检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Chakravarthi, Bharatesh Prasad, Prabhu B.M. Kumar, Pavan B.N. School of Computing and Augmented Intelligence Arizona State University TempeAZ United States Dept. of Computer Science and Engineering Indian Institute of Information Technology Dharwad India Dept. of Computer Science and Engineering Indian Institute of Information Technology Andhra Pradesh Sri City India

This paper comprehensively reviews hand gesture datasets based on Ultraleap’s leap motion controller, a popular device for capturing and tracking hand gestures in real-time. The aim is to offer researchers and practitioners a valuable resource for developing and evaluating gesture recognition algorithms. The review compares various datasets found in the literature, considering factors such as target domain, dataset size, gesture diversity, subject numbers, and data modality. The strengths and limitations of each dataset are discussed, along with the applications and research areas in which they have been utilized. An experimental evaluation of the leap motion controller 2 device is conducted to assess its capabilities in generating gesture data for various applications, specifically focusing on touchless interactive systems and virtual reality. This review serves as a roadmap for researchers, aiding them in selecting appropriate datasets for their specific gesture recognition tasks and advancing the field of hand gesture recognition using leap motion controller technology. © 2023, CC BY.

关键词： Palmprint recognition

来源：评论

学校读者我要写书评

暂无评论

Modified EfficientNetB3 Deep Learning Model to Classify Colour Fundus Images of Eye Diseases

Modified EfficientNetB3 Deep Learning Model to Classify Colo...

引用

Cybernetics, Cognition and Machine Learning Applications (ICCCMLA), IEEE International Conference on

作者： Riya Sharma Jayesh Gangrade Shweta Gangrade Ashish Mishra Gautam Kumar Vinit Kumar Gunjan Dept of Computer Science & Engineering Manipal University Jaipur Jaipur India Dept. of Artificial Intelligence & Machine Learning Manipal University Jaipur Jaipur India Dept of Information Technology Manipal University Jaipur Jaipur India Dept of Computer Science & Engineering Jaypee Institute of Information Technology Noida India Department of Computer Science & Engineering CMR Institute of Technology Hyderabad India

The only way to prevent blindness from eye problems is by early detection and prompt treatment. Although colour fundus photography (CFP) is useful for fundus inspection, there is a need for computer-assisted automated diagnosis tools due to the similarities between the early symptoms of many eye disorders. The suggested approach uses cutting-edge deep learning model to categorize images into several disease categories by learning distinguishing features from the input images. The high-resolution fundus photos from individuals with diabetic retinopathy (DR), glaucoma, cataract, and healthy eyes make up most of the dataset used for this research. The experimental findings show that the suggested system achieve 97% accuracy with modified efficientNetB3 model and surpasses current approaches for categorizing eye diseases. This approach may help doctors diagnose and treat eye conditions earlier, leading to better patient outcomes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Image Captioning-A Comprehensive Encoder-Decoder Approach on Flickr8K

Image Captioning-A Comprehensive Encoder-Decoder Approach on...

引用

Automation and Computation (AUTOCOM), International Conference on

作者： Mithilesh Mahajan Saanvi Bhat Madhav Agrawal Gitanjali Shinde Rutuja Rajendra Patil Gagandeep Kaur Dept. of Computer Science Engineering (Artificial Intelligence and Machine Learning) Vishwakarma Institute of Information Technology Pune India Symbiosis Institute of Technology Nagpur Campus Symbiosis International (Deemed University) Pune India

ISBN: (数字)9798331542375

ISBN: (纸本)9798331542382

This paper proposes a new encoder-decoder frame-work based on Convolutional Neural Networks (CNN) for feature extraction and Long Short-Term Memory (LSTM) networks for image caption generation. Image captioning, the process of generating descriptive textual content for images, has diverse applications, including accessibility tools for the visually impaired, virtual assistants, social media automation, and satellite imagery analysis in critical domains like defense and disaster management. The proposed model will leverage the power of CNNs for extracting high-level image features and LSTMs for generating semantically coherent captions. Rigorously evaluated on the Flickr8k dataset, the model achieved competitive BLEU scores, showing its efficacy in generating accurate and meaningful descriptions. The methodology integrates advanced preprocessing techniques and efficient training strategies, ensuring robustness and scalability. This re-search focuses on the need to use state-of-the-art machine learning techniques in the solving of real challenges, and some of the potential routes to improving image captioning systems are suggested to include multilingual capabilities and integration with larger datasets. Contributions include a clear explanation of the encoder-decoder approach, detailed metrics for evaluation, and insights into correcting limitations including dataset size and complex scene representation. This work is a step toward further progress in automated image-to-text systems, integrating precision, relevance, and adaptability.

