to detect both surface and near-surface defects,such as cracks and oscillation marks,with an enhanced signal-to-noise ratio(SNR)of 7.2 dB after signal processing.The method’s practicality was validated in a steel mill environment,where testing on casting slabs effectively detected defects,providing a foundation for improving industrial quality control.The proposed detection scheme offers a significant advancement in nondestructive testing(NDT)for high-temperature applications,contributing to more efficient and accurate monitoring of ferromagnetic material integrity.展开更多
In printed circuit board(PCB)manufacturing,surface defects can significantly affect product quality.To address the performance degradation,high false detection rates,and missed detections caused by complex backgrounds...In printed circuit board(PCB)manufacturing,surface defects can significantly affect product quality.To address the performance degradation,high false detection rates,and missed detections caused by complex backgrounds in current intelligent inspection algorithms,this paper proposes CG-YOLOv8,a lightweight and improved model based on YOLOv8n for PCB surface defect detection.The proposed method optimizes the network architecture and compresses parameters to reduce model complexity while maintaining high detection accuracy,thereby enhancing the capability of identifying diverse defects under complex conditions.Specifically,a cascaded multi-receptive field(CMRF)module is adopted to replace the SPPF module in the backbone to improve feature perception,and an inverted residual mobile block(IRMB)is integrated into the C2f module to further enhance performance.Additionally,conventional convolution layers are replaced with GSConv to reduce computational cost,and a lightweight Convolutional Block Attention Module based Convolution(CBAMConv)module is introduced after Grouped Spatial Convolution(GSConv)to preserve accuracy through attention mechanisms.The detection head is also optimized by removing medium and large-scale detection layers,thereby enhancing the model’s ability to detect small-scale defects and further reducing complexity.Experimental results show that,compared to the original YOLOv8n,the proposed CG-YOLOv8 reduces parameter count by 53.9%,improves mAP@0.5 by 2.2%,and increases precision and recall by 2.0%and 1.8%,respectively.These improvements demonstrate that CG-YOLOv8 offers an efficient and lightweight solution for PCB surface defect detection.展开更多
A more effective and accurate improved Sobel algorithm has been developed to detect surface defects on heavy rails. The proposed method can make up for the mere sensitivity to X and Y directions of the Sobel algorithm...A more effective and accurate improved Sobel algorithm has been developed to detect surface defects on heavy rails. The proposed method can make up for the mere sensitivity to X and Y directions of the Sobel algorithm by adding six templates at different directions. Meanwhile, an experimental platform for detecting surface defects consisting of the bed-jig, image-forming system with CCD cameras and light sources, parallel computer system and cable system has been constructed. The detection results of the backfin defects show that the improved Sobel algorithm can achieve an accurate and efficient positioning with decreasing interference noises to the defect edge. It can also extract more precise features and characteristic parameters of the backfin defect. Furthermore, the BP neural network adopted for defects classification with the inputting characteristic parameters of improved Sobel algorithm can obtain the optimal training precision of 0.0095827 with 106 iterative steps and time of 3 s less than Sobel algorithm with 146 steps and 5 s. Finally, an enhanced identification rate of 10% for the defects is also confirmed after the Sobel algorithm is improved.展开更多
Feature extraction is essential to the classification of surface defect images. The defects of hot-rolled steels distribute in different directions. Therefore, the methods of multi-scale geometric analysis （MGA） wer...Feature extraction is essential to the classification of surface defect images. The defects of hot-rolled steels distribute in different directions. Therefore, the methods of multi-scale geometric analysis （MGA） were employed to decompose the image into several directional subba^ds at several scales. Then, the statistical features of each subband were calculated to produce a high-dimensional feature vector, which was reduced to a lower-dimensional vector by graph embedding algorithms. Finally, support vector machine （SVM） was used for defect classification. The multi-scale feature extraction method was implemented via curvelet transform and kernel locality preserving projections （KLPP）. Experiment results show that the proposed method is effective for classifying the surface defects of hot-rolled steels and the total classification rate is up to 97.33%.展开更多
Inefficient charge separation and limited light absorption are two critical issues associated with high‐efficiency photocatalytic H2production using TiO2.Surface defects within a certain concentration range in photoc...Inefficient charge separation and limited light absorption are two critical issues associated with high‐efficiency photocatalytic H2production using TiO2.Surface defects within a certain concentration range in photocatalyst materials are beneficial for photocatalytic activity.In this study,surface defects(oxygen vacancies and metal cation replacement defects)were induced with a facile and effective approach by surface doping with low‐cost transition metals(Co,Ni,Cu,and Mn)on ultrafine TiO2.The obtained surface‐defective TiO2exhibited a3–4‐fold improved activity compared to that of the original ultrafine TiO2.In addition,a H2production rate of3.4μmol/h was obtained using visible light(λ>420nm)irradiation.The apparent quantum yield(AQY)at365nm reached36.9%over TiO2‐Cu,significantly more than the commercial P25TiO2.The enhancement of photocatalytic H2production activity can be attributed to improved rapid charge separation efficiency andexpanded light absorption window.This hydrothermal treatment with transition metal was proven to be a very facile and effective method for obtaining surface defects.展开更多
