检索结果-内蒙古大学图书馆

Disaster Management Systems: Utilizing YOLOv9 for Precise Monitoring of River Flood Flow Levels Using Video Surveillance

引用

SN computer science 2025年第3期6卷 1-14页

作者： Shankar, G. Geetha, M. Kalaiselvi Ezhumalai, P. Department of Computer Science and Engineering FEAT Annamalai University Chidambaram 608002 India Department of Computer Science and Engineering R.M.D. Engineering College Chennai 601206 India

Flood disasters pose significant threats to human lives and infrastructure, necessitating advanced methods for the timely and accurate monitoring of water levels in rivers. This study introduces an innovative approach utilizing the recent YOLOv9 (You Only Look Once, version 9) deep learning model to detect and monitor river water flow levels effectively. By leveraging YOLOv9’s robust object detection capabilities, the proposed system achieves precise segmentation and real-time analysis of river water regions. The method not only identifies river water areas with high accuracy but also quantifies flow levels, providing critical data for flood prediction and management. Comprehensive testing was conducted using diverse datasets representing various river conditions and flow scenarios. The results demonstrate that YOLOv9 significantly outperforms traditional methods in terms of detection speed and accuracy, making it a valuable tool for enhancing flood disaster response and management strategies. This work contributes to the development of more resilient and proactive flood disaster management systems through the integration of state-of-the-art machine learning techniques, offering significant improvements in predictive capabilities and response effectiveness. The proposed system was trained and evaluated using real-time river flood datasets, achieving a notable accuracy rate of 98%, a mean average precision (mAP) of 97.5%, precision of 98%, and recall of 98% in predicting flood level characteristics. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Deep learning Flood disaster management Machine learning Object detection River water flow detection Water level segmentation YOLOv9

来源：评论

学校读者我要写书评

暂无评论

Improved Music Genre Classification and Visualization using Conventional Neural Network 2

Improved Music Genre Classification and Visualization using ...

引用

2nd IEEE International Conference on Emerging Research in Computational science, ICERCS 2024

作者： Denslin Brabin, D.R. Swarna Lakshmi, P. Princy, V. DMI College of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798331534950

A genre in music is a style or kind of music. To make it easier to recognize favorite songs, music genres group related tracks together. Dissimilar machine learning approaches, for example Decision Tree, K-means Clustering, Support Vector etc., can be used to categorize diverse genres of music. This research proposes a genre classification system based on Conventional Neural Networks (CNNs). The acoustic signal is reformed into a spectrogram in the CNN model. Next, every audio file in a dataset has its Mel-Frequency Cepstral Coefficient (MFCC) retrieved, which is utilized to train the model. Any audio will be classified and its genre will be predicted by the developed model. Additionally, the suggested system is utilized to show the genres using various visualization techniques, including as stacked-bar charts, tables, pie charts, and stacked area charts. The audio is categorized using this system as Pop, Rock, Reggae, Hip-hop, Jazz, Blues, Country, Classical, and Disco. Using visualization techniques, the probability of each genre in an audio file are also displayed. Python is used to implement the suggested approach, and the model is educated by means of the GTZAN dataset. Pandas and matplotlib are used in the implementation of the visuals. On the GTZAN dataset, the precision of the suggested model is 0.875. © 2024 IEEE.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Deep Transfer Learning Based Diagnosis of Multiple Neurodegenerative Disorders 6

Deep Transfer Learning Based Diagnosis of Multiple Neurodege...

引用

6th IEEE International Conference on Emerging Smart Computing and Informatics, ESCI 2024

作者： Mahendran, Rakesh Kumar Aishwarya, R. Abinayapriya, R. Kumar, P. Rajalakshmi Engineering College Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350306613

