检索结果-内蒙古大学图书馆

3rd International Conference on Electronics and Renewable Systems, ICEARS 2025

作者： Narayanan, Sundarakrishnan Donthi, Yashas Mohana Ramakanth Kumar, P. Sreelakshmi, K. Rv College of Engineering® Center for Computer Vision Research Bengaluru India Rv College of Engineering® Center for Computer Vision Research Bengaluru India

ISBN: (纸本)9798331509675

In recent times, Artificial intelligence (AI) and machine learning (ML) technologies have demonstrated remarkable capabilities in addressing real time needs and challenges across multiple domains and fields. In the world of deep learning, various Neural Network architectures like DNN, CNN, and RNN have been instrumental in advancing solutions to complex problems spanning multiple industries, business environments and law enforcement use cases. This paper implemented a special type of CNN, YOLOv8(You Only Look Once), a SOTA architecture to detect the presence of rear-view mirrors in two-wheelers for traffic surveillance and law enforcement applications. The proposed research introduces a novel idea in the realm of road safety by presenting the first ever paper on vehicle mirror detection. This approach sheds light on an often-overlooked aspect, contributing significantly to the enhancement of road safety measures. A custom dataset of 1246 training images, 155 test images and 157 validation images has been considered. This dataset has been used to train and test a binary classification model with the label "mirror". Images with the label "mirror"indicate the presence of one or both rear-view mirrors on two-wheelers. On the other hand, images with no mirrors are treated as "background"images. The model achieved a maximum precision of 93.8%, recall of 87.9%, mAP50 of 94.1% and F1-score of 90.9%. These results stand as evidence of the robustness and proficiency of the model, ensuring that it can be potentially used in real world use cases, specifically for law enforcement applications. © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Efficient Smart Irrigation System using Internet of Things (IoT) for Agriculture Crops 3

Efficient Smart Irrigation System using Internet of Things (...

引用

3rd International Conference on Electronics and Renewable Systems, ICEARS 2025

作者： Ambika, G. Mohana Sudarshan, S. Kavitha, A. Sudhangowda, B.S. Nataraj, K. Rv College of Engineering® Center for Computer Vision Research Bengaluru India Rv College of Engineering® Center for Computer Vision Research Bengaluru India

ISBN: (数字)9798331509675

ISBN: (纸本)9798331509675

In this paper, application-based solution implemented for smart irrigation system using IoT and sensors. The proposed model provides a prediction of the user's time before which plant must need spraying, to compare water content present in soil and measure surrounding temperature and humidity in real-time. This application provides intelligent predictions which depend on environmental factors. If water is required, the tank valve is automatically opened and will be closed if water is not necessary. Nutrients dissolved in water are transported to each part of the plant. Water also protects plants from both frost and hot air currents. To maintain moisture of soil for efficient plant growth, it must be watered regularly. Instead of watering a little every day, opting for a more thorough watering at intervals is comparatively better. water needs to trickle below surface and reach end of pot. This ensures better root development. In this paper, as a first level plants of rose are considered, and first performed statistical analysis to find out the relationship between soil moisture with temperature and humidity. On a day, data is collected at three instances that is at morning, afternoon, and evening on both rainy and normal. This categorization is done as soil moisture and rate of evaporation differs at each different instance. On a rainy day, rate of evaporation will be nearly 0.0384614 due to more humidity and will have soil moisture in between 70% to 78% so plant may not need water whereas on a normal day, during afternoon there will be rate of evaporation nearly equal to 0.051156 which is more among all other cases and less soil moisture of 52% to 63% so watering is required. © 2025 IEEE.

关键词： Soil moisture

来源：评论

学校读者我要写书评

暂无评论

Navigating Heterogeneity and Privacy in One-Shot Federated Learning with Diffusion Models

Navigating Heterogeneity and Privacy in One-Shot Federated L...

