检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Zhang, Wenxiao Kong, Xiangrui Braunl, Thomas Hong, Jin B. Dept. of Computer Science and Software Engineering The University of Western Australia Perth Australia Dept. of Electrical Electronic and Computer Engineering The University of Western Australia Perth Australia

Embodied AI systems, including AI-powered robots that autonomously interact with the physical world, stand to be significantly advanced by Large Language Models (LLMs), which enable robots to better understand complex language commands and perform advanced tasks with enhanced comprehension and adaptability, highlighting their potential to improve embodied AI capabilities. However, this advancement also introduces safety challenges, particularly in robotic navigation tasks. Improper safety management can lead to failures in complex environments and make the system vulnerable to malicious command injections, resulting in unsafe behaviours such as detours or collisions. To address these issues, we propose SafeEmbodAI, a safety framework for integrating mobile robots into embodied AI systems. SafeEmbodAI incorporates secure prompting, state management, and safety validation mechanisms to secure and assist LLMs in reasoning through multi-modal data and validating responses. We designed a metric to evaluate mission-oriented exploration, and evaluations in simulated environments demonstrate that our framework effectively mitigates threats from malicious commands and improves performance in various environment settings, ensuring the safety of embodied AI systems. Notably, In complex environments with mixed obstacles, our method demonstrates a significant performance increase of 267% compared to the baseline in attack scenarios, highlighting its robustness in challenging conditions. © 2024, CC BY.

关键词： Mobile robots

来源：评论

学校读者我要写书评

暂无评论

Automation technology of seismic damage investigation for timber houses using deep learning

引用

AIJ Journal of Technology and Design 2021年第67期27卷 1578-1583页

作者： Yamada, Tomoyuki Takahashi, Noriyuki Chida, Hiroyuki Dept. of Civil Engineering and Architecture Tohoku Univ. Japan Dept. of Architecture and Building Science Tohoku Univ. Japan

In this paper, to achieve the fast and detailed evaluation of seismic damage, the novel diagnostic imaging applied multiple image recognition techniques (Classification, Detection, Segmentation) in deep learning is proposed. Dataset from images generated by Generative Adversarial Network (GAN) brings the high-accurate recognition of the real damaged images. It is revealed that the diagnostic imaging can extract precisely the damaged area with minimum noise. Furthermore, the result of the damage extraction can calculate the seismic damage rate of the exterior wall. © 2021 Architectural Institute of Japan. All rights reserved.

关键词： Timber

来源：评论

学校读者我要写书评

暂无评论

A Lightweight-CNN Model for Efficient Lung Cancer Detection and Grad-CAM Visualization

A Lightweight-CNN Model for Efficient Lung Cancer Detection ...

引用

Information and Communication Technology for Sustainable Development (ICICT4SD), International Conference on

作者： Tasmim Sagor Chandro Bakchy Hasibul Islam Peyal Md. Injamul Islam Golam Kibria Yeamin Shafiun Miraz Md. Nazmul Abdal Dept. of Electrical & Computer Engineering Rajshahi University of Engineering & Technology Rajshahi Bangladesh Dept. of Computer Science & Engineering Uttara University Dhaka Bangladesh Dept. of Electrical & Electronic Engineering Rajshahi University of Engineering & Technology Rajshahi Bangladesh Dept. of Computer Science & Engineering Rajshahi University of Engineering & Technology Rajshahi Bangladesh

The lungs' abnormal cell growth leads to the development of lung cancer. Early cancer identification could make treatment easier, potentially saving millions of lives annually. This study's main goal is to more rapidly and effectively classify various types of lung cancer by employing a lightweight, computationally efficient convolutional neural network (CNN) model to categorize three different types of lung cancer. With an outstanding validation accuracy of 99.48%, the suggested model surpasses the achievements of previous works. The 15,000 CT scan images in our dataset include three different forms of lung cancer. The proposed model performs exceptionally well, as evidenced by its astounding precision, recall, and F1-score, all above 99%, and by its flawless Area Under Curve (AUC) score of 100%. The proposed model has fewer parameters than the existing transfer learning models. Gradient Weighted Class Activation Mapping (Grad-CAM) was used to create class activation maps, which were then used to create a heatmap to display the classification zone.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-Target Integrated Sensing and Communication with MIMO In-Band Full-Duplex Systems 58

Multi-Target Integrated Sensing and Communication with MIMO ...

