检索结果-内蒙古大学图书馆

ROBDD-TrOCRBERTa: a novel robust-optimized blurred document text deblurring and completion with DCGAN-TrOCR and DistilRoBERTa

引用

International Journal of Information Technology (Singapore) 2024年第7期16卷 4611-4619页

作者： Ranjan, Arti Ravinder, M. Department of Computer Science and Engineering Indira Gandhi Delhi Technical University for Women (IGDTUW)

Blurred text documents such as historical documents, handwritten manuscripts, old newspapers, moist invoices or legal agreements, old books, hand written notes often present readability challenges because the quality of the text has deteriorated over time. The proposed Robust Optimized Blurred Document Text Deblurring and Text Recognition (ROBDD-TrOCRBERTa) method tackles the challenge of improving readability in deteriorated text documents such as historical documents, handwritten manuscripts, and old newspapers. This innovative approach is divided into two phases. First, it employs Deblurring using DCGAN to enhance image quality by reducing noise and blurriness. Subsequently, it leverages TrOCR integrated with DistilRoBERTa for efficient text recognition and completion. Experimental results show that this method is effective in various real-world scenarios, making it a promising solution for automated document analysis and digitization in challenging conditions. © Bharati Vidyapeeth's Institute of computer Applications and Management 2024.

关键词： DistilRoBERTa NLP-based post-processing ROBDD-TrOCRBERTa Robust-optimized blurred document deblurring Text deblurring TrOCR

来源：评论

学校读者我要写书评

暂无评论

Global video object segmentation with spatial constraint module

引用

Computational Visual Media 2023年第2期9卷 385-400页

作者： Yadang Chen Duolin Wang Zhiguo Chen Zhi-Xin Yang Enhua Wu Engineering Research Center of Digital Forensics Ministry of EducationSchool of Computer and SoftwareNanjing University of Information Science and TechnologyNanjing 210044China State Key Laboratory of Internet of Things for Smart City Department of Electromechanical EngineeringUniversity of MacaoMacao 999078China State Key Laboratory of Computer Science Institute of SoftwareUniversity of Chinese Academy of SciencesBeijing 100190China Faculty of Science and Technology University of MacaoMacao 999078China

We present a lightweight and efficient semisupervised video object segmentation network based on the space-time memory *** some extent,our method solves the two difficulties encountered in traditional video object segmentation:one is that the single frame calculation time is too long,and the other is that the current frame’s segmentation should use more information from past *** algorithm uses a global context(GC)module to achieve highperformance,real-time *** GC module can effectively integrate multi-frame image information without increased memory and can process each frame in real ***,the prediction mask of the previous frame is helpful for the segmentation of the current frame,so we input it into a spatial constraint module(SCM),which constrains the areas of segments in the current *** SCM effectively alleviates mismatching of similar targets yet consumes few additional *** added a refinement module to the decoder to improve boundary *** model achieves state-of-the-art results on various datasets,scoring 80.1%on YouTube-VOS 2018 and a J&F score of 78.0%on DAVIS 2017,while taking 0.05 s per frame on the DAVIS 2016 validation dataset.

关键词： video object segmentation semantic segmentation global context(GC)module spatial constraint

来源：评论

学校读者我要写书评

暂无评论

An Immutable Framework for Smart Healthcare Using Blockchain Technology

引用

computer Systems science & engineering 2023年第7期46卷 165-179页

作者： Faneela Muazzam A.Khan Suliman A.Alsuhibany Walid El-Shafai Mujeeb Ur Rehman Jawad Ahmad Department of Computer Science Quaid-i-Azam UniversityIslamabad45320Pakistan Department of Computer Science College of ComputerQassim UniversityBuraydah51452Saudi Arabia Department of Electronics and Electrical Communications Engineering Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt Department of Computer Science Security Engineering LaboratoryPrince Sultan UniversityRiyadh11586Saudi Arabia James Watt School of Engineering University of GlasgowG128QQUK School of Computing Edinburgh Napier UniversityEdinburghEH105DTUK

The advancements in sensing technologies,information processing,and communication schemes have revolutionized the healthcare *** Healthcare Records(EHR)facilitate the patients,doctors,hospitals,and other stakeholders to maintain valuable data and medical *** traditional EHRs are based on cloud-based architectures and are susceptible to multiple cyberattacks.A single attempt of a successful Denial of Service(DoS)attack can compromise the complete healthcare *** article introduces a secure and immutable blockchain-based framework for the Internet of Medical Things(IoMT)to address the stated *** proposed architecture is on the idea of a lightweight private blockchain-based network that facilitates the users and hospitals to perform multiple healthcare-related operations in a secure and trustworthy *** efficacy of the proposed framework is evaluated in the context of service execution time and *** experimental outcomes indicate that the proposed design attained lower service execution time and higher throughput under different control parameters.

