检索结果-内蒙古大学图书馆

A Review of image Super-Resolution Approaches Based on deep learning and Applications in Remote Sensing

REMOTE SENSING 2022年第21期14卷

作者： Wang, Xuan Yi, Jinglei Guo, Jian Song, Yongchao Lyu, Jun Xu, Jindong Yan, Weiqing Zhao, Jindong Cai, Qing Min, Haigen Yantai Univ Sch Comp & Control Engn Yantai 264005 Peoples R China Chinese Univ Hong Kong Sch Data Sci Shenzhen 518172 Peoples R China Univ Sci & Technol China Sch Informat Sci & Technol Hefei 230026 Peoples R China Changan Univ Sch Informat Engn Xian 710064 Peoples R China China Mobile Commun Corp Joint Lab Internet Vehicles Minist Educ Xian 710064 Peoples R China

At present, with the advance of satellite image processing technology, remote sensing images are becoming more widely used in real scenes. However, due to the limitations of current remote sensing imaging technology and the influence of the external environment, the resolution of remote sensing images often struggles to meet application requirements. In order to obtain high-resolution remote sensing images, image super-resolution methods are gradually being applied to the recovery and reconstruction of remote sensing images. The use of image super-resolution methods can overcome the current limitations of remote sensing image acquisition systems and acquisition environments, solving the problems of poor-quality remote sensing images, blurred regions of interest, and the requirement for high-efficiency image reconstruction, a research topic that is of significant relevance to image processing. In recent years, there has been tremendous progress made in image super-resolution methods, driven by the continuous development of deep learning algorithms. In this paper, we provide a comprehensive overview and analysis of deep-learning-based image super-resolution methods. Specifically, we first introduce the research background and details of image super-resolution techniques. Second, we present some important works on remote sensing image super-resolution, such as training and testing datasets, image quality and model performance evaluation methods, model design principles, related applications, etc. Finally, we point out some existing problems and future directions in the field of remote sensing image super-resolution.

关键词： image super-resolution deep learning remote sensing model design evaluation methods

来源：评论

学校读者我要写书评

暂无评论

GanoDIP - GAN Anomaly Detection through Intermediate Patches: a PCBA Manufacturing Case 3

GanoDIP - GAN Anomaly Detection through Intermediate Patches...

引用

3rd International Workshop on learning with Imbalanced Domains - Theory and Applications (LIDTA)

作者： Bougaham, Arnaud Bibal, Adrien Linden, Isabelle Frenay, Benot Univ Namur NADI Namur Belgium

Industry 4.0 and recent deep learning progress make it possible to solve problems that traditional methods could not. This is the case for anomaly detection that received a particular attention from the machine learning community, and resulted in a use of generative adversarial networks (GANs). In this work, we propose to use intermediate patches for the inference step, after aWGAN training procedure suitable for highly imbalanced datasets, to make the anomaly detection possible on full size Printed Circuit Board Assembly (PCBA) images. We therefore show that our technique can be used to support or replace actual industrial image processing algorithms, as well as to avoid a waste of time for industries.

关键词： Industry 4.0 AOI PCBA Anomaly Detection Imbalanced Dataset WGAN image processing real-World Dataset Unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Unveiling the Depths: A Comprehensive Analysis of Natural Language processing and Generative Adversarial Neural Networks for Text Generation Models in deep learning

Unveiling the Depths: A Comprehensive Analysis of Natural La...

引用

Circuits, Power and Intelligent Systems (CCPIS), International Conference on

作者： Rashi Agarwal Himanshu Agarwal Senam Pandey GEHU

deep learning comes under Machine learning that accomplishes more power and flexibility by learning to present different concepts or relations of real world to simpler concepts. We use deep learning fundaments in this paper because it has massive amount of data that helps in innovations. We include these neural networks of deep learning because it comes with a high accuracy rate with lower computations. Natural processing Language (NLP) and Generative Adversarial Network (GAN) are the methods that individually contribute to the text generation method. Although these are two different technologies giving the output for some common motive where Text generation plays a very important role in smart translations and dialogue systems. This review paper presents a model centered around text generation. This is done because combinedly we want to present what can be different approaches to look at a model like this. To solve the problem of unnecessarily used large texts, unsatisfactory feedback, NLP is used for text generation, GANN is used for text generation model, image generation etc. Finally, this is done to reduce time complexities, speed, efficiency in process because this is noticed that learning for a problem plays a vital role in education to enhance features.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A deep learning-Based Smartphone App for real-time Detection of Five Stages of Diabetic Retinopathy

A Deep Learning-Based Smartphone App for Real-Time Detection...

