检索结果-内蒙古大学图书馆

international conference of Electronics, Communication and Aerospace Technology (ICECA)

作者： Tummapudi Sunil Krishnagandhi Pachiappan S. Senthilrajan Y. Nagendar Renato R. Maaliw C. Pavin Department of CSE Malla Reddy Engineering College Malkajgiri Hyderabad India Electrical and Electronics Engineering Nandha Engineering College Erode Tamil Nadu India Department of Robotics& Automation Rajalakshmi Engineering College Chennai Tamil Nadu India Department of School of Computer Science and Artificial Intelligence SR University Warangal Telangana India Department of CSE Southern Luzon State University Lucban Quezon Philippines Biomedical Engineer YAR Tech Services Hyderabad India

ISBN: (数字)9798350367904

ISBN: (纸本)9798350367911

Convolutional Neural Networks (CNNs) represent a revolutionary breakthrough in improving crop productivity and sustainability when integrated into real-time monitoring systems for soil health in precision agriculture. CNNs are used in a novel way to analyze soil parameters directly from high-resolution photos taken by in-field sensors, such as moisture content, $\mathbf{p H}$ levels, and nutrient concentration. To forecast soil health measures, a CNN model was trained and validated using a collection of soil samples. the model achieved an accuracy rate of approximately $\mathbf{9 2 \%}$. the CNN-based technique shortens the time needed for soil health evaluation by almost $80 \%$ when compared to conventional approaches that involve manual sampling and laboratory analysis, allowing for quick corrective actions in the field. Additionally, more accurate and timely soil health monitoring was found to boost crop output prediction accuracy by $15 \%$. An up to $20 \%$ decrease in the amount of resources used, especially water and fertilizers, can be achieved by using real-time data processing to make educated decisions about irrigation, fertilizing, and pest management. the CNN model offers scalability and adaptability across a range of crop varieties and soil conditions, and it can be easily integrated into current agricultural IoT frameworks. All things considered, using CNNs for soil health monitoring offers an economical and effective way to improve the accuracy and output of contemporary farming methods.

关键词： Precision agriculture Irrigation Accuracy Soil Predictive models Real-time systems Data models Convolutional neural networks Monitoring Farming

来源：评论

学校读者我要写书评

暂无评论

Formally verifying consistency of sequence diagrams for safety critical systems

引用

SCIENCE OF COMPUTER PROGRAMMING 2022年 216卷 102777-102777页

作者： Chen, Xiaohong Liu, Qianqian Mallet, Frederic Li, Qin Cai, Shubin Jin, Zhi East China Normal Univ Shanghai Key Lab Trustworthy Comp Shanghai Peoples R China Univ Cote Azur CNRS Inria I3S Nancy France Shenzhen Univ Coll Comp Sci & Software Engn Shenzhen Peoples R China Peking Univ Sch Comp Sci Beijing Peoples R China

UML interactions, aka sequence diagrams, are frequently used by engineers to describe expected scenarios of good or bad behaviors of systems under design, as they provide allegedly a simple enough syntax to express a quite large variety of behaviors. this paper uses them to express safety requirements for safety critical systems in an incremental way, where the scenarios are progressively refined after checking the consistency of the requirements. the semantics of these scenarios are expressed by transforming them into an intermediate semantic model amenable to formal verification. In this paper, we rely on the Clock Constraint Specification Language (CCSL) as the intermediate semantic language. In some sense, sequence diagrams and CCSL constraints both express a family of acceptable infinite traces that must include the behaviors given by the finite set of finite execution traces against which we validate. We compare these requirements to actual execution traces to prove the validity of our transformation. As to check the consistency of the sequence diagrams, we present two verification methods based on SMT and clock graphs respectively. the SMT based method relies on our analysis tool called MyCCSL to analyze CCSL constraints. the clock graph based method transforms CCSL constraints into a clock graph, and does the analysis through traversing the clock graph. Finally, these two methods are evaluated against real cases from the railway transit systems. the results show that the SMT based method provides accurate but slow analysis, while the clock graph based method dramatically increases the verification efficiency aiming at two kinds of typical inconsistencies found by the SMT based method. (C) 2022 Elsevier B.V. All rights reserved.

