检索结果-内蒙古大学图书馆

Crop-paste and diffusion-based semi-supervised segmentation network for metal defect detection

KNOWLEDGE-BASED SYSTEMS 2025年 318卷

作者： Zhao, Lixiang Yu, Jianbo Tongji Univ Sch Mech Engn Caoan Rd 4800 Shanghai 201804 Peoples R China

Metal defect semantic segmentation is a crucial process for classifying and locating defects during the industrial production process, which holds paramount importance in elevating the quality of metal products. Recently, deep learning has exhibited impressive capabilities in identifying and segmenting defects on metal surfaces. However, the prevalent use of fully supervised segmentation techniques demands a substantial amount of annotated data for effective model training, which is hard to obtain in real scenarios. Additionally, most defects of metal products exhibit indistinct edge details, which hinders precise defect localization. In this study, a CropPaste and diffusion-based semi-supervised segmentation network (CPDNet) is proposed to identify pixel-level defects on metal surfaces by utilizing data that are both labeled and unlabeled. Firstly, a semi-supervised training method Crop-Paste is proposed to facilitate the learning of comprehensive semantic features from an extensive of unlabeled images and a restricted set of labeled images. Secondly, a frequency-directed diffusion model is proposed to recover high frequency features of defects to generate more accurate segmentation results. Lastly, an edge aware module is proposed in Sobel mean-teacher (M-T) UNet to improve the boundary information representation associated with defects. The experimental results on four datasets related to metal surface defects and a multimodal dataset show that CPDNet achieves a better performance in comparison with those state-of-the-art methods.

关键词： Metal defect detection defect segmentation Semi-supervised learning Diffusion model

来源：评论

学校读者我要写书评

暂无评论

segmentation of Inner Surface defects of Stainless Steel Pipes Based on Semi-bilateral Efficient Self-Attention Network

引用

JOURNAL OF NONDESTRUCTIVE EVALUATION 2025年第2期44卷 1-17页

作者： Wang, Hui Zhang, Chengbo Wang, Yangyu Ni, Pengcheng Wang, Yizhi Zhejiang Univ Technol Coll Mech Engn Hangzhou 310014 Zhejiang Peoples R China Zhejiang Univ Technol Key Lab Special Purpose Equipment & Adv Proc Techn Minist Educ Hangzhou 310014 Zhejiang Peoples R China Zhejiang Univ Technol Inst Taizhou Taizhou 318001 Zhejiang Peoples R China Shanghai Shangshang Stainless Steel Pipe Co Ltd Shanghai 201821 Peoples R China

Stainless steel pipes are one of the indispensable raw materials in industrial production, and they are widely used in fields such as aerospace, energy, and chemical industries. The service environment involves harsh conditions such as high temperature, high pressure, corrosion, and fatigue, placing high demands on structural safety. Due to the complex surface defects of steel pipes and the similarity of defect types, the assessment of the surface quality of steel pipes still relies on manual visual inspection. Based on these issues, this paper proposes a semi-bilateral efficient self-attention network (SBSANet) for quantifying the severity of internal surface defects in steel pipes. Firstly, a semi-bilateral and jump connection structure was designed to effectively compensate for information loss during the encoding process. Then, a multi-head attention encoding (MHAE) module was proposed to enable long-distance pixel interactions. Finally, a multi-scale residual context extraction (MSRCE) module was designed at the end of the encoder, effectively extracting multi-scale contextual information and enhancing the network's segmentation capability. Experimental results show that compared to other networks, the proposed method exhibits superior performance. SBSANet achieves a mean intersection over union (mIoU) of 78.48% on a self-made dataset and a running speed of 91.62 FPS, with mIoU of 81.28% and a running speed of 92.70 FPS on the NEU-Seg strip steel dataset.

关键词： Steel pipe detection defect segmentation Surface defect detection

来源：评论

学校读者我要写书评

暂无评论

A Lightweight Multiscale Feature Fusion Network for Solar Cell defect Detection

引用

Computers, Materials & Continua 2025年第1期82卷 521-542页

作者： Xiaoyun Chen Lanyao Zhang Xiaoling Chen Yigang Cen Linna Zhang Fugui Zhang School of Mechanical Engineering Guizhou UniversityGuiyang550025China School of Computer Science and Technology Beijing Jiaotong UniversityBeijing100044China

