检索结果-内蒙古大学图书馆

From Missing Pieces to Masterpieces: Image Completion with Context-Adaptive Diffusion

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年第7期47卷 6073-6087页

作者： Shamsolmoali, Pourya Zareapoor, Masoumeh Zhou, Huiyu Felsberg, Michael Tao, Dacheng Li, Xuelong University of York Department of Computer Science United Kingdom University of Leicester School of Computing and Mathematical Sciences United Kingdom Linkoping University Computer Vision Laboratory Sweden Nanyang Technological University College of Computing & Data Science Singapore China Northwestern Polytechnical University Key Laboratory of Intelligent Interaction and Applications Ministry of Industry and Information Technology Xi'an China

Image completion is a challenging task, particularly when ensuring that generated content seamlessly integrates with existing parts of an image. While recent diffusion models have shown promise, they often struggle with maintaining coherence between known and unknown (missing) regions. This issue arises from the lack of explicit spatial and semantic alignment during the diffusion process, resulting in content that does not smoothly integrate with the original image. Additionally, diffusion models typically rely on global learned distributions rather than localized features, leading to inconsistencies between the generated and existing image parts. In this work, we propose ConFill, a novel framework that introduces a Context-Adaptive Discrepancy (CAD) model to ensure that intermediate distributions of known and unknown regions are closely aligned throughout the diffusion process. By incorporating CAD, our model progressively reduces discrepancies between generated and original images at each diffusion step, leading to contextually aligned completion. Moreover, ConFill uses a new Dynamic Sampling mechanism that adaptively increases the sampling rate in regions with high reconstruction complexity. This approach enables precise adjustments, enhancing detail and integration in restored areas. Extensive experiments demonstrate that ConFill outperforms current methods, setting a new benchmark in image completion. © 1979-2012 IEEE.

关键词： computer aided design

来源：评论

学校读者我要写书评

暂无评论

Granular Network Traffic Classification on Mobile Application Service for Better Parental Control 2

Granular Network Traffic Classification on Mobile Applicatio...

引用

2nd IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation, IATMSI 2024

作者： Dhivvya, J.P. Aswini Dath, Guru Chinta, Mahesh Kumar Karki, Ankit Amrita Vishwa Vidyapeetham Amrita School of Computing Center for Wireless Networks and Applications Department of Computer Science and Engineering Kollam Amritapuri India Amrita Vishwa Vidyapeetham Amrita School of Computing Department of Computer Science and Engineering Kollam Amritapuri India

ISBN: (纸本)9798350360523

The widespread use of mobile applications among children had its limitations in excessive screen usage in entertainment. Traditional parental control techniques mostly depend on blocking applications based on their name or categories and are ineffective at identifying potentially new dangerous applications. This study addresses this issue by introducing a paradigm shift for protecting kids' digital experiences through granular network traffic classifications. This system classifies mobile app network traffic into various groups by implementing machine learning or deep learning models. Machine learning models like linear regression, Adaboost Classifier, Support Vector Classification (SVC), Ridge Classifier and deep learning models like Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), and Long Short Term Memory (LSTM) were used to identify and classify mobile applications. Initially, the deep learning RNN model results had 80% accuracy in the classification of mobile apps, and the machine learning logistic SVC and Adaboost Classifier had 72% accuracy. After adding new applications to the existing dataset, the accuracy of the RNN model has remained the same 80% accuracy. Meanwhile, the machine learning model accuracy reduced to 70%, proving the flexibility of the system. This work can assist parents in monitoring their kids to use mobile applications more for education. This work has a huge scope for information technology management of social change using hybrid learning models in software-defined networking in local networks for the administrators to empower their users to wisely use the internet. © 2024 IEEE.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Automated semantic segmentation and Diagnosis of Coconut Tree Infection Using Drone Imagery and Machine Learning approach for smart agriculture

Automated semantic segmentation and Diagnosis of Coconut Tre...