关键词： Blockchain technology healthcare applications cybersecurity services IoMT DoS EHR

来源：评论

学校读者我要写书评

暂无评论

A Study on Cypher Assignment Based on Secrecy Capacity

引用

International Journal of Network Security 2024年第2期26卷 167-172页

作者： Yang, Cheng-Ying Chen, Jong-Shin Yao, Jenq-Foung J F Hwang, Min-Shiang Department of Computer Science University of Taipei Taipei Taiwan Department of Information and Communication Engineering Chaoyang University of Science and Technology Taiwan Computer Science Georgia College and State University Milledgeville United States Department of Computer Science & Information Engineering Asia University Taiwan Fintech and Blockchain Research Center Asia University Taiwan 500 Lioufeng Rd. Wufeng Taichung41354 Taiwan

Accessing wireless services is convenient but not secure. The issue of privacy and security is highly demanded on the internet. The information is sent with the undesired threats of eavesdroppers. Secure communication enables the desired receiver to correctly and successfully receive the information without interruption. In the meantime, a secure system protects the transmitted information from the eavesdroppers. Traditionally, encryption, authorization, and authentication schemes could be employed to protect from eavesdroppers. However, the unanticipated attacks have been developing. According to Shannon’s perfect communication, the undesired receivers, such as eavesdroppers, could receive an error-free cryptogram. Based on the physical-layer security, it could be expected that secure communication could be implemented without complex computing at the higher level. The physical-layer security coding scheme aims to achieve an extremely low decoding error probability and obtain a critical security level against attackers. The positive secrecy capacity (secrecy rate) could provide secure communication. Hence, a positive secrecy capacity is the primary concern in the system. Besides, physical-layer security coding is not only to help the system achieve a critical level of security but also to obtain an extremely low decoding error probability. However, it decreases the bandwidth efficiency and suffers the system performance. Hence, the tradeoff between the security level and the coding rate is important to lead a secure communication. © (2023), (Femto Technique Co.). All Rights Reserved.

关键词： Bandwidth

来源：评论

学校读者我要写书评

暂无评论

Correlation-Based Weight Algorithm for Diagnosing Parkinson’s Induced Voice Disorder

引用

SN computer science 2024年第7期5卷 841页

作者： Indu, Richa Dimri, Sushil Chandra Department of Computer Science and Engineering Graphic Era Deemed to be University Uttarakhand Dehradun India

Dysarthria, dysphonia, and hypophonia are the early symptoms of Parkinson's disease. Thus, vocal attributes are frequently used for the initial diagnosis of Parkinson's disease, but the role of gender in these features is usually overlooked. So, the present study proposes a multi-tasking algorithm that attempts to examine the relevance of gender in Parkinson's disease detection and accomplishes feature selection and identification. This study used three correlation coefficients, Pearson R, Spearman's Rho, and Kendall Tau, to generate a weight factor for feature classification. Based on these weights, each attribute in the dataset has been categorized into three categories: accordant, assorted, and divergent. For feature selection, combinations of these categories are utilized and the performance is assessed using Random Forest, Support Vector Machine, K-Nearest Neighbors, and Decision Tree classifiers. According to the experimental results, on gender-diverse data, features acquired under the combination of Accordant-Assorted categories are the best identifiers of Parkinson's induced voice impairments, which attained a maximum accuracy of 99.99% on gender-diverse data (Datasets 1 and 2). Similarly, in gender-based classification (Dataset 3) features in the Divergent category are most effective, achieving a maximum of 99.98% accuracy for female voice samples, and 92.31% for male voice features. This study identified Random Forest as the finest classifier across both gender-diverse and gender-based datasets. Surpassing the existing methods, the suggested algorithm effectively assesses gender as a vital factor in identifying Parkinson-related vocal abnormalities. However, further investigation is needed in better diagnosis of Parkinson's disorder using male voice traits. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Correlation coefficients Decision tree K-nearest neighbors Parkinson’s disease Random forest Support vector machine

来源：评论

学校读者我要写书评

暂无评论

Byte Frequency Based Indicators for Crypto-Ransomware Detection from Empirical Analysis