引用

2025 IEEE/CVF Winter Conference on Applications of computer vision, WACV 2025

作者： Mendieta, Matías Sun, Guangyu Chen, Chen Center for Research in Computer Vision University of Central Florida United States

ISBN: (纸本)9798331510831

Federated learning (FL) enables multiple clients to train models collectively while preserving data privacy. However, FL faces challenges in terms of communication cost and data heterogeneity. One-shot federated learning has emerged as a solution by reducing communication rounds, improving efficiency, and providing better security against eavesdropping attacks. Nevertheless, data heterogeneity remains a significant challenge, impacting performance. This work explores the effectiveness of diffusion models in oneshot FL, demonstrating their applicability in addressing data heterogeneity and improving FL performance. Additionally, we investigate the utility of our diffusion model approach, FedDiff, compared to other one-shot FL methods under differential privacy (DP). Furthermore, to improve generated sample quality under DP settings, we propose a pragmatic Fourier Magnitude Filtering (FMF) method, enhancing the effectiveness of the generated data for global model training. Code available at https://***/mmendiet/FedDiff. © 2025 IEEE.

关键词： diffusion models federated learning

来源：评论

学校读者我要写书评

暂无评论

Semantic segmentation of agricultural images:A survey

引用

Information Processing in Agriculture 2024年第2期11卷 172-186页

作者： Zifei Luo Wenzhu Yang Yunfeng Yuan Ruru Gou Xiaonan Li School of Cyber Security and Computer Hebei UniversityHebei 071002China Hebei Machine Vision Engineering Research Center BaodingHebei 071002China College of Education Hebei UniversityHebei 071002China

As an important research topic in recent years,semantic segmentation has been widely applied to image understanding problems in various *** the successful application of deep learning methods in machine vision,the superior performance has been transferred to agricultural image processing by combining them with traditional *** segmentation methods have revolutionized the development of agricultural automation and are commonly used for crop cover and type analysis,pest and disease identification,*** frst give a review of the recent advances in traditional and deep learning methods for semantic segmentation of agricultural images according to different segmentation *** we introduce the traditional methods that can effectively utilize the original image information and the powerful performance of deep learningbased ***,we outline their applications in agricultural image *** our literature,we identify the challenges in agricultural image segmentation and summarize the innovative developments that address these *** robustness of the existing segmentation methods for processing complex images still needs to be improved urgently,and their generalization abilities are also *** particular,the limited number of labeled samples is a roadblock to new developed deep learning methods for their training and *** this,segmentation methods that augment the dataset or incorporate multimodal information enable deep learning methods to further improve the segmentation *** review provides a reference for the application of image semantic segmentation in the field of agricultural informatization.

关键词： Semantic segmentation Agricultural images Deep learning Convolution neural networks

来源：评论

学校读者我要写书评

暂无评论

Learning Multi-Modal Scale-Aware Attentions for Efficient and Robust Road Segmentation

引用

无人系统（英文） 2024年第2期12卷 201-213页

作者： Yunjiao Zhou Jianfei Yang Haozhi Cao Zhaoyang Zeng Han Zou Lihua Xie School of Electrical and Electronic Engineering Nanyang Technological UniversitySingapore Computer Vision and Robot Research Center International Digital Economy AcademyShenzhenGuangdongP.R.China

Road segmentation is essential to unmanned systems,contributing to road perception and navigation in the field of autonomous *** multi-modal road segmentation methods have shown promising results by leveraging the complementary data of RGB and Depth to provide robust 3D geometry information,existing methods suffer from severe efficiency problems that hinder their practical appli-cation in autonomous *** direct concatenation of multi-modal features with a densely-connected network leads to increased semantic gaps among modalities and scales,causing high computational and time *** address these issues,we propose a Multi-modal Scale-aware Attention Network(MSAN)to fuse RGB and Depth data effectively via a novel transformer-based cross-attention module,namely Multi-modal Scare-aware Transformer(MST),which fuses RGB-D features from a global perspective across multiple *** better consolidate different scales of features,we further propose a Scale-aware Attention Module(SAM)that captures channel-wise attention efficiently for cross-scale *** two attention-based modules explore the complementarity of modalities and scales,narrowing the gaps and avoiding complex structures for road *** experiments demonstrate MSAN achieves competitive performance at a low computational cost,suitable for real-time implementation on edge-devices in autonomous driving systems.

