检索结果-内蒙古大学图书馆

DHNet: a surface defect detection model utilizing multi-scale convolutional kernels

JOURNAL OF real-time image processing 2025年第1期22卷 1-15页

作者： Zhang, Yingying Wang, Shuo Wang, Jinhai Zhao, Yu Chen, Zhiwei Hefei Univ Technol Sch Elect Engn & Automat Hefei Peoples R China

Detecting micro-defects in densely populated printed circuit boards (PCBs) with complex backgrounds is a critical challenge. To address the problem, the DHNet, a small object detection network based on YOLOv8 employing multi-scale convolutional kernels is proposed for feature extraction and fusion. The lightweight VOVGSHet module is designed for feature fusion and a pyramid structure to efficiently leverage feature map relationships while minimizing model complexity and parameters. Otherwise, to optimize the original extraction structure and enhance multi-scale defect detection, convolutional kernels of varying sizes process the same input channels. Additionally, the incorporation of the Wise-IoU loss function improves small defect detection accuracy and efficiency. Moreover, extensive experiments on a custom PCB dataset demonstrate DHNet's effectiveness, achieving an outstanding mean Average Precision (mAP) of 84.5%, surpassing the original YOLOv8 network by 4.0%, with parameters only of 2.85 M. Model demonstrates a latency of 3.6 ms on NVIDIA 4090. However, YOLOv8n has a latency of 4.4 ms. Validation on public deepPCB and NEU datasets further confirms DHNet's superiority, which can reach 99.1% and 79.9% mAP, respectively. Finally, successful deployment on the NVIDIA Jetson Nano platform validates DHNet's suitability for real-time defect detection in industrial applications.

关键词： Printed circuit boards (PCBs) Defect detection deep learning Multi-scale feature fusion Jetson Nano

来源：评论

学校读者我要写书评

暂无评论

Feature Extraction of Lathe Tool Crater Wear image Using Auto-Encoder

引用

TRANSACTIONS OF THE KOREAN SOCIETY OF MECHANICAL ENGINEERS A 2023年第3期47卷 273-281页

作者： Choi, Jae Uk Heo, Hyo Beom Park, Seung Hwan Extraction, Feature Chungnam Natl Univ Dept Mech Engn Daejeon South Korea

To monitor tool wear during cutting processing, tool wear is mainly measured indirectly through sensor signals that are most correlated with wear. The direct measurement method of tool wear using image and optical sensors is more accurate than indirect measurement, but it is mainly used to measure the amount of wear because it is difficult to apply in real time. Existing studies have been conducted mainly on flank wear caused by friction with workpiece. On the other hand, crater wear is an important monitoring factor because it is caused by friction with chips generated during processing and causes sudden tool breakage. However, for crater wear, it is difficult to measure the amount of wear because the indicator of the amount of wear is depth. Therefore, although image processing-based studies have been conducted to measure the amount of crater wear, there is a clear limit to accurately measure the depth only with the image on the top of the tool. In this work, we propose a method to extract unique features of crater wear images through autoencoder, a deep learning technique, and use them as a new measure of wear.

关键词： Tool Condition Monitoring Tool Wear Feature Extraction Machine learning deep learning image processing

来源：评论

学校读者我要写书评

暂无评论

Inter - intra observer variability using deep learning and traditional image processing for breast cancer 13

Inter - intra observer variability using deep learning and t...

引用

13th International Conference on Developments in eSystems Engineering (DeSE)

作者： Almazroa, Ahmed Alsomaie, Barrak Alluhaydan, Najd Alhaidary, Amal Fahim, Mohammed Abdul, Wadood Alawad, Mohammed Khan, Altaf Alenezi, Ebtihal Almotairi, Taghreed Alyahya, AlJowharah Alfulayj, Raghad Althobaiti, Manar bin Maythir, Jawaher King Saud bin Abdulaziz Univ Hlth Sci King Abdullah Int Med Res Ctr Med Imaging Dept Riyadh Saudi Arabia King Abdul Aziz Med City Med Imaging Dept Riyadh Saudi Arabia King Saud Univ Coll Comp & Informat Sci Dept Comp Engn Riyadh Saudi Arabia King Saud bin Abdulaziz Univ Hlth Sci King Abdullah Int Med Res Ctr Biostat & Bioinformat Dept Riyadh Saudi Arabia King Saud bin Abdulaziz Univ Hlth Sci Radiol Sci Dept Riyadh Saudi Arabia

