检索结果-内蒙古大学图书馆

Advances in Modern Age Technologies for Health and Engineering science (AMATHE), International Conference on

作者： Shree Hari B Emmanuel Joy Jenice Aroma Division of Data Science and Cyber Security Karunya Institute of Technology and Sciences Coimbatore India Division of Artificial Intelligence and Machine Learning Karunya Institute of Technology and Sciences Coimbatore India CTO and AI Researcher MGK Logistics Madurai India

ISBN: (数字)9798350371567

ISBN: (纸本)9798350371574

Aerial images are photos taken from above, usually from planes or drones, displaying a landscape as seen from above. High-quality aerial images are crucial in various tasks, like precision agriculture, post-disaster monitoring, and land use classification. However, acquiring these images can be challenging due to the constraints of underlying image sensors and several other factors. This paper introduces a deep generative approach to enhance the spatial resolution of aerial imagery. The proposed method uses an Enhanced Deep Super-Resolution (EDSR) model architecture with a multi-stage residual refinement strategy. This particular model is comprised of residual blocks with two convolutional layers and residual connections. Through a succession of these residual blocks, the model processes low-resolution, 3-channel aerial images and finally upscales the images with a convolutional layer and a depth-to-space transformation. By increasing the spatial resolution of aerial imagery, this deep generative method seeks to overcome the problems that low-quality images present in real-world applications. Findings from the experiment on the VEDAI dataset demonstrate the effectiveness of the proposed method in producing high-quality aerial images. The improvements of PSNR and SSIM metrics by 1.05% and 0.34% respectively and the reduction in MAE and NRMSE by 0.0045 and 1.23% highlight its potential impact on various fields that depend on accurate and comprehensive aerial information. The outcomes of this work will contribute to the future development of picture super-resolution techniques for use in remote sensing and geospatial analysis.

关键词： Precision agriculture Visualization Accuracy Superresolution Geospatial analysis Spatial resolution Task analysis

来源：评论

学校读者我要写书评

暂无评论

Finding the right XAI method - A Guide for the Evaluation and Ranking of Explainable AI Methods in Climate science

arXiv

引用

arXiv 2023年

作者： Bommer, Philine Kretschmer, Marlene Hedström, Anna Bareeva, Dilyara Höhne, Marina M.-C. Department of Machine Learning Technische Universität Berlin Berlin10587 Germany Understandable Machine Intelligence Lab Department of Data Science ATB Potsdam14469 Germany Institute for Meteorology University of Leipzig Leipzig Germany Department of Meteorology University of Reading Reading United Kingdom BIFOLD - Berlin Institute for the Foundations of Learning and Data Berlin10587 Germany Machine Learning Group UiT the Arctic University of Norway Tromso9037 Norway Department of Computer Science University of Potsdam Potsdam14476 Germany

Explainable artificial intelligence (XAI) methods shed light on the predictions of machine learning algorithms. Several different approaches exist and have already been applied in climate science. However, usually missing ground truth explanations complicate their evaluation and comparison, subsequently impeding the choice of the XAI method. Therefore, in this work, we introduce XAI evaluation in the climate context and discuss different desired explanation properties, namely robustness, faithfulness, randomization, complexity, and localization. To this end, we chose previous work as a case study where the decade of annual-mean temperature maps is predicted. After training both a multi-layer perceptron (MLP) and a convolutional neural network (CNN), multiple XAI methods are applied and their skill scores in reference to a random uniform explanation are calculated for each property. Independent of the network, we find that XAI methods Integrated Gradients, layer-wise relevance propagation, and input times gradients exhibit considerable robustness, faithfulness, and complexity while sacrificing randomization performance. Sensitivity methods - gradient, SmoothGrad, NoiseGrad, and FusionGrad, match the robustness skill but sacrifice faithfulness and complexity for randomization skill. We find architecturedependent performance differences regarding robustness, complexity and localization skills of different XAI methods, highlighting the necessity for research task-specific evaluation. Overall, our work offers an overview of different evaluation properties in the climate science context and shows how to compare and benchmark different explanation methods, assessing their suitability based on strengths and weaknesses, for the specific research problem at hand. By that, we aim to support climate researchers in the selection of a suitable XAI method. © 2023, CC BY.

