检索结果-内蒙古大学图书馆

Enhancing the efficiency of lung cancer screening: predictive models utilizing deep learning from CT scans

Neural Computing and Applications 2025年 1-19页

作者： Tawfeek, Medhat A. Alrashdi, Ibrahim Alruwaili, Madallah Shaban, Warda M. Talaat, Fatma M. Department of Computer Science Faculty of Computers and Information Menoufia University Shebin Elkom32511 Egypt Department of Computer Science College of Computer and Information Sciences Jouf University Sakakah Saudi Arabia Department of Computer Engineering and Networks College of Computer and Information Sciences Jouf University Sakakah Saudi Arabia Department of Communication and Electronics Engineering Nile Higher Institute for Engineering and Technology Mansoura Egypt Faculty of Artificial Intelligence Kafrelsheikh University Kafrelsheikh Egypt Faculty of Computer Science and Engineering New Mansoura University Gamasa35712 Egypt

Lung cancer is the most lethal form of cancer. This paper introduces a novel framework to discern and classify pulmonary disorders such as pneumonia, tuberculosis, and lung cancer by analyzing conventional X-ray and CT scan images called lung cancer risk prediction (LCRP) model. LCRP has four modules, namely data collection and preprocessing, data augmentation module, image segmentation module, and prediction module. Actually, LCRP employs three deep learning models;sequential model, functional model, and transfer model on publicly available training datasets. Convolutional neural networks (CNNs) have emerged as a highly effective field in machine learning, particularly for image datasets in the field of biomedical applications. The primary goal is to validate these models by comparing their performance with other models in order to determine their effectiveness in addressing challenging datasets. Our research has revealed a noteworthy enhancement in the efficiency of binary and multi-class classification using mask R-CNN image segmentation. During the model training process, a combination of Adam and stochastic gradient descent dual optimizers has been used to improve performance. LCRP have outperformed current pre-trained models by minimizing training parameters, computational costs, and overhead. It introduces 98.5% accuracy, 88.7% specificity, 89% sensitivity, 89.2% precision, and 89.09% F-measure. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

IoT based Leakage Detection and Emergency Shutdown 6

IoT based Leakage Detection and Emergency Shutdown

引用

6th International Conference on Mobile Computing and Sustainable Informatics, ICMCSI 2025

作者： Abarna, P.S. Priya, S. Padma Pavalarajan, S. Sivatharani, S. Manshi, M. Department of Computer Science and Business Systems PSNA College of Engineering and Technology Dindigul India

ISBN: (纸本)9798331522667

The risk of gas leaks has grown significantly as a life threatening issue in industrial activities, cooking, and heating. This system integrates automatic reaction mechanisms, real-time monitoring capabilities, and advanced detection technologies in a multifaceted manner to address various dangers. The comprehensive gas leak detection and automatic cut off system is intended to improve safety in a variety of situations, including industrial and residential ones. To precisely identify the existence of dangerous gases like methane and propane, the proposed system combines infrared, electrochemical, and semiconductor types of gas sensors integrated with a mobile application, users may remotely administer and monitor the system in real-time. The system initiates a multi-tiered response upon detecting a gas leak;it sounds an alarm, notifies user-specified The mobile devices about the leak, and turns on an exhaust fan to release concentrated gas. A solenoid valve is used by the system to quickly shut off the gas supply and electricity from any location. The need of user-friendly controls and prompt notifications in emergency situations is highlighted in this paper's discussion of software architecture, sensor integration, and user interface design. Experimental results show that the system is reliable and capable of responding quickly under different circumstances. In future the application of machine learning algorithms for predictive maintenance and the possibility of integrating IoT to further strengthen security protocols. The results emphasize importance to deploy these kinds of extensive safety systems widely in order to reduce the risks associated with gas exposure. © 2025 IEEE.

关键词： Solenoid valves

来源：评论

学校读者我要写书评

暂无评论

OCR-CRNN (WBS): an optical character recognition system based on convolutional recurrent neural network embedded with word beam search decoder for extraction of text

引用

International Journal of Information technology (Singapore) 2025年 1-8页

作者： Kumar, Mahesh Shalu Dureja, Ajay Narula, Rachna Shyla Arora, Ravi Department of Information Technology Bharati Vidyapeeth’s College of Engineering (BVCOE) New Delhi India Department of Computer Science and Technology Manav Rachna University Haryana Faridabad India Department of Computer Science & Engineering Bharati Vidyapeeth’s College of Engineering New Delhi India