关键词： Measurement Training Adaptation models Automation Accuracy Social networking (online) Virtual assistants Feature extraction Convolutional neural networks Long short term memory

来源：评论

学校读者我要写书评

暂无评论

Exploration of Bonafide and Spoofed Audio Classification Using Machine Learning Models

Exploration of Bonafide and Spoofed Audio Classification Usi...

引用

IEEE India Conference (INDICON)

作者： Naseem Babu Prattipati Kumar Jimson Mathew Udit Satija Dept. of Computer Science and Engineering Indian Institute of Technology Patna Bihar India Dept. of Electrical Engineering Indian Institute of Technology Patna Bihar India Dept. of Computer Science and Engineering National Institute of Technology Patna Bihar India

ISBN: (纸本)9781665452724

Recently, it has been shown that most automatic speaker verification systems (ASV) are vulnerable to various spoofing attacks. These spoofing attacks can be divided into three major categories: those deriving from text to speech (TTS), voice conversion (VC), and replay spoofing attacks. In the last couple of years, many researchers have proposed many algorithms to detect these spoofing attacks in ASV systems and have made considerable improvements. This paper presents a novel bonafide and spoofed audio detection approach using machine learning and deep learning classifier models with resampling and data augmentation techniques. This paper uses seven machine learning classifier models with different entropy features and three deep learning classifier models with various acoustic features. The performance of these classifier models is analyzed using metrics like precision, recall, accuracy, equal error rate, loss, and confusion metrics. State-of-the-art authors used equal error rate as their performance measure metric; our primary focus is on equal error rate in this paper. And we observed that in our proposed methodology, the deep learning classifier models showed considerable performance improvement. The training phase shows approximate 100% accuracy with a loss of 0.0090 and an equal error rate of 0.06%. In the evaluation phase, it offers approximate 98% accuracy with a loss of 0.0906 and an equal error rate of 1.72%.

关键词： Deep learning Training Machine learning algorithms Error analysis Measurement uncertainty Signal processing algorithms Transforms

来源：评论

学校读者我要写书评

暂无评论

Exploiting Change Blindness for Video Coding: Perspectives from a Less Promising User Study

Exploiting Change Blindness for Video Coding: Perspectives f...

引用

International Workshop on Quality of Multimedia Experience, QoMEx

作者： Mitra Amiri Steven Le Moan Christian Herglotz Dept. of Computer Science ColorLab Norwegian University of Science and Technology (NTNU) Gjøvik Norway Chair of Computer Engineering Brandenburgisch-Technische Universität Cottbus-Senftenberg Cottbus Germany

ISBN: (数字)9798350361582

ISBN: (纸本)9798350361599

What the human visual system can perceive is strongly limited by the capacity of our working memory and attention. Such limitations result in the human observer’s inability to perceive large-scale changes in a stimulus, a phenomenon known as change blindness. In this paper, we started with the premise that this phenomenon can be exploited in video coding, especially HDR-video compression where the bitrate is high. We designed an HDR-video encoding approach that relies on spatially and temporally varying quantization parameters within the framework of HEVC video encoding. In the absence of a reliable change blindness prediction model, to extract compression candidate regions (CCR) we used an existing saliency prediction algorithm. We explored different configurations and carried out a subjective study to test our hypothesis. While our methodology did not lead to significantly superior performance in terms of the ratio between perceived quality and bitrate, we were able to determine potential flaws in our methodology, such as the employed saliency model for CCR prediction (chosen for computational efficiency, but eventually not sufficiently accurate), as well as a very strong subjective bias due to observers priming themselves early on in the experiment about the type of artifacts they should look for, thus creating a scenario with little ecological validity.