Neurodegenerative disorders challenge healthcare systems globally, impacting millions of lives and diminishing quality of life. Machine learning, particularly deep learning, offers promise in improving disease classification. Using a two-step methodology, This study looks into classifying three distinct neurodegenerative diseases using transfer learning: cerebral ataxia, Alzheimer's disease, and Parkinson's disease. First, we use neural network architectures that have already been trained, including VGG16 and ResNet101, which were first trained on large image datasets. This enables us to extract high-level features that capture important disease patterns and characteristics from medical images. By adapting these pre-Trained networks to the unique attributes of each disease, we enhance the model's ability to discriminate between them effectively. We use optimization techniques like SGD, Rmsprop, and Adam algorithm to get the optimized results, we employ a range of evaluation metrics, including accuracy and loss. These metrics provide a comprehensive understanding of the model's competence in accurately classifying patients with diverse neurodegenerative diseases. To make our model more accessible and user-friendly, we deploy it through the Flask web framework. Our research findings conclusively demonstrate that the amalgamation of transfer learning, Flask deployment, and stringent evaluation metrics substantially enhances the accuracy of neurodegenerative disease classification when compared to traditional diagnostic methods. This novel strategy promises to transform the identification and management of neurodegenerative illnesses, resulting in more prompt and efficient patient interventions. © 2024 IEEE.

关键词： biomedical MRI CNN Deep learning Inception v3 ResNet101 transfer learning VGG16

来源：评论

学校读者我要写书评

暂无评论

Heart disease prediction using machine learning

Heart disease prediction using machine learning

引用

International Conference on Artificial Intelligence, Blockchain, Computing and Security, ICABCS 2023

作者： Julian, Anitha Deepika, R. Geetha, B. Jeslin Sweety, V. Department of Computer Science and Engineering Saveetha Engineering College Chennai India

ISBN: (纸本)9781032669663

Anticipating and diagnosing coronary illness is the greatest test in the clinical business and depends on variables like the actual assessment, side effects and indications of the patient. Factors that impact coronary illness are body cholesterol levels, smoking propensity and corpulence, family background of diseases, pulse, and workplace. Coronary illness can be anticipated in light of different side effects, for example, age, orientation, pulse, and so on and lessens the demise pace of heart patients. In this task, we will be intently working with the coronary illness expectation. The coronary illness dataset from that dataset we will determine different experiences that assist us with knowing the weightage of each element and how they are interrelated to one another. Our only point is to recognize the likelihood of individual that will be impacted by a heart issue or not. We are utilizing K-Closest Neighbor (KNN) to foresee the exactness. © 2024, CRC Press/Balkema. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

The Agro Predictor - A Smart farming approach using Deep Learning

The Agro Predictor - A Smart farming approach using Deep Lea...

引用

2024 International Conference on Recent Innovation in Smart and Sustainable Technology, ICRISST 2024

作者： Sheela, K. Santha Threegha, S. Yazhmozhe, G.M. Velammal College of Engineering and Technology Computer Science Engineering Madurai India

ISBN: (纸本)9798350309140

The prime focus of this Agro-Predictor system which relies on the key learnings of the Internet of Things and Deep Learning Approach, to benefit the agricultural sector, placing its undivided attention on critical areas. Analysis of soil types and different weather conditions affecting, recommendations based on all the predictions made for the best production and farming. The integration of advanced technologies like Deep Learning, Data Mining, IOT and etc, has influenced agricultural sector transformation in various ways. The potential of deep learning models has been demonstrated to revolutionize farming practices by collecting the required data set, training and segregating the following, and predicting the required outcomes with the best recommendations. The work highlights the need for innovative solutions to address the growing global demand by identifying problem areas that require further research for sustainable agriculture by predicting various factors to reach the required solutions in farming. © 2024 IEEE.

关键词： Smart agriculture

来源：评论

学校读者我要写书评

暂无评论

Integration of Cross-Chain Technology-Based High-Performance Blockchain Network for Industrial Internet of Things 4

Integration of Cross-Chain Technology-Based High-Performance...

引用

4th International Conference on Emerging Trends in Networks and computer Communications, ETNCC 2024

作者： Kothari, Rakshit Jain, Kalpana Choudhary, Naveen College of Technology and Engineering MPUAT Computer Science and Engineering Udaipur India

