检索结果-内蒙古大学图书馆

Performance of pre-trained deep learning models for land use land cover classification using remote sensing imaging datasets

引用

Environmental Earth sciences 2025年第11期84卷 1-36页

作者： Haider, Irfan Khan, Muhammad Attique Masood, Saleha Algamdi, Shabbab Ali Alasiry, Areej Marzougui, Mehrez Nam, Yunyoung Department of Computer Science HITEC University Taxila Pakistan Department of AI College of Computer Engineering and Science Prince Mohammad Bin Fahd University Al-Khobar Saudi Arabia IRC for Finance and Digital Economy King Fahd University of Petroleum and Minerals Dhahran Saudi Arabia Department of Software Engineering College of Computer Science and Engineering Prince Sattam Bin Abdulaziz University Al Kharj Saudi Arabia College of Computer Science King Khalid University Abha61413 Saudi Arabia Department of Computer Science and Engineering Soonchunhyang University Asan31538 Korea Republic of

This study presents a comparative analysis of ten pre-trained convolutional neural network (CNN) models, evaluated across three remote sensing datasets: EuroSat, NWPU, and Earth Hazards (Land Sliding). We investigate the interplay between model architecture and classifier selection by incorporating five different neural network (NN) classifiers, emphasizing their impact on predictive accuracy and computational efficiency. Due to its densely connected architecture, DenseNet201 achieved the highest accuracy—97% on EuroSat, 99.40% on NWPU, and 97.80% on Earth Hazards. In contrast, MobileNetV2, while slightly less accurate, demonstrated superior computational efficiency, recording the shortest prediction times of 39.943 s on EuroSat, 27.482 s on NWPU, and 2.8986 s on Earth Hazards. Additionally, classifier choice significantly influenced performance, with the Wide NN classifier excelling in diverse datasets and the Medium NN classifier optimizing speed. Our findings underscore the importance of balancing accuracy and efficiency in selecting CNN models for remote sensing applications, suggesting future research should explore ensembling techniques and lightweight models to enhance performance. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Software design

来源：评论

学校读者我要写书评

暂无评论

Deep ensemble model for skin cancer classification with improved feature set

引用

Multimedia Tools and Applications 2025年第10期84卷 7599-7626页

作者： Burada, Sreedhar Manjunathswamy, B.E. Kumar, M. Sunil Department of Computer Science and Engineering Don Bosco Institute of Technology Karnataka Bengaluru560074 India Department of Computer Science and Engineering School of Computing Mohan Babu University Sree Vidyanikethan Engineering College Andhra Pradesh Tirupati517102 India

Early identification of skin cancer is mandatory to minimize the worldwide death rate as this disease is covering more than 30% of mortality rates in young and adults. Researchers are in the move of proposing advanced models for the accurate classification of skin disease. This paper introduces a novel skin cancer detection model under the stages of preprocessing, feature extraction, classification as well as segmentation. The given input is first preprocessed via Gaussian filtering. Subsequently, the Deep learning model termed as ‘U-net’ is used for the segmentation process. Afterward, the extraction of the feature set including features like Improved Center symmetric Local Binary Pattern (ICS-LBP), Local Gradient Pattern (LGP), Median binary pattern (MBP) is performed. Classification takes place using Deep Ensemble Model based on the extracted features, that includes BiGRU, Deep Maxout, and Improved Deep Convolutional Neural Network (CNN), respectively. Also, as the hyperparameter tuning of the classification model enhances the performance, this work introduces a new Hybrid Atom and Arithmetic Operation Algorithm (HAAOA) algorithm for tuning the optimal weights of the Bi-GRU (Bi-directional Gated Recurrent Unit) and Deep Maxout model. At last, the validation of the proposed work is performed through different performance measures by comparing it with different algorithms. Accordingly, accuracy of the adopted method at 90th learning percentage is 93.56%, which is (~) 82.63%, 85.15%, 84.98%, 83.66% and 83.42% better than AOA, ASO, BWO, BMO and DO, respectively. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Fake Face Detection Based on Fusion of Spatial Texture and High-Frequency Noise