引用

Journal of computer science & Technology 2022年第2期37卷 423-442页

作者： Geun Yong Kim Joon-Young Paik Yeongcheol Kim Eun-Sun Cho Department of Computer Science and Engineering Chungnam National UniversityDaejeon 34134South Korea School of Computer Science and Technology Tiangong UniversityTianjin 300387China

File entropy is one of the major indicators of crypto-ransomware because the encryption by ransomware increases the randomness of file ***,entropy-based ransomware detection has certain limitations;for example,when distinguishing ransomware-encrypted files from normal files with inherently high-level entropy,misclassification is very *** addition,the entropy evaluation cost for an entire file renders entropy-based detection impractical for large *** this paper,we propose two indicators based on byte frequency for use in ransomware detection;these are termed EntropySA and DistSA,and both consider the interesting characteristics of certain file subareas termed“sample areas”(SAs).For an encrypted file,both the sampled area and the whole file exhibit high-level randomness,but for a plain file,the sampled area embeds informative structures such as a file header and thus exhibits relatively low-level randomness even though the entire file exhibits high-level *** and DistSA use“byte frequency”and a variation of byte frequency,respectively,derived from sampled *** indicators cause less overhead than other entropy-based detection methods,as experimentally proven using realistic ransomware *** evaluate the effectiveness and feasibility of our indicators,we also employ three expensive but elaborate classification models(neural network,support vector machine and threshold-based approaches).Using these models,our experimental indicators yielded an average Fl-measure of 0.994 and an average detection rate of 99.46%for file encryption attacks by realistic ransomware samples.

关键词： computer security cryptography machine learning ransomware

来源：评论

学校读者我要写书评

暂无评论

Sign Language Translation and Voice Impairment Support System using Deep Learning 5

Sign Language Translation and Voice Impairment Support Syste...

引用

5th International Conference on Data Intelligence and Cognitive Informatics, ICDICI 2024

作者： Shajeena, J. Shiny, R.M. Mary Vespa, M. Kavitha, R. Srm Institute of Science and Technology Department of Computer Science and Engineering Tiruchirapalli Campus India St Joseph's College of Engineering Department of Computer Science and Engineering Chennai India Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Department of Computer Science and Engineering Chennai India Vel Tech High Tech Dr. Rangarajan Dr. Sakunthala Engineering College Department of Computer Science and Engineering Chennai India

ISBN: (数字)9798350389609

ISBN: (纸本)9798350389609

The Sign Language Translation and Voice Impairment Support System (SLT-VISS) represents a groundbreaking application of deep learning methodology aimed at facilitating communication for individuals with hearing impairments or speech disabilities. By utilizing Convolutional Neural Network (CNN), OpenCV for image processing, and Pyttsx3 for text-to-voice conversion, SLT-VISS offers real-time translation of sign language gestures into both text and speech output. Using a camera as input, the CNN algorithm accurately identifies and classifies sign language gestures displayed in front of it. OpenCV complements this process by preprocessing the input images for gesture segmentation and feature extraction, enhancing the accuracy of gesture recognition. Once a gesture is recognized, Pyttsx3 converts the translated text into synthesized speech, providing immediate auditory feedback to the user. SLT-VISS is designed to bridge the communication gap faced by individuals with communication disabilities, enabling them to express themselves effectively in various social and professional settings. By seamlessly translating sign language into both text and speech, this system promotes inclusivity and accessibility, empowering users to communicate with confidence and independence. The integration of CNN, OpenCV, and Pyttsx3 technologies within SLT-VISS signifies a significant advancement in assistive technology. As technology continues to evolve, further enhancements to SLT-VISS holds the potential to revolutionize communication accessibility and enable greater inclusivity in society. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Credit card default prediction using ML and DL techniques

引用

Internet of Things and Cyber-Physical Systems 2024年第1期4卷 293-306页

作者： Wahab, Fazal Khan, Imran Sabada, Sneha Corvit Systems Peshawar Pakistan Department of Informatics University of Sussex United Kingdom Department of Engineering and Computer Science University of Hertfordshire United Kingdom

