检索结果-内蒙古大学图书馆

Deep Learning-Based CNN Multi-Modal Camera Model Identification for Video Source Identification

Informatica (Slovenia) 2023年第3期47卷 417-430页

作者： Singh, Surjeet Sehgal, Vivek Kumar Department of Computer Science and Engineering and Information Technology Jaypee University of Information Technology Solan Himachal Pradesh Waknaghat India

Here is a high demand for multimedia forensics analysts to locate the original camera of photographs and videos that are being taken nowadays. There has been considerable progress in the technology of identifying the source of data, which has enabled conflict resolutions involving copyright infringements and identifying those responsible for serious offenses to be resolved. Video source identification is a challenging task nowadays due to easily available editing tools. This study focuses on the issue of identifying the camera model used to acquire video sequences used in this research that is, identifying the type of camera used to capture the video sequence under investigation. For this purpose, we created two distinct CNN-based camera model recognition techniques to be used in an innovative multi-modal setting. The proposed multi-modal methods combine audio and visual information in order to address the identification issue, which is superior to mono-modal methods which use only the visual or audio information from the investigated video to provide the identification information. According to legal standards of admissible evidence and criminal procedure, Forensic Science involves the application of science to the legal aspects of criminal and civil law, primarily during criminal investigations, in line with the standards of admissible evidence and criminal procedure in the law. It is responsible for collecting, preserving, and analyzing scientific evidence in the course of an investigation. It has become a critical part of criminology as a result of the rapid rise in crime rates over the last few decades. Our proposed methods were tested on a well-known dataset known as the Vision dataset, which contains about 2000 video sequences gathered from various devices of varying types. It is conducted experiments on social media platforms such as YouTube and WhatsApp as well as native videos directly obtained from their acquisition devices by the means of their acquisiti

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Noise4Denoise:Leveraging noise for unsupervised point cloud denoising

引用

Computational Visual Media 2024年第4期10卷 659-669页

作者： Weijia Wang Xiao Liu Hailing Zhou Lei Wei Zhigang Deng Manzur Murshed Xuequan Lu Deakin University GeelongVIC 3216Australia Department of Mechanical Engineering and Product Design Engineering Swinburne University of TechnologyMelbourneVIC 3122Australia Department of Computer Science University of HoustonHoustonTX 77204-3027USA Department of Computer Science and Information Technology La Trobe UniversityMelbourneVIC 3086Australia

Existing deep learning-based point cloud denoising methods are generally trained in a supervised manner that requires clean data as ground-truth ***,in practice,it is not always feasible to obtain clean point *** this paper,we introduce a novel unsupervised point cloud denoising method that eliminates the need to use clean point clouds as groundtruth labels during *** demonstrate that it is feasible for neural networks to only take noisy point clouds as input,and learn to approximate and restore their clean *** particular,we generate two noise levels for the original point clouds,requiring the second noise level to be twice the amount of the first noise *** this,we can deduce the relationship between the displacement information that recovers the clean surfaces across the two levels of noise,and thus learn the displacement of each noisy point in order to recover the corresponding clean *** experiments demonstrate that our method achieves outstanding denoising results across various datasets with synthetic and real-world noise,obtaining better performance than previous unsupervised methods and competitive performance to current supervised methods.

关键词： point clouds denoising geometry processing deep learning

来源：评论

学校读者我要写书评

暂无评论

Selective feature-based ovarian cancer prediction using MobileNet and explainable AI to manage women healthcare

引用

Multimedia Tools and Applications 2024年 1-22页

作者： Almujally, Nouf Abdullah Alzahrani, Abdulrahman Hakeem, Abeer M. Attiah, Afraa Umer, Muhammad Alsubai, Shtwai Polsinelli, Matteo Ashraf, Imran Department of Information Systems College of Computer and Information Sciences Princess Nourah bint Abdulrahman University P.O. Box 84428 Riyadh11671 Saudi Arabia Department of Computer Science and Engineering College of Computer Science and Engineering University of Hafr Al Batin Hafr Al Batin39524 Saudi Arabia Department of Information Technology Faculty of Computing and Information Technology King Abdulaziz University Jeddah21589 Saudi Arabia Department of Computer Science The Islamia University of Bahawalpur Bahawalpur Pakistan Department of Computer Science College of Computer Engineering and Sciences Prince Sattam bin Abdulaziz University P.O. Box 151 Al-Kharj11942 Saudi Arabia Department of Management & amp Innovation Systems University of Salerno Fisciano84084 Italy Department of Information and Communication Engineering Yeungnam University Gyeongsan38541 Korea Republic of