关键词： Accuracy Biological system modeling Semantics Bit rate Blindness Predictive models Observers

来源：评论

学校读者我要写书评

暂无评论

An Empirical Investigation on Electronic Government Services Continuance and Trust

An Empirical Investigation on Electronic Government Services...

引用

International Symposium on technology and Society (ISTAS)

作者： Muhammad Ovais Ahmad Iftikhar Ahmad Iqra Sadaf Khan Dept. of Computer Science Karlstad University Karlstad Sweden Department of Computer Science and Information Technology University of Engineering and Technology Peshawar Pakistan Dept. of Management Studies Aalto University Espoo Finland

Trust is an important factor that contributes to citizens' willingness to continuance use of e-gov services. However, there is a lack of prior investigation about trust and continuance use of e-gov services in Pakistan - a developing country. We propose a model to investigate citizens' trust and e-gov services' continuous use intention to fill this research gap. Our study collected data from an online survey of 558 Pakistani citizens. Using partial least squares analysis, we found that disposition to trust positively correlates with both internet and government trust. Moreover, citizen satisfaction, trust, perceived usefulness, confirmation, and perceived risk all have significant impacts on the continuous use intention of e-gov services. This research extends and validates the Expectation-Confirmation Model by exploring key factors that influence e-gov continuance use intention. As such, our study offers valuable insights for policymakers and practitioners involved in e-gov service delivery in developing countries like Pakistan. The paper also discusses our findings' implications and identifies future research directions.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Developing and Implementing An Effective Big Data Classification System

Developing and Implementing An Effective Big Data Classifica...

引用

Integrated Circuits and Communication Systems (ICICACS), IEEE International Conference on

作者： Manjunath Managuli Sagar Pujar Sangamesh C Managuli Krishnamurthy K. T Rajesh Nandalike Pavan Kunchur Dept. of Electronics & Communication KLS Gogte Institute of Technology Belagavi India Dept. of Computer Science KLS Gogte Institute of Technology Belagavi India Dept. of Mechanical Engineering PES University Bangalore India Dept. of Electronics & Instrumentation B M S College of Engineering Bangalore India Dept. of Electronics & Communication Nitte Meenakshi Institute of Technology Bangalore India

Relevant data technology advancements have led to a massive growth in classifications. Deep learning systems, however, need help handling such a massive volume of data. This work aims to use the Evaluation of different classification algorithms for big data, Comparison of accuracy, precision, recall, and F1 score among various algorithms, and Identification of algorithms suitable for handling large-scale datasets and conventional deep learning to analyze a large amount of data. Classifiers allowed researchers to examine how various random sampling strategies affected the model's performance. Two methods of feature selection were employed in the tests. The Bayesian Network likewise employs a single deep learning classifier. By lowering the dimensionality of the data, this study seeks to maximize model performance. A recording dataset larger than 60 GB was made less dimensional to conduct the trials. The findings demonstrate that, when employing both sampled and non-sample data, overall performance varies between 78% accuracy and 82% accuracy, a minor variation in performance while utilizing both sampled and non-sample (all the data) datasets. The random subset approach and the Bayesian network classifier were employed in the trials, and the overall findings show that the variety experiment had the best performance out of all the combinations of experiments. Test Cradle Experimentation: Variation length, delay variation, and packet loss are used to assess the predicted scheme's efficiency.

关键词：

来源：评论

学校读者我要写书评

暂无评论

AI-Driven Forecasting Mechanism for Cardiovascular Diseases: A Hybrid Approach using MLP and K-NN Models 2

AI-Driven Forecasting Mechanism for Cardiovascular Diseases:...