关键词： Autonomous driving road segmentation multi-modal fusion cross attention scale-aware fusion real-time application

来源：评论

学校读者我要写书评

暂无评论

LEARNING SEMANTIC PROXIES FROM VISUAL PROMPTS FOR PARAMETER-EFFICIENT FINE-TUNING IN DEEP METRIC LEARNING 12

LEARNING SEMANTIC PROXIES FROM VISUAL PROMPTS FOR PARAMETER-...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Ren, Li Chen, Chen Wang, Liqiang Hua, Kien Department of Computer Science University of Central Florida United States Center for Research in Computer Vision University of Central Florida United States

Deep Metric Learning (DML) has long attracted the attention of the machine learning community as a key objective. Existing solutions concentrate on fine-tuning the pre-trained models on conventional image datasets. As a result of the success of recent pre-trained models trained from larger-scale datasets, it is challenging to adapt the model to the DML tasks in the local data domain while retaining the previously gained knowledge. In this paper, we investigate parameter-efficient methods for fine-tuning the pre-trained model for DML tasks. In particular, we propose a novel and effective framework based on learning Visual Prompts (VPT) in the pre-trained vision Transformers (ViT). Based on the conventional proxy-based DML paradigm, we augment the proxy by incorporating the semantic information from the input image and the ViT, in which we optimize the visual prompts for each class. We demonstrate that our new approximations with semantic information are superior to representative capabilities, thereby improving metric learning performance. We conduct extensive experiments to demonstrate that our proposed framework is effective and efficient by evaluating popular DML benchmarks. In particular, we demonstrate that our fine-tuning method achieves comparable or even better performance than recent state-of-the-art full fine-tuning works of DML while tuning only a small percentage of total parameters. Code is available at https://***/Noahsark/ParameterEfficient-DML. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Omni Contextual Aggregation Networks for High-Fidelity Image Inpainting

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年第6期35卷 6129-6144页

作者： Peng, Jinjia Li, Mengkai Wang, Bingyan Wang, Huibing Hebei University School of Cyber Security and Computer Hebei China Hebei Machine Vision Engineering Research Center China Dalian Maritime University College of Information Science and Technology Liaoning China

Image inpainting aims to restore a realistic image from a damaged or incomplete version. Although Transformer-based methods have achieved impressive results by modeling long-range dependencies, the inherent quadratic complexity of canonical self-attention has typically led to these approaches adopting uni-dimensional modeling, which limits the model's ability to capture complex relationships from both spatial and channel dimensions. To this end, this paper exploits a novel attention paradigm termed Dynamic Omni-Attention Mechanism (DOAM) for simultaneously modeling pixel-interaction from both spatial and channel dimensions, and implements the information interaction across the omni-axis (i.e., spatial and channel) with linear computational complexity. In addition, to handle large-scale degradation, this paper proposes a Multi-band Feature Enhancement (MFE) module to enhance feature representation in downsampling, thus unlocking the potential of subsequent attentional interactions. Moreover, motivated by recent advances in image restoration, this paper incorporates a domain-related prior representation from CNN-based Network to modulate the features during proposed attention mechanism and feed-forward networks. Integrating the above designs into an encoder-decoder architecture, the proposed Omni Contextual Aggregation Networks (OCANet) achieve superior performance at lower parameters and time costs than the competitive baselines. Extensive experiments on CelebA-HQ, Paris Street View, FFHQ and Dunhuang datasets validate the efficacy of the proposed method. © 2025 IEEE.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Rethinking Data Heterogeneity in Federated Learning: Introducing a New Notion and Standard Benchmarks

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第3期5卷 1386-1397页

作者： Vahidian, Saeed Morafah, Mahdi Chen, Chen Shah, Mubarak Lin, Bill Department of Electrical and Computer Engineering University of California La Jolla San DiegoCA92093 United States Center for Research in Computer Vision University of Central Florida OrlandoFL32816-2365 United States Center for Research in ComputerVision University of Central Florida OrlandoFL32816 United States