Ovarian cancer is a global health concern due to the unavailability of an effective screening strategy and is often diagnosed at a late stage with approximately 70% of the case which reduces the survival chances of patients. Initial diagnosis is challenging due to inconspicuous symptoms in its initial stages which complicate its timely diagnosis. Regular screenings, such as pelvic exams, ultrasounds, and blood tests targeting specific biomarkers can be helpful for early diagnosis. In addition, the use of machine learning models can help automate this process thereby assisting medical experts for accurate diagnosis. This study aims at timely and accurate detection of ovarian cancer using a transfer learning approach that uses the MobileNet model. Moreover, the Chi-square technique is used to extract the most impactful features for better accuracy. For experiments, the Soochow University ovarian cancer dataset is employed. Extensive experiments are performed using all features from the dataset, as well as, using Chi-square-based selective features. The best accuracy of 90.88% is achieved with the Xception model when all features are used. Experimental results show a substantial increase when selective features are used indicating a 98.49% accuracy using the MobileNet model with 20 most important features. In addition, precision, recall, and F1 scores of 99.13%, 99.27%, and 99.20% are obtained showing the model’s robustness and generalization. This study also employs Shapley additive explanations to explain the importance of various features toward the model’s output thereby providing transparency for the model’s decision-making process. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Approach for Detecting Malware Using Autoencoder

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer Science 2024年第8期51卷 1051-1059页

作者： Panchagnula, Venu Madhav Satya Keerthi, N.V.L.Ch. Surekha, S. Sujatha, R. Veeraiah, Duggineni Ramesh, Eluri Lakshmi, B. Electronics and Communication Engineering Department Prasad V Potluri Siddhartha Institute of Technology Andhra Pradesh Vijayawada India Computer Science and Engineering Department Koneru Lakshmaiah Education Foundation Andhra Pradesh Guntur India AIML Department Vignana Bharathi Institute of Technology Ghatkesar Hyderabad India Information Technology Department M. Kumarasamy College of Engineering Karur India Computer Science and Engineering Department Lakireddy Bali Reddy College of Engineering Andhra Pradesh Mylavaram India Computer Science and Engineering Department R.V.R & J. C. College of Engineering Guntur India Electronics and Communication Engineering Department Rajiv Gandhi University of Knowledge Technologies Andhra Pradesh Nuzvid India

Research in the field of malware detection is currently leaning toward methods based on artificial intelligence algorithms due to the increasing limitations of traditional detection methods and the increasing accuracy of these methods. This study therefore introduces a novel approach to malware detection. This method integrates a deep learning model's autoencoder network with a grayscale graphic depiction of malware. By looking at the autoencoder's reconstruction error, we may assess if the grey-scale picture approach is feasible to use for malware detection. Additionally, it differentiates between malicious and safe software by utilizing the autoencoder's dimensionality reduction capabilities. Using the Android dataset, our newly introduced detection model attained a 95% accuracy and maintained a consistently high F-score of approximately 95%. This outperformed more conventional methods of machine learning detection. © (2024), (International Association of Engineers). All Rights Reserved.

关键词： Malware

来源：评论

学校读者我要写书评

暂无评论

A KUSHO WORDS RECOGNITION METHOD FOR CONTACTLESS information MANAGEMENT IN RESTAURANT KITCHENS 51

A KUSHO WORDS RECOGNITION METHOD FOR CONTACTLESS INFORMATION...

引用

51st International Conference on computers and Industrial engineering, CIE 2024

作者： Usuki, Risa Usuki, Jun Department of Computer Science and Engineering Waseda University Tokyo Japan Department of Information Network and Communication Kanagawa Institute of Technology Kanagawa Japan