引用

58th Asilomar Conference on Signals, Systems and computers, ACSSC 2024

作者： Chih, Yung-Chih Shen, Chung-An Celik, Abdulkadir Fouda, Mohammed E. Eltawil, Ahmed National Taiwan University of Science and Technology Electronic and Computer Engineering Dept. Taipei Taiwan King Abdullah University of Science and Technology Thuwal Saudi Arabia Rain Neuromorphics Inc. San FranciscoCA United States

ISBN: (纸本)9798350354058

In recent years, the integration of Multi-Input Multi-Output (MIMO) technology with In-Band Full-Duplex (IBFD) systems has emerged as a promising approach for multi-targets Integrated Sensing and Communication (ISAC), particularly in the context of 6G and vehicle-to-everything (V2X) communications. However, the simultaneous transmission and reception of signals in IBFD systems face significant challenges, mainly due to self-interference (SI), which severely impacts system performance. This paper presents a novel framework that combines self-interference cancellation (SIC) techniques with the MIMO-IBFD-ISAC system to enhance multi-target sensing capabilities. The framework utilizes an independent component analysis (ICA) algorithm to effectively separate self-interference from the desired signals. Additionally, a specialized multi-targets detection method is employed to recover radar signals reflected from targets accurately. Simulation results demonstrate that the proposed approach improves target detection accuracy and communication efficiency, making it a viable solution for next-generation ISAC systems. © 2024 IEEE.

关键词： Radar interference

来源：评论

学校读者我要写书评

暂无评论

Optimizing Multi-Class Classification in Educational Data with Ensemble Learning and Data Balancing Techniques

Optimizing Multi-Class Classification in Educational Data wi...

引用

Intelligent Data science Technologies and Applications (IDSTA), International Conference on

作者： Mohammad F. Al-Hammouri Ziad Akram Ali Hammouri Islam T. Almalkawi Ansam Lafee Dept. of Computer Engineering The Hashemite University Zarqa Jordan Dept. of Computer Science Middle East University Amman Jordan

ISBN: (数字)9798350354751

ISBN: (纸本)9798350354768

The field of education is increasingly embracing AI tools to improve student outcomes. This work aims to reduce academic failure in higher education by employing machine learning techniques to identify at-risk students early in their educational journey, enabling the implementation of supportive strategies to assist them. This study examines a dataset from a higher education institution and utilizes it to develop a classification model for predicting students’ academic performance. The problem is formulated as a multi-class classification task with three categories: Graduate, Enrolled, and Dropout, with a significant imbalance skewed toward the Graduate. To improve prediction accuracy toward the minority class, the data balancing technique SMOTE with Edited Nearest Neighbor (SMOTE-ENN) is applied. Three popular classification models—Random Forest, XGBOOST and CatBoost—are employed. The findings show that SMOTE-ENN significantly improves classification results. Moreover, XGBOOST demonstrated the highest accuracy (94.6%) in correctly identifying all classes, as evidenced by the confusion matrix evaluation, achieving the highest results compared to previous work in the literature. Implementing these models allows for accurate predictions of students’ performance and helps reduce dropout rates.

关键词： Performance evaluation Accuracy Machine learning algorithms Education Forestry Predictive models Data science Prediction algorithms Data models Ensemble learning

来源：评论

学校读者我要写书评

暂无评论

Technological Innovations in IoT-Based Smart Agriculture Systems for Sustainable Crop Management 4

Technological Innovations in IoT-Based Smart Agriculture Sys...

