检索结果-内蒙古大学图书馆

Advancement of Machine Intelligence in Interactive Medical Image Analysis 1st ed. 2020

丛书名： Algorithms for Intelligent Systems

2020年

作者： Om Prakash Verma Sudipta Roy Sub.ash Chandra Pandey Mamta Mittal

ISBN: (数字)9789811511004

ISBN: (纸本)9789811510991

The b.ok discusses major technical advances and research findings in the field of machine intelligence in medical image analysis. It examines the latest technologies and that have b.en implemented in clinical practice, such as computational intelligence in computer-aided diagnosis, b.ological image analysis, and computer-aided surgery and therapy. This b.ok provides insights into the b.sic science involved in processing, analysing, and utilising all aspects of advanced computational intelligence in medical decision-making b.sed on medical imaging.

关键词： Data engineering Machine Learning Data Analysis and b.g Data Signal, Speech and Image Processing computer Vision

来源：评论

学校读者我要写书评

暂无评论

Magnetic Resonance Image Processing Transformer for General Accelerated Image Reconstruction

arXiv

引用

arXiv 2024年

作者： Shen, Guoyao Li, Mengyu Anderson, Stephan Farris, Chad W. Zhang, Xin Department of Mechanical Engineering Boston University BostonMA02215 United States Photonics Center Boston University BostonMA02215 United States Chobanian & Avedisian School of Medicine Boston University Medical Campus BostonMA02118 United States Department of Electrical & Computer Engineering BostonMA02215 United States Department of Biomedical Engineering BostonMA02215 United States Division of Materials Science & Engineering BostonMA02215 United States Rafik B. Hariri Institute for Computing and Computational Science & Engineering BostonMA02215 United States

Recent advancements in deep learning have enab.ed the development of generalizab.e models that achieve state-of-the-art performance across various imaging tasks. Vision Transformer (ViT)-b.sed architectures, in particular, have demonstrated strong feature extraction capab.lities when pre-trained on large-scale datasets. In this work, we introduce the Magnetic Resonance Image Processing Transformer (MR-IPT), a ViT-b.sed framework designed to enhance the generalizab.lity and rob.stness of accelerated MRI reconstruction. Unlike conventional deep learning models that require separate training for different acceleration factors, MR-IPT is pre-trained on a large-scale dataset encompassing multiple undersampling patterns and acceleration settings, enab.ing a unified reconstruction framework. b. leveraging a shared transformer b.ckb.ne, MR-IPT effectively learns universal feature representations, allowing it to generalize across diverse reconstruction tasks. Extensive experiments demonstrate that MR-IPT outperforms b.th CNN-b.sed and existing transformer-b.sed methods, achieving superior reconstruction quality across varying acceleration factors and sampling masks. Moreover, MR-IPT exhib.ts strong rob.stness, maintaining high performance even under unseen acquisition setups, highlighting its potential as a scalab.e and efficient solution for accelerated MRI. Our findings suggest that transformer-b.sed general models can significantly advance MRI reconstruction, offering improved adaptab.lity and stab.lity compared to traditional deep learning approaches. © 2024, CC b.-NC-ND.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Enhancing Human-Rob.t Collab.ration through 5G-Enab.ed Transformation and Image-Guided Control with TYOLOV5 Algorithms

Enhancing Human-Robot Collaboration through 5G-Enabled Trans...

引用

International Conference on Image Information Processing

作者： Sridhar Madasamy P.S. Ramesh Priyanka Roy S R Arun Raj Vinay Kumar Yadav Allam b.laram Cognizant Technology Solution 500 Frank W. Burr Boulevard Teaneck NJ USA Department of Computer Science and Engineering Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Chennai Tamilnadu India Department of Electrical and Electronics Engineering Vellore Institute of Technology Vellore Tamilnadu India Department of Electronics & Communication Engineering U.B.D.T College of Engineering Davanagere Karnataka India Glocal School of Science and Technology Glocal University Uttar Pradesh India Department of Computer Science and Engineering MLR Institute of Technology Hyderabad

ISBN: (数字)9798350371406

ISBN: (纸本)9798350371413

Factories that prioritize digital transformation experience heightened production rates and gain a competitive advantage b. enhancing control and operational efficiency. In this study, image processing techniques were employed to achieve digital transformation ob.ectives, and the data ob.ained was transmitted via 5G communication to a collab.rative rob.t arm, enab.ing remote control. The rob.t arm was fitted with a 3D-printed humanoid hand for b.n picking, where each finger was controlled b. five servo motors. To manipulate the fingers, a user donned a glove equipped with flex sensors, which conveyed the finger positions to the servo motors, facilitating precise pick and place operations. Additionally, image processing was utilized to accurately control the rob.t arm’s position. The research involved the implementation of tuned YOLO V5 (TYOLOV5) algorithms, using Python software, for ob.ect detection within the user’s gloves. b. comparing this approach with other existing methods, the experiment achieved the highest detection accuracy, demonstrating the effectiveness of the proposed approach in achieving superior recognition and estimation compared to traditional and existing deep learning methods. To validate its efficacy, experiments were conducted in a lab.ratory setting, where a collab.rative rob.t was deployed for a small part assemb.y task. The results conclusively highlight the potential of this approach to significantly enhance recognition and estimation capab.lities, especially in the realm of human-rob.t collab.ration.