关键词： Multilayer neural networks

来源：评论

学校读者我要写书评

暂无评论

Intelligent Posture Detection System for Improved Ergonomics

Intelligent Posture Detection System for Improved Ergonomics

引用

Intelligent Control, Computing and Communications (IC3), International Conference on

作者： Jay Patil Suvarna Joshi Rahesha Mulla Kushal Yadav Department of Computer Science and Eng. MIT School of Computing MIT ADT University Pune India Department of Artificial Intelligence and Machine Learning Symbiosis Institute of Technology Symbiosis International (Deemed) University Pune India Department of Data Science Indian Institute of Technology Madras India

ISBN: (数字)9798331527495

ISBN: (纸本)9798331527501

This paper introduces a cost-effective smart posture detection system aimed at promoting ergonomic sitting habits and preventing musculoskeletal disorders. The system integrates ultrasonic sensors for backrest proximity measurement and load cells for seat weight distribution analysis. By analyzing these inputs, it identifies posture irregularities and provides real-time feedback for corrective action. Testing results demonstrate the system's reliability, achieving a detection accuracy of 99% across users with varying body types and postures. The proposed hybrid approach ensures simplicity, affordability, and scalability, making it a viable solution for homes, workplaces, and schools. This research contributes to the growing field of smart health monitoring systems, offering a practical alternative to traditional posture correction methods.

关键词： Weight measurement Accuracy Ultrasonic variables measurement Ergonomics Real-time systems Acoustics Intelligent sensors Monitoring Wearable sensors Testing

来源：评论

学校读者我要写书评

暂无评论

The challenges of the nonlinear regime for physics-informed neural networks 24

The challenges of the nonlinear regime for physics-informed ...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Andrea Bonfanti Giuseppe Bruno Cristina Cipriani BMW AG Digital Campus Munich Basque Center for Applied Mathematics University of the Basque Country BMW AG Digital Campus Munich Technical University of Munich Munich Center for Machine Learning Munich Data Science Institute

ISBN: (纸本)9798331314385

The Neural Tangent Kernel (NTK) viewpoint is widely employed to analyze the training dynamics of overparameterized Physics-Informed Neural Networks (PINNs). However, unlike the case of linear Partial Differential Equations (PDEs), we show how the NTK perspective falls short in the nonlinear scenario. Specifically, we establish that the NTK yields a random matrix at initialization that is not constant during training, contrary to conventional belief. Another significant difference from the linear regime is that, even in the idealistic infinite-width limit, the Hessian does not vanish and hence it cannot be disregarded during training. This motivates the adoption of second-order optimization methods. We explore the convergence guarantees of such methods in both linear and nonlinear cases, addressing challenges such as spectral bias and slow convergence. Every theoretical result is supported by numerical examples with both linear and nonlinear PDEs, and we highlight the benefits of second-order methods in benchmark test cases.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An Enhanced Traffic Incident Detection using Factor Analysis and Weighted Random Forest Algorithm 5

An Enhanced Traffic Incident Detection using Factor Analysis...

引用

5th International Conference on IoT Based Control Networks and Intelligent Systems, ICICNIS 2024

作者： Kanna, P. Rajesh Vanithamani, S. Karunakaran, P. Pandiaraja, P. Tamilarasi, N. Nithin, P. Bannari Amman Institute of Technology Department of Computer Science and Engineering Tamil Nadu Erode India M. Kumarasamy College of Engineering Department of Master of Computer Applications Tamil Nadu Karur India Nandha Engineering College Department of Artificial Intelligence and Data Science Tamil Nadu Erode India Vel Tech Rangarajan Dr. Sagunthala Department of Computer Science and Engineering R&D Institute of Science and Technology Tamil Nadu Chennai India Nehru Institute of Technology Department of Electronics and Communication Engineering Tamil Nadu Coimbatore India Bannari Amman Institute of Technology Department of Artificial Intelligence and Machine Learning Tamil Nadu Erode India

ISBN: (纸本)9798331518097

Efficient and precise traffic incident detection is essential to reduce casualties and property damage. To address the issue of unbalanced event data, this work offers a novel methodology known as FA-WRF (Factor Analysis and Weighted Random Forest). This approach combines dimensionality reduction through factor analysis with classification using weighted random forests, data preparation through the Synthetic Minority Over-Sampling Technique (SMOTE), and data preparation with SMOTE. The included feature of severity detection is highlighted, as are the performance metrics of the FA-WRF model using well-established metrics such as detection level rate, number of false count alarm rate, classification type rate, and area specification under the receiver operating characteristic curve (AUC). The superiority of the FA-WRF model is illustrated using real-world expressway traffic data represented by imbalanced incidents through thorough comparisons with various machine learning-based algorithms such as Support Vector machine, k-nearest neighbours algorithms, Logistic Regression algorithms, and decision-making trees. In addition to advancing incident detection, this technique has encouraging prospects for enhancing traffic management procedures and well-informed decision-making procedures in transportation networks. © 2024 IEEE.