引用

2024 IEEE International Conference on Intelligent Systems and Advanced applications, ICISAA 2024

作者： Vasumathy, M. Devakanth, J. Jude Moses Anto Thirugnanam, Mythili Roselinkiruba, R. Hamsaveni, R. Nandhini, R. Kingston Engineering College Computer Science and Engineering Tamilnadu Vellore India Madanapalle Institute of Technology & Science Department ofComputer Applications Andhra Pradesh Madanapalle India Vellore Institute of Technology School of Computer Science and Engineering Tamilnadu Vellore India Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Avadi Computer Science and Engineering Tamilnadu Chennai India Computer Science and Engineering Andhra Pradesh Chittoor India D.K.M College for Women Computer Science and Applications Tamilnadu Vellore632001 India

ISBN: (纸本)9798331539948

The coconut industry of our Indian nation contributes an extensive range of economy to the national GDP. About 19,247 million coconuts, or 31.45% of the world's production, were produced in India in the year 2021-2022 making it the top producer. However, the coconut farmers face major challenges from a plethora of pests and diseases of coconut trees, and yield loss, which affects the farmers economically, standing around at 10-15%. Major pests, such as Rhinoceros beetles and red palm weevils, are highly destructible, in some scenarios leading even to the mortality of crops. In most of the cases, when trees that are affected need a cure, farmers have to climb the trees to put chemicals in crowns infested with the pests, exposing the farmers to harmful chemicals and risking severe injuries or permanent disabilities. As a solution, the proposed methodology involves the collection of extensive drone imagery from coconut plantations, followed by the annotation and preprocessing of the data. This work utilizes state-of-the-art deep learning models, including Unet, convolutional neural networks (CNNs) for semantic segmentation, to differentiate coconut trees from the background and other vegetation. Subsequently, machine learning classifiers are employed to diagnose diseases by analyzing visual symptoms present in the segmented trees. The proposed approach offers significant improvements in accuracy and efficiency compared with other traditional manual methods, providing real-time insights into tree health and disease management. This system has the potential to enhance crop monitoring, optimize resource allocation, and ultimately increase agricultural productivity. © 2024 IEEE.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Low Light Image Enhancement: A Special Click 2

Low Light Image Enhancement: A Special Click

引用

2nd International Conference on Computational Intelligence, Communication Technology and Networking, CICTN 2025

作者： Kapoor, Shalini Arora, Yojna Bansal, Nidhi Virdi, Khushwant Ismail, Firdous Sadaf M. Malik, Suraj School of Computer Science Engineering and Technology Bennett University Uttar Pradesh Greater Noida India Department of Computer Science and Engineering Sharda University Uttar Pradesh Greater Noida India Haryana Faridabad India Department of Computer Science and Applications School of Engineering and Technology Sharda University Uttar Pradesh Greater Noida India Nagpur Campus Pune India Department of Computer Science & Engineering IIMT University Uttar Pradesh Meerut India

ISBN: (纸本)9798331530389

Photography and images are an important and integral part of today's world, resulting in the development of multimedia and many other subjects comparable to image processing. The specifications of the camera and camera lens are just not enough hence image processing is always required. Low light photography, the problem gets even bigger because the camera lens fails to capture enough light and hence low light photography depends mostly upon image processing algorithms which help to reduce noise in the image, brighten the image for better overall visual quality and enlighten the hidden dark areas. Many efforts have been put into advancing image quality in little light, but none have given satisfactory results. This research report aims to analyses the techniques used for low-light image improvement and what advancements can be made to advance the image condition further. © 2025 IEEE.

