检索结果-内蒙古大学图书馆

CLIMS++: Cross Language Image Matching with Automatic Context Discovery for Weakly Supervised Semantic Segmentation

International Journal of computer Vision 2025年 1-20页

作者： Xie, Jinheng Deng, Songhe Hou, Xianxu Luo, Zhaochuan Shen, Linlin Huang, Yawen Zheng, Yefeng Shou, Mike Zheng Computer Vision Institute School of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China Department of Computer Science Wenzhou-Kean University Wenzhou325060 China Guangdong Provincial Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen China Show Lab National University of Singapore Singapore639798 Singapore School of AI and Advanced Computing Xi’an Jiaotong-Liverpool University Suzhou215400 China Jarvis Research Center Tencent YouTu Lab Shenzhen518703 China

While promising results have been achieved in weakly-supervised semantic segmentation (WSSS), limited supervision from image-level tags inevitably induces discriminative reliance and spurious relations between target classes and background regions. Thus, Class Activation Map (CAM) usually tends to activate discriminative object regions and falsely includes lots of class-related backgrounds. Without pixel-level supervisions, it could be very difficult to enlarge the foreground activation and suppress those false activation of background regions. In this paper, we propose a novel framework of Cross Language Image Matching with Automatic Context Discovery (CLIMS++), based on the recently introduced Contrastive Language-Image Pre-training (CLIP) model, for WSSS. The core idea of our framework is to introduce natural language supervision to activate more complete object regions and suppress class-related background regions in CAM. In particular, we design object, background region, and text label matching losses to guide the model to excite more reasonable object regions of each category. In addition, we propose to automatically find spurious relations between foreground categories and backgrounds, through which a background suppression loss is designed to suppress the activation of class-related backgrounds. The above designs enable the proposed CLIMS++ to generate a more complete and compact activation map for the target objects. Extensive experiments on PASCAL VOC 2012 and MS COCO 2014 datasets show that our CLIMS++ significantly outperforms the previous state-of-the-art methods. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Modeling to Estimate Used Car Prices 7th

Machine Learning Modeling to Estimate Used Car Prices

引用

7th International Conference on Smart City Applications, SCA 2022

作者： Hankar, Mustapha Birjali, Marouane Beni-Hssane, Abderrahim LAROSERI Laboratory Computer Science Department Faculty of Sciences University of Chouaib Doukkali El Jadida Morocco

ISBN: (纸本)9783031268519

Due to the increased prices of new cars nowadays, many customers prefer to buy a used car. Several online websites are made to offer their third-party services to contact used car sellers with the potential buyers. However, many customers who log into these websites in order to buy or sell a used car don’t have prior knowledge about its pricing value. The price of used cars often varies based on many features like production year, mileage, fuel type, fiscal power, mark, etc. In this paper, we designed a machine learning (ML) system that enables both potential buyers and sellers of used cars to estimate their car price before making a decision to buy or sell. In all tested ML models, gradient boosting regressor showed a high R-squared score (82%), a low root mean square error (RMSE), and a low mean absolute percentage error (MAPE). The system is finally deployed in a web application for a further usage of end users. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Data-Driven Graph Construction of Power Flow Graphs for Electric Power Transmission Networks 23

Data-Driven Graph Construction of Power Flow Graphs for Elec...

引用

23rd IEEE International Conference on Machine Learning and Applications, ICMLA 2024

作者： Poole, Benjamin Ratnakumar, Rajan Madurasinghe, Dulip Tharaka Kummerle, Christian Venayagamoorthy, Ganesh Kumar Lee, Minwoo University of North Carolina Department of Computer Science Charlotte United States University of Pretoria Department of Electrical Electronic and Computer Engineering South Africa Clemson University Real-Time Power and Intelligent Systems Laboratory United States