Considering that the surface defects of cold rolled strips are hard to be recognized by human eyes under high-speed circumstances, an automatic recognition technique was discussed. Spectrum images of defects can be go...Considering that the surface defects of cold rolled strips are hard to be recognized by human eyes under high-speed circumstances, an automatic recognition technique was discussed. Spectrum images of defects can be got by fast Fourier transform （FFF） and sum of valid pixels （SVP）, and its optimized center region, which concentrates nearly all energies, are extracted as an original feature set. Using genetic algorithm to optimize the feature set, an optimized feature set with 51 features can be achieved. Using the optimized feature set as an input vector of neural networks, the recognition effects of LVQ neural networks have been studied. Experiment results show that the new method can get a higher classification rate and can settle the automatic recognition problem of surface defects on cold rolled strips ideally.展开更多
Hypo-peritectic steels are widely used in various industrial fields because of their high strength,high toughness,high processability,high weldability,and low material cost.However,surface defects are liable to occur ...Hypo-peritectic steels are widely used in various industrial fields because of their high strength,high toughness,high processability,high weldability,and low material cost.However,surface defects are liable to occur during continuous casting,which includes depression,longitudinal cracks,deep oscillation marks,and severe level fluctuation with slag entrapment.The high-efficiency production of hypo-peritectic steels by continuous casting is still a great challenge due to the limited understanding of the mechanism of peritectic solidification.This work reviews the definition and classification of hypo-peritectic steels and introduces the formation tendency of common surface defects related to peritectic solidification.New achievements in the mechanism of peritectic reaction and transformation have been listed.Finally,countermeasures to avoiding surface defects of hypo-peritectic steels duiring continuous casting are summarized.Enlightening certain points in the continuous casting of hypo-peritectic steels and the development of new techniques to overcome the present problems will be a great aid to researchers.展开更多
The effect of solution heat treatment （SHT） on mechanical properties, microstructure and surface quality of Al-1.2Mg-0.6Si-0.2Cu-0.6Zn alloy was investigated by tensile test, Erichsen test, surface topography, scann...The effect of solution heat treatment （SHT） on mechanical properties, microstructure and surface quality of Al-1.2Mg-0.6Si-0.2Cu-0.6Zn alloy was investigated by tensile test, Erichsen test, surface topography, scanning electron microscope （SEM） and electron back-scattered diffraction （EBSD）. The results indicate that with the increase in SHT temperature, yield strength and cupping test value （IE） of the sheets increase greatly and reach a peak value, then decrease. Meanwhile, intermetallic com- pounds dissolve into matrix gradually. The grains grow up as SHT temperature increases, and abnormal grain growth leads to the surface defects after solution-treated above 560 ~C. Considering mechanical properties, IE value, residual phases, grain size and surface quality of the sheets, SHT temperature for the alloy should not be higher than 550 ℃.展开更多
Development of active and stable catalysts for low-temperature CO oxidation has long been regarded as a hot topic.In this contribution,we used CeO_(2) with high-density surface pits as support to prepare an active and...Development of active and stable catalysts for low-temperature CO oxidation has long been regarded as a hot topic.In this contribution,we used CeO_(2) with high-density surface pits as support to prepare an active and stable Au/CeO_(2) catalyst by an adsorption-deposition method.The obtained 0.05 wt%Au/CeO_(2)-TD(where TD represents thermal decomposition)can maintain its activity at 80℃ for more than 20 h or even after calcination at 800℃ for 2 h.The characterization results showed that the high-density surface pits on CeO_(2)-TD play a decisive role in the stabilization of Au and enhancement of the redox property.This work may provide a new strategy to improve the stability of supported metal catalysts by a simple and conventional method.展开更多
Carbon dioxide adsorbed on different kinds of CaO surfaces has been investigated with the help of the first principle density functional theory plane wave calculations. Various possible configurations have been consid...Carbon dioxide adsorbed on different kinds of CaO surfaces has been investigated with the help of the first principle density functional theory plane wave calculations. Various possible configurations have been considered and the calculated results showed that CO2 was strongly adsorbed by C atom bonded with the CaO （001） and （110） surfaces with adsorption energies of 1.38 and 3.22 eV, respectively. The adsorption of CO2 molecule on defect surfaces is complicated compared with that on the pristine surfaces. The adsorption energy of CO2 absorbed on the CaO（110） surface is larger than that of CaO（001） surface when the type of defect surface is the same.展开更多
基金funded by Woosong University Academic Research 2024.