引用

Chinese Journal of Electronics 2025年第1期34卷 212-221页

作者： Dengyong Zhang Feifan Qi Jiahao Chen Jiaxin Chen Rongrong Gong Yuehong Tian Lebing Zhang Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation School of Computer and Communication Engineering Changsha University of Science and Technology School of Computer and Communication Engineering Changsha University of Science and Technology Changsha Social Work College Changkuangao Beijing Technology Co. Ltd. School of Computer and Artificial Intelligence Huaihua University

The rapid development of the Internet has led to the widespread dissemination of manipulated facial images, significantly impacting people's daily lives. With the continuous advancement of Deepfake technology, the generated counterfeit facial images have become increasingly challenging to distinguish. There is an urgent need for a more robust and convincing detection method. Current detection methods mainly operate in the spatial domain and transform the spatial domain into other domains for analysis. With the emergence of transformers, some researchers have also combined traditional convolutional networks with transformers for detection. This paper explores the artifacts left by Deepfakes in various domains and, based on this exploration, proposes a detection method that utilizes the steganalysis rich model to extract high-frequency noise to complement spatial features. We have designed two main modules to fully leverage the interaction between these two aspects based on traditional convolutional neural networks. The first is the multi-scale mixed feature attention module, which introduces artifacts from high-frequency noise into spatial textures, thereby enhancing the model's learning of spatial texture features. The second is the multi-scale channel attention module, which reduces the impact of background noise by weighting the features. Our proposed method was experimentally evaluated on mainstream datasets, and a significant amount of experimental results demonstrate the effectiveness of our approach in detecting Deepfake forged faces, outperforming the majority of existing methods.

关键词： Deepfakes Adaptation models Attention mechanisms Noise Transforms Transformers Feature extraction Internet Convolutional neural networks Faces

来源：评论

学校读者我要写书评

暂无评论

Insider Attack Prevention:LAPUP—Lightweight Authentication Protocol Using PUF

引用

Journal of Communications and Information Networks 2024年第2期9卷 192-206页

作者： Siranjeevi Rajamanickam Satyanarayana Vollala N.Ramasubramanian Department of Computer Engineering Government Polytechnic CollegeTiruchirappalli 620022India Department of Computer Science and Engineering Dr.Shyama Prasad Mukherjee International Institute of Information Technology-Naya RaipurChhattisgarh 493661India Department of Computer Science and Engineering National Institute of TechnologyTiruchirappalli 620015India

The incredible progress in technologies has drastically increased the usage of Web *** share their credentials like userid and password or use their smart cards to get authenticated by the application *** cards are handy to use,but they are susceptible to stolen smart card attacks and few other notable security *** prefer to use Web applications that guarantee for security against several security attacks,especially insider attacks,which is *** of several existing schemes prove the security pitfalls of the protocols from preventing security attacks,specifically insider *** paper introduces LAPUP:a novel lightweight authentication protocol using physically unclonable function(PUF)to prevent security attacks,principally insider *** PUFs are used to generate the security keys,challenge-response pair(CRP)and hardware signature for designing the *** transmitted messages are shared as hash values and encrypted by the keys generated by *** messages are devoid of all possible attacks executed by any attacker,including insider *** is also free from stolen verifier attacks,as the databases are secured by using the hardware signature generated by *** analysis of the protocol exhibits the strength of LAPUP in preventing insider attacks and its resistance against several other security *** evaluation results of the communication and computation costs of LAPUP clearly shows that it achieves better performance than existing protocols,despite providing enhanced security.

关键词： physically unclonable function(PUF) insiderattacks authenticationkeymanagementserver(AKMS) security

来源：评论

学校读者我要写书评

暂无评论

Inductive Lottery Ticket Learning for Graph Neural Networks

引用

Journal of computer science & Technology 2024年第6期39卷 1223-1237页

作者： Yong-Duo Sui Xiang Wang Tianlong Chen Meng Wang Xiang-Nan He Tat-Seng Chua School of Data Science University of Science and Technology of ChinaHefei 230027China Department of Electrical and Computer Engineering The University of Texas at AustinAustin 78712U.S.A School of Computer Science and Information Engineering Hefei University of TechnologyHefei 230009China School of Computing National University of SingaporeSingapore