ISBN: (纸本)9798350374889

A comprehensive understanding of the topology of the electric power transmission network (EPTN) is essential for reliable and robust control of power systems. While existing research primarily relies on domain-specific methods, it lacks data-driven approaches that have proven effective in modeling the topology of complex systems. To address this gap, this paper explores the potential of data-driven methods for more accurate and adaptive solutions to uncover the true underlying topology of EPTNs. First, this paper examines Gaussian Graphical Models (GGM) to create an EPTN network graph (i.e., undirected simple graph). Second, to further refine and validate this estimated network graph, a physics-based, domain-specific refinement algorithm is proposed to prune false edges and construct the corresponding electric power flow network graph (i.e., directed multi-graph). The proposed method is tested using a synchrophasor dataset collected from a two-area, four-machine power system simulated on the real-time digital simulator (RTDS) platform. Experimental results show both the network and flow graphs can be reconstructed using various operating conditions and topologies with limited failure cases. © 2024 IEEE.

关键词： Undirected graphs

来源：评论

学校读者我要写书评

暂无评论

Rice Cultivation and Its Disease Classification in Precision Agriculture

Rice Cultivation and Its Disease Classification in Precision...

引用

2023 International Conference on Artificial Intelligence and Smart Communication, AISC 2023

作者： Kumar, Amit Bhowmik, Biswajit National Institute of Technology Brics Laboratory Department of Computer Science and Engineering Karnataka Bharat Mangalore India

ISBN: (纸本)9798350322309

Bharat is a cultivation-based country that cultivates various crops, including rice, in colossal capacity. Rice is one of the daily used crops across the country and other parts of the globe. Rice is cultivated in almost every state in Bharat. However, rice plant diseases severely degrade the quality and quantity of the crop. Rice plants can be affected by different conditions, e.g., foot rot, sheath blight, etc., causing a loss in the farming output. Various steps, including academic research, are practiced to detect rice diseases in their early stage to minimize the loss and increase the productivity of the agriculture sector. The schemes involve traditional and technology-based innovative solutions like agricultural cyber-physical systems (ACPS) or precision agriculture. This paper provides a state-of-the-art on rice crop cultivation, its disease classifications, etc. The paper depicts the importance of rice in daily life and how it plays a vital role in health. Later, this paper provides statistics on rice production in Bharat, followed by the export of rice to different countries. Subsequently, we offer a detailed analysis of rice cultivation, various diseases that can occur in rice plants and other prevention mechanisms for these diseases in the early stages of their life cycle. Finally, recent developments in precision agriculture are provided. © 2023 IEEE.

关键词： Precision agriculture

来源：评论

学校读者我要写书评

暂无评论

Real-Time Implementation of Quadrotor UAV Control System Based on a Deep Reinforcement Learning Approach

引用

computers, Materials & Continua 2024年第12期81卷 4757-4786页

作者： Taha Yacine Trad Kheireddine Choutri Mohand Lagha Souham Meshoul Fouad Khenfri Raouf Fareh Hadil Shaiba Aeronautical Sciences Laboratory Aeronautical and Spatial Studies InstituteBlida 1 UniversityBlida0900Algeria Department of Information Technology College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversityRiyadh11671Saudi Arabia Energy and Embedded Systems for Transportation Research Department ESTACA-LABMontigny-Le-Bretonneux78066France Department of Electrical Engineering University of SharjahSharjah27272United Arab Emirates Department of Computer Science College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh11671Saudi Arabia

The popularity of quadrotor Unmanned Aerial Vehicles(UAVs)stems from their simple propulsion systems and structural ***,their complex and nonlinear dynamic behavior presents a significant challenge for control,necessitating sophisticated algorithms to ensure stability and accuracy in *** strategies have been explored by researchers and control engineers,with learning-based methods like reinforcement learning,deep learning,and neural networks showing promise in enhancing the robustness and adaptability of quadrotor control *** paper investigates a Reinforcement Learning(RL)approach for both high and low-level quadrotor control systems,focusing on attitude stabilization and position tracking tasks.A novel reward function and actor-critic network structures are designed to stimulate high-order observable states,improving the agent’s understanding of the quadrotor’s dynamics and environmental *** address the challenge of RL hyper-parameter tuning,a new framework is introduced that combines Simulated Annealing(SA)with a reinforcement learning algorithm,specifically Simulated Annealing-Twin Delayed Deep Deterministic Policy Gradient(SA-TD3).This approach is evaluated for path-following and stabilization tasks through comparative assessments with two commonly used control methods:Backstepping and Sliding Mode Control(SMC).While the implementation of the well-trained agents exhibited unexpected behavior during real-world testing,a reduced neural network used for altitude control was successfully implemented on a Parrot Mambo mini *** results showcase the potential of the proposed SA-TD3 framework for real-world applications,demonstrating improved stability and precision across various test scenarios and highlighting its feasibility for practical deployment.