ISBN: (纸本)9781665422383

Breast cancer is one of the most common life-threatening diseases that affects women globally. Saudi Arabia is also one of the countries that suffer from a serious number of this disease among women. In terms of diagnosis modalities, a mammogram is the first line for detecting breast cancer. In addition, breast cancer can be screened by real-time ultrasound images, which are of relatively less quality and have more impact (noninvasive) images. Therefore, the purpose of this study is to develop image enhancement techniques using deep learning and image processing techniques. The main goal is to improve the ultrasound images in order to help radiologists screen the disease more accurately. For this study, ninety female patients of ages between 15 - 77 years are considered. These patients were already diagnosed using ultrasound with breast lesions. The images are visually graded and evaluated by two trained radiologists, both pre- and post- enhancement. In particular, two parameters were considered;1) BI-RAD categories, 2) Breast cancer classification. The agreement between radiologists and post-enhancement was assessed using simple kappa and weighted kappa statistics. Moreover, sensitivity and specificity are also calculated.

关键词： Breast cancer image processing deep learning Computer-aided diagnosis systems

来源：评论

学校读者我要写书评

暂无评论

image segmentation for blind lanes based on improved SegNet model

引用

JOURNAL OF ELECTRONIC IMAGING 2023年第1期32卷 013038-013038页

作者： Xia, Yongquan Li, Yiqing Ye, Qianqian Dong, Jianhua Zhengzhou Univ Light Ind Sch Comp & Commun Engn Zhengzhou Peoples R China

The position of blind lanes must be correctly determined in order for blind people to travel safely. Aiming at the low accuracy and slow speed of traditional blind lanes image segmentation algorithms, a semantic segmentation method based on SegNet and MobileNetV3 is proposed. The main idea is to replace the coding part of the original SegNet model with the feature extraction part of MobileNetV3 and remove the pooling layer. Blind lanes images were collected through online search and self-shooting, and then the data were manually marked by LabelMe software and trained on TensorFlow deep learning framework. The experimental results show that the improved model has high segmentation accuracy and recognition speed. The pixel accuracy of blind lanes segmentation is 98.21%, the mean intersection over union is 96.29%, and the average time for processing a 416 x 416 image is 0.057 s, which meets the real-time requirements of the blind guidance system.

关键词： blind lanes segmentation image semantic segmentation deep learning image segmentation MobileNetV3 SegNet

来源：评论

学校读者我要写书评

暂无评论

Optimizing Federated learning With deep Reinforcement learning for Digital Twin Empowered Industrial IoT

引用

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2023年第2期19卷 1884-1893页

作者： Yang, Wei Xiang, Wei Yang, Yuan Cheng, Peng Xian Univ Technol Sch Automat & Informat Engn Xian 710048 Peoples R China La Trobe Univ Sch Comp Engn & Math Sci Melbourne Vic 3086 Australia James Cook Univ Coll Sci & Engn Cairns Qld 4870 Australia

The accelerated development of the Industrial Internet of Things (IIoT) is catalyzing the digitalization of industrial production to achieve Industry 4.0. In this article, we propose a novel digital twin (DT) empowered IIoT (DTEI) architecture, in which DTs capture the properties of industrial devices for real-time processing and intelligent decision making. To alleviate data transmission burden and privacy leakage, we aim to optimize federated learning (FL) to construct the DTEI model. Specifically, to cope with the heterogeneity of IIoT devices, we develop the DTEI-assisted deep reinforcement learning method for the selection process of IIoT devices in FL, especially for selecting IIoT devices with high utility values. Furthermore, we propose an asynchronous FL scheme to address the discrete effects caused by heterogeneous IIoT devices. Experimental results show that our proposed scheme features faster convergence and higher training accuracy compared to the benchmark.