引用

Conference on real-time image processing and deep learning

作者： Majumder, S. Elloumi, Y. Akil, M. Kachouri, R. Kehtarnavaz, N. Univ Texas Dallas Embedded Machine Learning Lab Richardson TX 75080 USA Univ Gustave Eiffel ESIEE Paris CNRS LIGM F-77454 Marne La Vallee France Univ Monastir Fac Med LabTIM Monastir Tunisia

ISBN: (纸本)9781510635807

This paper presents the real-time implementation of deep neural networks on smartphone platforms to detect and classify diabetic retinopathy from eye fundus images. This implementation is an extension of a previously reported implementation by considering all the five stages of diabetic retinopathy. Two deep neural networks are first trained, one for detecting four stages and the other to further classify the last stage into two more stages, based on the EyePACS and APTOS datasets fundus images and by using transfer learning. Then, it is shown how these trained networks are turned into a smartphone app, both Android and iOS versions, to process images captured by smartphone cameras in real-time. The app is designed in such a way that fundus images can be captured and processed in real-time by smartphones together with lens attachments that are commercially available. The developed real-time smartphone app provides a cost-effective and widely accessible approach for conducting first-pass diabetic retinopathy eye exams in remote clinics or areas with limited access to fundus cameras and ophthalmologists.

关键词： real-time implementation of deep neural networks on smartphones real-time smartphone app for detection and classification of diabetic retinopathy first-pass eye exam by smartphone app

来源：评论

学校读者我要写书评

暂无评论

A real-time pothole detection based on deep learning approach

A real-time pothole detection based on deep learning approac...

引用

2020 International Symposium on Automation, Information and Computing, ISAIC 2020

作者： Yik, Yeoh Keng Alias, Nurul Ezaila Yusof, Yusmeeraz Isaak, Suhaila Division of Electronic and Computer Engineering and Environmental Engineering Faculty of Engineering Universiti Teknologi Malaysia Johor Bahru Johor81310 Malaysia

Today, the number of vehicles using the road including highways and single carriage way is increasing. road structure safety monitoring system that is safe for road users and also important to ensure long-term vehicle safety and prevent accidents due to road damage such as potholes, landslides and uneven roads. Most news reports of road accidents are also caused by potholes that are almost 10-30 cm deep, coupled with heavy rainfall that reduces visibility among drivers, significant damage to the suspension system to the vehicle or unnecessary traffic congestion. In this paper, deep learning detection with YOLOv3 algorithm is proposed apart from researches ranging from accelerometer detection, image processing or machine learning based detection as it is easier to develop and provide more accurate results. After pothole has been detected in real-time webcam, the location will be logged and displayed using Google Maps API for visualization. a total of 330 sets of data were sampled for the implementation of the pothole detection training model. As the results, the model provided 65.05 mAP and 0.9 % precision rate and 0.41 recall rate. The limitation of YOLOv3 algorithm detection can be improve further using GPU with higher specification performances and can sample 1000 to 10,000 datasets. The proposed algorithm provides acceptably high precision and efficient pothole monitoring solution under different scenarios for the users and may benefit the public and the government to monitor pothole in real-time. © 2021 Institute of Physics Publishing. All rights reserved.