关键词： Camera lenses

来源：评论

学校读者我要写书评

暂无评论

Implementing Smart Grids via Cognitive Radio: An Overview

Implementing Smart Grids via Cognitive Radio: An Overview

引用

2023 International Conference on Power Energy, Environment and Intelligent Control, PEEIC 2023

作者： Kalra, Bharti Prakash, Febin Sharma, Himanshi Tula's Institute Department of Computer Science and Engineering Dehradun India Department of Computer Bangalore India Noida Institute of Engineering and Technology Master of Computer Applications Uttar Pradesh Greater Noida India

ISBN: (纸本)9798350357769

Calidad of service issues related to antiquated communication frameworks in conventional electrical grids are addressed by the cognition smart grid connectivity paradigm. By incorporating cutting-edge communication and information technology into electric power lines, this novel strategy promotes a two-way exchange of information. However, there is a significant need for SG communications systems to manage large data volumes because to the rise in wireless services and apps, which is being driven by the growing adoption of Internet interconnected Things smart devices. Moreover, our work surpasses the state of the art in that it not only tackles the problems caused by current communication systems but also provides a proactive solution that foresees and handles the growing data needs resulting from the changing wireless application and Internet of Things (loT) landscape. The suggested hybrid design fosters adaptation and resilience in the face of rapidly advancing technology breakthroughs, marking a key step to future-proofing SG communication systems. © 2023 IEEE.

关键词： 5G Technology and Fog Computing Cognitive Radio Internet of Things (loT) Quality of Service (QoS) Smart Grid

来源：评论

学校读者我要写书评

暂无评论

A Reinforcement Learning Approach for Routing in Marine Communication Network of Fishing Vessels

引用

SN computer science 2025年第1期6卷 1-12页

作者： Surendran, Simi Montresor, Alberto Vinodini Ramesh, Maneesha Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Amritapuri Kollam India Department of Information Engineering and Computer Science University of Trento Trento Italy Amrita Center for Wireless Networks & Applications (AmritaWNA) Amrita School of Engineering Amrita Vishwa Vidyapeetham Amritapuri Kollam India

The lack of affordable communication facilities to the shore remains a fundamental problem for fishermen engaged in deep-sea fishing. The Offshore Communication Network (OCN) is a wireless network of fishing vessels, whose goal is to provide Internet over the ocean. However, the dynamic nature of OCNs characterized by extreme weather, the difficulty of deploying additional infrastructure, wave-induced vessel movements, and high mobility causes significant challenges for traditional routing protocols. This paper proposes OCN-AR, a Q-learning-based adaptive routing strategy for ocean networks. The quality of the learning process relies on the reward function, which has been carefully designed to incorporate the most important features, including real-time forecasts of connectivity quality, path probability, link availability duration, and distance to the destination. The routing performance is evaluated through extensive simulations conducted under diverse conditions, including varying mobility scenarios, transmission rates, vessel rocking intensities, and node densities, and is compared against traditional protocols. The results demonstrate that OCN-AR significantly outperforms existing routing approaches, making it a reliable solution for maritime communication. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Fishing vessel network Marine communication network Reinforcement learning Routing protocol

来源：评论

学校读者我要写书评

暂无评论

Psychology based Recommender System for Visual E-Learning

Psychology based Recommender System for Visual E-Learning

引用

2023 International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2023

作者： Manimozhi, S. Jeyachidra, J. Umamaheswari, G. Periyar Maniammai Institute of Science and Technology Department of Computer Science and Applications Vallam Tamilnadu Thanjavur613 403 India

ISBN: (纸本)9781665474511

The Visual E-learning Recommender System (VERS) recommends the learning materials based on the personality and knowledge of the learners. This paper focuses on recommendation system which has the ability to find out personality of the learners and the knowledge level. The relevant video content can be provided by the system based on the personality and knowledge. This task has been achieved through two mining techniques namely clustering and decision tree. In clustering technique, associated data are categorized based on the personality and knowledge of the learner. The rules are derived based on knowledge level of the learner using Decision Tree and are used for recommendation. © 2023 IEEE.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of the Machine Learning Algorithms for the Early Detection of Breast Carcinoma 2

Performance Analysis of the Machine Learning Algorithms for ...