Advances in IT technology have led to research into KUSHO recognition, which enables the non-contact operation of devices without the use of devices such as keyboards or controllers. This is a technology that transmits information about the characters written in space towards the camera (hereafter this action is referred to as KUSHO) to the devices. KUSHO has advantages, such as the ability to transfer information without training. In restaurant kitchens, workers transmit various types of information using devices with input functions, however, hygiene considerations are essential when operating equipment in kitchens. In addition, in noisy kitchens, it is also difficult to use voice input, and there are limits to paper-based information transmission. Under these circumstances, in this research, the introduction of KUSHO input technology for non-contact information transfer in restaurant kitchens is being investigated. Therefore, the content handled by kitchen workers for information management is first analyzed and shown to be representable in three input formats. Subsequently, an additional hand signal to be used while doing KUSHO and an improved algorithm for determining them are proposed. Finally, an algorithm for KUSHO recognition of character strings using standard input time is proposed. Experiments on the implementation of these proposals have shown the effectiveness of the new decision algorithm for the addition of a hand signal, with a 100% correct rate. For strings of two or more uppercase alphabetic characters, the recognition rate was 92.5% for all eight types of "KUSHO operation" and the recognition rate was 70.0% for ten randomly selected from 100 different "product names." For "quantity," it was difficult to recognize multiple digits, however 100% recognition rate was achieved by dividing the digits into each single digit 0~9 using an improved hand signal algorithm. Thus, algorithm some problems remain, the effectiveness of all the proposed methods ha

关键词： information management

来源：评论

学校读者我要写书评

暂无评论

Fractional Heat Kernel Smoothing Using Fractional Order Graph Laplacian Matrix 13

Fractional Heat Kernel Smoothing Using Fractional Order Grap...

引用

13th IEEE Global Conference on Consumer Electronic, GCCE 2024

作者： Tseng, Chien-Cheng Lee, Su-Ling National Kaohsiung University of Science and Technology Department of Computer and Communication Engineering Kaohsiung Taiwan Chang Jung Christian University Department of Computer Science and Information Engineering Tainan Taiwan

ISBN: (纸本)9798350355079

Heat kernel smoothing (HKS) over graph is a valuable tool for reducing noise in irregular graph signals. In the conventional HKS method, the discrete heat equation (DHE) is derived by replacing the continuous Laplacian operator with the graph Laplacian matrix. The solution to the DHE is then used to compute the denoised signal from the measured noisy signal. This paper studies the fractional heat kernel smoothing (FHKS) method. Firstly, the fractional discrete heat equation (FDHE) is derived by replacing the continuous Laplacian operator with a fractional-order graph Laplacian matrix. The solution to the FDHE is then employed to calculate the denoised signal. The conventional HKS method is a special case of the proposed FHKS method when the order is chosen as one. Secondly, a graph filtering implementation structure of the FHKS method is investigated. Thirdly, the relationship between the improvement of the signal-to-noise ratio (SNR) and the fractional order is studied to determine the optimal order for noise reduction. Finally, temperature data from a sensor network collected in Taiwan is used to demonstrate the effectiveness of the proposed FHKS method. © 2024 IEEE.

关键词： Signal denoising

来源：评论

学校读者我要写书评

暂无评论

Deep learning intelligence for influencer-based topological classification for online social networks

引用

International Journal of Innovative Computing and Applications 2023年第4期14卷 211-219页

作者： Jain, Somya Sinha, Adwitiya Department of Computer Science and Engineering and Information Technology Jaypee Institute of Information Technology 62 Uttar Pradesh Noida India

Social network analysis provides quantifiable methods and topological metrics to examine the networked structure for several interdisciplinary applications. In our research, a social network of GitHub community is constructed that forms a dense network of 37,700 developers with 289,003 associations amongst them. The research involves finding the central developers in the GitHub network using graph analytics and benchmark centrality metrics, including degree, betweenness, closeness, PageRank and eigenvector;which is based upon network structural information. Our research methodology provides a breakthrough towards predicting the classification of GitHub users using artificial intelligence-based learning model trained with derived topological network centrality metrics. The proposed approach performs feature extraction for the developers by computing centrality score of each user followed by building correlation matrix using centrality parameters based on network topology. Further, the derived topological centrality scores were used as input features to train and build artificial intelligence-based models for classification. Our experimentation is better performance of artificial neural network over autoencoders, logistic regression and hyper-parameter tuned support vector machine. Certain intermediate outcomes include correlation, principal component analysis, loss monitoring, etc. The performance evaluation was performed in terms of macro and weighted F1-score, recall, precision, and accuracy. Copyright © 2023 Inderscience Enterprises Ltd.

关键词： Social networking (online)

来源：评论

学校读者我要写书评

暂无评论

Ransomware Detection Utilizing Ensemble Based Interpretable Deep Learning Model 2

Ransomware Detection Utilizing Ensemble Based Interpretable ...