关键词： deep reinforcement learning (DRL) digital twin (DT) federated learning (FL) Industrial Internet of Things (IIoT) learning efficiency real time

来源：评论

学校读者我要写书评

暂无评论

Condition Monitoring of Railway Pantograph Using R-CNN and image processing

引用

JOURNAL OF ELECTRICAL ENGINEERING & TECHNOLOGY 2023年第3期18卷 2407-2416页

作者： Na, Kyung-Min Lee, Kiwon Kim, Hyungchul Railway Components Evaluat Ctr Korea Testing Lab Seoul South Korea Korea Railrd Res Inst Smart Elect & Signaling Div Uiwang South Korea

A railway pantograph supplies a vehicle with the electric power from the OCL (Overhead Contact Line). The contact strip, which directly contacts the OCL, is vulnerable to wear and should be periodically replaced to prevent accidents from power outages. To that end, the pantograph is regularly checked visually during routine maintenance, along with the panhead which includes the horn and the contact strip. However, real-time monitoring is still difficult, and a reliable method needs to be developed. The existing methods use image processing to detect the actual condition of the contact strip. In this paper, we suggest a method for detecting the wear size of the contact strip using image processing and deep learning. In addition, we demonstrate how to assess the condition of the panhead and the horn in the contact. The monitoring equipment is built on the test bed and is automatically configured to acquire images of the moving pantograph. The tilt of the panhead is estimated from image recognition and the condition of the horn is assessed with deep learning. In the future, this method can be used not only for efficient pantograph maintenance but also for determining the performance according to the contact condition.

关键词： Electric railway Pantograph Contact strip Maintenance

来源：评论

学校读者我要写书评

暂无评论

Estimating Water Levels through Smartphone-imaged Gauges: A Comparative Analysis of ANN, DL, and CNN Models

引用

WATER RESOURCES MANAGEMENT 2025年第4期39卷 1639-1654页

作者： Santos, Celso Augusto Guimaraes Ghorbani, Mohammad Ali Abdi, Erfan Patel, Utkarsh Sadeddin, Siria Univ Fed Paraiba Dept Civil & Environm Engn BR-58051900 Joao Pessoa PB Brazil Univ Tabriz Dept Water Engn Tabriz Iran Indian Inst Technol IIT Bhilai Main Campus Durg 491001 Chhattisgarh India Univ Nacl Colombia Math Dept Bogota Colombia

Understanding and monitoring water levels are essential for various applications, including environmental protection, public safety, and resource management. Water level estimation, a critical aspect of hydrological monitoring, is often constrained by challenges such as resource scarcity, high costs, and time-intensive processes. This research addresses these limitations by developing a machine learning-based system for automatic and real-time water level control. Specifically, it investigates the effectiveness of a non-contact, image-based water level measurement approach, leveraging recent advancements in mobile imaging technology. images were captured using a standard smartphone equipped with an RGB camera for water level analysis. Through precise image alignment processing under both clear and turbid conditions, the water's edge on a gauge was accurately detected. The study centers on the development and comparison of three computational models: Artificial Neural Networks (ANN), deep learning (DL), and Convolutional Neural Networks (CNN). These models were trained to estimate water levels from processed image data. Results demonstrated varying levels of accuracy across models, with the CNN model outperforming others, achieving the lowest error rate of 24.36 mm and the highest correlation of 0.986. In contrast, the ANN model yielded the highest error rate at 30.76 mm and the lowest correlation of 0.968, highlighting the relative effectiveness of CNN in this application. Given the high accuracy (92.6%) of the image processing method and CNN model in detecting water surface edges and determining water levels, this system has substantial potential to enhance water resource management and control efficiency.

关键词： Computational Hydrology Convolutional Neural Networks Hydrological Monitoring image Analysis Non-contact Measurement

来源：评论

学校读者我要写书评

暂无评论

Understanding the black-box: towards interpretable and reliable deep learning models

引用

PEERJ COMPUTER SCIENCE 2023年 9卷 e1629页

作者： Qamar, Tehreem Bawany, Narmeen Zakaria Jinnah Univ Women Ctr Comp Res Dept Comp Sci & Software Engn Karachi Pakistan

deep learning (DL) has revolutionized the field of artificial intelligence by providing sophisticated models across a diverse range of applications, from image and speech recognition to natural language processing and autonomous driving. However, deep learning models are typically black-box models where the reason for predictions is unknown. Consequently, the reliability of the model becomes questionable in many circumstances. Explainable AI (XAI) plays an important role in improving the transparency and interpretability of the model thereby making it more reliable for real-time deployment. To investigate the reliability and truthfulness of DL models, this research develops image classification models using transfer learning mechanism and validates the results using XAI technique. Thus, the contribution of this research is twofold, we employ three pre-trained models VGG16, MobileNetV2 and ResNet50 using multiple transfer learning techniques for a fruit classification task consisting of 131 classes. Next, we inspect the reliability of models, based on these pre-trained networks, by utilizing Local Interpretable Model-Agnostic Explanations, the LIME, a popular XAI technique that generates explanations for the predictions. Experimental results reveal that transfer learning provides optimized results of around 98% accuracy. The classification of the models is validated on different instances using LIME and it was observed that each model predictions are interpretable and understandable as they are based on pertinent image features that are relevant to particular classes. We believe that this research gives an insight for determining how an interpretation can be drawn from a complex AI model such that its accountability and trustworthiness can be increased.