With the growing need for accurate digital text retrieval from images, OCR (optical character recognition) plays a critical role in fields such as document digitization, automated data processing, and information retrieval. The given paper introduces optical OCR system, comprising Convolutional Recurrent Neural Network (CRNN) embedded with a Word Beam Search (WBS) decoder for extraction of text. The framework performs preprocessing, feature extraction, and post-processing strategies that enhance the accuracy and robustness of text extraction. Preprocessing techniques and post-processing adjustments significantly improved the model's ability to handle diverse image formats, including scanned documents and screenshots. The proposed system demonstrates an overall accuracy of 92%, with macro and weighted averages across precision, recall, and F1-scores also at 92%. Individual class evaluations reveal notable performance consistency, with precision ranging from 0.79 to 0.98, recall between 0.79 and 0.96, and F1-scores from 0.79 to 0.94 across 26 alphabetic characters. This research highlights the system's potential for real-world applications, demonstrating its scalability, precision, and efficiency in extracting text data across various formats, ensuring high OCR accuracy in automated data processing. © Bharati Vidyapeeth's Institute of computer Applications and Management 2025.

关键词： CRNN Document Digitization Optical character recognition (OCR) Text extraction WBS

来源：评论

学校读者我要写书评

暂无评论

Recognition of American sign language using modified deep residual CNN with modified canny edge segmentation

引用

Multimedia Tools and Applications 2025年 1-28页

作者： Hariharan, U. Devarajan, Devikanniga Kumar, P. Santhosh Rajkumar, K. Meena, M. Akilan, T. Department of Computer Science and Engineering Apex Institute of Technology Chandigarh University Punjab Mohali140413 India Department of Computer Science and Engineering GITAM University Karnataka Bengaluru561203 India Department of Information Technology Faculty of Engineering and technology SRM Institute of Science and Technology Ramapuram Tamil Nadu Chennai600089 India Department of Data Science and Business Systems School of Computing Faculty of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Tamil Nadu Chennai603203 India Department of Computer Science and Engineering Sri Sairam Engineering College Tamil Nadu Tambaram602109 India Research Coordinator Dept. of IQAC Amity University Kolkata West Bengal 700135 India

American Sign Language (ASL) recognition aims to recognize hand gestures, and it is a crucial solution to communicating between the deaf community and hearing people. However, existing sign language recognition algorithms still have some drawbacks, such as difficulty recognizing hand movements and low recognition accuracy for most sign language recognition. A Modified Convolutional Neural Network (MCNN) deep residual 101 classifier-based American Sign Language identification system has been developed to address this problem. There are three main parts presents in the method. The first part is pre-processing the images to remove the noise, enhance the contrast of the picture, adjust the contrast level and smoothen the picture using various filters. The second part is the segmentation, and it’s used to partition the image using a modified canny edge detection method by removing all weak edges present in the image. Finally, classification will be done using the Modified CNN deep residual 101 classifiers. By classifying the image, the American Sign Language is accurately identified. This process is conducted through images. The outcome shows that the suggested approach has a 0.97% accuracy and a 0.05% False Positive Rate. Other CNN architectures such as resNet 50 and resNet 18 reached 0.90% and 0.80% accuracy, respectively, which is lower than our proposed method. In addition, the resNet 101 classifier effectively recognizes the difficult hand gestures through the image data. It obtains high recognition accuracy for 36 signs is 0 to 9 numbers and a to z alphabets from American Sign Language. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Injecting Visual Features into Whisper for Parameter-Efficient Noise-Robust Audio-Visual Speech Recognition

Injecting Visual Features into Whisper for Parameter-Efficie...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Yang, Zhao Yeo, Yue Heng Jiang, Rui Fu, Xiao Chen, Weiguang Xi, Wei Zhao, Jizhong School of Computer Science and Technology Xi'an Jiaotong University China School of Computer Science and Engineering Nanyang Technological University Singapore College of Computer Science and Electronic Engineering Hunan University China

ISBN: (纸本)9798350368741

Audio-visual speech recognition (AVSR) aims to enhance the robustness of an automatic speech recognition (ASR) systems by incorporating visual information from lip movements, especially in challenging noisy environments. Nevertheless, most current approaches either involve training from scratch or fully finetuning a pre-trained model, both of which incur significant computational costs and are often impractical for large-scale speech foundation models. This gap highlights the need for more efficient methods to leverage visual and acoustic information in AVSR tasks. To address this challenge, we propose AVWhisper, a parameter-efficient model that integrates visual and acoustic representations by injecting visual features from the AV-HuBERT encoder into the pre-trained Whisper model. Our approach leverages the existing attention mechanisms in Whisper to facilitate cross-modal interaction and integrates auxiliary visual information through lightweight adapters based on Low-Rank Adaptation (LoRA) and prompt-based techniques. Furthermore, a two-phase training strategy is adopted to effectively handle cross-domain differences and visual information injection problems respectively. Extensive experiments on the LRS3-TED dataset demonstrate that AVWhisper consistently outperforms state-of-the-art methods across various noise conditions, offering a more efficient and scalable solution for audio-visual speech recognition. © 2025 IEEE.