关键词： Deep reinforcement learning hyper-parameters optimization path following quadrotor twin delayed deep deterministic policy gradient and simulated annealing

来源：评论

学校读者我要写书评

暂无评论

Generating prediction clustering graph network analysis data from masked time series with GANs

引用

Neural Computing and Applications 2025年第14期37卷 8699-8729页

作者： Toujani, Radhia Hendaoui, Fatma Ferchichi, Ahlem Chihaoui, Mejda Ferchichi, Aya Bakari, Wided Computer Science Department Applied College University of Ha’il Hail55424 Saudi Arabia BESTMOD Department Higher Institute of management University of Tunis Bardo 2000 Tunisia RIADI-GDL Laboratory ENSI University of Manouba Manouba2010 Tunisia

Research in social community mining from social media faces challenges due to the constantly evolving nature of online networks. Deep learning methods often struggle with noisy data caused by sudden changes and fail to capture essential temporal information for community evolution. Additionally, data security concerns are frequently overlooked. This paper presents a novel probabilistic generative adversarial network (GAN) model designed for dynamic social community detection, addressing challenges posed by the evolving nature of networks and noisy data. Our approach integrates a hybrid probabilistic convolutional neural network (PCNN) and gated recurrent unit (GRU) architecture, capturing temporal community dynamics while preserving data confidentiality through deep generative masking techniques. The model generates synthetic time series data that effectively mimics real data while safeguarding sensitive information and improving forecasting accuracy. Experimental results demonstrate the model’s superior performance in terms of RMSE, MAPE, CE, CRPS*, and ARR, outperforming traditional deep learning models and providing valuable insights for decision-making in spatiotemporal community detection. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

IoT-Enabled Smart Applications and Challenges 8

IoT-Enabled Smart Applications and Challenges

引用

8th International Conference on Communication and Electronics Systems, ICCES 2023

作者： Prathyusha, M.R. Bhowmik, Biswajit National Institute of Technology Karnataka Brics Laboratory Department of Computer Science and Engineering Mangalore575025 India

ISBN: (纸本)9798350396638

Significant advances in the development of information and communication technology lead to collaborating computational entities into cyber-physical systems (CPS). A CPS serves as the foundation for many latest technology, e.g., the internet of things (IoT). Both IoT and CPS technologies provide several advantages that make our daily lives more simple and smart. These technologies are employed for a wide range of smart applications like smart city, smart home, smart transport, smart healthcare, etc., that can improve business, security, and health-related experiences. This paper primarily explores various IoT-enabled smart applications and examines their necessity, characteristics, and related issues. The paper also highlights various challenges of these smart applications followed by the state-of-the-art to address them. Each smart application promotes a sustainable, healthier, and comfortable life, including their future in the smart world. © 2023 IEEE.

