检索结果-内蒙古大学图书馆

3rd International Sustainability and Resilience Conference: Climate Change, ISRC 2021

作者： Chepa, Shravya Singh, Shivani Dutt, Himanish Sharma, Aashish Naik, Santosh Mahajan, Hema Hyderabad Institute of Technology and Management Computer Science Engineering Hyderabad India Hyderabad Institute of Technology and Management Electronics and Communication Engineering Hyderabad India Hyderabad Institute of Technology and Management Mechanical Engineering Hyderabad India

ISBN: (纸本)9781665416320

Fallacious waste management is an enormous enemy towards a healthy environment. Currently, the world's waste is categorized into municipal solid waste, industrial waste, agricultural waste, and hazardous waste. In earlier days instead of using different methodologies for waste segregation and management, people used similar methods for assorted kinds of waste produced in varying categories. But in recent years there has been a huge leap in techniques that involve segregation of waste and then proper disposal process. The two major technologies that have been developed for eradicating the complications in the field of waste management are IoT (Internet of Things) and deep learning. Apart from segregation and managing the garbage, the collection of garbage is also an important task. When the garbage isn't collected at regular intervals of time, it affects the health of citizens residing in that surrounding area. Both the technologies have done great work in eradicating this communication gap and made sure that proper segregation of waste takes place leaving no stone unturned. This survey paper discusses various strategies of waste segregation and management. © 2021 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

A driver-in-the-loop co-simulation framework for testing predictive EDAS for commercial vehicles in urban environments

A driver-in-the-loop co-simulation framework for testing pre...

引用

6th International Commercial Vehicle Technology Symposium, CVT 2020

作者： Chada, Sai Krishna Gőrges, Daniel Ebert, Achim Teutsch, Roman Division of Electromobility Department of Electrical and Computer Engineering University of Kaiserslautern Kaiserslautern67663 Germany Computer Graphics and HCI Group Department of Computer Science University of Kaiserslautern Kaiserslautern67663 Germany Institute for Mechanical and Automotive Engineering Department of Mechanical and Process Engineering University of Kaiserslautern Kaiserslautern67663 Germany

ISBN: (纸本)9783658297169

To experience realistic driving scenarios in the virtual simulation environment, strengths of both the traffic simulation and the vehicle simulation tools must be combined. In this paper, a driver-in-the-loop co-simulation framework for testing predictive eco-driving assistance systems (EDAS) for commercial vehicles in urban environments is proposed. The co-simulation is performed by exchanging information between IPG TruckMaker, a vehicle simulation software tailored for commercial vehicles, and SUMO, an open-source software for microscopic traffic simulation. To avoid delays during the data exchange process, time-steps of both the tools are synchronized. The co-simulation environment replicates a bus route in the city of Kaiserslautern along with the corresponding bus stops, traffic lights, road speed limits, pedestrian crossings and road elevation. The stochastic traffic and pedestrian flow in the road network is generated using SUMO. In this work, the co-simulation framework is extended by introducing the driver into the simulation loop. This is achieved by coupling the vehicle simulation tool with a driving simulator, in which the driver controls the ego-vehicle. The proposed framework will serve to test various use-cases of predictive eco-driving assistant systems (EDAS). © Der/die Autor(en), exklusiv lizenziert durch.

关键词： Open source software

来源：评论

学校读者我要写书评

暂无评论

Beyond Uncertainty: Risk-Aware Active View Acquisition for Safe Robot Navigation and 3D Scene Understanding with FisherRF

arXiv

引用

arXiv 2024年

作者： Liu, Guangyi Jiang, Wen Lei, Boshu Pandey, Vivek Daniilidis, Kostas Motee, Nader Department of Mechanical Engineering and Mechanics Lehigh University BethlehemPA18015 United States Department of Computer and Information Science University of Pennsylvaniac PhiladelphiaPA19104 United States