引用

4th International Conference on Technological Advancements in Computational sciences, ICTACS 2024

作者： Padmapriya, S. Todmal, Ashok Veeraiah, Vivek Rajput, Bhawna Janghel Muthugurunathan, G. Rakshitha Dhanalakshmi Srinivasan University Dept. of Cse Tamil Nadu India Mit College of Management Mit Art Design and Technology University Maharashtra Pune India Sri Siddhartha Academy of Higher Education Dept. of Computer Science Karnataka India Rungta College of Engineering and Technology Dept. of Cse Chhattisgarh Bhilai India Madanapalle Institute of Technology & Science Dept. of Cse AP Madanapalle India Nmamit Dept. of Aiml Nitte India

ISBN: (纸本)9798350387490

Modern farming is now achievable because to the IoT, which has revolutionised agricultural management. investigates the most current IoT based smart agriculture system technologies in view of sustainable crop management. Combining IoT sensors with data analytics found in studies that precision farming, automated irrigation systems, real-time soil monitoring, coupled with resource optimisation may be accomplished. IoT technology might help farmers to better regulate their water, pest management, along with soil health, thus improving their yield with less impact on the environment. Exposing upcoming trends along with research opportunities investigates cost difficulties, system integration, data security. This paper provides understanding of how IoT technology is changing sustainable agriculture by way of a thorough review of present developments and case studies. © 2024 IEEE.

关键词： Smart agriculture

来源：评论

学校读者我要写书评

暂无评论

Coupled Electric-Thermal Damage Modeling and Machine Learning for Lightning Strikes on Soil

Coupled Electric-Thermal Damage Modeling and Machine Learnin...

引用

International Conference on Big Data and Information Analytics (BigDIA)

作者： Xiao Jin Pingping Rao Wencan Guar Dept. Department of Civil Engineering University of Shanghai for Science and Technology Shanghai China

ISBN: (数字)9798350354621

ISBN: (纸本)9798350354638

To study the damage mechanism of soil when struck by lightning, this paper proposes an electro-thermal coupled numerical model of soil for solving the problem. The double exponential function is used to represent the naturally occurring lightning waveforms. The novel numerical computational model indicates the electrical disasters and damage mechanism of soil under lightning action. The calculation results show that under the action of lightning, the soil produces electrical penetration and causes thermal damage. In addition, a machine learning strategy has been devised to evaluate changes in soil damage conditions, and a formula for soil damage tensor related to the basic electrical parameters of soil is provided. The innovative numerical modeling reveals the mechanism of soil failure and electrical penetration during lightning strikes and and provides theoretical support for mine mitigation.

关键词： Temperature distribution Analytical models Tensors Prevention and mitigation Lightning Machine learning Soil Numerical models Plasmas Plasma temperature

来源：评论

学校读者我要写书评

暂无评论

Texture Features Based Hybrid Multi Support Vector Machine Model for the diagnosis of Alzheimer's disease through Brain MRI Images

Texture Features Based Hybrid Multi Support Vector Machine M...

引用

2023 International Conference on Computational Intelligence, Networks and Security, ICCINS 2023

作者： Mallika, R.M. Rani, K Usha Hemalatha, K. Giri, M. Dept. of Cse Siddarth Institute of Engineering &technology Andhra Pradesh Puttur India Sri Padmavati Mahila Visva Vidyalayam Dept. of Computer Science Andhra Pradesh Tirupati India Madanapalle Institute of Technology & Science Dept. of CSE-AI Andhra Pradesh Madanapalle India Siddarth Institute of Engineering & Technology Department of Cse Andhra Pradesh Puttur India