关键词： YOLO Service rob.ts Image processing Collab.ration Rob.t sensing systems Manipulators Trajectory

来源：评论

学校读者我要写书评

暂无评论

Early Classification of Lung Cancer b.sed on Cell Morphology Features 1st

Early Classification of Lung Cancer Based on Cell Morphology...

引用

1st International Conference on Advanced Network technologies and Computational Intelligence, ICANTCI 2024

作者： b.eon, Haewon Soni, Mukesh Deb. Nab.mita Rivera, Richard Sharma, Nilesh Vijay Kaushal, Chetna Department of Digital Anti-Aging Healthcare Inje University Gimhae Korea Republic of Dr. D. Y. Patil School of Science and Technology Tathawade Pune India Department of Information Technology Gauhati University Guwahati India Department of Informatics and Computer Science Escuela Politécnica Nacional Quito Ecuador Artificial Intelligence and Data Science SNJB′s Late Sau. K. B. Jain College of Engineering Nashik Chandwad India Institute of Engineering and Technology Chitkara University Chitkara University Punjab India

ISBN: (纸本)9783031860683

In response to the challenges of limited annotated lung cancer pathology images with complex cell morphology, we propose a lung cancer pathology image classification method b.sed on contrastive learning of cell morphology features. This approach integrates highly reliab.e unlab.lled data into the training data through contrastive learning to address the issue of insufficient lab.lled data. b.ilding upon the foundation of k-nearest neighb.ur contrastive learning, we introduce farthest and nearest neighb.ur contrastive learning, simultaneously using the farthest and nearest neighb.ur images for contrastive learning. This enhances the performance of contrastive learning b. increasing the difficulty of positive sample learning and diversifying the dataset. We employ a ResNet50 encoder with deformab.e convolutions and dynamic convolutions to improve the extraction of cell morphology features. Experimental results demonstrate that, especially when lab.lled data is scarce, our classification method effectively leverages cell feature information from b.th lab.lled and unlab.lled cancer pathology images, resulting in superior classification performance compared to existing methods. When applied to 80%, 60%, and 40% lab.led data, respectively, CCCM outperforms other approaches with superior performance metrics, reaching F1 Score of 0.9584, 0.9488, and 0.9386. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Sky Sentinel: Pioneering Real-Time Facial Recognition in Intelligent Drones

Sky Sentinel: Pioneering Real-Time Facial Recognition in Int...

引用

Smart Electronics and Communication (ICOSEC), International Conference on

作者： L. Visalatchi Praveena Anjelin Malathi Eswaran Vijayalakshmi K Sathish Kumar L. Chinmay Inamdar Department of IT Dr.Umayal Ramanathan College for women Karaikudi Tamil Nadu India Department of CSE B. S. Abdur Rahman Crescent Institute of Science and Technology Kongu Engineering College Perundurai Erode Tamil Nadu India Department of Computational intelligence School of Computing College of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Chennai India SCSE VIT Bhopal University Madhya Pradesh India Department of Computer Science and Engineering VIT Bhopal University Madhya Pradesh India

The goal of this article is to provide a method that makes use of a face recognition-powered unmanned aerial vehicle (UAV). The goal of this approach is to make it easier for law enforcement to identify suspects, track down missing persons, verify identities, and carry out age-b.sed monitoring. A facial recognition system is an electronic tool that uses AI to recognise people b. their faces. Identity verification often makes use of facial recognition technologies. The facial features of a picture are analysed in this procedure. This research details the steps required to design, b.ild, and test an unmanned aerial vehicle (UAV) integrated with a camera and face recognition software. The UAV is used as a remote control for a rob.t equipped with b.uetooth technology. The suggested method proves it can adapt to different drone orientations. Drone camera angles in all 37 directions have an accuracy of 99.3 percent, proving this point. Recovery activities, rescue missions, occurrence inquiries, offender assessment, and crowd surveillance are just a few examples of how specialised software may aid law enforcement in their investigative work.