Solar cell defect detection is crucial for quality inspection in photovoltaic power generation *** the production process,defect samples occur infrequently and exhibit random shapes and sizes,which makes it challenging to collect defective ***,the complex surface background of polysilicon cell wafers complicates the accurate identification and localization of defective *** paper proposes a novel Lightweight Multiscale Feature Fusion network(LMFF)to address these *** network comprises a feature extraction network,a multi-scale feature fusion module(MFF),and a segmentation ***,a feature extraction network is proposed to obtain multi-scale feature outputs,and a multi-scale feature fusion module(MFF)is used to fuse multi-scale feature information *** order to capture finer-grained multi-scale information from the fusion features,we propose a multi-scale attention module(MSA)in the segmentation network to enhance the network’s ability for small target ***,depthwise separable convolutions are introduced to construct depthwise separable residual blocks(DSR)to reduce the model’s parameter ***,to validate the proposed method’s defect segmentation and localization performance,we constructed three solar cell defect detection datasets:SolarCells,SolarCells-S,and *** and SolarCells-S are monocrystalline silicon datasets,and PVEL-S is a polycrystalline silicon *** results show that the IOU of our method on these three datasets can reach 68.5%,51.0%,and 92.7%,respectively,and the F1-Score can reach 81.3%,67.5%,and 96.2%,respectively,which surpasses other commonly usedmethods and verifies the effectiveness of our LMFF network.

关键词： defect segmentation multi-scale feature fusion multi-scale attention depthwise separable residual block

来源：评论

学校读者我要写书评

暂无评论

Hard-Normal Example-Aware Template Mutual Matching for Industrial Anomaly Detection

引用

INTERNATIONAL JOURNAL OF COMPUTER VISION 2025年第5期133卷 2927-2949页

作者： Chen, Zixuan Xie, Xiaohua Yang, Lingxiao Lai, Jian-Huang Sun Yat Sen Univ Sch Comp Sci & Engn Guangzhou 510006 Guangdong Peoples R China Guangdong Prov Key Lab Informat Secur Technol Guangzhou Peoples R China Minist Educ Key Lab Machine Intelligence & Adv Comp Guangzhou Peoples R China

Anomaly detectors are widely used in industrial manufacturing to detect and localize unknown defects in query images. These detectors are trained on anomaly-free samples and have successfully distinguished anomalies from most normal samples. However, hard-normal examples are scattered and far apart from most normal samples, and thus they are often mistaken for anomalies by existing methods. To address this issue, we propose Hard-normal Example-aware Template Mutual Matching (HETMM), an efficient framework to build a robust prototype-based decision boundary. Specifically, HETMM employs the proposed Affine-invariant Template Mutual Matching (ATMM) to mitigate the affection brought by the affine transformations and easy-normal examples. By mutually matching the pixel-level prototypes within the patch-level search spaces between query and template set, ATMM can accurately distinguish between hard-normal examples and anomalies, achieving low false-positive and missed-detection rates. In addition, we also propose PTS to compress the original template set for speed-up. PTS selects cluster centres and hard-normal examples to preserve the original decision boundary, allowing this tiny set to achieve comparable performance to the original one. Extensive experiments demonstrate that HETMM outperforms state-of-the-art methods, while using a 60-sheet tiny set can achieve competitive performance and real-time inference speed (around 26.1 FPS) on a Quadro 8000 RTX GPU. HETMM is training-free and can be hot-updated by directly inserting novel samples into the template set, which can promptly address some incremental learning issues in industrial manufacturing.

关键词： Anomaly detection defect segmentation Template matching Real-time inference Hot update

来源：评论

学校读者我要写书评

暂无评论

An online surface defect detection method for weld seam based on SAE model and background extraction method

引用

OPTICS AND LASER TECHNOLOGY 2025年 187卷

作者： Shu, Leshi Zou, Gang Meng, Zhaoxu Wang, Yilin Huazhong Univ Sci & Technol Sch Mech Sci & Engn State Key Lab Digital Mfg Equipment & Technol Wuhan 430074 Peoples R China

The surface defect detection method of weld seam based on line-structured light has the advantages of noncontact measurement, high accuracy, and strong anti-interference capability, which has received increasing attention. How to improve the efficiency of defect detection to meet the needs of online detection in actual industry, while ensuring detection accuracy remains a challenge. This study proposed a surface defect detection method for weld seam based on Stacked Auto Encoder (SAE) model and background extraction method. In the proposed method, the raw weld contour data obtained from the structured light sensor is preprocessed to reduce the influence of environmental noise and sensor movement. Then defect detection is divided into two steps: defect recognition and defect segmentation. The former applies the SAE model to identify defective areas in the entire weld seam to avoid analyzing defect free areas and improve efficiency, while the latter uses the background extraction method to segment defects from the contour of the weld seam containing defects to reduce the complexity of defect segmentation. The proposed method has been applied to the typical defect detection of aluminum alloy samples of high-speed rail vehicle bodies, such as surface pore, arc pits, overlap, undercut, and surface collapse. The results show that the accuracy of the defect recognition model in recognizing continuous weld defects exceeds 97 %. The segmentation error of typical weld seam defects is within 0.2 mm.