引用

2nd International Conference on Self Sustainable Artificial Intelligence Systems, ICSSAS 2024

作者： Chaganti, Koushik Reddy Krishnamurty, Pokuri Venkataradha Kumar, Addala Hemantha Gowd, G. Sudha Balakrishna, Chinnala Naresh, P. University of the Cumberlands Dept. of Computer and Information Systems KY United States Audisankara College of Engineering and Technology Cse Department Andhra Pradesh Gudur India Srinivasa Ramanujan Institute Of Technology Cse Department Anantapuram India Cse Department Hyderabad India Dayananda Sagar University Dept of Cse Bangalore India

ISBN: (纸本)9798350368413

Cardiovascular disease (CVD) results in malfunctioning of the heart and also the related blood vessels and regularly primes to death or corporal paralysis. Initial and automated discovery of CVD is crucial for saving many of the common humanoid lives. While several Investigations have been carried out to achieve this goal, and there is still room for development in terms of performance and reliability. This new work provides one more step in this area, since it utilizes two reliable ML algorithms, 'multi-layer perceptron (MLP)' and 'K-nearest neighbour (K-NN)'to detect CVD using publicly available data from the University of engineering California Irvine source. The models' concerts are enhanced optimally and by eradicating the possible outliers and common existing attributes with values that are treated as null. They performed general Experimental and effective findings show that the MLP model outperforms the KNN model in terms of reporting accuracy (82.47%) and area below the curve (86.41%). As a result, the proposed MLP model is recommended for automated CVD identification. Furthermore, this approach may be used to identify new illnesses, and the suggested model's routine can be tested against other standard datasets. © 2024 IEEE.

关键词： Nearest neighbor search

来源：评论

学校读者我要写书评

暂无评论

Dark Patterns Detection on E-Commerce Websites

Dark Patterns Detection on E-Commerce Websites

引用

Automation and Computation (AUTOCOM), International Conference on

作者： Aarya Sawant Amrisha Gamane Shraddha Sonawane Omkaresh Kulkarni Rutuja Rajendra Patil Gagandeep Kaur Dept. of Computer Science Engineering (Artificial Intelligence and Machine Learning) Vishwakarma Institute of Information Technology Pune India Symbiosis Institute of Technology Nagpur Campus Symbiosis International (Deemed University) Pune India

ISBN: (数字)9798331542375

ISBN: (纸本)9798331542382

Dark patterns are techniques employed on e-commerce website interfaces for the purpose of influencing users' actions to engage in undesirable behavior such as making unintentional purchases or submitting personal information when they did not want to. This is where this study seeks to address through identification of such patterns using a machine learning method. To cover false and genuine content, a dataset with 10,000 textual samples, obtained from several e-commerce sites, was compiled. Thus, for analysis, the easily interpretable classification model, Logistic Regression, was used. The results showed that the detection accuracy of the model is 92%, where precision and Recall metrics' point to the model's abilities to detect manipulative language. The results point to the ability of machine learning methods in detecting and combating fraud in cyberspace and thus promoting an ethical e-commerce market.

关键词： Measurement Logistic regression Accuracy Computational modeling Mathematical models Fraud Electronic commerce Problem-solving Protection Random forests

来源：评论

学校读者我要写书评

暂无评论

Adjustment of Interval of Voices in Cognitive Shuffle Method by Body Movement Feedback with Binaural Sound

Adjustment of Interval of Voices in Cognitive Shuffle Method...

引用

Summer Virtual Conference on Software engineering, Artificial Intelligence, Networking and Parallel/Distributed Computing (SNPD-Summer), ACIS International

作者： Makoto Fukumoto Hirohisa Konishi Dept. of Computer Science and Engineering Fukuoka Institute of Technology Fukuoka Japan

ISBN: (纸本)9798350396386

Insomnia remains a severe problem in social and personal. Sound and music have been used to solve the problem. Cognitive shuffle is one of the insomnia treatments by using voices for shuffling the user’s thoughts. The purpose of this study is to propose a method for adjusting the interval of voices used in the cognitive shuffle method. Based on the hypothetic arousal level of the user estimated by body movement, the interval is determined. When the arousal level was lower, the intervals were set longer. A feedback system was constructed to investigate the proposed method’s fundamental efficiency on sleep. A sheet-type pressure sensor was used to measure the user’s body movement. Moreover, the binaural sound was used to create voices in the cognitive shuffle method. The comparing experiment was conducted with a sound condition with the proposed method and control with no sound. As experimental results, a marginal difference was observed between the two conditions in the number of body movements. No significant difference in subjective sleepiness after the experiment was observed.

关键词： Pressure sensors Sleep Music Pressure measurement Artificial intelligence Software engineering

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：