关键词： Dimensionality reduction

来源：评论

学校读者我要写书评

暂无评论

Multilingual Facial Emotion Decoding and Posture Recognition using Random Forest Classifier

Multilingual Facial Emotion Decoding and Posture Recognition...

引用

Intelligent Technologies (ACOIT), Asian Conference on

作者： K Sangeetha Vs Balaji M Manju S Renuka Devi P Velayutham Department of Artificial Intelligence and Machine Learning Rajalakshmi Engineering College Chennai India Department of Artificial Intelligence and Data Science Rajalakshmi Engineering College Chennai India Department of Computer Science and Engineering Aarupadai Veedu Institute of Technology Chennai India

ISBN: (数字)9798350374957

ISBN: (纸本)9798350374964

Effective communication is crucial for success in various interactions, including personal and online interviews. The work proposed is to refine the communication effectiveness and extend the understanding of interactions by incorporating new system functions of real-time emotion and body language capture. Facial and body landmarks are analyzed using Google’s Mediapipe to ensure accurate identification of the frequently performed emotions and frequently posed postures. The proposed system uses three classifiers, namely Random Forest, Gradient Boosting, and Ridge Classifier, all integrated in one pipeline; this makes our work distinct. In the testing phase the classifiers are checked to avoid misidentification of emotions or incorrect posture in different situations. One of the major novelties of this work is its ability to address users of different languages, thus it is relevant anywhere in the world. This dynamic and robust system provides accurate emotion analysis achieving the accuracy of 92%, enhancing understanding in various communication contexts and setting a new standard for interview and interaction analysis.

关键词： Emotion recognition Accuracy Face recognition Pipelines Real-time systems Multilingual Interviews Random forests Standards Testing

来源：评论

学校读者我要写书评

暂无评论

Internet of Animal Health Things (IoAHT) Framework for Clinical Mastitis Detection in Dairy Cows

Internet of Animal Health Things (IoAHT) Framework for Clini...

引用

Recent Advances in Information Technology for Sustainable Development (ICRAIS), International Conference on

作者： Ankitha K Venugopala P S Harish Kunder Ankitha Shetty Dept. of Artificial Intelligence and Data Science NMAMIT Nitte Karkala India Dept. of Artificial Intelligence and Machine Learning Alvas Institute of Engineering and Technology Moodbidri India

The Earth serves as a shared habitat not just for humans but also for an array of animal species and organisms. Throughout history, animals have played a crucial role in preserving the natural ecosystem balance and meeting diverse human needs. They contribute significantly to various sectors, such as Agriculture, Dairy, Poultry, and animal husbandry, thus bolstering a country's economy. Human responsibility lies in ensuring the well-being of these animals, which is overseen by Veterinary science. In India, a major milk and milk product producer, the dairy industry actively contributes to the nation's economy. Among the primary milk sources-cows, buffaloes, goats, and sheep-cows hold a revered place in different faiths. Particularly in the dairy industry, cows are vital for milk production. However, monitoring cow health poses challenges. Clinical Mastitis, an inflammation affecting cows' mammary glands, hampers milk production. Automatic milking machines are commonly used, but identifying and segregating mastitis-infected cows from healthy ones remains difficult. The contagious nature of the disease increases the risk of widespread infection among cows. To combat this issue, we propose a solution called the Internet of Animal Health Things (IoAHT), based on the Internet of Things (IoT). IoAHT employs IoT technology for animal health, aiming to accurately identify cases of clinical mastitis in cows. The implementation of IoAHT has the potential to revolutionize health monitoring in the dairy industry, safeguarding cow well-being and ensuring sustainable milk production.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Noisy Recovery in Unlimited Sampling via Adaptive Modulo Representations

Noisy Recovery in Unlimited Sampling via Adaptive Modulo Rep...

