检索结果-内蒙古大学图书馆

Automatic Image Annotation Using Adaptive Convolutional Deep Learning Model

Intelligent Automation & Soft Computing 2023年第4期36卷 481-497页

作者： R.Jayaraj S.Lokesh Department of Computer Science and Engineering Hindusthan College of Engineering and TechnologyCoimbatore641032TamilnaduIndia Department of Computer Science and Engineering Hindusthan Institute of TechnologyCoimbatore641032TamilnaduIndia

Every day,websites and personal archives create more and more *** size of these archives is *** comfort of use of these huge digital image gatherings donates to their ***,not all of these folders deliver relevant indexing *** the outcomes,it is dif-ﬁcult to discover data that the user can be absorbed ***,in order to determine the signiﬁcance of the data,it is important to identify the contents in an informative *** annotation can be one of the greatest problematic domains in multimedia research and computer ***,in this paper,Adap-tive Convolutional Deep Learning Model(ACDLM)is developed for automatic image ***,the databases are collected from the open-source system which consists of some labelled images(for training phase)and some unlabeled images{Corel 5 K,MSRC v2}.After that,the images are sent to the pre-processing step such as colour space quantization and texture color class *** pre-processed images are sent to the segmentation approach for efﬁcient labelling technique using J-image segmentation(JSEG).Theﬁnal step is an auto-matic annotation using ACDLM which is a combination of Convolutional Neural Network(CNN)and Honey Badger Algorithm(HBA).Based on the proposed classiﬁer,the unlabeled images are *** proposed methodology is imple-mented in MATLAB and performance is evaluated by performance metrics such as accuracy,precision,recall and F1_*** the assistance of the pro-posed methodology,the unlabeled images are labelled.

关键词： Deep learning model J-image segmentation honey badger algorithm convolutional neural network image annotation

来源：评论

学校读者我要写书评

暂无评论

Enhancing COVID-19 and Pneumonia Detection with a Hybrid ACGAN-CNN Model 15

Enhancing COVID-19 and Pneumonia Detection with a Hybrid ACG...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Sasipriyaa, N. Ravikumar, R. Sharmila, C. Gothai, E. Krishnakumar, B. Nithya Sri, R.K. Kongu Engineering College Department of Computer Science and Engineering Tamilnadu India The Erode College of Pharmacy Pharmaceutical Analysis Tamilnadu India SRM IST Trichy Department of Computer Science and Engineering India Kongu Engineering College Department of Computer Science and Engineering Tamilnadu India

ISBN: (纸本)9798350370249

Corona Virus caused a pandemic outbreak all over the world during 2020-2021. Identification of such diseases in the X-ray images needs help of deep learning methodologies involving classifiers. Achieving proficiency in training convolutional neural networks requires a considerable volume of data. Higher accuracy is needed while dealing with medical fieldrelated problems especially. When dealing with small datasets, networks tend to exhibit overfitting problems. In order to get larger dataset, techniques for data augmentation serve to enhance the neural networks' capacity for generalization by expanding existing training data more efficiently. While conventional data augmentation methods do enhance performance, their scope remains restricted. Generative Adversarial Networks (GANs) have proven instrumental in producing fresh data, thereby elevating the performance of CNNs. Methodologies for augmenting data with the help of GAN's is not as much notable in Convolutional neural networks. In this research study, a GAN architecture is used for augmenting X -ray images of chest in order to expand the dataset. Auxiliary Classifier Generative Adversarial Network model has been used to produce images. It consists of a generator that produces new images with random noise and a discriminator whose job is to identify fake & real images. The results showcase that the new image generated is fine-tuned over each epoch and the significance can be seen in the output images. We compare the quality of our ACGAN images with the conventionally augmented images. © 2024 IEEE.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Feature engineering Methods for Analyzing Blood Samples for Early Diagnosis of Hepatitis Using Machine Learning Approaches

引用

computer Modeling in engineering & sciences 2025年第3期142卷 3229-3254页

作者： Mohamed A.G.Hazber Ebrahim Mohammed Senan Hezam Saud Alrashidi Department of Information and Computer Science College of Computer Science and EngineeringUniversity of Ha’ilHail81481Saudi Arabia Department of Computer Science College of Applied SciencesHajjah UniversityHajjah9677Yemen Department of Artificial Intelligence Faculty of Computer Science and Information TechnologyAl-Razi UniversitySana’a9671Yemen