关键词： Cyber Physical System

来源：评论

学校读者我要写书评

暂无评论

SmokerViT: A Transformer-Based Method for Smoker Recognition

引用

computers, Materials & Continua 2023年第10期77卷 403-424页

作者： Ali Khan Somaiya Khan Bilal Hassan Rizwan Khan Zhonglong Zheng College of Mathematics and Computer Science Zhejiang Normal UniversityJinhua321004China School of Electronics Engineering Beijing University of Posts and TelecommunicationsBeijing100876China Department of Electrical Engineering and Computer Science Khalifa University of Science and TechnologyAbu Dhabi127788United Arab Emirates Key Laboratory of Intelligent Education of Zhejiang Province Zhejiang Normal UniversityJinhua321004China

Smoking has an economic and environmental impact on society due to the toxic substances it *** Neural Networks(CNNs)need help describing low-level features and can miss important ***,accurate smoker detection is vital with minimum false *** answer the issue,the researchers of this paper have turned to a self-attention mechanism inspired by the ViT,which has displayed state-of-the-art performance in the classification *** effectively enforce the smoking prohibition in non-smoking locations,this work presents a Vision Transformer-inspired model called SmokerViT for detecting ***,this research utilizes a locally curated dataset of 1120 images evenly distributed among the two classes(Smoking and NotSmoking).Further,this research performs augmentations on the smoker detection dataset to have many images with various representations to overcome the dataset size *** convolutional operations used in most existing works,the proposed SmokerViT model employs a self-attention mechanism in the Transformer block,making it suitable for the smoker classification ***,this work integrates the multi-layer perceptron head block in the SmokerViT model,which contains dense layers with rectified linear activation and linear kernel regularizer with L2 for the recognition *** work presents an exhaustive analysis to prove the efficiency of the proposed SmokerViT *** performance of the proposed SmokerViT performance is evaluated and compared with the existing methods,where it achieves an overall classification accuracy of 97.77%,with 98.21%recall and 97.35%precision,outperforming the state-of-the-art deep learning models,including convolutional neural networks(CNNs)and other vision transformer-based models.

关键词： Smoker recognition SmokerViT deep learning transformer for vision

来源：评论

学校读者我要写书评

暂无评论

Games for Artificial Intelligence Research: A Review and Perspectives

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第12期5卷 5949-5968页

作者： Hu, Chengpeng Zhao, Yunlong Wang, Ziqi Du, Haocheng Liu, Jialin Research Institute of Trustworthy Autonomous System Guangdong Provincial Key Laboratory of Brain-Inspired Intelligent Computation Department of Computer Science and Engineering Shenzhen518055 China LightSpeed Studios Tencent Shenzhen518057 China

Games have been the perfect test beds for artificial intelligence (AI) research for the characteristics that widely exist in real-world scenarios. Learning and optimization, decision-making in dynamic and uncertain environments, game theory, planning and scheduling, design, and education are common research areas shared between games and real-world problems. Numerous open-source games or game-based environments have been implemented to study AI. In addition to single- or multiplayer, collaborative, and adversarial games, there has also been growing interest in implementing platforms for creative design in recent years. Those platforms provide ideal benchmarks for exploring and comparing AI ideas and techniques. This article reviews the games and game-based platforms for AI research, provides guidance on matching particular types of AI with suitable games for testing and matching particular needs in games with suitable AI techniques, discusses the research trend induced by the evolution of those games and platforms, and gives an outlook. © 2024 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A State of Art Review on Testing Open Multi-Agent Systems 3rd

A State of Art Review on Testing Open Multi-Agent Systems

引用

3rd International Conference on Novel and Intelligent Digital Systems, NiDS 2023

作者： Guassmi, Djaber Dehimi, Nour El Houda Derdour, Makhlouf Computer Science Department LIAOA Laboratory University of Oum El Bouaghi Oum El Bouaghi Algeria

ISBN: (纸本)9783031440960

Open Multi-Agent Systems (MASs) represent a paradigm in which multiple autonomous agents communicate and collaborate to achieve shared objectives in an open and dynamic environment. These systems require specific software engineering capabilities. In the realm of software validation, testing plays a crucial role, but it can be challenging to conceptualize and execute. Designing systems under the Open MAS paradigm poses even greater challenges due to characteristics such as autonomy, reactivity, pro-activity, and social abilities. This paper presents state-of-the-art research on testing Open MAS. The goal is to ultimately provide a new test method that addresses the unique characteristics of Open MAS, along with software and efficient test tools that facilitate the automation of testing activities for these systems. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Multi agent 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：