Though successful, federated learning (FL) presents new challenges for machine learning, especially when the issue of data heterogeneity, also known as Non-IID data, arises. To cope with the statistical heterogeneity, previous works incorporated a proximal term in local optimization or modified the model aggregation scheme at the server side or advocated clustered federated learning approaches, where the central server groups agent population into clusters with jointly trainable data distributions to take the advantage of a certain level of personalization. While effective, they lack a deep elaboration on what kind of data heterogeneity and how the data heterogeneity impacts the accuracy performance of the participating clients. In contrast to many of the prior FL approaches,wedemonstrate not only the issue of data heterogeneity in current setups is not necessarily a problem but also in fact it can be beneficial for *** are intuitive: 1) Dissimilar labels of clients (label skew) are not necessarily considered data heterogeneity, and 2) the principal angle between the clients' data subspaces spanned by their corresponding principal vectors of data is a better estimate of the data heterogeneity. Impact Statement-FL is becoming a compelling learning paradigm in the artificial intelligence (AI) area. However, FL suffers from a notorious issue which is the existence of statistical Non-IID data across different distributed clients. Due to diverse participants, severe data heterogeneity can be present in different clients data, which has been demonstrated to result in unstable and slow convergence. For instance, training a global model across hospitals to identify brain/cancer tumors where every hospitals? images come froma different domain/distribution. To simulate this statistical data heterogeneity, data heterogeneity has been simply modeled as Non-IID label skewwhich tends to be a rigid data partitioning and is hardly representative and th

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Masked Generative Light Field Prompting for Pixel-Level Structure Segmentations

引用

research 2024年第4期2024卷 533-544页

作者： Mianzhao Wang Fan Shi Xu Cheng Shengyong Chen The Engineering Research Center of Learning-Based Intelligent System(Ministry of Education) Tianjin University of TechnologyTianjin 300384China Key Laboratory of Computer Vision and System(Ministry of Education) Tianjin University of TechnologyTianjin 300384China School of Computer Science and Engineering Tianjin University of TechnologyTianjin 300384China

Pixel-level structure segmentations have attracted considerable attention,playing a crucial role in autonomous driving within the metaverse and enhancing comprehension in light field-based machine ***,current light field modeling methods fail to integrate appearance and geometric structural information into a coherent semantic space,thereby limiting the capability of light field transmission for visual *** this paper,we propose a general light field modeling method for pixel-level structure segmentation,comprising a generative light field prompting encoder(LF-GPE)and a prompt-based masked light field pretraining(LF-PMP)*** LF-GPE,serving as a light field backbone,can extract both appearance and geometric structural cues *** aligns these features into a unified visual space,facilitating semantic ***,our LF-PMP,during the pretraining phase,integrates a mixed light field and a multi-view light field *** prioritizes considering the geometric structural properties of the light field,enabling the light field backbone to accumulate a wealth of prior *** evaluate our pretrained LF-GPE on two downstream tasks:light field salient object detection and semantic *** results demonstrate that LF-GPE can effectively learn high-quality light field features and achieve highly competitive performance in pixel-level segmentation tasks.

关键词： prompt backbone integrate

来源：评论

学校读者我要写书评

暂无评论

LSBNet: Lightweight Symmetrically Balanced Network for Real-Time Semantic Segmentation 41st

LSBNet: Lightweight Symmetrically Balanced Network for Real...

引用

41st computer Graphics International Conference, CGI 2024

作者： Zhang, Guodong Yang, Wenzhu School of Cyber Security and Computer Hebei University Baoding China Machine Vision Engineering Research Center Hebei University Baoding China

ISBN: (纸本)9783031818059

To solve the balance between the excessive computational overhead and the real-time inference speed of the existing semantic segmentation models on mobile devices, a Lightweight Symmetrically Balanced Network (LSBNet) for real-time semantic segmentation is proposed. LSBNet achieves its goals through three key modules: the Detail Reinforcement Module (DRM), the Information Complementarity Module (ICM), and the Effective Aggregation Pyramid Pooling Module (EAPPM). DRM is designed to compensate for the loss of spatial detail information in the feature extraction process to improve segmentation accuracy. ICM solves the imbalance problem in the process of fusion of semantic features and detail features, and enhances the detail information in semantic features through channel attention and spatial attention mechanisms. In addition, EAPPM is used to aggregate global multi-scale contextual information. Experimental results show that LSBNet achieves excellent performance of 73.4% mIoU and 73.7% mIoU on Cityscapes and Camvid datasets, respectively. It is worth mentioning that LSBNet contains only 8.7M parameters, which greatly reduces the number of parameters while maintaining high segmentation accuracy compared to most existing advanced models, demonstrating its superiority in lightweight model design. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Semantic Segmentation

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：