文摘This study investigates the application of Learnable Memory Vision Transformers(LMViT)for detecting metal surface flaws,comparing their performance with traditional CNNs,specifically ResNet18 and ResNet50,as well as other transformer-based models including Token to Token ViT,ViT withoutmemory,and Parallel ViT.Leveraging awidely-used steel surface defect dataset,the research applies data augmentation and t-distributed stochastic neighbor embedding(t-SNE)to enhance feature extraction and understanding.These techniques mitigated overfitting,stabilized training,and improved generalization capabilities.The LMViT model achieved a test accuracy of 97.22%,significantly outperforming ResNet18(88.89%)and ResNet50(88.90%),aswell as the Token to TokenViT(88.46%),ViT without memory(87.18),and Parallel ViT(91.03%).Furthermore,LMViT exhibited superior training and validation performance,attaining a validation accuracy of 98.2%compared to 91.0%for ResNet 18,96.0%for ResNet50,and 89.12%,87.51%,and 91.21%for Token to Token ViT,ViT without memory,and Parallel ViT,respectively.The findings highlight the LMViT’s ability to capture long-range dependencies in images,an areawhere CNNs struggle due to their reliance on local receptive fields and hierarchical feature extraction.The additional transformer-based models also demonstrate improved performance in capturing complex features over CNNs,with LMViT excelling particularly at detecting subtle and complex defects,which is critical for maintaining product quality and operational efficiency in industrial applications.For instance,the LMViT model successfully identified fine scratches and minor surface irregularities that CNNs often misclassify.This study not only demonstrates LMViT’s potential for real-world defect detection but also underscores the promise of other transformer-based architectures like Token to Token ViT,ViT without memory,and Parallel ViT in industrial scenarios where complex spatial relationships are key.Future research may focus on enhancing LMViT’s computational efficiency for deployment in real-time quality control systems.
基金Program for Young Excellent Talents in the University of Fujian Province,Grant/Award Number:201847The National Key Research and Development Program of China,Grant/Award Number:2022YFB3206605Natural Science Foundation of Xiamen Municipality,Grant/Award Number:3502Z20227189。
文摘Efficient detection of surface defects is primary for ensuring product quality during manufacturing processes.To enhance the performance of deep learning-based methods in practical applications,the authors propose Dense-YOLO,a fast surface defect detection network that combines the strengths of DenseNet and you only look once version 3(YOLOv3).The authors design a lightweight backbone network with improved densely connected blocks,optimising the utilisation of shallow features while maintaining high detection speeds.Additionally,the authors refine the feature pyramid network of YOLOv3 to increase the recall of tiny defects and overall positioning accuracy.Furthermore,an online multi-angle template matching technique is introduced based on normalised cross-correlation to precisely locate the detection area.This refined template matching method not only accelerates detection speed but also mitigates the influence of the background.To validate the effectiveness of our enhancements,the authors conduct comparative experiments across two private datasets and one public dataset.Results show that Dense-YOLO outperforms existing methods,such as faster R-CNN,YOLOv3,YOLOv5s,YOLOv7,and SSD,in terms of mean average precision(mAP)and detection speed.Moreover,Dense-YOLO outperforms networks inherited from VGG and ResNet,including improved faster R-CNN,FCOS,M2Det-320 and FRCN,in mAP.