关键词： Surface defect detection of welds Line-structured light Deep learning Background extraction method defect segmentation

来源：评论

学校读者我要写书评

暂无评论

SCSNet: a novel transformer-CNN fusion architecture for enhanced segmentation and classification on high-resolution semiconductor micro-scale defects

引用

APPLIED INTELLIGENCE 2025年第6期55卷 1-18页

作者： Luo, Yuening Mei, Zhouzhouzhou Qiao, Yibo Chen, Yining Zhejiang Univ Sch Micronano Elect 733 Jianjian 3rd Rd Hangzhou 311200 Zhejiang Peoples R China

In the domain of semiconductor integrated circuit manufacturing, accurately identifying the root causes of defects is critical for enhancing yield rates. Traditionally, this analytical process has been both time-intensive and challenged by inaccuracies, primarily due to the intricate and varied morphology of wafer defects. While convolutional neural networks (CNNs) with encoder-decoder architectures have made significant strides in the segmentation of defects, they inherently struggle to capture distant interactions and achieve high performance in classification tasks. Conversely, recent advancements in transformers have showcased their proficiency in learning global image dependencies. However, transformers often lack the specific graphical priors and the adaptability typically associated with CNNs. Addressing these limitations, we introduce SCSNet, an innovative architecture that merges the strengths of transformers and CNNs. This fusion network is designed to enhance both segmentation and classification of scanning electron microscopy (SEM) images of wafer defects. SCSNet incorporates a conventional encoder-decoder framework, supplemented by shape flow branches and multi-cross-attention (MCF) modules within a skip connection architecture. Rigorous experimentation on a dataset of 4425 high-resolution wafer defects, sourced from our operational wafer fabrication facility, demonstrates SCSNet's superior performance. Notably, SCSNet surpasses existing advanced CNNs, transformers, and their hybrid counterparts, achieving a classification accuracy of 97.62% and a segmentation Intersection over Union (IoU) of 84.09%. Currently implemented on our local server for engineering use, SCSNet represents a major advancement in semiconductor manufacturing, offering a more precise and efficient tool for wafer defect analysis.

关键词： CNN Transformer Semiconductor yield improvement defect classification defect segmentation

来源：评论

学校读者我要写书评

暂无评论

Fine-Grained Tiny defect Detection in Spiral Welds: A Joint Framework Combining Semantic Discrimination and Contrast Transformation

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Cui, Wenqi Song, Kechen Zhang, Yu Zhang, Yanning Lv, Guotong Yan, Yunhui Northeastern Univ Natl Frontiers Sci Ctr Ind Intelligence & Syst Opt Sch Mech Engn & Automat Shenyang 110819 Liaoning Peoples R China Northeastern Univ Key Lab Data Analyt & Optimizat Smart Ind Minist Educ Shenyang 110819 Liaoning Peoples R China Shenyang Univ Technol Sch Software Shenyang 110870 Peoples R China Linggang Beipiao Steel Pipe Co Ltd Chaoyang 122100 Peoples R China

X-ray-based nondestructive testing (NDT) is crucial for quality monitoring and maintenance decision-making of spiral steel pipes. Deep neural networks have shown impressive results in weld defect detection. However, in real-world spiral welding systems, the differences in appearance and distribution between spiral weld images and other weld images lead to a decrease in the detection accuracy of these models. There are still challenges in accurately localizing tiny defects in pixel-redundant and low-contrast spiral weld images. Additionally, little attention has been given to the practical issue of the inherent blurring in spiral welding images due to X-ray imaging. To this end, we propose a joint framework combining semantic discrimination and contrast transformation (SDCT). Specifically, we design a semantic discriminative architecture that generates adaptively initialized features guided by semantics and alleviates defect scale differences. Subsequently, a contrast transformation mechanism (CTM) is introduced to dynamically optimize defect features. Finally, a progressive iterative decoding (PID) strategy is employed to generate fine-grained pixel-level outputs. Comprehensive experiments conducted on the newly constructed tiny spiral weld X-ray (T-SWX-ray) dataset and two public benchmarks demonstrate that SDCT achieves competitive performance using the state-of-the-art techniques, thereby providing a robust solution for fine-grained detection of tiny defects in spiral welds. The source code and T-SWX-ray dataset are publicly available at https://***/SD-Weld/SDCT.