关键词： deep learning Explainable AI Transfer learning Pre-trained models

来源：评论

学校读者我要写书评

暂无评论

Artifact reduction in photoacoustic images by generating virtual dense array sensor from hemispheric sparse array sensor using deep learning

引用

JOURNAL OF MEDICAL ULTRASONICS 2024年第2期51卷 169-183页

作者： Yamakawa, Makoto Shiina, Tsuyoshi Shibaura Inst Technol SIT Res Labs 3-7-5 ToyosuKoto Ku Tokyo 1358548 Japan

PurposeVascular distribution is important information for diagnosing diseases and supporting surgery. Photoacoustic imaging is a technology that can image blood vessels noninvasively and with high resolution. In photoacoustic imaging, a hemispherical array sensor is especially suitable for measuring blood vessels running in various directions. However, as a hemispherical array sensor, a sparse array sensor is often used due to technical and cost issues, which causes artifacts in photoacoustic images. Therefore, in this study, we reduce these artifacts using deep learning technology to generate signals of virtual dense array *** 2D virtual array sensor signals using a 3D convolutional neural network (CNN) requires huge computational costs and is impractical. Therefore, we installed virtual sensors between the real sensors along the spiral pattern in three different directions and used a 2D CNN to generate signals of the virtual sensors in each direction. Then we reconstructed a photoacoustic image using the signals from both the real sensors and the virtual *** evaluated the proposed method using simulation data and human palm measurement data. We found that these artifacts were significantly reduced in the images reconstructed using the proposed method, while the artifacts were strong in the images obtained only from the real sensor *** the proposed method, we were able to significantly reduce artifacts, and as a result, it became possible to recognize deep blood vessels. In addition, the processing time of the proposed method was sufficiently applicable to clinical measurement.

关键词： Photoacoustic imaging Sparse array sensor Hemispherical array sensor deep learning Artifact reduction

来源：评论

学校读者我要写书评

暂无评论

SLTM Network: Efficient Application of Lightweight image Segmentation Technology in Detecting Drivable Areas for Unmanned Line-Marking Machines

引用

IEEE ACCESS 2024年 12卷 169001-169012页

作者： Wang, Chao Chen, Xiangkai Wang, Bingtao Zhang, Liang Liu, Bing Shandong Univ Sch Mech Elect & Informat Engn Weihai 264209 Peoples R China Shandong Univ Off Acad Affairs Weihai 264209 Peoples R China

image segmentation plays a crucial role in the roadwork operations of autonomous line-painting machines. However, the limited resources of mobile platforms in intelligent line-painting applications pose a dual challenge of ensuring both accuracy and real-time performance in road segmentation. To address this issue, this study introduces a lightweight yet efficient image segmentation model, termed the SLTM Network. Central to this network is the lightweight SLTM module, which significantly reduces the model's parameter count and lowers the computational overhead of the decoder. To enhance the interplay of information at different spatial resolutions, the network incorporates an SE attention-enhanced upsampling module (SAUM) and employs a Spatial Attention Sequence (SAS) unit to improve global environment perception at a low computational cost. Comprehensive experimental evaluations on the Cityscapes dataset demonstrate that the SLTM Network excels in balancing speed and accuracy, achieving an mIoU of 70.5% with only 4.07M parameters and an impressive inference speed of 267.1 FPS. On the embedded device Jetson Xavier NX, it achieves an inference speed of 34.2 FPS. Compared to existing lightweight image segmentation models, the SLTM Network exhibits significant advantages in both processing speed and accuracy, making it particularly suitable for real-time autonomous line-painting machine applications.

关键词： Roads image segmentation Feature extraction Computer architecture Computational efficiency Synthetic aperture sonar Semantics Lightweight image segmentation unmanned line-marking machines deep learning in autonomous driving embedded systems optimization SLTM network real-time road segmentation

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：