关键词： Safety critical systems Sequence diagram Behavior models Clock graph Formal verification

来源：评论

学校读者我要写书评

暂无评论

Enhanced Breast Cancer Detection using ResNet50V2-based Convolutional Neural Networks

Enhanced Breast Cancer Detection using ResNet50V2-based Conv...

引用

Soft Computing for Security Applications (ICSCSA), international conference on

作者： Pratham Kaushik Sunila Choudhary Centre for Research Impact & Outcome Chitkara University Institute of Engineering and Technology Chitkara University Rajpura Punjab India Chitkara Centre for Research and Development Chitkara University Himachal Pradesh India

ISBN: (数字)9798331515720

ISBN: (纸本)9798331515737

Breast cancer is one of the leading causes of death among the female population globally; hence, there is a need for diagnostic tools with high accuracy and reliability. Conventionally used diagnostic techniques, which include mammography, ultrasound, and histopathological examination, are proven methods; however, they have disadvantages such as interpretative variability and require highly skilled pathologists. Recently developed deep learning algorithms, especially Convolutional Neural Networks, have shown huge potential for the automation of medical image analysis. this research study proposes a new model for breast cancer detection based on the ResNet50V2 architecture, which was pre-trained for image classification tasks on the ImageNet dataset but is now being used to classify histopathological images from the BreakHis dataset. the proposed model uses data augmentation techniques like random brightness adjustment, flipping, and rotation for optimum generalization before extraction of features using the ResNet50V2 base. It is then fine-tuned with fully connected layers and dropout to prevent overfitting. the performance of the model for the 25 epochs ran to a ROC-AUC of 0.84413, an accuracy of 0.79634, and a loss of 0.47527. While very promising, this showed a high number of false negatives for the model, very indicative that it requires further calibration and testing on larger and more diverse datasets. this paper, therefore, represents the potential of ResNet50V2-based models to improve aid to pathologists for the automated detection of breast cancer, consequently opening avenues toward better prognosis in patients.

关键词： Accuracy Computational modeling Feature extraction Data augmentation Breast cancer Data models Calibration Convolutional neural networks Testing Image classification

来源：评论

学校读者我要写书评

暂无评论

Modified Bilateral Control by Using Intervention Impedance Based on Passivity of Flexible Master-Slave Manipulators and Its Design methods

Modified Bilateral Control by Using Intervention Impedance B...

引用

international conference on Control, automation and Systems

作者： Takahiko Mori Department of Electronic Control Engineering Gifu National College of Technology

ISBN: (纸本)9781467322478

Recently, study on a control of flexible manipulators have been receiving increasing attension in the domain of master-slave system. the master-slave system is a control system to operate a manipulator for objects in a remote environment which is at a distance from the presence of a human operator. In order to consider the influence by the elasticity of flexible manipulators, it is necessary to study on the control of vibration of flexible master-slave manipulators and the passivity. thus in this paper, we discuss about a new modified bilateral controller considering the elasticity by using intervention impedance in basis on the passivity and a control performance of the system. this paper is organized in the following way: First, we introduce mechanical models of manipulators. Second, we propose bilateral controls by using intervention impedance based on passivity. third, we transeform each models represented by time domain into models represented by s-plain, find the scattering matrix, and prove the passivity of system. Fourth, we propose a new modified controller controller for the purpose of realizaion. Lastly, we show some results of experiments in order to verify the usefulness of proposed controller.

关键词： Tele-robotics Flexible Structure Manipulator Bilateral Control Passivity

来源：评论

学校读者我要写书评

暂无评论

IEC 61131-3 Model for Model-Driven Development

IEC 61131-3 Model for Model-Driven Development

引用

Annual conference on IEEE Industrial Electronics Society

作者： Monika Wenger Alois Zoitl Vienna University of Technology Automation and Control Institute

ISBN: (纸本)9781467324199

Existing industrial automation projects require automated methods for faster changes and updates as well as continuous improvement and maintenance. Software Engineering uses Model Driven Development to shorten the development time by keeping documentation, models and code consistent during the whole development time and by using code generators to create many similar code artifacts and code for different platforms. To make this concept also available for automation projects we present an IEC 61131 model based on the Extended Backus Naur From provided by the standard. For the proposed IEC 61131 model also related application areas are outlined.