The active view acquisition problem has been extensively studied in the context of robot navigation using NeRF and 3D Gaussian Splatting. To enhance scene reconstruction efficiency and ensure robot safety, we propose the Risk-aware Environment Masking (RaEM) framework. RaEM leverages coherent risk measures to dynamically prioritize safety-critical regions of the unknown environment, guiding active view acquisition algorithms toward identifying the next-best-view (NBV). Integrated with FisherRF, which selects the NBV by maximizing expected information gain, our framework achieves a dual objective: improving robot safety and increasing efficiency in risk-aware 3D scene reconstruction and understanding. Extensive high-fidelity experiments validate the effectiveness of our approach, demonstrating its ability to establish a robust and safety-focused framework for active robot exploration and 3D scene understanding. © 2024, CC BY.

关键词： Robots

来源：评论

学校读者我要写书评

暂无评论

Combining Weakly Supervised Segmentation with Multitask Learning for Improved 3D MRI Brain Tumour Classification 2nd

Combining Weakly Supervised Segmentation with Multitask Lea...

引用

2nd Workshop on Medical Image Learning with Noisy and Limited Data, MILLanD 2023

作者： Rajapaksa, Sajith Namdar, Khashayar Khalvati, Farzad Neurosciences and Mental Health Research Program The Hospital for Sick Children Toronto Canada Institute of Medical Science University of Toronto Toronto Canada Department of Medical Imaging University of Toronto Toronto Canada Department of Computer Science University of Toronto Toronto Canada Department of Mechanical and Industrial Engineering University of Toronto Toronto Canada Vector Institute Toronto Canada

ISBN: (纸本)9783031471964

Segmentation plays a crucial role in computer-aided medical image diagnosis, as it enables the models to focus on the region of interest (ROI) and improve classification performance. However, medical image datasets often lack segmentation masks due to cost. In this work, we have introduced a novel end-to-end pipeline that enhances classification by incorporating ROIs generated by a weakly supervised method as an auxiliary task in a deep multitask learning framework. We have demonstrated that our approach outperforms conventional methods that only rely on classification. By using weakly supervised segmentation, we are able to leverage pixel-wise information, which ultimately leads to improved classification performance. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Explaining complex systems: a tutorial on transparency and interpretability in machine learning models (part II)

引用

IFAC-PapersOnLine 2024年第15期58卷 497-501页

作者： Donatello Materassi Sean Warnick Cristian Rojas Maarten Schoukens Elizabeth Cross Department of Electrical and Computer Engineering University of Minnesota US Department of Homeland Security Science and Technology Directorate EECS/Division of Decision and Control Systems KTH Royal Institute of Technology Control Systems Group Eindhoven University of Technology Eindhoven The Netherlands Department of Mechanical Engineering University of Sheffield

In the second segment of the tutorial, we transition from the granularity of local interpretability to a broader exploration of eXplainable AI (XAI) methods. Building on the specific focus of the first part, which delved into Local Interpretable Model-Agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP), this section takes a more expansive approach. We will navigate through various XAI techniques of more global nature, covering counterfactual explanations, equation discovery, and the integration of physics-informed AI. Unlike the initial part, which concentrated on two specific methods, this section offers a general overview of these broader classes of techniques for explanation. The objective is to provide participants with a comprehensive understanding of the diverse strategies available for making complex machine learning models interpretable on a more global scale.

关键词： eXplainable AI

来源：评论

学校读者我要写书评

暂无评论

Data Mining and AI for Early Diagnosis and Treatment Optimization in Autoimmune Encephalitis

Data Mining and AI for Early Diagnosis and Treatment Optimiz...