The banking sector is widely acknowledged for its intrinsic unpredictability and susceptibility to risk. Bank loans have emerged as one of the most recent services offered over the past several decades. Banks typically serve as intermediaries for loans, investments, short-term loans, and other types of credit. The usage of credit cards is experiencing a steady increase, thereby leading to a rise in the default rate that banks encounter. Although there has been much research investigating the efficacy of conventional Machine Learning (ML) models, there has been relatively less emphasis on Deep Learning (DL) techniques. The application of DL approaches to credit card default prediction has not been extensively researched despite their considerable potential in numerous fields. Moreover, the current literature frequently lacks particular information regarding the DL structures, hyperparameters, and optimization techniques employed. To predict credit card default, this study evaluates the efficacy of a DL model and compares it to other ML models, such as Decision Tree (DT) and Adaboost. The objective of this research is to identify the specific DL parameters that contribute to the observed enhancements in the accuracy of credit card default prediction. This research makes use of the UCI ML repository to access the credit card defaulted customer dataset. Subsequently, various techniques are employed to preprocess the unprocessed data and visually present the outcomes through the use of exploratory data analysis (EDA). Furthermore, the algorithms are hypertuned to evaluate the enhancement in prediction. We used standard evaluation metrics to evaluate all the models. The evaluation indicates that the AdaBoost and DT exhibit the highest accuracy rate of 82 % in predicting credit card default, surpassing the accuracy of the ANN model, which is 78 %. © 2024 The Authors

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

A Novel Method Based on CNN-LSTM to Characterize Knee Osteoarthritis from Radiography

引用

Proceedings of the National Academy of sciences India Section B - Biological sciences 2024年第2期94卷 423-438页

作者： Malathi, S.Y. Bharamagoudar, Geeta R. Department of Computer Science and Engineering KLE Institute of Technology Hubballi Karnataka Hubballi India

This is particularly true for the senior population, whose quality of life has been drastically reduced as a result of the increasing incidence of several health problems. Over 27 million people in the United States suffer from osteoarthritis of the knee (OAK), a painful condition that may severely limit mobility. When the articular cartilage between the tibia and femur in the knee is damaged, osteoarthritis of the knee develops. OAK symptoms include a loss of mobility and the inability to walk normally due to knee pain, detected using an X-ray. We detail here a novel method of evaluating the severity of knee osteoarthritis (OA) by X-ray analysis. Modern methods are comprised of pre-processing, feature extraction using a convolutional neural network (CNN), and classification with latent semantic modeling (LSM) (LSTM). Data from the osteoarthritis initiatives (OAI) database, which is available to the public, was utilized to evaluate the methodology proposed. The current method has been shown to be effective, and the OAI database has information on KL grade assessment for both knees. OAK is the subject of state-of-the-art, global observational investigation by experts using a program called OAI. This collection was created to serve as a one-stop shop for researchers seeking the scholarly materials they need to systematically examine OA indicators as a possible endpoint for the advanced stages of the illness. The statistics reveal a mean accuracy of 100%. When compared to earlier deep learning approaches, these outcomes are much superior. © The Author(s), under exclusive licence to The National Academy of sciences, India 2024.

关键词： Deep learning Knee osteoarthritis Radiography X-ray

来源：评论

学校读者我要写书评

暂无评论

A Novel Divide and Conquer Solution for Long-term Video Salient Object Detection

引用

Machine Intelligence Research 2024年第4期21卷 684-703页

作者： Yun-Xiao Li Cheng-Li-Zhao Chen Shuai Li Ai-Min Hao Hong Qin State Key Laboratory of Virtual Reality Technology and Systems Beihang UniversityBeijing 100191China College of Computer Science and Technology China University of Petroleum(East China)Qingdao 266580China Department of Computer Science Stony Brook UniversityNew York 11794USA

Recently,a new research trend in our video salient object detection(VSOD)research community has focused on enhancing the detection results via model self-fine-tuning using sparsely mined high-quality keyframes from the given *** such a learning scheme is generally effective,it has a critical limitation,i.e.,the model learned on sparse frames only possesses weak generalization *** situation could become worse on“long”videos since they tend to have intensive scene ***,in such videos,the keyframe information from a longer time span is less relevant to the previous,which could also cause learning conflict and deteriorate the model ***,the learning scheme is usually incapable of handling complex pattern *** solve this problem,we propose a divide-and-conquer framework,which can convert a complex problem domain into multiple simple ***,we devise a novel background consistency analysis(BCA)which effectively divides the mined frames into disjoint *** for each group,we assign an individual deep model on it to capture its key attribute during the fine-tuning *** the testing phase,we design a model-matching strategy,which could dynamically select the best-matched model from those fine-tuned ones to handle the given testing *** experiments show that our method can adapt severe background appearance variation coupling with object movement and obtain robust saliency detection compared with the previous scheme and the state-of-the-art methods.

关键词： Video salient object detection background consistency analysis weakly supervised learning long-term information background shift.

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：