关键词： development time international Electrotechnical Commission Code generation Codes Automated methods Continuous improvement automation Industrial automation

来源：评论

学校读者我要写书评

暂无评论

ARSFineTune: On-the-Fly Tuning of Vision models for Unmanned Ground Vehicles

ARSFineTune: On-the-Fly Tuning of Vision Models for Unmanned...

引用

international conference on Distributed Computing in Sensor Systems (DCOSS)

作者： Masud Ahmed Zahid Hasan Abu Zaher Md Faridee Mohammad Saeid Anwar Kasthuri Jayarajah Sanjay Purushotham Suya You Nirmalya Roy University of Maryland Baltimore County USA New Jersey Institute of Technology USA DEVCOM Army Research Laboratory USA

ISBN: (数字)9798350369441

ISBN: (纸本)9798350369458

the performance of semantic segmentation (SS) can degrade when the data distribution in the deployed environment is different from what the model initially learned during their training. While domain adaptation (DA) and continual learning (CL) methods have been proposed to improve performance in new or unseen domains over time, the effort required for annotating large swathes of training data during deployment is non-trivial; acquiring new data for training incurs both significant network, device memory costs, and manual effort for labeling. To address this, we propose ARSFineTune, a novel framework that actively selects the most informative regions of visuals encountered by a mobile robot for the CL network to learn from, greatly minimizing the data transfer overhead related to annotations. We first propose a proficient entropy-driven ranking mechanism to identify candidate regions and rank challenging images at the edge node. We then facilitate a cyclical feedback loop between the server and edge, continuously refining the accuracy of semantic segmentation by fine-tuning the model with minimal transferred data to/from the field deployed device. We implement ARSFineTune in a real-time setting using the robotics Operating System (ROS), where a Jackal (an unmanned ground vehicle - UGV) collaborates with the central server. through extensive experiments, we found that ARSFineTune delivers competitive performance, closely aligning with existing state-of-the-art techniques, while requiring substantially less data for fine-tuning. Specifically, with only 5% of the total labeled regions (25% challenging regions of the most 20% problematic samples) of the entire dataset for fine-tuning, ARSFineTune reaches a performance level nearly identical (≈ 97%) to the previous state-of-the-art model, boasting mIoU scores of 59.5% on the Cityscape dataset and 41% on the CAD-EdgeTune dataset, which is a challenging dataset due to varying lighting conditions over time. the reduction in

关键词： Training Performance evaluation Annotations Semantic segmentation Operating systems Image edge detection Memory management

来源：评论

学校读者我要写书评

暂无评论

Computational approaches to Explainable Artificial Intelligence: Advances in theory, applications and trends