引用

International Conference on ICT in Business Industry & Government (ICTBIG)

作者： L. Natrayan H. Mohammed Ali S. Kaliappan Gotte Ranjith Kumar Department of Mechanical Engineering Saveetha School of Engineering SIMATS Chennai TN India Department Mechanical Engineering SRM Institute of Science and Technology- Ramapuram Campus Chennai TN India Division of Research and Development Lovely Professional University Punjab India School of Computer Science & Artificial Intelligence SR University Warangal Telangana India

To guarantee high-quality and trustworthy data, we first highlight the need of preprocessing. The dataset is prepared for analysis by using standard methods such as data cleansing, normalization, and feature engineering. The technique relies heavily on the feature selection process, which picks out the most informative aspects of a large collection of clinical, imaging, and laboratory data. ReliefF (Relief Feature Selection), Recursive Feature Elimination (RFE), and Principal Component Analysis (PCA) are three of the most popular feature selection techniques utilized here. These techniques make it easier to find unique characteristics in data, which in turn makes machine learning models more effective. A logistic regression model is used to predict the likelihood of autoimmune encephalitis based on certain characteristics to make an early diagnosis. In addition, linear programming is used to incorporate treatment optimization as an optimization issue. This guarantees that the suggested strategy maximizes therapy effectiveness within the bounds of possible side effects, available resources, and individual patient preferences. By adhering to a code of ethics, we preserve patient privacy and meet all applicable regulations. The suggested technique provides a holistic answer for autoimmune encephalitis evaluation and management. It greatly enhances diagnostic precision, therapeutic efficacy, and computational efficiency by merging data mining, AI, and mathematical modeling. These developments, together with our dedication to ethical compliance, make our method a very attractive and effective option for the diagnosis and individualized treatment of autoimmune encephalitis.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Prescribed Performance-Based Finite-Time Neural Adaptive Backstepping Control for the Chaotic PMSM

Prescribed Performance-Based Finite-Time Neural Adaptive Bac...

引用

2022 International Conference on mechanical and Electronics engineering, ICMEE 2022

作者： Wu, Fengbin Zhang, Junxing Li, Shaobo Li, Shaoyang Wu, Xiao Wang, Shuai College of Computer Science and Technology Guizhou University State Key Laboratory of Public Big Data Guiyang China School of Mechanical Engineering Guizhou University State Key Laboratory of Public Big Data Guiyang China School of Mechanical Engineering Guizhou University Guiyang China Guizhou University State Key Laboratory of Public Big Data Guiyang China

ISBN: (纸本)9781665492201

This paper presents a prescribed performance-based finite-time neural adaptive backstepping control scheme for the chaotic permanent magnet synchronous motor (PMSM). Specifically, an error transformation coupled with a prescribed performance function is introduced to guarantee that the tracking error keeps within a defined bound. The finite-time stability theory and backstepping framework are further combined to design finite-time adaptive laws and controllers. Then, it is shown that all signals are ultimately bounded in finite time and the tracking error can converge to a defined region in finite time. Finally, simulation results are presented to verify the feasibility of the proposed controller. © 2022 IEEE.

关键词： Backstepping

来源：评论

学校读者我要写书评

暂无评论

Optimal Design of Shock Absorption for Hexapod Robot based on Magnetorheological Fluid 8

Optimal Design of Shock Absorption for Hexapod Robot based o...

引用

2022 8th Asia Conference on mechanical engineering and Aerospace engineering, MEAE 2022

作者： Huang, Sihao Chen, Boyang Yin, Maoxin Ren, Yiming Xie, Zikang Department of Materials Science and Engineering Materials Science and Engineering Shenzhen MSU-BIT University Guangdong Shenzhen518172 China School of Intelligent Manufacturing and Equipment Mechanical Design Manufacture Shenzhen Institute of Information Technology Guangdong Shenzhen518000 China Department of Computational Mathematics and Cybernetics Mathematics and Applied Mathematics Shenzhen MSU-BIT University Guangdong Shenzhen518172 China Department of Engineering Electrical and Computer Engineering Shenzhen MSU-BIT University Guangdong Shenzhen518172 China

To achieve the best shock absorption effect of the hexapod robot, the influence of three leg structures on the stability of the hexapod robot under different path environments was studied, and a set of highly implementable solutions was designed. Use SolidWorks to build a hexapod robot model, import it into ADAMS to build a dynamic model and build different path environments, and conduct forward and inverse kinematics analysis at the same time. Using the Simulink plug-in in MATLAB and ADAMS joint kinematics simulation, the spring damping and magnetorheological fluid components are established in Simulink, which are connected in parallel with the model in ADMAS, and the stability of the hexapod robot under three-leg structures in different environments is obtained. According to the results, the optimal shock-absorbing leg structure is obtained. © Published under licence by IOP Publishing Ltd.