Graph neural networks (GNNs) have gained increasing popularity, while usually suffering from unaffordable computations for real-world large-scale applications. Hence, pruning GNNs is of great need but largely unexplored. The recent work Unified GNN Sparsification (UGS) studies lottery ticket learning for GNNs, aiming to find a subset of model parameters and graph structures that can best maintain the GNN performance. However, it is tailed for the transductive setting, failing to generalize to unseen graphs, which are common in inductive tasks like graph classification. In this work, we propose a simple and effective learning paradigm, Inductive Co-Pruning of GNNs (ICPG), to endow graph lottery tickets with inductive pruning capacity. To prune the input graphs, we design a predictive model to generate importance scores for each edge based on the input. To prune the model parameters, it views the weight’s magnitude as their importance scores. Then we design an iterative co-pruning strategy to trim the graph edges and GNN weights based on their importance scores. Although it might be strikingly simple, ICPG surpasses the existing pruning method and can be universally applicable in both inductive and transductive learning settings. On 10 graph-classification and two node-classification benchmarks, ICPG achieves the same performance level with 14.26%–43.12% sparsity for graphs and 48.80%–91.41% sparsity for the GNN model.

关键词： lottery ticket hypothesis graph neural networks neural network pruning

来源：评论

学校读者我要写书评

暂无评论

Weighted PageRank Algorithm Search Engine Ranking Model for Web Pages

引用

Intelligent Automation & Soft Computing 2023年第4期36卷 183-192页

作者： S.Samsudeen Shaffi I.Muthulakshmi Department of Computer Science and Engineering PET Engineering CollegeVallioorTamil Nadu627117India Department of Computer Science and Engineering V.V.College of EngineeringTisaiyanvilaiTamil Nadu627657India

As data grows in size,search engines face new challenges in extracting more relevant content for users’*** a result,a number of retrieval and ranking algorithms have been employed to ensure that the results are relevant to the user’s ***,most existing indexes and ranking algo-rithms crawl documents and web pages based on a limited set of criteria designed to meet user expectations,making it impossible to deliver exceptionally accurate *** a result,this study investigates and analyses how search engines work,as well as the elements that contribute to higher *** paper addresses the issue of bias by proposing a new ranking algorithm based on the PageRank(PR)algorithm,which is one of the most widely used page ranking algorithms We pro-pose weighted PageRank(WPR)algorithms to test the relationship between these various *** Weighted Page Rank(WPR)model was used in three dis-tinct trials to compare the rankings of documents and pages based on one or more user preferences ***ﬁndings of utilizing the Weighted Page Rank model showed that using multiple criteria to rankﬁnal pages is better than using only one,and that some criteria had a greater impact on ranking results than others.

关键词： Weighted pagerank algorithms search engines web pages web crawlers World Wide Web

来源：评论

学校读者我要写书评

暂无评论

Container Based Nomadic Vehicular Cloud Using Cell Transmission Model

引用

computer Systems science & engineering 2023年第1期44卷 423-440页

作者： Devakirubai Navulkumar Menakadevi Thangavelu Department of Computer Science and Engineering Jayam College of Engineering and TechnologyDharmapuri636813India Department of Electronics and Communication Engineering Adhiyaman College of EngineeringHosur635109India

Nomadic Vehicular Cloud(NVC)is envisaged in this *** predo-minant aspects of NVC is,it moves along with the vehicle that initiates it and functions only with the resources of moving vehicles on the heavy traffic road without relying on any of the static infrastructure and NVC decides the initiation time of container migration using cell transmission model(CTM).Containers are used in the place of Virtual Machines(VM),as containers’features are very apt to NVC’s dynamic *** specifications of 5G NR V2X PC5 interface are applied to NVC,for the feature of not relying on the network ***-days,the peak traffic on the road and the bottlenecks due to it are inevitable,which are seen here as the benefits for VC in terms of resource availability and residual in-network *** speed range of high-end vehicles poses the issue of dis-connectivity among VC participants,that results the container migration *** the entire VC participants are on the move,to maintain proximity of the containers hosted by them,estimating their movements plays a vital *** infer the vehicle movements on the road stretch and initiate the container migration prior enough to avoid the migration failure due to vehicles dynamicity,this paper proposes to apply the CTM to the container based and 5G NR V2X enabled *** simulation results show that there is a significant increase in the success rate of vehicular cloud in terms of successful container migrations.

关键词： Vehicular cloud container migration cell transmission model 5G NR V2X PC5 interface

来源：评论

学校读者我要写书评

暂无评论

Prediction of Student Academic Performance Using Regression Method 4th

Prediction of Student Academic Performance Using Regression ...