引用

INFORMATION FUSION 2023年第1期100卷

作者： Gorriz, J. M. Avarez-Illan, I. Avarez-Marquina, A. Arco, J. E. Atzmueller, M. Ballarini, F. Barakova, E. Bologna, G. Bonomini, P. Castellanos-Dominguez, G. Castillo-Barnes, D. Cho, S. B. Contreras, R. Cuadra, J. M. Dominguez, E. Dominguez-Mateos, F. Duro, R. J. Elizondo, D. Fernandez-Caballero, A. Fernandez-Jover, E. Formoso, M. A. Gallego-Molina, N. J. Gamazo, J. Gonzalez, J. Garcia Garcia-Rodriguez, J. Garre, C. Garrigos, J. Gomez-Rodellar, A. Gomez-Vilda, P. Grana, M. Guerrero-Rodriguez, B. Hendrikse, S. C. F. Jimenez-Mesa, C. Jodra-Chuan, M. Julian, V. Kotz, G. Kutt, K. Leming, M. de Lope, J. Macas, B. Marrero-Aguiar, V. Martinez, J. J. Martinez-Murcia, F. J. Martinez-Tomas, R. Mekyska, J. Nalepa, G. J. Novais, P. Orellana, D. Ortiz, A. Palacios-Alonso, D. Palma, J. Pereira, A. Pinacho-Davidson, P. Pinninghoff, M. A. Ponticorvo, M. Psarrou, A. Ramirez, J. Rincon, M. Rodellar-Biarge, V. Rodriguez-Rodriguez, I. Roelofsma, P. H. M. P. Santos, J. Salas-Gonzalez, D. Salcedo-Lagos, P. Segovia, F. Shoeibi, A. Silva, M. Simic, D. Suckling, J. Treur, J. Tsanas, A. Varela, R. Wang, S. H. Wang, W. Zhang, Y. D. Zhu, H. Zhu, Z. Ferrandez-Vicente, J. M. Univ Granada Data Sci & Computat Intelligence Inst SiPBA Granada Spain Univ Granada Data Sci & Computat Intelligence Inst DaSCII Granada Spain Univ Cambridge Dept Psychiat Cambridge England Univ Malaga Dept Commun Engn Malaga Spain Univ Leicester Sch Comp & Math Sci Leicester Leics England Univ Osnabruck Semant Informat Syst Grp Osnabruck Germany German Res Ctr AI DFKI Osnabruck Germany Univ Politecn Cartagena Dept Elect Comp Technol & Projects Cartagena Spain Univ Politecn Valencia Valencian Res Inst AI VRAIN Valencia Spain Jagiellonian Univ Jagiellonian Human Ctr AI Lab JAHCAI Krakow Poland Jagiellonian Univ Inst Appl Comp Sci Krakow Poland Univ Murcia Dept Informat & Commun Engn Murcia Spain Univ A Coruna Dept Comp Sci & Informat Technol Ctr Informat & Commun Technol Res CITIC La Coruna Spain Univ Oviedo Dept Comp Sci Oviedo Spain Inst Argentino Matemat Alberto Calderon Buenos Aires DF Argentina Consejo Nacl Invest Cient & Tecn CONICET Buenos Aires DF Argentina Inst Tecnol Buenos Aires ITBA Buenos Aires DF Argentina Eindhoven Univ Technol NL-5612 AZ Eindhoven Netherlands Univ Malaga Dept Comp Sci Malaga Spain De Montfort Univ Leicester Leics England Univ Concepcion Dept Comp Sci Fac Engn Concepcion Chile Univ Castilla La Mancha Albacete Res Inst Informat Neurocognit & Emot Res Unit Albacete Spain Univ Concepcion Dept Educ Informat Concepcion Chile Univ Nacl Educac Distancia UNED Dept AI Madrid Spain Univ Naples Federico II Dept Humanist Studies Naples Italy Univ Rey Juan Carlos Escuela Tecn Super Ingn Informat Campus Mostoles Madrid 28933 Spain Univ Politecn Madrid Neuromorph Speech Proc Lab Ctr Biomed Technol Campus Montegancedo Madrid 28223 Spain Inst Salud Carlos III Biomed Res Networking Ctr Mental Hlth Madrid 28016 Spain Univ Edinburgh Usher Inst Fac Med Edinburgh Midlothian Scotland Univ Miguel Hernandez Inst Bioingn Alicante Spain Univ Alicante Dept Comp Technol Alicante Spain Univ Cent Ecuador Quito E

Deep Learning (DL), a groundbreaking branch of Machine Learning (ML), has emerged as a driving force in both theoretical and applied Artificial Intelligence (AI). DL algorithms, rooted in complex and non-linear artificial neural systems, excel at extracting high-level features from data. DL has demonstrated human -level performance in real-world tasks, including clinical diagnostics, and has unlocked solutions to previously intractable problems in virtual agent design, robotics, genomics, neuroimaging, computer vision, and industrial automation. In this paper, the most relevant advances from the last few years in Artificial Intelligence (AI) and several applications to neuroscience, neuroimaging, computer vision, and robotics are presented, reviewed and discussed. In this way, we summarize the state-of-the-art in AI methods, models and applications within a collection of works presented at the 9th international conference on the Interplay between Natural and Artificial Computation (IWINAC). the works presented in this paper are excellent examples of new scientific discoveries made in laboratories that have successfully transitioned to real-life applications.

关键词： Explainable Artificial Intelligence Data science Computational approaches Machine learning Deep learning Neuroscience robotics Biomedical applications Computer-aided diagnosis systems

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：