基金supported by the Jiangsu Province Science and Technology Policy Guidance Program(Industry-University-Research Cooperation)/Forward-Looking Joint Research Project(BY2016005-05).
文摘In response to themissed and false detections that are easily caused by the large variety of and significant differences among aluminum surface defects,a detection algorithm based on an improved You Only Look Once(YOLO)v8n network is proposed.First,a C2f_DWR_DRB module is constructed by introducing a dilation-wise residual(DWR)module and a dilated reparameterization block(DRB)to replace the C2f module at the high level of the backbone network,enriching the gradient flow information and increasing the effective receptive field(ERF).Second,an efficient local attention(ELA)mechanism is fused with the high-level screening-feature pyramid networks(HS-FPN)module,and an ELA_HSFPN is designed to replace the original feature fusion module,enhancing the ability of the network to cope with multiscale detection tasks.Moreover,a lightweight shared convolutional detection head(SCDH)is introduced to reduce the number of parameters and the computational complexity of the module while enhancing the performance and generalizability of the model.Finally,the soft intersection over union(SIoU)replaces the original loss function to improve the convergence speed and prediction accuracy of the model.Experimental results show that compared with that of the original YOLOv8n model,the mAP@0.5 of the improved algorithm is increased by 5.1%,the number of parameters and computational complexity are reduced by 33.3%and 32.1%,respectively,and the FPS is increased by 4.9%.Compared with other mainstream object detection algorithms,the improved algorithm still leads in terms of core indicators and has good generalizability for surface defects encountered in other industrial scenarios.
基金supported by the Key Research and Development Program of Shaanxi Province-International Science and Technology Cooperation Program Project (No.2020KW-001)the Contract for Xi'an Municipal Science and Technology Plan Project-Xi'an City Strong Foundation Innovation Plan (No.21XJZZ0074)the Key Project of Graduate Student Innovation Fund at Xi'an University of Posts and Telecommunications (No.CXJJZL2023013)。
文摘The detection of surface defects in concrete bridges using deep learning is of significant importance for reducing operational risks,saving maintenance costs,and driving the intelligent transformation of bridge defect detection.In contrast to the subjective and inefficient manual visual inspection,deep learning-based algorithms for concrete defect detection exhibit remarkable advantages,emerging as a focal point in recent research.This paper comprehensively analyzes the research progress of deep learning algorithms in the field of surface defect detection in concrete bridges in recent years.It introduces the early detection methods for surface defects in concrete bridges and the development of deep learning.Subsequently,it provides an overview of deep learning-based concrete bridge surface defect detection research from three aspects:image classification,object detection,and semantic segmentation.The paper summarizes the strengths and weaknesses of existing methods and the challenges they face.Additionally,it analyzes and prospects the development trends of surface defect detection in concrete bridges.
基金supported in part by the National Key R&D Program of China(Grant No.2023YFB3307604)the Shanxi Province Basic Research Program Youth Science Research Project(Grant Nos.202303021212054 and 202303021212046)+3 种基金the Key Projects Supported by Hebei Natural Science Foundation(Grant No.E2024203125)the National Science Foundation of China(Grant No.52105391)the Hebei Provincial Science and Technology Major Project(Grant No.23280101Z)the National Key Laboratory of Metal Forming Technology and Heavy Equipment Open Fund(Grant No.S2308100.W17).
文摘A novel dual-branch decoding fusion convolutional neural network model(DDFNet)specifically designed for real-time salient object detection(SOD)on steel surfaces is proposed.DDFNet is based on a standard encoder–decoder architecture.DDFNet integrates three key innovations:first,we introduce a novel,lightweight multi-scale progressive aggregation residual network that effectively suppresses background interference and refines defect details,enabling efficient salient feature extraction.Then,we propose an innovative dual-branch decoding fusion structure,comprising the refined defect representation branch and the enhanced defect representation branch,which enhance accuracy in defect region identification and feature representation.Additionally,to further improve the detection of small and complex defects,we incorporate a multi-scale attention fusion module.Experimental results on the public ESDIs-SOD dataset show that DDFNet,with only 3.69 million parameters,achieves detection performance comparable to current state-of-the-art models,demonstrating its potential for real-time industrial applications.Furthermore,our DDFNet-L variant consistently outperforms leading methods in detection performance.The code is available at https://github.com/13140W/DDFNet.