引用

2nd IEEE International Conference on Interdisciplinary Approaches in Technology and Management for Social Innovation, IATMSI 2024

作者： Vanitha, K. Karpagavalli, S.M. Mahesh, T.R. Ali, A. Althaf Sridhar, T. Anitha, K. Faculty of Engineering Department of Computer Science and Engineering Coimbatore India Al Ameen Engineering College Department of Computer Science and Engg Tamilnadu Erode India Faculty of Engineering and Technology Department of Computer Science and Engineering Bangalore India Madanapalle Institute of Technology & Science Department of Computer Applications Andhra Pradesh India Alliance University Alliance College of Engineering and Design Bangalore India Csi College of Engineering Department of Information Technology Tamilnadu Ooty India

ISBN: (纸本)9798350360523

Breast Carcinoma, generally known as breast cancer, primarily affects women, though men can develop it as well. Because of the existence of breast tissue and exposure to female hormones, notably oestrogen, women are at a higher risk It's critical to diagnose breast tumors early. Several techniques based on machine learning (ML were used in this study to classify breast cancer using a dataset that was made available to the public. F-score, recall, precision, preciseness, and other performance metrics were used to evaluate these ML algorithms. Previous research and experimental findings indicate that Random Forest achieved the highest accuracy, with a remarkable accuracy rate of 99.12%. © 2024 IEEE.

关键词： Hormones

来源：评论

学校读者我要写书评

暂无评论

Optimal Algorithms for Covering a Rectangle with Identical Circles 14

Optimal Algorithms for Covering a Rectangle with Identical C...

引用

14th International Conference on Computing Communication and Networking Technologies, ICCCNT 2023

作者： Ghoshal, Ritwick Raj, Arjun Jayakumar, P. Amrita Vishwa Vidyapeetham Amrita School of Computing Department of Computer Science and Applications Kollam Amritapuri India

ISBN: (纸本)9798350335095

The optimization problem of covering a given geometrical region, such as a rectangle, with the minimum number of identical circles has been extensively studied in various disciplines. The paper introduces two quadratic time algorithms to solve two variants of the problem. The first algorithm computes the configurations or positions of the circles to cover the rectangle provided the circles cannot overlap. The second algorithm computes the same for the problem variant where the circles can overlap. The algorithms proved effective in handling diverse rectangle sizes and circle radii, efficiently solving the problem of covering a given rectangular region with the minimum number of identical circles and finding their positions. The empirical evaluation shows that the two algorithms compute the configuration of circles with minimum overlap or loss respectively. © 2023 IEEE.

关键词： Computational geometry

来源：评论

学校读者我要写书评

暂无评论

A Transformer Embedding-Based Model for Malicious DGA-Generated Domain Detection 16

A Transformer Embedding-Based Model for Malicious DGA-Genera...

引用

16th IEEE International Conference on Computational Intelligence and Communication Networks, CICN 2024

作者： Yerima, Suleiman Y. Vinod, P. Shaalan, Khaled The British University in Dubai Faculty of Engineering and It Dubai United Arab Emirates Cochin University of Science and Technology Department of Computer Applications Kerala Kochi682022 India The British University in Dubai Department of Computer Science Dubai United Arab Emirates

ISBN: (纸本)9798331505264

Domain Generation Algorithms (DGA) techniques are frequently used by cybercriminals to make malware more evasive by generating thousands of fake domain names that prevent established security solutions from effectively discovering or disrupting attacks. The proliferation of DGA has increased the difficulty of distinguishing genuine domains from fake ones, thus necessitating the investigation of new approaches to improve detection efficiency. Therefore, in this paper we propose a method for detecting malicious DGA domain names based on the Generative Pre-Trained Transformer (GPT) Large Language Model (LLM). Our approach utilizes GPT to learn dense vector representations with the aim of enabling more effective detection of the malicious domains by machine learning algorithms. The system is implemented and evaluated using a publicly available dataset and compared to widely-used techniques such TF-IDF, Bag-of-Words, N-grams, word2vec as well as to two variants of BERT. The results of our experiments demonstrates the superiority of the GPT-based technique over the others, with the best performance recorded by the XGboost classifier achieving the highest accuracy of 93.1% compared to the other machine learning algorithms. © 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：