Hepatitis is an infection that affects the liver through contaminated foods or blood transfusions,and it has many types,from normal to *** is diagnosed through many blood tests and factors;Artificial Intelligence(AI)techniques have played an important role in early diagnosis and help physicians make *** study evaluated the performance of Machine Learning(ML)algorithms on the hepatitis data *** dataset contains missing values that have been processed and outliers *** dataset was counterbalanced by the Synthetic Minority Over-sampling Technique(SMOTE).The features of the data set were processed in two ways:first,the application of the Recursive Feature Elimination(RFE)algorithm to arrange the percentage of contribution of each feature to the diagnosis of hepatitis,then selection of important features using the t-distributed Stochastic Neighbor Embedding(t-SNE)and Principal Component Analysis(PCA)***,the SelectKBest function was applied to give scores for each attribute,followed by the t-SNE and PCA ***,the classification algorithms K-Nearest Neighbors(KNN),Support Vector Machine(SVM),Artificial Neural Network(ANN),Decision Tree(DT),and Random Forest(RF)were fed by the dataset after processing the features in different methods are RFE with t-SNE and PCA and SelectKBest with t-SNE and PCA).All algorithms yielded promising results for diagnosing hepatitis data *** RF with RFE and PCA methods achieved accuracy,Precision,Recall,and AUC of 97.18%,96.72%,97.29%,and 94.2%,respectively,during the training *** the testing phase,it reached accuracy,Precision,Recall,and AUC by 96.31%,95.23%,97.11%,and 92.67%,respectively.

关键词： Hepatitis machine learning PCA RFE SelectKBest t-SNE

来源：评论

学校读者我要写书评

暂无评论

Tri-UnityNet: A Multifaceted Ensemble Model for Pneumonia Detection

Tri-UnityNet: A Multifaceted Ensemble Model for Pneumonia De...

引用

2024 International Conference on Advances in Data engineering and Intelligent Computing Systems, ADICS 2024

作者： Dhivyaa, C.R. Nithya, K. Shanmugapriya, P. Deebika, R. Kumar, K. Sathis Sudhakar, R. School Of Computer Science And Engineering Vellore Institute Of Technology Vellore India Srm Institute Of Science And Technology Computer Science And Engineering Trichy India Computer Science And Engineering Erode India K.S.R College Of Engineering Computer Science And Engineering Tiruchengode India Bannari Amman Institute Of Technology Computer Science And Engineering Sathyamangalam India Nandha College Of Technology Computer Science And Engineering Erode India

ISBN: (纸本)9798350364828

Pneumonia is an infection often caused by several viral infections and prediction of pneumonia requires expertise from radiotherapists, posing challenges, especially in remote areas. Developing an automatic pneumonia detection system is crucial to detect and treat pneumonia disease. Hence, an innovative approach is proposed to detect pneumonia by utilizing the ensemble of deep learning models, specifically VGG16, ResNet50, and InceptionV3. The ensemble method integrates predictions from these various architectures to enhance overall classification performance. Rather than starting the training of each model from the beginning, transfer learning techniques are utilized to fine-tune Convolutional Neural Network (CNN) models using a pre-existing knowledge base. This involves leveraging pre-trained models and adapting them specifically for the disease diagnose on Chest X-ray image dataset. The features are extracted from VGG16, ResNet50, and InceptionV3 models and the classes of disease are predicted by using softmax layers. Finally, ensemble learning is employed to fuse the predictions of multiple neural network architectures within Tri-UnityNet, enhancing its accuracy and robustness in pneumonia detection. By leveraging the complementary strengths of diverse models, the ensemble approach achieves superior performance, contributing to more reliable diagnostic outcomes. © 2024 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

A Strong PUF-Based Security Protocol to Protect AI Model Parameters Against Privacy Information Leakage

引用

IEEE Internet of Things Journal 2025年第12期12卷 20815-20827页

作者： Zhou, Ziyu Li, Gang Zhang, Yuejun Zheng, Ziyang Yuan, Tengfei Wang, Pengjun Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo315211 China Wenzhou University College of Electrical and Electronic Engineering Wenzhou325000 China

In the era of intelligent computing, with the aid of Internet of Things (IoT) technology, artificial intelligence (AI) chips can be embedded at the terminal, object, edge, and cloud levels, ultimately achieving the vision where there is computation, there is AI intelligence. This not only enhances the efficiency of production and daily life but also exponentially increases the risk of privacy information leakage within AI models. This article leverages the characteristics of strong physical unclonable functions (PUFs), in which the inherent feature information is hidden in physical variations and difficult to steal, to design a security protocol based on strong PUFs that provides effective protection for AI model parameters in the IoT environment. The protocol treats AI model parameters as responses and selects challenges capable of generating these responses. Since the responses generated by the challenges can be considered as randomly generated, transmitting the challenges does not disclose the response information, thus avoiding the risk of parameter hacking. Additionally, the protocol utilizes machine-learning modeling techniques and lightweight encryption technologies to reduce the storage costs for identity information and the computational overhead of encryption operations. Through a security analysis of the protocol, it demonstrates that even under ideal attack conditions, the proposed protocol can resist various attacks. By using formal verification with the ProVerif tool, it confirms the security of the protocol flow and the effective protection of private information. © 2014 IEEE.