ISBN: (纸本)9798350313796

Alzheimer's Disease (AD) is a complex neurodegenerative disorder that severely affects cognitive functions and poses significant challenges in early and accurate diagnosis. In recent years, machine learning techniques have shown remarkable potential in medical image analysis, particularly when applied to Brain MRI Images for AD diagnosis. This study proposes a novel approach utilizing a Texture Features based Hybrid Multi-Support Vector Machine Model to enhance the precision and early detection of Alzheimer's disease through Brain MRI Images. The proposed model combines the strength of texture features extracted from Brain MRI Images with a Hybrid Multi- Support Vector Machine framework. Texture features capture subtle patterns and structural variations within brain images, making them valuable in distinguishing different stages of Alzheimer's disease. By leveraging the optimization capabilities of the Hybrid Multi-Support Vector Machine, the model effectively handles high-dimensional data and complex patterns, contributing to improved diagnostic accuracy. To evaluate the model's performance, a comprehensive dataset of Brain MRI Images, encompassing AD patients, Mild Cognitive Impairment cases, and non-AD individuals, was employed. The experimental results demonstrate the effectiveness of the proposed Texture Features based Hybrid Multi-Support Vector Machine Model in accurately classifying Alzheimer's disease cases, outperforming traditional diagnostic methods. © 2023 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Design and Implementation of Smart Classroom Using Cisco Packet Tracer

Design and Implementation of Smart Classroom Using Cisco Pac...

引用

Advances in Electronics, Communication, Computing and Intelligent Information Systems (ICAECIS), International Conference on

作者： Kavya Duvvuri Harshitha Kanisettypalli Jaswanth Kunisetty Manju Khanna Dept. of Computer Science Engineering Amrita School of Engineering Bengaluru India

Schools and colleges play a very crucial role in everyone's life. It is very important to impart good morals and safe education to future generations. As days go by, the necessity for faster job completion and smarter devices is increasing. The safety of students becomes important as the number of students enrolling in schools is gradually increasing. Discipline and time management skills must be imparted to students from a tender age in the school environment. Both safety and discipline can be ensured by making use of devices that are smarter and remotely controllable. Through this work, we have introduced the design of a smart classroom using a network simulation called Cisco Packet Tracer. The work is based on the concept of the Internet of Things. It has multiple subsystems that together make the class a wonderful place for students to build their future.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Land Use Land Cover Classification Using Machine Learning and Remote Sensing Data: A Case Study of Karnaprayag, Uttarakhand, India

Land Use Land Cover Classification Using Machine Learning an...

引用

Electronics, Communication and Signal Processing (ICECSP), International Conference on

作者： Rashmi Saini Suraj Singh Dept. of Computer Science and Engineering Govind Ballabh Pant Institute of Engineering and Technology Pauri Garhwal India

ISBN: (数字)9798350364590

ISBN: (纸本)9798350375381

Land use and land cover (LULC) classification is crucial for planning and sustainable development, especially in mountainous regions. This study focuses on the Tehsil Karnaprayag, situated in the vicinity of the Greater Himalaya. With a total area of 1858.903 km 2 and elevations ranging from 647m to 7641m. Using multispectral Sentinel-2 imagery and Machine Learning techniques i.e. Extreme gradient boosting (XGBoost) and K-nearest neighbor (KNN). This study aims to classify seven distinct land cover classes: snow, forest, water bodies, grassland, agricultural land, fallow land, and built-up areas. The classification results demonstrate the effectiveness of XGBoost classifier achieving an accuracy of 90.22% and a kappa value of 0.882, while KNN achieved an accuracy of 88.59% with a kappa value of 0.862. The analysis reveals that forest and fallow land are the predominant land cover types in the study area, constituting nearly 60% and 15% of the total area respectively. Notably, XGBoost exhibits the highest class-specific accuracy of 97.60 % for the snow class, demonstrating its utility in discriminating between challenging land cover types. This research highlights the importance of LULC classification in informing land management strategies and facilitating sustainable development initiatives in mountainous regions. The findings provide valuable insights into the spatial distribution and composition of land cover in Tehsil Karnaprayag, essential for effective planning and conservation efforts in the Himalayan ecosystem.

关键词： Accuracy Snow Soft sensors Land surface Forestry Nearest neighbor methods Signal processing Planning Sustainable development Remote sensing

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：