引用

2nd International Conference on Power, Electrical, Electronics and Industrial Applications, PEEIACON 2024

作者： Chisty, Tanjir Alam Rahman Rahman, Md. Mahbubur University of Professionals Department of Information and Communication Technology Bangladesh Dhaka Bangladesh Military Institute of Science and Technology Department of Computer Science and Engineering Dhaka Bangladesh

ISBN: (纸本)9798331517984

Ransomware, a form of malicious software, encrypts victim data and renders it inaccessible until a ransom is paid. Different ransomware variants employ various tactics to avoid detection during their attacks. Understanding the techniques used in ransomware development and deployment is crucial to counter these threats. This study presents a stacked ensemble-based machine learning (ML) approach for ransomware detection. The ensemble model consists of Naive Bayes (NB), Random Forest (RF), and K-Nearest Neibors (KNN) as base classifiers and Artificial Neural Network (ANN) as meta-classifiers. Additionally, Explainable AI (XAI) utilizing LIME and SHAP has been applied to gain insights into prediction outcomes and validate model correctness. The use of LIME and SHAP ensures the inter-pretability of the model, which is a crucial factor in real-world implications. The research employs a publicly available dataset and evaluates the model using multiple metrics. Experimental results reveal that the proposed model achieves an impressive accuracy of 98.57%. The model is benchmarked against existing ML techniques, demonstrating its superior performance. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Modeling and Simulation of Virtual Reality Assistive Tool for Children with Autism Spectral Disorder 2nd

Modeling and Simulation of Virtual Reality Assistive Tool fo...

引用

2nd International Conference on Deep Sciences for Computing and communications, IconDeepCom 2023

作者： NirmalaPriya, G. Shree Aravindh, S.S. Tharun, D. Deepak, M. Hermalin, Cynthiya S. Department of Electronics and Communication Engineering Rajalakshmi Institute of Technology Chennai India Department of Computer and Communication Engineering Rajalakshmi Institute of Technology Chennai India

ISBN: (纸本)9783031689079

Autism spectrum disorder (ASD) is a prevalent issue contributing to deprivation among people in terms of their capability to maintain social relationships and interact normally with fellow members of society. Teens with ASD battle to exercise proper sociable behavior as well as mingle with others sharing the same age range. It affects all sorts of races, ethnicities and budgets alike. Despite its long- term nature, ASD has therapies and treatments that can help address issues associated with it such as restlessness, aggression, repetitive behavior, focus difficulties, worry and depression through the use of medicine. The use of virtual reality-based training, in addition to medical intervention, may be used to address various behavioral tradeoffs through repeated conditioning and learning. A variety of learning scenarios that would be unsafe, challenging, or time-consuming to set up or frequently recreate in the real world can now be simulated using virtual reality (VR) technology. In this paper, a virtual reality simulator for with the aid of social educators, achieve this goal. This study used the gaming program Social Activity Trainer to simulate and construct a virtual reality (VR)-based assistive aid (SAT). The assistive tool was created to help youngsters with ASD interact and use social skills. Based on three key skills—facial expression recognition, reading comprehension, and task delivery—seven tasks were created for the VR assistive tool. There are seven modules in a virtual reality scenario, including one for music therapy. The EEG signals are looked at to assess the degree of autism in 10 ASD children who participated in the test. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

Intelligent Radio Spectrum Management and Dynamic Spectrum Allocation Algorithm

Intelligent Radio Spectrum Management and Dynamic Spectrum A...

引用

2024 International Conference on Power, Electrical engineering, Electronics and Control, PEEEC 2024

作者： Fan, Xuechao Suzhou College of Information Technology Department of Communication and Information Engineering Jiangsu 215200 China

ISBN: (纸本)9798350378917

With the surge in wireless communication services, spectrum resources have grown scarce, posing a significant hindrance to the industry's expansion. Traditional static spectrum allocation falls short in addressing the escalating demands from communication terminals, particularly under conditions of fluctuating network loads and intricate communication environments. Consequently, exploring intelligent radio spectrum management and dynamic allocation algorithms carries substantial practical importance and application potential. This paper introduces an intelligent radio spectrum management and dynamic allocation approach leveraging deep learning (DL). This approach utilizes a DL model to perceive and forecast the wireless environment in real-time, enabling dynamic adjustments to spectrum resource allocation based on real-time factors like network load and communication quality. By analyzing vast historical data, the DL model identifies patterns and trends in spectrum usage, thereby facilitating precise decision-making for allocation. Experimental results indicate that, in contrast to conventional static methods, this algorithm exhibits greater flexibility and adaptability, effectively navigating complex, evolving wireless landscapes and diverse communication requirements. © 2024 IEEE.

关键词： Decision making

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：