基金funded by the National Natural Science Foundation of China(grant number 62306186)the Technology Plan Joint Foundation of Liaoning Province(grant number 2023-MSLH-246)the Technology Plan Joint Foundation of Liaoning Province(grant number 2023-BSBA-238).
文摘Detecting surface defects on unused rails is crucial for evaluating rail quality and durability to ensure the safety of rail transportation.However,existing detection methods often struggle with challenges such as complex defect morphology,texture similarity,and fuzzy edges,leading to poor accuracy and missed detections.In order to resolve these problems,we propose MSCM-Net(Multi-Scale Cross-Modal Network),a multiscale cross-modal framework focused on detecting rail surface defects.MSCM-Net introduces an attention mechanism to dynamically weight the fusion of RGB and depth maps,effectively capturing and enhancing features at different scales for each modality.To further enrich feature representation and improve edge detection in blurred areas,we propose a multi-scale void fusion module that integrates multi-scale feature information.To improve cross-modal feature fusion,we develop a cross-enhanced fusion module that transfers fused features between layers to incorporate interlayer information.We also introduce a multimodal feature integration module,which merges modality-specific features from separate decoders into a shared decoder,enhancing detection by leveraging richer complementary information.Finally,we validate MSCM-Net on the NEU RSDDS-AUG RGB-depth dataset,comparing it against 12 leading methods,and the results show that MSCM-Net achieves superior performance on all metrics.
基金supported by the Scientific and technological key project in Henan Province 22210224002the Natural Science Foundation of Henan Polytechnic University B2021-38.
文摘The service life of internal combustion engines is significantly influenced by surface defects in cylinder liners.To address the limitations of traditional detection methods,we propose an enhanced YOLOv8 model with Swin Transformer as the backbone network.This approach leverages Swin Transformer's multi-head self-attention mechanism for improved feature extraction of defects spanning various scales.Integrated with the YOLOv8 detection head,our model achieves a mean average precision of 85.1%on our dataset,outperforming baseline methods by 1.4%.The model's effectiveness is further demonstrated on a steel-surface defect dataset,indicating its broad applicability in industrial surface defect detection.Our work highlights the potential of combining Swin Transformer and YOLOv8 for accurate and efficient defect detection.
基金the Key Project of Basic Research of Yunnan Province(No.202101AS070016)。
文摘In order to meet the requirements of accurate identification of surface defects on copper strip in industrial production,a detection model of surface defects based on machine vision,CSC-YOLO,is proposed.The model uses YOLOv4-tiny as the benchmark network.First,K-means clustering is introduced into the benchmark network to obtain anchor frames that match the self-built dataset.Second,a cross-region fusion module is introduced in the backbone network to solve the difficult target recognition problem by fusing contextual semantic information.Third,the spatial pyramid pooling-efficient channel attention network(SPP-E)module is introduced in the path aggregation network(PANet)to enhance the extraction of features.Fourth,to prevent the loss of channel information,a lightweight attention mechanism is introduced to improve the performance of the network.Finally,the performance of the model is improved by adding adjustment factors to correct the loss function for the dimensional characteristics of the surface defects.CSC-YOLO was tested on the self-built dataset of surface defects in copper strip,and the experimental results showed that the mAP of the model can reach 93.58%,which is a 3.37% improvement compared with the benchmark network,and FPS,although decreasing compared with the benchmark network,reached 104.CSC-YOLO takes into account the real-time requirements of copper strip production.The comparison experiments with Faster RCNN,SSD300,YOLOv3,YOLOv4,Resnet50-YOLOv4,YOLOv5s,YOLOv7,and other algorithms show that the algorithm obtains a faster computation speed while maintaining a higher detection accuracy.
基金supported by the National Natural Science Foundation of China(U23B2060,62088102)the Key Research and Development Program of China(2020AAA0108305).