引用

International Workshop on Compressed Sensing Theory and its Applications to Radar, Sonar and Remote Sensing (CoSeRa)

作者： Felipe Pagginelli Patricio Paul Catala Felix Krahmer Dept. of Mathematics TU Munich Garching Germany IBMI Helmholtz Munich Neuherberg Germany Dept. of Mathematics&Munich Data Science Institute TU Munich and Munich Center for Machine Learning Garching

ISBN: (数字)9798350365504

ISBN: (纸本)9798350365511

Recent works put forth the Unlimited Sensing Framework (USF), a novel approach to analog-to-digital conversion for high dynamic range sensing. It addresses the saturation phenomenon that commonly arises when physical measurements exceed the dynamic range of a sensor, yielding permanent loss of the input data. However, the USF still has some limitations when dealing with random noise. In the present paper, we propose a novel iterative method to tackle unlimited sensing in a noisy setting. In one step, our approach applies local transformations of the range to remove strong artifacts caused by the noise on local subdivisions of the domain. In the following step, the signal is then approximated via a least squares method. These two types of steps are then alternated. We illustrate the performances of our algorithm in high noise regime.

关键词： Multimodal sensors Conferences Noise Imaging Radar imaging Loss measurement High dynamic range Noise measurement Iterative methods Least mean squares methods

来源：评论

学校读者我要写书评

暂无评论

Solving multiscale dynamical systems by deep learning

arXiv

引用

arXiv 2024年

作者： Xu, Zhi-Qin John Yao, Junjie Yi, Yuxiao Hang, Liangkai Weinan, E. Zhang, Yaoyu Zhang, Tianhan Institute of Natural Sciences School of Mathematical Sciences MOE-LSC and Qing Yuan Research Institute Shanghai Jiao Tong University China Shanghai Center for Brain Science and Brain-Inspired Technology China Center for Machine Learning Research School of Mathematical Sciences Peking University China AI for Science Institute Beijing China Department of Mechanics and Aerospace Engineering Southern University of Science and Technology China

Multiscale dynamical systems, modeled by high-dimensional stiff ordinary differential equations (ODEs) with wide-ranging characteristic timescales, arise across diverse fields of science and engineering, but their numerical solvers often encounter severe efficiency bottlenecks. This paper introduces a novel DeePODE method, which consists of a global multiscale sampling method and a fitting by deep neural networks to handle multiscale systems. DeePODE’s primary contribution is to address the multiscale challenge of efficiently uncovering representative training sets by combining the Monte Carlo method and the ODE system’s intrinsic evolution without suffering from the "curse of dimensionality". The DeePODE method is validated in multiscale systems from diverse areas, including a predator-prey model, a power system oscillation, a battery electrolyte auto-ignition, and turbulent flames. Our methods exhibit strong generalization capabilities to unseen conditions, highlighting the power of deep learning in modeling intricate multiscale dynamical processes across science and engineering domains. © 2024, CC BY.

关键词： Ordinary differential equations

来源：评论

学校读者我要写书评

暂无评论

Development of Deep learning Model for Detection of Abdominal Diseases using Radiological Images

Development of Deep Learning Model for Detection of Abdomina...

引用

Artificial Intelligence in Education and Industry 4.0 (IDICAIEI), DMIHER International Conference on

作者： Vidhi Wankhade Soyal Nimbalkar Kajal Dakhare Rohan Jangamwar Meher Langote Prateek Verma Artificial Intelligence and Data Science Faculty of Engineering and Technology Datta Meghe Institute of Higher Education and Research Maharashtra India Artificial Intelligence and Machine Learning Faculty of Engineering and Technology Datta Meghe Institute of Higher Education and Research Maharashtra India

ISBN: (数字)9798331528713

ISBN: (纸本)9798331528720

As the abdominal and chest injuries are high morbidity and mortality around the globe, these pose an acute challenge in emergency care. Current data from recently released sources indicate that more than half of the patients visiting ER and also being admitted into hospital account for trauma; and the burden of both abdominal and chest injuries is quite high. Imaging plays an important role in diagnosis and treatment planning in the management of abdominal and chest injuries. The most widely practiced modality is radiograph imaging simply because it produces quality and easy results. Improving Automated pathology object detection of things in radiographs with deep learning algorithms promises much for the precision of diagnosis and optimization in health care. This paper discusses and provides a very detailed review of optimized CNNs, primarily ResNet18, ResNet34, along with a custom-built neural network for the identification of disease from X-ray pictures. Such models are trained and validated to reach excellent sensitivity and specificity on a wide variety of diseases by using the annotated RSNA 2023 dataset by professionals in radiology. It extended the scope of earlier research in this field by looking at other imaging conditions and diseases as well. This puts its contribution into the larger, ongoing discussion of AI in radiology. The results of this study help to contribute to the thesis that the promise of deep learning in radiological diagnosis is substantial, and it could be used to better enhance patient outcomes in the practice of emergency radiology.

关键词： Deep learning Pathology Ethics Neural networks Imaging Radiology Artificial intelligence Injuries X-ray imaging Diseases

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：