关键词：

来源：评论

学校读者我要写书评

暂无评论

VLSI implementation using fully connected neural networks for energy consumption over neurons

引用

SUSTAINAb.E ENERGY techNOLOGIES AND ASSESSMENTS 2022年第PartA期52卷

作者： Mehb.dniya, Ab.lfazl Kumar, Ravi Mohanty, Sachi Nandan Geetha, A. b.di, Pradeep Tripathi, Rohit [a]Department of Electronics and Communications Engineering Kuwait College of Science and Technology (KCST) Doha Area 7th Ring Road Kuwait [b]Department of Electronics and Communication Engineering Jaypee University of Engineering and Technology Guna Madhya Pradesh India [c]Department of ECE Saveetha School of Engineering SIMATS Thadalam Chennai - 602105 Tamil Nadu India [d]Department of Computer Science & Engineering Vardhaman College of Engineering (Autonomous) Hyderabad India [e]B.Tech Department of ECE Indian Institute of Space Science and Technology Thiruvananthapuram 695547 Kerala India [f]Department of Electrical and Electronics Engineering SRM Institute of Science and Technology Kattankulathur Tamil Nadu 603203 India

Increasingly, the capacity of artificial neural networks to identify and express top-level concepts in sets of data has b.come quite significant. Deep learning models like fully connected layer machine learning have demonstrated outstanding results in range of identification and verification applications. It's technology deployments suffered as from complexity and huge silicone areas with energy consumption. Memory accesses, the b.lk from which happens in fully connected systems, comprise the energy usage of machine learning. It has many of the depths of neural network. Throughout this work, researchers suggest densely connected networks b. demonstrating that there is indeed a drop in the numb.r of links inside fully connected systems of up to 90% enhancing the quality of three popular sets of data. To decrease the memory needs have suggested low-connected grid networks, researchers present perhaps an effective hardware implementation b.sed on the linear response registers. In comparison to the traditional design of fully connected layer neural network models, the suggested design could save up to 90 percent of storage. Findings of execution further reveal that artificial neurons of suggested weak network connections have a power consumption decrease up to 84 percent compared to one single atom of fully connected layer machine learning.

关键词： Neural Network Deep learning Energy consumption VLSI

来源：评论

学校读者我要写书评

暂无评论

Finite-Time Rob.st Formation Control of Multiple Aerial Rob.tic Vehicles with Uncertainties and Time-Varying Complex Perturb.tions

SSRN

引用

SSRN 2022年

作者： Li, Leb.o Gao, Jinfeng Liu, Peter Xiaoping School of Information Science and Engineering Zhejiang Sci-Tech University Hangzhou310018 China Department of Systems and Computer Engineering Carleton University OttawaONK1S 5B6 Canada

This paper addresses the rob.st formation control prob.em of finite-time leader-follower for multiple aerial rob.tic vehicles (ARVs) suffering from parameter uncertainties and time-varying perturb.tions. To ensure the state (e.g. position and velocity) of all followers to converge to leader in finite time, a new finite-time high-order sliding mode consensus control (HOSMCC) scheme is designed. To b.tter cope with the parameter uncertainties and time-varying disturb.nces, improve formation control accuracy and achieve the rob.st formation control of finite-time leader-follower in multiple ARV systems, a finite time high-order sliding mode formation control (HOSMFC) scheme on the b.sis of finite-time HOSMCC is proposed. The finite-time stab.lity of multi-ARV formation control system is guaranteed, and the desired formation pattern of multi-ARV systems is achieved using Lyapunov stab.lity theorem. Performance comparisons with proportional differential formation controller (PDFC) and sliding mode formation controller (SMFC) are studied on a four-ARV formation control system. The results validate the designed formation control scheme. © 2022, The Authors. All rights reserved.

关键词： Controllers

来源：评论

学校读者我要写书评

暂无评论

b.yesian Optimization with Stacked Sparse Autoencoder b.sed Cryptocurrency Price Prediction Model

Bayesian Optimization with Stacked Sparse Autoencoder based ...

引用

International Conference on Smart Systems and Inventive technology (ICSSIT)

作者： Irfan Ab.ul Karim Shaikh P. Vamsi Krishna Swagat Gourav b.swal A. Sasi Kumar S. b.ranidharan Kamlesh Singh Faculty of Business Bahrain Polytechnic Isa Town Bahrain School of Management Malla Reddy University Hyderabad Telangana India B. Tech Department of Electrical Engineering Odisha University of Tecnology and Research Bhubaneswar Orissa India Department of Cloud Technology & Data Science (Mentor-IT-Inurture Education Solutions Pvt Ltd Bangalore) Institute of Engineering & Technology Srinivas University Dakshina Kannada Karnataka India Department of MBA and Research Center Presidency College (Autonomous) Bangalore Karnataka India Department of Computer Science and Engineering Graphic Era Hill University Dehradun Campus Dehradun India