文摘Automatic surface defect detection is a critical technique for ensuring product quality in industrial casting production.While general object detection techniques have made remarkable progress over the past decade,casting surface defect detection still has considerable room for improvement.Lack of sufficient and high-quality data has become one of the most challenging problems for casting surface defect detection.In this paper,we construct a new casting surface defect dataset(CSDD)containing 2100 high-resolution images of casting surface defects and 56356 defects in total.The class and defect region for each defect are manually labeled.We conduct a series of experiments on this dataset using multiple state-of-the-art object detection methods,establishing a comprehensive set of baselines.We also propose a defect detection method based on YOLOv5 with the global attention mechanism and partial convolution.Our proposed method achieves superior performance compared to other object detection methods.Additionally,we also conduct a series of experiments with multiple state-of-the-art semantic segmentation methods,providing extensive baselines for defect segmentation.To the best of our knowledge,the CSDD has the largest number of defects for casting surface defect detection and segmentation.It would benefit both the industrial vision research and manufacturing applications.Dataset and code are available at https://github.com/Kerio99/CSDD.
基金supported by the National Natural Science Foundation of China(No.62103298)the Natural Science Foundation of Hebei Province(No.F2018209289)。
文摘Aiming at the problems of low detection efficiency and difficult positioning of traditional steel surface defect detection methods,a lightweight steel surface defect detection model based on you only look once version 7(YOLOv7)is proposed.First,a cascading style sheets(CSS)block module is proposed,which uses more lightweight operations to obtain redundant information in the feature map,reduces the amount of computation,and effectively improves the detection speed.Secondly,the improved spatial pyramid pooling with cross stage partial convolutions(SPPCSPC)structure is adopted to ensure that the model can also pay attention to the defect location information while predicting the defect category information,obtain richer defect features.In addition,the convolution operation in the original model is simplified,which significantly reduces the size of the model and helps to improve the detection speed.Finally,using efficient intersection over union(EIOU)loss to focus on high-quality anchors,speed up convergence and improve positioning accuracy.Experiments were carried out on the Northeastern University-defect(NEU-DET)steel surface defect dataset.Compared with the original YOLOv7 model,the number of parameters of this model was reduced by 40%,the frames per second(FPS)reached 112,and the average accuracy reached 79.1%,the detection accuracy and speed have been improved,which can meet the needs of steel surface defect detection.
文摘T he residual stray magnetic fields present in ferromagnetic casting slabs were investigated in this work,which result from the magnetic fields generated during the steel casting process.Existing optical detection methods face challenges owing to surface oxide scales,and conventional high-precision magnetic sensors are ineffective at high temperatures.To overcome these limitations,a small coil sensor was employed to measure the residual magnetism strength in oscillation traces,using metal magnetic memory and electromagnetic induction methods,which can carry out detection without an external excitation source.Using this technology,the proposed scheme successfully detects defects at high tempe-ratures(up to 670℃)without a cooling device.The key findings include the ability to detect both surface and near-surface defects,such as cracks and oscillation marks,with an enhanced signal-to-noise ratio(SNR)of 7.2 dB after signal processing.The method’s practicality was validated in a steel mill environment,where testing on casting slabs effectively detected defects,providing a foundation for improving industrial quality control.The proposed detection scheme offers a significant advancement in nondestructive testing(NDT)for high-temperature applications,contributing to more efficient and accurate monitoring of ferromagnetic material integrity.
基金funded by the Joint Funds of the National Natural Science Foundation of China(U2341223)the Beijing Municipal Natural Science Foundation(No.4232067).
文摘In printed circuit board(PCB)manufacturing,surface defects can significantly affect product quality.To address the performance degradation,high false detection rates,and missed detections caused by complex backgrounds in current intelligent inspection algorithms,this paper proposes CG-YOLOv8,a lightweight and improved model based on YOLOv8n for PCB surface defect detection.The proposed method optimizes the network architecture and compresses parameters to reduce model complexity while maintaining high detection accuracy,thereby enhancing the capability of identifying diverse defects under complex conditions.Specifically,a cascaded multi-receptive field(CMRF)module is adopted to replace the SPPF module in the backbone to improve feature perception,and an inverted residual mobile block(IRMB)is integrated into the C2f module to further enhance performance.Additionally,conventional convolution layers are replaced with GSConv to reduce computational cost,and a lightweight Convolutional Block Attention Module based Convolution(CBAMConv)module is introduced after Grouped Spatial Convolution(GSConv)to preserve accuracy through attention mechanisms.The detection head is also optimized by removing medium and large-scale detection layers,thereby enhancing the model’s ability to detect small-scale defects and further reducing complexity.Experimental results show that,compared to the original YOLOv8n,the proposed CG-YOLOv8 reduces parameter count by 53.9%,improves mAP@0.5 by 2.2%,and increases precision and recall by 2.0%and 1.8%,respectively.These improvements demonstrate that CG-YOLOv8 offers an efficient and lightweight solution for PCB surface defect detection.