关键词： Traffic congestion

来源：评论

学校读者我要写书评

暂无评论

基于深度学习的恶劣战场环境图像恢复方法

引用

控制与决策 2024年第4期39卷 1297-1304页

作者：孙传猛陈嘉欣裴东兴马铁华祖静任一峰中北大学省部共建动态测试技术国家重点实验室太原030051 中北大学电气与控制工程学院太原030051

为实现恶劣战场环境下降质图像的有效恢复、降低环境因素对战场态势感知的干扰,构建一种全新的、端到端的图像恢复方法——门控采样网络(GSNet).该网络以编码块-解码块为基本架构,以CNNs与门控卷积为编码与解码机制,以压缩和激励网络为... 详细信息

为实现恶劣战场环境下降质图像的有效恢复、降低环境因素对战场态势感知的干扰,构建一种全新的、端到端的图像恢复方法——门控采样网络(GSNet).该网络以编码块-解码块为基本架构,以CNNs与门控卷积为编码与解码机制,以压缩和激励网络为编码块与解码块的连接机制,以高阶信息重要程度的重标定区分目标与背景特征,以通道粒度因子压缩方法为轻量化策略,实现对战场恶劣环境图像的快速恢复.相关实验结果表明,GSNet模型可使PSNR达到19.35 dB,并且SSIM达到0.724,无论是客观指标评价还是主观视觉效果,性能均优于对比的主流图像恢复算法;轻量级GSNet模型在较小提升PSNR、SSIM等指标的情况下,其参数量、FLOPs以及单张图像处理时间分别降低56.6%、54.6%和55.56%.

关键词：图像恢复恶劣战场环境深度学习门控卷积压缩和激励网络轻量化

来源：评论

学校读者我要写书评

暂无评论

Crowd Social Distance and Mask Detection Using Classical Machine learning

Crowd Social Distance and Mask Detection Using Classical Mac...

引用

Sustainable Emerging Innovations in Engineering and Technology (ICSEIET), International Conference on

作者： Abhishek Suya Ankit Negi Mahima Bisht Shubham Parihar Mukesh Kumar Chandradeep Bhatt CSE Department Graphic Era Hill University Dehradun India Department of Computer Science Graphic Era Hill University Dehradun Uttarakhand Department of Computer Science & Engineering Graphic Era Hill University Dehradun Uttarakhand

Due to the ongoing COVID-19 pandemic's impact on public health and safety, there is an immediate requirement for creative measures to limit the transmission of the virus. In the paper, we present a computer vision-based system for detecting social distancing violations and mask-wearing compliance in crowded public spaces. The system uses a combination of deep learning algorithms and image processing techniques to analyze camera feeds and identify violations in real-time. We describe the architecture of the system, which includes a camera network, edge devices for image processing and analysis, and a central server for data management and reporting. We also evaluate the accuracy and efficiency of the system using a dataset of simulated crowd scenarios and real-world tests in public spaces.

关键词：

来源：评论

学校读者我要写书评

暂无评论

deep learning visual analysis in laparoscopic surgery: a systematic review and diagnostic test accuracy meta-analysis

引用

SURGICAL ENDOSCOPY AND OTHER INTERVENTIONAL TECHNIQUES 2021年第4期35卷 1521-1533页

作者： Anteby, Roi Horesh, Nir Soffer, Shelly Zager, Yaniv Barash, Yiftach Amiel, Imri Rosin, Danny Gutman, Mordechai Klang, Eyal Tel Aviv Univ Fac Med Tel Aviv Israel Chaim Sheba Med Ctr Dept Surg Ramat Gan Israel Chaim Sheba Med Ctr Dept Diagnost Imaging Ramat Gan Israel Chaim Sheba Med Ctr Deep Vis Lab Ramat Gan Israel Icahn Sch Med Mt Sinai Inst Healthcare Delivery Sci Dept Populat Hlth Sci & Policy New York NY 10029 USA

Background In the past decade, deep learning has revolutionized medical image processing. This technique may advance laparoscopic surgery. Study objective was to evaluate whether deep learning networks accurately analyze videos of laparoscopic procedures. Methods Medline, Embase, IEEE Xplore, and the Web of science databases were searched from January 2012 to May 5, 2020. Selected studies tested a deep learning model, specifically convolutional neural networks, for video analysis of laparoscopic surgery. Study characteristics including the dataset source, type of operation, number of videos, and prediction application were compared. A random effects model was used for estimating pooled sensitivity and specificity of the computer algorithms. Summary receiver operating characteristic curves were calculated by the bivariate model of Reitsma. Results Thirty-two out of 508 studies identified met inclusion criteria. Applications included instrument recognition and detection (45%), phase recognition (20%), anatomy recognition and detection (15%), action recognition (13%), surgery time prediction (5%), and gauze recognition (3%). The most common tested procedures were cholecystectomy (51%) and gynecological-mainly hysterectomy and myomectomy (26%). A total of 3004 videos were analyzed. Publications in clinical journals increased in 2020 compared to bio-computational ones. Four studies provided enough data to construct 8 contingency tables, enabling calculation of test accuracy with a pooled sensitivity of 0.93 (95% CI 0.85-0.97) and specificity of 0.96 (95% CI 0.84-0.99). Yet, the majority of papers had a high risk of bias. Conclusions deep learning research holds potential in laparoscopic surgery, but is limited in methodologies. Clinicians may advance AI in surgery, specifically by offering standardized visual databases and reporting.