关键词： Authentication Protocol

来源：评论

学校读者我要写书评

暂无评论

Electronic health records protection strategy by using blockchain approach

引用

Multimedia Tools and Applications 2024年第39期83卷 86883-86894页

作者： Kathole, Atul B. Patil, Sonali D. Kumbhare, Savita Vhatkar, Kapil Kshirsagar, Aniruddha P. Kimbahune, Vinod V. Computer Engineering Dr. D. Y. Patil Institute of Technology Pune India IT Dept. Pimpri Chinchwad College of Engineering Pune India Computer Science and Engineering Department K.B.P College of Engineering Satara Satara India

Electronic health records (EHRs) provide significant challenges for medical organizations due to the digitalization of traditional medical information. When using EHRs, both doctors and patients are accustomed to spending considerable time conducting the necessary queries;nevertheless, the acquired data is not always related to the sought one, and access may be revoked if this is the case. The exchange and maintenance of medical health records are among the most crucial tasks in the healthcare system. The integrity of health records is crucial, as is maintaining patient anonymity, and either would have devastating effects in the event of a breach. Therefore, it is one of the most important reasons to keep electronic health records safe. Modern healthcare infrastructure is notoriously difficult to use, wasteful, and fraught with problems related to data integrity, security, and privacy. Better EHR management and monitoring, however, has the potential to lessen the severity of the problems with security and complexity. The blockchain's potential, with its decentralized and trustworthy nature in the healthcare industry, makes this a reality. Problems with data management and the way information is validated and disseminated are at the heart of the healthcare delivery system's inherent difficulties. Improved access to medication tracking, hospital assets, drug systems, patient information, and so on is the primary benefit of employing blockchain technology in healthcare data management. Because of how important it is to have access to a patient's medical history when prescribing medication, blockchain technology has the potential to significantly enhance the current healthcare delivery system. Therefore, it is critical to develop a blockchain-based system for verifying and protecting medical records. The primary objective of this article is to create a permissioned blockchain-based Ciphertext Policy-Attribute-based encryption system to protect privacy and regulate acces

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Medicinal Plant Identification and Information Provision Using AI

Medicinal Plant Identification and Information Provision Usi...

引用

2025 International Conference on Multi-Agent Systems for Collaborative Intelligence, ICMSCI 2025

作者： Anslam Sibi, S. Sundaresan, C. Surya, S. Sujith, J.P. Srm Valliammai Engineering College Department of Computer Science and Engineering Chennai India

ISBN: (数字)9798331509828

ISBN: (纸本)9798331509828

Medicinal plants have been integral to traditional medicine, offering a wide range of health benefits and natural remedies. However, accurate identification is essential for safe use and the preservation of this valuable knowledge. This project proposes an AI-based system that combines EfficientNet and XGBoost models to enhance the accuracy of medicinal plant identification from images. While traditional systems rely on manual identification or basic models like CNN and SVM, they often fail to account for variations in plant appearance due to growth stages or environmental conditions. To address these limitations, the proposed system leverages EfficientNet for robust feature extraction from plant images, utilizing its superior ability to handle image variations across diverse conditions. These features are then passed to an XGBoost classifier, which excels in handling structured data and fine-tuning predictions. Together, this combination ensures high identification accuracy across a wide variety of plant species. Additionally, the system integrates a Hugging Face Transformer for natural language understanding and generation, allowing it to provide comprehensive information about each plant's medicinal properties and uses. This approach bridges the gap between accurate identification and practical application, promoting safe usage and preserving traditional medicinal knowledge. © 2025 IEEE.