基金Project(51174151)supported by the National Natural Science Foundation of ChinaProject(2010Z19003)supported by the Major Scientific Research Program of Hubei Provincial Department of Education,ChinaProject(2010CDB03403)supported by the Natural Science Foundation of Science and Technology Department of Hubei Province,China
文摘A more effective and accurate improved Sobel algorithm has been developed to detect surface defects on heavy rails. The proposed method can make up for the mere sensitivity to X and Y directions of the Sobel algorithm by adding six templates at different directions. Meanwhile, an experimental platform for detecting surface defects consisting of the bed-jig, image-forming system with CCD cameras and light sources, parallel computer system and cable system has been constructed. The detection results of the backfin defects show that the improved Sobel algorithm can achieve an accurate and efficient positioning with decreasing interference noises to the defect edge. It can also extract more precise features and characteristic parameters of the backfin defect. Furthermore, the BP neural network adopted for defects classification with the inputting characteristic parameters of improved Sobel algorithm can obtain the optimal training precision of 0.0095827 with 106 iterative steps and time of 3 s less than Sobel algorithm with 146 steps and 5 s. Finally, an enhanced identification rate of 10% for the defects is also confirmed after the Sobel algorithm is improved.
基金supports by the Program for New Century Excellent Talents in Chinese Universities (No.NCET-08-0726)Beijing Nova Program (No. 2007B027)the Fundamental Research Funds for the Central Universities (No. FRF-TP-09-027B)
文摘Feature extraction is essential to the classification of surface defect images. The defects of hot-rolled steels distribute in different directions. Therefore, the methods of multi-scale geometric analysis （MGA） were employed to decompose the image into several directional subba^ds at several scales. Then, the statistical features of each subband were calculated to produce a high-dimensional feature vector, which was reduced to a lower-dimensional vector by graph embedding algorithms. Finally, support vector machine （SVM） was used for defect classification. The multi-scale feature extraction method was implemented via curvelet transform and kernel locality preserving projections （KLPP）. Experiment results show that the proposed method is effective for classifying the surface defects of hot-rolled steels and the total classification rate is up to 97.33%.
基金supported by the Double First‐rate Subject‐Food Science and Engineering Program of Hebei Province (2018SPGCA18)Young Tip‐top Talents Plan of Universities and Colleges in Hebei Province of China (BJ2017026)the Specific Foundation for Doctor in Hebei Agriculture University of China (ZD201709)~~
文摘Inefficient charge separation and limited light absorption are two critical issues associated with high‐efficiency photocatalytic H2production using TiO2.Surface defects within a certain concentration range in photocatalyst materials are beneficial for photocatalytic activity.In this study,surface defects(oxygen vacancies and metal cation replacement defects)were induced with a facile and effective approach by surface doping with low‐cost transition metals(Co,Ni,Cu,and Mn)on ultrafine TiO2.The obtained surface‐defective TiO2exhibited a3–4‐fold improved activity compared to that of the original ultrafine TiO2.In addition,a H2production rate of3.4μmol/h was obtained using visible light(λ>420nm)irradiation.The apparent quantum yield(AQY)at365nm reached36.9%over TiO2‐Cu,significantly more than the commercial P25TiO2.The enhancement of photocatalytic H2production activity can be attributed to improved rapid charge separation efficiency andexpanded light absorption window.This hydrothermal treatment with transition metal was proven to be a very facile and effective method for obtaining surface defects.
基金This work was financially supported by the National High Technology Research and Development Program of China (No.2003AA331080 and 2001AA339030)the Talent Science Research Foundation of Henan University of Science & Technology (No.09001121).