引用

4th Doctoral Symposium on Computational Intelligence, DoSCI 2023

作者： Lamba, Sonia Gupta, Aakanksha Singh, Sarvjeet Saraswat, Birendra Kumar Department of Computer Science ABES Engineering College UP Ghaziabad India Department of Computer Science and Engineering Raj Kumar Goel Institute of Technology Ghaziabad India

ISBN: (纸本)9789819937158

Nowadays, in a world full of competition, students have so many extracurricular activities for the sake of their overall development, and these activities go parallel with their academics. Sometimes, in this restless environment, students lose track of their weak subjects. It would really help them if they can know in advance in which matter, they are more likely to get fewer marks so that they can improve their overall performance. Although it is harder said than done, the main obstacle is that the prediction models are not so efficient. The main aim of this paper is to provide the idea of an effective and efficient regression model using which a teacher or the staff of an institute can predict its students’ results and improve the aspects in which they are lacking. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

An Upgraded Approach for Identifying Partially Reduplicated Forms in Bengali Text

引用

SN computer science 2024年第7期5卷 892页

作者： Barman, Abhijit Saha, Diganta Pal, Alok Ranjan Department of Computer Science and Engineering Jadavpur University Kolkata 700032 India Department of Computer Science and Engineering College of Engineering and Management Kolaghat 721171 India

This paper presents a concise methodology for the detection of partially reduplicated Multi-Word Expressions (MWEs) in Bengali texts. The entire process of identifying such reduplicated forms is carried out in two distinct phases, each contributing to the accuracy and effectiveness of the overall approach. In the first phase, a Levenshtein distance based algorithm is employed to identify partial reduplicated forms within Bengali text. This algorithm assesses the similarity between words and determines whether a partially reduplicated structure exists, thereby flagging relevant instances of reduplicated forms. Moving to the second phase, the performance of the first phase is enhanced through the application of a noble technique known as Word Expansion. By doing so, the performance of the reduplication identification process are significantly improved, leading to more accurate results. Evaluation metrics include Precision (90.00%), Recall (85.71%), and F1-Score (87.80%). These high scores underscore the system’s capability to successfully identify and categorize partially reduplicated MWEs within Bengali text. Moreover, the performance metrics also demonstrate that this approach surpasses the current state-of-the-art methods for identifying reduplicated expressions in Bengali text, reaffirming its efficacy and relevance. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Bi-gram Levenshtein distance Multi-word expression (MWE) Natural language processing (NLP) Reduplication Tokenization Word expansion

来源：评论

学校读者我要写书评

暂无评论

IntelliCrop: Major Cropping Pattern Prediction using Machine Learning Techniques 7

IntelliCrop: Major Cropping Pattern Prediction using Machine...

引用

7th International Conference on Inventive Computation Technologies, ICICT 2024

作者： Attavar, Aditi Manohar, Akshatha Leela Himoshra, D. Harshiya Banu, B. Pooja, M.R. Vidyavardhaka College of Engineering Computer Science And Engineering Mysuru India Vidyavardhaka College of Engineering Department of Computer Science And Engineering Mysuru India

ISBN: (纸本)9798350359299

The sustainable livelihoods of farmers in India hinge upon effective crop management and optimal yield. However, the existing manual system for crop selection is time-consuming, relies heavily on experience, and often leads to suboptimal yields and financial losses. To address this challenge, we propose an automated cropping pattern prediction system that leverages advanced machine learning techniques. Our system aims to empower farmers by providing insights into the diverse range of crops suitable for their specific land. By understanding the intricate relationships between different crop types, farmers can make informed decisions to maximize yield quality. The absence of a robust system for identifying cropping patterns results in random crop choices, leading to yield fluctuations. This study focuses on developing a system that predicts the optimal crop priority, considering various factors such as land characteristics, soil conditions, and climate data. This prioritization minimizes losses due to suboptimal crop combinations. The machine learning algorithms that are used include Naïve Bayes, and Eclat, which aim at disseminating information from the data at once, will enable the agricultural departments to work effectively. This approach allows for a more informed and effective decision-making process in agricultural practices. © 2024 IEEE.

关键词： Forecasting

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：