关键词： defect segmentation nondestructive testing (NDT) quality monitoring quality monitoring small target detection small target detection weld defect detection weld defect detection weld defect detection

来源：评论

学校读者我要写书评

暂无评论

Approach to weld segmentation and defect classification in radiographic images of pipe welds

引用

NDT & E INTERNATIONAL 2022年第0期127卷 102597-102597页

作者： Golodov, V. A. Maltseva, A. A. Almetyevsk State Oil Inst 2 Lenina St Almetyevsk Russia South Ural State Univ 76 Lenin Prospekt Chelyabinsk Russia

Detecting indications (points suspicious for defects) in weld radiographs is an important research topic in the field of industrial non-destructive testing. Many computer-aided detection techniques have been designed for such applications as detecting indications occurrence, segmentation of the indication area, classification of the indications. However, these techniques are mainly focused on only one of the listed problems. Different defects may exhibit different visual properties in shapes, sizes, textures, contrasts, and positions, that often leading to ad hoc solutions. The paper investigates to the fine tuning of the machine learning approach to high resolution radiograph processing produced by real-time radiography using digital detector arrays (DDA) method. The main contributions of this work are the preprocessing feature for human readability of the radiographic images and the proposed neural network-based solutions for all stages, from weld detection on the radiographic images and its segmentation to indication segmentation and classification. The designed approach was implemented as a web service with web site front-end. The demo version of the software is available with this instruction https://gith ***/NastyaMittseva/defectRecognitionSystem.

关键词： defect detection Radiographic images Neural networks Pipe welds segmentation defect segmentation defect classification

来源：评论

学校读者我要写书评

暂无评论

Deep learning-based detection of internal defect types and their grades in high-pressure aluminum castings

引用

MEASUREMENT 2025年 242卷

作者： Parlak, Ismail Enes Emel, Erdal Bursa Tech Univ Dept Ind Engn TR-16310 Bursa Turkiye Bursa Uludag Univ Dept Ind Engn TR-16240 Bursa Turkiye

With the increasing use of light alloy castings in automobiles, ensuring quality control is essential for safety. Xray imaging offers a practical approach to detecting internal defects in cast components. This study proposes a method to automatically and in real-time identify the location, type, and size of internal defects in aluminum parts produced via high-pressure casting. The proposed two-stage method can detect, segment, and grade defects without expensive hardware in less than a second. Using the YOLOv5 algorithm for defect detection in the first stage, a mean Average Precision (mAP) of 0.971 was achieved. In the second stage, defect grading is performed through segmentation, enabling classification in accordance with international standards without requiring additional training. The methodology provides real-time and highly accurate internal defect quality control and can be applied to different metals and standards. The dataset used in this study contains over 5,000 labelled X-ray images of aluminum cast parts, and it is made available as open access to support the NDT community and researchers.

关键词： Artificial intelligence Deep learning defect detection defect segmentation X-ray Aluminum

来源：评论

学校读者我要写书评

暂无评论

A multi-task segmentation and classification network for remote ship hull inspection

引用

OCEAN ENGINEERING 2024年 301卷

作者： Lin, Bosen Dong, Xinghui Ocean Univ China Sch Comp Sci & Technol 238 Songling Rd Qingdao 266100 Shangdong Peoples R China

In contrast to manual close-up ship hull inspection methods, Remote Inspection Technology (RIT) offers the potential to improve performance while minimising costs. Nevertheless, the effectiveness of the RIT is still subject to the experience and expertise of the inspectors. As a result, development has focused on applying automated data processing methods to RIT itself. However, in the hull inspection scenario, several challenges remain, including suboptimal hull imaging conditions, disparities in the distribution of defect categories, and significant variations in defect size. To overcome these challenges, we introduce a multi -task hull inspection network (MTHI-Net) that leverages the principles of multi -task learning to improve hull inspection accuracy. This network addresses two problems associated with RIT: image -level defect classification and pixel -level defect segmentation. Specifically, MTHI-Net exploits the advantages of spatial and channel self -attention mechanisms, a residual refinement module, and a feature fusion module to enhance the feature representation capability for defect classification and mitigate the probability of mis-segmentation. In addition, a lightweight MTHI-Net called MTHI-Net-Lite was developed. The experimental results show that the proposed networks fulfil the hull inspection task better than the baselines in both real underwater and in -dock RIT scenarios.

关键词： Ship hull inspection Remote Inspection Technology (RIT) Multi-task learning defect segmentation defect classification

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：