文摘Considering that the surface defects of cold rolled strips are hard to be recognized by human eyes under high-speed circumstances, an automatic recognition technique was discussed. Spectrum images of defects can be got by fast Fourier transform （FFF） and sum of valid pixels （SVP）, and its optimized center region, which concentrates nearly all energies, are extracted as an original feature set. Using genetic algorithm to optimize the feature set, an optimized feature set with 51 features can be achieved. Using the optimized feature set as an input vector of neural networks, the recognition effects of LVQ neural networks have been studied. Experiment results show that the new method can get a higher classification rate and can settle the automatic recognition problem of surface defects on cold rolled strips ideally.
基金financially supported by the Fundamental Research Funds for the Central Universities(No.FRF-TP-19-017A3)the National Natural Science Foundation of China(No.51874026)。
文摘Hypo-peritectic steels are widely used in various industrial fields because of their high strength,high toughness,high processability,high weldability,and low material cost.However,surface defects are liable to occur during continuous casting,which includes depression,longitudinal cracks,deep oscillation marks,and severe level fluctuation with slag entrapment.The high-efficiency production of hypo-peritectic steels by continuous casting is still a great challenge due to the limited understanding of the mechanism of peritectic solidification.This work reviews the definition and classification of hypo-peritectic steels and introduces the formation tendency of common surface defects related to peritectic solidification.New achievements in the mechanism of peritectic reaction and transformation have been listed.Finally,countermeasures to avoiding surface defects of hypo-peritectic steels duiring continuous casting are summarized.Enlightening certain points in the continuous casting of hypo-peritectic steels and the development of new techniques to overcome the present problems will be a great aid to researchers.
基金supported by the National Program on Key Basic Research Project of China(No. 2012CB619504)the National Natural Science Foundation of China(No.51271037)
文摘The effect of solution heat treatment （SHT） on mechanical properties, microstructure and surface quality of Al-1.2Mg-0.6Si-0.2Cu-0.6Zn alloy was investigated by tensile test, Erichsen test, surface topography, scanning electron microscope （SEM） and electron back-scattered diffraction （EBSD）. The results indicate that with the increase in SHT temperature, yield strength and cupping test value （IE） of the sheets increase greatly and reach a peak value, then decrease. Meanwhile, intermetallic com- pounds dissolve into matrix gradually. The grains grow up as SHT temperature increases, and abnormal grain growth leads to the surface defects after solution-treated above 560 ~C. Considering mechanical properties, IE value, residual phases, grain size and surface quality of the sheets, SHT temperature for the alloy should not be higher than 550 ℃.
基金financially supported by the National Key Research and Development Program of China(No.2016YFC0204300)the National Natural Science Foundation of China(Nos.21571061,21333003 and 21908079)Pujiang Program of the Shanghai Municipal Human Resources and Social Security Bureau(No.18PJD011)。
文摘Development of active and stable catalysts for low-temperature CO oxidation has long been regarded as a hot topic.In this contribution,we used CeO_(2) with high-density surface pits as support to prepare an active and stable Au/CeO_(2) catalyst by an adsorption-deposition method.The obtained 0.05 wt%Au/CeO_(2)-TD(where TD represents thermal decomposition)can maintain its activity at 80℃ for more than 20 h or even after calcination at 800℃ for 2 h.The characterization results showed that the high-density surface pits on CeO_(2)-TD play a decisive role in the stabilization of Au and enhancement of the redox property.This work may provide a new strategy to improve the stability of supported metal catalysts by a simple and conventional method.
基金supported by the National Natural Science Foundation of China(21203027,21073035)Natural Science Foundation of Fujian Province for Distinguished Young Investigator Grant(2013J06004)Funds of Fujian Province(2012J01032,2012J01041)
文摘Carbon dioxide adsorbed on different kinds of CaO surfaces has been investigated with the help of the first principle density functional theory plane wave calculations. Various possible configurations have been considered and the calculated results showed that CO2 was strongly adsorbed by C atom bonded with the CaO （001） and （110） surfaces with adsorption energies of 1.38 and 3.22 eV, respectively. The adsorption of CO2 molecule on defect surfaces is complicated compared with that on the pristine surfaces. The adsorption energy of CO2 absorbed on the CaO（110） surface is larger than that of CaO（001） surface when the type of defect surface is the same.