ISBN: (纸本)9798350353266

The Internet of Things (IoT) has emerged as a promising technology for connecting physical objects and enabling communication between them. However, IoT faces several challenges such as security, privacy, and interoperability that need to be addressed before large-scale deployment. Blockchain has the potential to address these challenges due to its decentralized architecture, cryptographic security, immutability and consensus mechanisms. This paper explores the integration of blockchain technology with IoT for industrial applications. A cross-chain architecture is proposed to build a high-performance blockchain network customized for the needs of Industrial IoT (IIoT). The limitations of existing blockchain platforms like low transaction throughput are overcome by improvements in the consensus algorithm and employing a multi-chain structure. A proof-of-stake based consensus is implemented to achieve fast consensus and improve scalability. Cross-chain communication protocols are designed to enable interoperability between heterogeneous blockchains. The experimental evaluation demonstrates that the resulting blockchain network can achieve over 1000 transactions per second throughput per chain while maintaining decentralization, security and low latency. The cross-chain structure enables this performance to scale as the network expands. The paper provides comprehensive technical details of the various components of the blockchain platform including consensus, cross-chain communication and network structure. The designed blockchain architecture can potentially enable several IIoT use cases by enhancing security, accessibility, automation and coordination of industrial systems. © 2024 IEEE.

关键词： Blockchain Network Cross-Chain Technology Industrial Internet of Things Practical Byzantine Fault Tolerance Proof-of-Work (PoW) Proof-ofStake

来源：评论

学校读者我要写书评

暂无评论

Surveillance of species at risk by unarmed aerial vehicles

Surveillance of species at risk by unarmed aerial vehicles

引用

International Conference on Artificial Intelligence, Blockchain, Computing and Security, ICABCS 2023

作者： Vijaya Chamundeeswari, V. Sujith Varshan, G. Sanjay Kumar, Y. Sudarson, A.M. Department of Computer Science and Engineering Saveetha Engineering College Chennai India

ISBN: (纸本)9781032669663

A Living being or an organism that is about to vanish from this world will come under the category of Endangered *** Explanation for their state vary from time to time but few important reasons are loss of habitat and loss of genetic variation. These organism will attain the state of endangered when their population is below 30% and it no proper source for its lack of growth. The reason for being endangered species are – Population reduction rate, Geographical range, Population size and Population restriction, Poaching. Many different species are adversely affected by poaching. Different Laws have been sanctioned in order to protect the endangered species in many countries. Forbid hunting, Restrict land development, or Create protected areas are some of the measures for their growth. Some species are even taken away from their habitat in order to ensure their existence, This is called as extensive conservation *** breeding and habitat restoration are the few examples of the above. Many efforts have been made in order to restrict the *** Cameras, Drones and DNA tracking are the some of the methods and their probablitiy of success differs with each other. These systems are static in nature whereas the solution this provides is that with the help of Drone system with high resolution image capturing system the endangered species can be monitored along with their habitat. The advantage of using this system is the range is dynamic and precision is accurate. With the help of image processing technique endangered species is segregated and monitored from the rest of the species. © 2024, CRC Press/Balkema. All rights reserved.

关键词： Antennas

来源：评论

学校读者我要写书评

暂无评论

Fake News Detection on Social Media Using Ensemble Methods

引用

computers, Materials & Continua 2024年第12期81卷 4525-4549页

作者： Muhammad Ali Ilyas Abdul Rehman Assad Abbas Dongsun Kim Muhammad Tahir Naseem Nasro Min Allah Department of Computer Science COMSATS UniversityIslamabad45550Pakistan School of Computer Science and Engineering Kyungpook National UniversityDaegu41566Republic of Korea Department of Computer Science and Engineering Korea UniversitySeoul02841Republic of Korea Department of Electronic Engineering Yeungnam UniversityGyeongsan-si38541Republic of Korea Department of Computer Science College of Computer Science and Information TechnologyImam Abdulrahman Bin Faisal UniversityDammam34223Saudi Arabia

In an era dominated by information dissemination through various channels like newspapers,social media,radio,and television,the surge in content production,especially on social platforms,has amplified the challenge of distinguishing between truthful and deceptive *** news,a prevalent issue,particularly on social media,complicates the assessment of news *** pervasive spread of fake news not only misleads the public but also erodes trust in legitimate news sources,creating confusion and polarizing *** the volume of information grows,individuals increasingly struggle to discern credible content from false narratives,leading to widespread misinformation and potentially harmful *** numerous methodologies proposed for fake news detection,including knowledge-based,language-based,and machine-learning approaches,their efficacy often diminishes when confronted with high-dimensional datasets and data riddled with noise or *** study addresses this challenge by evaluating the synergistic benefits of combining feature extraction and feature selection techniques in fake news *** employ multiple feature extraction methods,including Count Vectorizer,Bag of Words,Global Vectors for Word Representation(GloVe),Word to Vector(Word2Vec),and Term Frequency-Inverse Document Frequency(TF-IDF),alongside feature selection techniques such as Information Gain,Chi-Square,Principal Component Analysis(PCA),and Document *** comprehensive approach enhances the model’s ability to identify and analyze relevant features,leading to more accurate and effective fake news *** findings highlight the importance of a multi-faceted approach,offering a significant improvement in model accuracy and ***,the study emphasizes the adaptability of the proposed ensemble model across diverse datasets,reinforcing its potential for broader application in real-world *** introduce a pioneering ensemble