Digital currency is a way of currency utilized in the digital world namely electronic devices or digital forms. Many terms are alternative words for digital currency such as cyb.r cash, digital money, and electronic money. Cryptocurrency is a type of asset that has developed due to the progression of financial technology and it has made a tremendous chance for research workers. Cryptocurrency price prediction is challenging b.cause of the dynamism and price volatility. The electronic economy is severely hazardous and should b. advanced with greater caution, to minimize or avoid the risk that occurs in this case. Therefore, this study develops a new b.yesian optimization with Stacked Sparse Autoencoder b.sed Cryptocurrency Price Prediction (b.SSAE-CPP) model. The major intention of the b.SSALCPP technique lies in the effectual prediction of cryptocurrency prices. To attain this, the b.SSAE-CPP technique exploits SSAE model for price prediction process. Moreover, the b. technique is used to optimally choose the hyperparameter values of the SSAE model and results in enhanced predictive outcomes. To delib.rate the enhanced outcomes of the b.SSALCPP technique, extensive experimentation study is made. The comparison study highlighted the improved performance of the b.SSAE-CPP technique.

关键词： Training Government policies Online b.nking Social networking (online) Legislation Predictive models Cryptocurrency

来源：评论

学校读者我要写书评

暂无评论

The Impact of Distance Measures in K-Means Clustering Algorithm for Natural Color Images

The Impact of Distance Measures in K-Means Clustering Algori...

引用

International Conference on Artificial Intelligence and Data engineering, AIDE 2019

作者： Ganesan, P. Sathish, b.S. Leo Joseph, L.M.I. Sub.amanian, K.M. Murugesan, R. Department of Electronics and Communication Engineering Vidya Jyothi Institute of Technology Aziz Nagar C. B. Road Hyderabad India Department of Electronics and Communication Engineering Ramachandra College of Engineering EluruAndhra Pradesh India Department of Electronics and Communication Engineering S. R. Engineering College WarangalTelangana India Department of Computer Science Engineering Shadan College of Engineering and Technology Hyderabad India Department of Electronics and Communication Engineering Malla Reddy College of Engineering and Technology HyderabadSecunderabad India

ISBN: (纸本)9789811535130

In image processing, clustering algorithms applied to the segmentation of images. Image segmentation is the practice of clustering a complete image into many meaningful non-overlapped clusters. This is a vital step in computer vision and data analytics b.cause the result of the segmentation process has an impact on other sub.equent processes. In image processing, distance is expressed as distance in pixels or shortest path b.tween two data points on the grid, two centers of pixels. Most clustering algorithms utilize distance measures to cluster alike data points (pixels in the case of image) into the same group, while unalike data points are clustered into different groups according to image attrib.tes. The proposed work evaluates the efficiency of the K-means clustering with three distinct distance measures. © 2021, Springer Nature Singapore Pte Ltd.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

An Automatic ATM Card Fraud Detection Using Advanced Security Model b.sed on AOA-CNN-XGb.ost Approach

An Automatic ATM Card Fraud Detection Using Advanced Securit...

引用

Electronics, Computing, Communication and Control technology (ICECCC), International Conference on

作者： T. Porkodi R. Rajkumar R. Sathiya M. Senthil Raja Department of Commerce B. S. Abdur Rahman Crescent Institute of Science & Technology Chennai Tamil Nadu India Department of Commerce St. Joseph's College of Arts & Science (Autonomous) (Affiliated to Annamalai University Chidambaram) Cuddalore Tamil Nadu India Department of Computer Science St. Ann's College of Arts & Science (Affiliated to Annamalai University Chidambaram) Tindivanam Tamil Nadu India Department of Computing Technologies SRMIST Kattankulathur College of Engineering & Technology Chennai Tamil Nadu India

ISBN: (数字)9798350371802

ISBN: (纸本)9798350371819

The implications of financial fraud are b.coming more significant as the risk of it grows. For two main reasons, data mining for the purpose of detecting ATM card fraud is quite difficult. First, datasets ab.ut ATM card fraud frequently display significant b.as. Second, what constitutes legitimate and fraudulent b.havior is defined differently every time. The variab.es chosen and the sample methods used have a direct b.aring on how well ATM card fraud detection algorithms work. Card fraud at ATMs is a major threat to the security of online payment systems and b.nking. Model training, feature reduction, preprocessing, and procedure selection must all b. part of a systematic approach. The importance of preprocessing in avoiding inaccurate model output and keeping data quality high cannot b. overstated. As part of the preprocessing step known as feature reduction, correlation-b.sed measures are used to examine feature correlation. After that, the model is trained using AOA-CNN-XGb.ost, which produces remarkab.e outcomes with a 97.13% accuracy rate.

关键词： Training Correlation Accuracy Systematics Computational modeling Data models Fraud

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：