关键词： Audio-Visual Speech Recognition LoRA Noise-Robust Parameter-Efficient Prompt Whisper

来源：评论

学校读者我要写书评

暂无评论

An Efficient Multi-View Heterogeneous Hypergraph Convolutional Network for Heterogeneous Information Network Representation Learning

引用

IEEE Transactions on Big Data 2025年第3期11卷 1144-1157页

作者： Bing, Rui Yuan, Guan Zhang, Yanmei Wang, Senzhang Li, Bohan Zhou, Yong China University of Mining and Technology School of Computer Science and Technology Xuzhou221116 China Engineering Research Center of Mine Digitalization Xuzhou221116 China Central South University School of Computer Science and Engineering Changsha410083 China Nanjing University of Aeronautics and Astronautics College of Artificial Intelligence and Computer Science and Technology Nanjing211106 China Ministry of Education Key Laboratory of Brain-Machine Intelligence Technology Nanjing211106 China

Heterogeneous hypergraph neural networks are powerful tools to capture complex correlations among various nodes in Heterogeneous Information Networks (HINs). Despite satisfied performances of them, they are still plagued by the following problems: 1) They cannot capture the correlations in structural and semantic view at once, leading to topological information loss. 2) Due to the number of nodes being greater than the number of node types, node-level self-attention they used causes massive parameters and leads to high time consumption. 3) Interactions in meta-paths may be redundant, resulting in the correlations bias. To address the three issues, we propose an efficient Multi View Heterogeneous Hypergraph Convolutional Network (MVH, Member, IEEE, 2 GCN). It first constructs relational andsemantic hypergraphs based on different types of edges and meta-paths respectively, to represent the complex correlations in structural view and semantic view. Meanwhile, the clean semantic hypergraphs are generated by structure learning network to avoid redundancy. Then, an efficient hypergraph convolutional network is designed to learn node embeddings. By doing so, correlations in the two views are captured. Finally, the learned node embeddings from two views are aggregated via a gated embedding fusion module for downstream *** results demonstrate that MVH2GCN is effective and efficient. © 2015 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhanced Triple Layered Approach for Mitigating Security Risks in Cloud

引用

computers, Materials & Continua 2025年第4期83卷 719-738页

作者： Tajinder Kumar Purushottam Sharma Xiaochun Cheng Sachin Lalar Shubham Kumar Sandhya Bansal Computer Science&Engineering Department Jai Parkash Mukand Lal Innovative Engineering&Technology InstituteRadaur135133HaryanaIndia School of Computer Science&Engineering Galgotias UniversityGreater Noida203201Uttar PradeshIndia Computer Science Department Bay Campus Fabian WaySwansea UniversitySwanseaSA18ENUK Department of Engineering and Technology Gurugram UniversityGurugram122003India Presidency School of Computer Science Presidency UniversityBangalore560089India Maharishi Markendeshwar Engineering College Maharishi Markandeshwar(Deemed to be)UniversityMullanaAmbala133203India

With cloud computing,large chunks of data can be handled at a small ***,there are some reservations regarding the security and privacy of cloud data *** solving these issues and enhancing cloud computing security,this research provides a Three-Layered Security Access model(TLSA)aligned to an intrusion detection mechanism,access control mechanism,and data encryption *** TLSA underlines the need for the protection of sensitive *** proposed approach starts with Layer 1 data encryption using the Advanced Encryption Standard(AES).For data transfer and storage,this encryption guarantees the data’s authenticity and ***,the solution employs the AES encryption algorithm to secure essential data before storing them in the Cloud to minimize unauthorized ***-based access control(RBAC)implements the second strategic level,which ensures specific personnel access certain data and *** RBAC,each user is allowed a specific role and *** implies that permitted users can access some data stored in the *** layer assists in filtering granular access to data,reducing the risk that undesired data will be discovered during the *** 3 deals with intrusion detection systems(IDS),which detect and quickly deal with malicious actions and intrusion *** proposed TLSA security model of e-commerce includes conventional levels of security,such as encryption and access control,and encloses an insight intrusion detection *** method offers integrated solutions for most typical security issues of cloud computing,including data secrecy,method of access,and *** extensive performance test was carried out to confirm the efficiency of the proposed three-tier security *** have been made with state-of-art techniques,including DES,RSA,and DUAL-RSA,keeping into account Accuracy,QILV,F-Measure,Sensitivity,MSE,PSNR,SSIM,and computation time,encryption time,and decryption *** proposed

关键词： Cloud security:data encryption AES access control intrusion detection systems(IDS) role-based access control(RBAC)

来源：评论

学校读者我要写书评

暂无评论

Personalized Recommendations System for Mental Health Support using NLP 3

Personalized Recommendations System for Mental Health Suppor...