关键词： Fake news detection Machine Learning(ML) Deep Learning(DL) Chi-Square ensembling

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation of Two-Tier Hybrid Cryptographic Models for Secure Data Transactions in Cloud Computing 2

Performance Evaluation of Two-Tier Hybrid Cryptographic Mode...

引用

2nd IEEE International Conference on Contemporary Computing and Communications, InC4 2024

作者： Debnath, Pankoj Kar, Tonoy Chandra Ashraf, Din Md Arif, Redwanul Islam Lysuzzaman, Md Sylhet Engineering College Department of Computer Science and Engineering Sylhet Bangladesh

ISBN: (纸本)9798350383652

Cloud computing is a trending and enormous technology. With the improvements in the field of Cloud computing, the main challenge arises in the transaction of data securely from one platform to another. Various cryptographic algorithms are used to provide a secure transaction of data. However, conventional cryptographic algorithms might not be sufficient enough for modern threats. To face these potential vulnerabilities, hybrid cryptographic algorithms can be the solution. Hybrid cryptography integrates multiple cryptographic algorithms to enhance performance and overall safety. RSA(Rivest-Shamir-Adleman) is the most common public key algorithm for encryption and decryption. This paper analyzes a modified form of RSA known as Modified-RSA that enhances security in hybrid cryptographic algorithms by incorporating three prime numbers instead of two. The Modified-RSA(MRSA) hybrid models are compared with the original RSA hybrid models with some theoretical and practical aspects. In this research, a comparative study of two-tier hybrid cryptographic models applied to secure data on the cloud is presented. Practical implementations were conducted with hybrid models of RSA and Modified-RSA using Python simulation. Encryption time, decryption time, average throughput, and efficiency were found using relatively more extensive data. The main motive was to find out the performance of every cryptographic hybrid and see which hybrid model is more efficient. © 2024 IEEE.

关键词： Cloud platforms

来源：评论

学校读者我要写书评

暂无评论

Deep Fusion of Vision Transformers, Graph Neural Networks, and LayoutLM for Enhanced Multimodal Detection of Lung Cancer: A Novel Approach in Computational Oncology 1

Deep Fusion of Vision Transformers, Graph Neural Networks, a...

引用

1st IEEE International Conference on Computing and Data science, ICCDS 2024

作者： Gowri, B. Shyamala Sharan, M. Gehlot, Yemini Varma, Veeralanka Chandrahas Easwari Engineering College Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350365337

An innovative AI framework has been designed to improve early lung cancer detection by combining Vision Transformers (ViT), Graph Neural Networks (GNNs), and LayoutLM. Utilizing both the IQ-OTH/NCCD and Prostate, Lung, Colorectal, and Ovarian (PLCO) cancer screening datasets offers a varied and thorough data basis for training and validating models. ViT enables advanced image processing of CT scans by utilizing self-attention processes to capture intricate visual characteristics. Graph neural networks analyze intricate spatial patterns in images to enhance diagnostic precision. Simultaneously, LayoutLM analyzes unstructured text data from patient records, enhancing the depth of the study. This method aims to enhance the diagnostic skills of current models by thoroughly leveraging multimodal data for comprehensive cancer detection. By using two separate datasets, comprehensive testing and improvement are achieved, leading to enhanced performance across diverse patient demographics and conditions. The framework sets a new standard in precision oncology by incorporating advanced AI technologies, leading to significant advancements in early detection, patient care, and personalized treatment approaches. This contributes to transformative improvements in clinical diagnostic methods and patient results in lung cancer treatment. © 2024 IEEE.

关键词： computerized tomography

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：