关键词： Magnetorheological fluids

来源：评论

学校读者我要写书评

暂无评论

CLISC: BRIDGING CLIP AND SAM BY ENHANCED CAM FOR UNSUPERVISED BRAIN TUMOR SEGMENTATION

arXiv

引用

arXiv 2025年

作者： Ma, Xiaochuan Fu, Jia Liao, Wenjun Zhang, Shichuan Wang, Guotai School of Computer Science and Engineering China School of Mechanical and Electrical Engineering University of Electronic Science and Technology of China Chengdu China Department of Radiation Oncology Sichuan Cancer Hospital & Institute Sichuan Cancer Center School of Medicine University of Electronic Science and Technology of China Chengdu China Shanghai Artificial Intelligence Laboratory Shanghai China

Brain tumor segmentation is important for diagnosis of the tumor, and current deep-learning methods rely on a large set of annotated images for training, with high annotation costs. Unsupervised segmentation is promising to avoid human annotations while the performance is often limited. In this study, we present a novel unsupervised segmentation approach that leverages the capabilities of foundation models, and it consists of three main steps: (1) A vision-language model (i.e., CLIP) is employed to obtain image-level pseudo-labels for training a classification network. Class Activation Mapping (CAM) is then employed to extract Regions of Interest (ROIs), where an adaptive masking-based data augmentation is used to enhance ROI identification. (2) The ROIs are used to generate bounding box and point prompts for the Segment Anything Model (SAM) to obtain segmentation pseudo-labels. (3) A 3D segmentation network is trained with the SAM-derived pseudo-labels, where low-quality pseudo-labels are filtered out in a self-learning process based on the similarity between the SAM’s output and the network’s prediction. Evaluation on the BraTS2020 dataset demonstrates that our approach obtained an average Dice Similarity Score (DSC) of 85.60%, outperforming five state-of-the-art unsupervised segmentation methods by more than 10 percentage points. Besides, our approach outperforms directly using SAM for zero-shot inference, and its performance is close to fully supervised learning. © 2025, CC BY.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Recognition of Handwritten Punjabi Characters Using Machine Learning Techniques 1

Recognition of Handwritten Punjabi Characters Using Machine ...

引用

1st IEEE International Conference on Advances in Computing, Communication and Networking, ICAC2N 2024

作者： Kaushik, P. Kaushik, Manas Manglik Chauhan, Yuvraj Singh Kumar, Narendra Kumar, Ashwani Rathore, Rachna Chandigarh University Dept. of CSE Chandigarh India Manipal University Dept. of Mechanical and Industrial Eng Karnataka India Chandigarh University Dept. of Computer Science Engineering Chandigarh India Usha Martin University Dept. of Computing and IT Jharkhand India Mangalayatan University Dept. of Computer Eng. and Applications Aligarh India Shrimant Madhav Rao University Dept. of Mathematics Gwalior India

ISBN: (纸本)9798350356816

Detecting handwritten Punjabi alphabets (PbAD) presents significant challenges for text detection systems due to the similarity among many characters and their complex curves and edges. Existing automatic detection systems often struggle to deliver high accuracy with real-world data. In this study, we propose an approach for PbAD that synthesizes various techniques reviewed from existing literature to optimize accuracy. Our methodology involves evaluating and combining multiple classification and feature extraction techniques, as identified through an extensive literature review. Unlike many studies that rely on manually created datasets, our approach will leverage both available online datasets and those developed in previous research. Previous research has reported accuracies ranging from 80% to 99%, and our goal is to advance these results by integrating and refining the best practices from the literature. © 2024 IEEE.

关键词： Adversarial machine learning

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：