引用

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

作者： Akash, G. Gopinath, S. Mohamed Zachariah, S.M. Mithuna, S. K S Rangasamy College of Technology Department of Computer Science and Engineering Namakkal India

ISBN: (纸本)9798331509675

The prevalence of mental health conditions is increasing so much that detecting them early will make interventions more effective. I introduce a recommendation system which benefits from Natural Language Processing combined with machine learning approaches to analyze users' mental health conditions through their written and spoken content. Users can transmit writing or speech data directly to the system for processing. Among all available algorithms Machine learning algorithms including Logistic Regression, Random Forest, Decision Tree and Multinomial Naive Bayes exhibit the best performance. The system performs user classification based on their diagnosed mental disorder which enables it to present appropriate mental health video content specifically designed to support their needs. Their goal was to create a system which provides real-time feedback while raising accessibility to mental health care and starting interventions early. Integration of advanced technology into personal mental health experiences enters a critical phase because of these developments. © 2025 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

A Customizable Virtual Assistant Enhanced with Vision and Speech Capabilities 12th

A Customizable Virtual Assistant Enhanced with Vision and Sp...

引用

12th International Conference on Recent Trends in Computing, ICRTC 2024

作者： Christy Jeba Malar, A. Deva Priya, M. Naveen Kumaran, R. Sanjay Abhishek, J. Bhuvana, S. Karthick, S. Department of Information Technology Sri Krishna College of Technology Tamilnadu Coimbatore India Department of Computer Science and Engineering Karpagam College of Engineering Tamilnadu Coimbatore India School of Computer Science Engineering and Information Systems Vellore Institute of Technology Vellore India Department of Computer Science and Engineering SRM Institute of Science and Technology Uttar Pradesh Delhi–NCR Campus Ghaziabad India

ISBN: (纸本)9789819788354

Recent advancements in Smart Assistants (SAs) as well as home automation have captured the attention of both researchers and consumers. Virtual Assistants (VAs) that are speech-enabled are commonly referred to as smart speakers. They offer a wide range of network-based services. In some cases, they seamlessly link to smart environments, introducing innovative and efficient user interfaces. Nevertheless, these devices present their own set of challenges and limitations. The distinctive features of this technology, particularly its hand-free operation handled by voice and implementation of voice-user interface create a unique landscape. Existing technology adoption models, however, fall short in offering a comprehensive explanation for the adoption of this innovative technology. To gain deeper insights into the motivation behind approving and using in-home voice assistants, this work shifts its focus towards examining voice interactions. Utilization of speech recognition for converting speech into text enables constructing an attractive personalized assistant. This development facilitates seamless tasks such as sending emails without manual typing, conducting Google searches without opening the browser, and accomplishing various daily activities including playing music or launching a preferred IDE, all with a single voice command. To assess its performance, the system is seamlessly integrated with home automation (open-source) environment and operated continuously for some days. During this period, users are encouraged to interact with it, and the results demonstrate increased accuracy, reliability and appeal. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Offline Signature Verification: An Extensive Survey of Deep Learning Methods 4

Offline Signature Verification: An Extensive Survey of Deep ...

引用

4th International Conference on Sentiment Analysis and Deep Learning, ICSADL 2025

作者： Pandey, Gaurav Kumar Raj, Vansh Agarwal, Ayush Dixit, Mayank Chauhan, Sachin Singh Srivastava, Sumit Birla Institute of Technology Mesra Computer Science and Engineering Department Ranchi India Hindustan College of Science and Technology Computer Science and Engineering Department Mathura India Hindustan College of Science and Technology Cse Department Mathura India Birla Institute of Technology Mesra Cse Department Ranchi India

ISBN: (纸本)9798331523923

Offline signature verification is essential for identity authentication and fraud mitigation in many applications, such as banking and legal documentation. Conventional approaches to signature verification predominantly depended on manually designed features and statistical methodologies, although they frequently encountered difficulties with intricate intra-class variances and inter-class similarities. In recent years, deep learning has arisen as a potent instrument to address these difficulties through autonomous feature extraction and resilient categorization. This study offers an extensive examination of cutting-edge deep learning methodologies for offline signature verification. We examine fundamental designs like Convolutional Neural Networks (CNNs), hybrid models, and transfer learning methodologies, and evaluate their efficacy on renowned signature datasets such as CEDAR and GPDS. Furthermore, we examine the difficulties related to imbalanced datasets, counterfeit detection, and the choice of suitable assessment metrics, including accuracy, False Acceptance Rate (FAR), and False Rejection Rate (FRR). This paper identifies existing limits in DL methodologies and proposes future research routes to tackle challenges including dataset scarcity, model generalization, and computational efficiency. Our findings provide studies with insights into the advantages and disadvantages of diverse deep learning methodologies, thereby aiding in advancing more precise and efficient signature verification systems. © 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

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：