关键词： Medicinal chemistry

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Enabled Social Media Recommendation Based on User Comments

引用

computer Systems science & engineering 2023年第2期44卷 1691-1702页

作者： K.Saraswathi V.Mohanraj Y.Suresh J.Senthilkumar Department of Computer Science and Engineering Government College of EngineeringSalem636011India Department of Information Technology Sona College of TechnologySalem636005India

Nowadays,review systems have been developed with social media Recommendation systems(RS).Although research on RS social media is increas-ing year by year,the comprehensive literature review and classiﬁcation of this RS research is limited and needs to be *** previous method did notﬁnd any user reviews within a time,so it gets poor accuracy and doesn’tﬁlter the irre-levant comments efﬁ*** Recursive Neural Network-based Trust Recom-mender System(RNN-TRS)is proposed to overcome this method’s *** it is efﬁcient to analyse the trust comment and remove the irrelevant sentence ***ﬁrst step is to collect the data based on the transactional reviews of social *** second step is pre-processing using Imbalanced Col-laborative Filtering(ICF)to remove the null values from the *** the features from the pre-processing step using the Maximum Support Grade Scale(MSGS)to extract the maximum number of scaling features in the dataset and grade the weights(length,count,etc.).In the Extracting features for Training and testing method before that in the feature weights evaluating the softmax acti-vation function for calculating the average weights of the ***,In the classiﬁcation method,the Recursive Neural Network-based Trust Recommender System(RNN-TRS)for User reviews based on the Positive and negative scores is analysed by the *** simulation results improve the predicting accuracy and reduce time complexity better than previous methods.

关键词： Recommendation systems(RS) social media recursive neural network-based trust recommender system(RNN-TRS) user reviews

来源：评论

学校读者我要写书评

暂无评论

A Base Station Deployment Algorithm for Wireless Positioning Considering Dynamic Obstacles

引用

computers, Materials & Continua 2025年第3期82卷 4573-4591页

作者： Aiguo Li Yunfei Jia Department of Network Engineering Xi’an University of Science and TechnologyXi’an710600China College of Computer Science and Technology Xi’an University of Science and TechnologyXi’an710600China

In the context of security systems,adequate signal coverage is paramount for the communication between security personnel and the accurate positioning of *** studies focus on optimizing base station deployment under the assumption of static obstacles,aiming to maximize the perception coverage of wireless RF(Radio Frequency)signals and reduce positioning blind ***,in practical security systems,obstacles are subject to change,necessitating the consideration of base station deployment in dynamic ***,research in this area still needs to be *** paper proposes a Dynamic Indoor Environment Beacon Deployment Algorithm(DIE-BDA)to address this *** algorithm considers the dynamic alterations in obstacle locations within the designated *** determines the requisite number of base stations,the requisite time,and the area’s practical and overall signal coverage *** experimental results demonstrate that the algorithm can calculate the deployment strategy in 0.12 s following a change in obstacle *** results show that the algorithm in this paper requires 0.12 s to compute the deployment strategy after the positions of obstacles *** 13 base stations,it achieves an effective coverage rate of 93.5%and an overall coverage rate of 97.75%.The algorithm can rapidly compute a revised deployment strategy in response to changes in obstacle positions within security systems,thereby ensuring the efficacy of signal coverage.

关键词： Wireless positioning base station deployment dynamic obstacles dynamic obstacle wireless positioning

来源：评论

学校读者我要写书评

暂无评论

HDR Reconstruction from LDR Scenes Using Basic Image Formation Procedure 14

HDR Reconstruction from LDR Scenes Using Basic Image Formati...

引用

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

作者： Sisira, C.S. Ann Thomas, Princy Government Engineering College Computer Science & Engineering Department of Computer Science & Engineering Thrissur India Government Engineering College Department of Computer Science & Engineering Thrissur India

ISBN: (纸本)9798350335095

High dynamic range (HDR) images are the one's which includes all the objects from the original scene with correct exposure. The HDR content can be very helpful in analysis purposes as it possess all features of the scene as it is. Although it is useful the creation of HDR is very much expensive. Low dynamic range (LDR) images are less expensive but may include saturation regions leads to the absence of some content. For this reason there is a necessity to think of an alternative. The reconstruction of HDR from LDR content can be the alternative solution for the problem. This process can be done using different *** studies mainly represents different kinds of machine learning techniques. The methods include machine learning , deep learning , different kinds of CNN architectures specially designed for particular purposes , GAN models , reverse tone mapping operators etc. This survey also analyses evaluation techniques for assessing quality of images and techniques like image declipping, tone mapping operator etc. Some of the analyzed methods also deals with reconstruction of HDR videos from LDR videos rather than from raw *** in the survey reveals accuracy of different methods. © 2023 IEEE.

关键词： Inverse problems

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：