关键词： Artificial intelligence Neural networks deep learning Computer vision Laparoscopy

来源：评论

学校读者我要写书评

暂无评论

ViMPose: Human Pose Estimation Based on Vision Mamba

ViMPose: Human Pose Estimation Based on Vision Mamba

引用

Chinese Automation Congress (CAC)

作者： Bingchuan Yang Wenyuan Cun Gang Peng Jingjing Guo Chuangye Li Jiong Zhao School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China AVIC Chengdu Aircraft Industrial (Group) Co. Ltd. Chengdu China

ISBN: (数字)9798350368604

ISBN: (纸本)9798350368611

Recently, state space models (SSM) based on efficient hardware-aware design, such as the Mamba deep learning model, have demonstrated exceptional efficacy in visual feature recognition functions. However, few studies have explored the potential of this novel architecture for pose estimation tasks. In this paper, we propose ViMPose, a baseline model for human pose estimation based on Vision Mamba. We demonstrate the model's excellent performance in pose estimation from multiple aspects, including model simplicity, inference speed, and lightweight parameters. Specifically, ViMPose employs a new backbone with bidirectional Mamba blocks to extract features from given human instances and uses a lightweight decoder for human pose estimation. Employing the scalable capacity and lightweight nature of Vision Mamba, ViMPose achieves high recognition accuracy with fewer parameters, striking a new balance between real-time efficiency and performance. Furthermore, ViMPose exhibits lower memory usage when processing high-resolution image inputs. Findings of the COCO dataset experiments highlight the ViMPose model's effectiveness and considerable promise for human pose estimation tasks.

关键词： deep learning Visualization Computer vision Accuracy Computational modeling Pose estimation Memory management Feature extraction real-time systems Decoding

来源：评论

学校读者我要写书评

暂无评论

Training deep neural networks for wireless sensor networks using loosely and weakly labeled images

引用

NEUROCOMPUTING 2021年 427卷 64-73页

作者： Zhou, Qianwei Chen, Yuhang Li, Baoqing Li, Xiaoxin Zhou, Chen Huang, Jingchang Hu, Haigen Zhejiang Univ Technol Coll Comp Sci & Technol Hangzhou 310023 Peoples R China Key Lab Visual Media Intelligent Proc Technol Zhe Hangzhou 310023 Peoples R China Chinese Acad Sci Shanghai Inst Microsyst & Informat Technol Shanghai 200050 Peoples R China

Although deep learning has achieved remarkable successes over the past years, few reports have been published about applying deep neural networks to Wireless Sensor Networks (WSNs) for image targets recognition where data, energy, computation resources are limited. In this work, a Cost-Effective Domain Generalization (CEDG) algorithm has been proposed to train an efficient network with minimum labor requirements. CEDG transfers networks from a publicly available source domain to an application specific target domain through an automatically allocated synthetic domain. The target domain is isolated from parameters tuning and used for model selection and testing only. The target domain is significantly different from the source domain because it has new target categories and is consisted of low quality images that are out of focus, low in resolution, low in illumination, low in photographing angle. The trained network has about 7 M (ResNet-20 is about 41 M) multiplications per prediction that is small enough to allow a digital signal processor chip to do real-time recognitions in our WSN. The category level averaged error on the unseen and unbalanced target domain has been decreased by 41.12%. (c) 2020 Published by Elsevier B.V.

关键词： deep neural networks Wireless sensor networks Automated data labeling image recognition Transfer learning Model compression

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：