检索结果-内蒙古大学图书馆

Efficient video summarization through MobileNetSSD: a robust deep learning-based framework for efficient video summarization focused on objects of interest

引用

Multimedia Tools and Applications 2024年 1-26页

作者： Yarrarapu, Manasa Leelavathy, Narkedamilly Haritha, Dasari Department of Computer Science and Engineering Jawaharlal Nehru Technological University Andhra Pradesh Kakinada533003 India Department of Computer Science and Engineering Godavari Institute of Engineering and Technology Andhra Pradesh Rajahmundry533296 India Department of Computer Science and Engineering University College of Engineering Jawaharlal Nehru Technological University Andhra Pradesh Kakinada533003 India

Now-a-days, the generation of videos has increased dramatically due to the quick growth of multimedia and the internet. The need for effective ways to store, manage, and index the massive numbers of videos has become imperative due to this expansion. As a result, a method needs to be proposed that collects only the necessary data from the original recording. In computer vision, Video summarization is a significant task, and its primary goal is to give a quick summary of the video by removing irrelevant information and capturing key frames from the video. Many approaches have developed over the last several decades, using the most recent deep neural network architectures that represent the current state-of-the-art. Our method involves extracting vital key frames from the input video using the MobileNetSSD model, which is well-known for its efficient recognition and localization of objects of interest. These highlighted frames are essential in creating a detailed video summary. Furthermore, a method of temporal analysis is applied to guarantee that the summary accurately reflects the relevant events in the order in which they occurred, contributing to a coherent and meaningful representation of the information. We evaluated the proposed approach on TV Sum and SUM me video datasets, comparing the results against cutting-edge video summarization techniques. Our approach works effectively to produce clear and meaningful video summaries. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

AeroDetect: Real-Time Small Object Detection in Aerial Images 2nd

AeroDetect: Real-Time Small Object Detection in Aerial Imag...

引用

2nd International Conference on Advances in Data-driven Computing and Intelligent Systems, ADCIS 2023

作者： Singhal, Nikita Pandey, Anuranjan Kumar Kumar, Ankit Department of Computer Engineering Army Institute of Technology Pune India

ISBN: (纸本)9789819995233

Real-time small object detection in aerial images is a challenging task in computer vision because of the complex environmental conditions and fewer pixels occupied by the objects. Along with this getting real-time inference makes this much more difficult. To tackle these challenges we propose an improvement in the classical YOLOv3 network which is capable of removing noise from the images and performing small object detection to while maintaining a high frame frequency. We are using a drone to capture real-time footage from the AIT campus, which is further used for classifying objects like vehicles, pedestrians, etc. Our proposed model "AeroDetect" was able to beat various state-of-the-art models like YOLOv3 and YOLOv5 in various parameters by having higher mAP, and a lesser number of layers and parameters while maintaining a high frame rate which is crucial for our use case. We achieved mAP of 0.402 on the Visdrone dataset as compared to 0.4 and 0.38 by YOLOv3 and YOLOv5, respectively. On the VEDAI dataset, we got mAP of 0.632, whereas YOLOv3 and YOLOv5 achieved 0.578 and 0.604, respectively. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Machine Learning Models for Stock Price Prediction 2

A Comparative Study of Machine Learning Models for Stock Pri...

引用

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

作者： Ashtagi, Rashmi Deshmukh, Amogh Dhomse, Siddhesh Dhumal, Gayatri Dhurve, Siddhartha Vishwakarma Institute of Technology Department of Computer Engineering Pune India

ISBN: (纸本)9798350368413

The stock market has always been evolutionary. It has been an area of interest for researchers to forecasting the stock prices too guide investment decisions and avoiding financial risk since many years. This paper highlights the significance of precise stock price prediction and investigates the use of several machine learning techniques to strengthen the predictive accuracy. The proposed system makes uses two different approaches: Regression, which predicts the company's stock closing price, and Classification predicts whether the closing price of the stocks will increase or decrease the next day. Historical stock data is examined carefully and preprocessed to ensure high quality of data. Using feature engineering appropriate features are extracted to identify the intrinsic patterns and trends within the stock market. Algorithms like Linear Regression, ARIMA, LSTM and Random Forest are implemented and thoroughly evaluated for their efficiency in stock price prediction. After conducting a comparative analysis of the two algorithms, it becomes evident that while LSTM performs exceptionally in regression tasks, identifying the temporal associations and patterns that guide the stock market data in its development, Random Forest provides more firm predictions of the stock price movement. This study provides the extensive analysis of the real-life application of machine learning for discussions on the stock market, thus allowing investors to derive a more reliable resource to refer to when dealing with the challenges of the stock market. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Learning Non-cooperative Driving Practices and its Impact on Road Traffic Dynamics 17

Learning Non-cooperative Driving Practices and its Impact on...

引用

17th International Conference on COMmunication Systems and NETworkS, COMSNETS 2025

作者： Osho Chakraborty, Suchetana Indian Institute of Technology Department of Computer Science and Engineering Jodhpur India

ISBN: (纸本)9798331531195

Cooperative driving that prioritizes global safety and efficiency over self-interest is crucial in ensuring smooth traffic flow and enhanced road safety. However, experiencing noncooperative driving is common under heterogeneous and chaotic on-road traffic conditions. While non-cooperativeness is often context-dependent and assessed during direct vehicle interactions, it can be inferred through longitudinal analysis of driving patterns from vehicular kinematics data. This behavior not only adversely affects the vehicle itself but also has a cascading impact on nearby vehicles, ultimately disrupting overall traffic flow. In this study, we quantify a vehicle's non-cooperativeness using kinematic data and further model its cascading effects on nearby cooperative vehicles. Experimental analysis was conducted using the SUMO traffic simulator and the TraCI interface to assess the impact of non-cooperativeness on overall traffic dynamics. This research is among the first to examine the cascading effects of a vehicle's non-cooperativeness on its cooperative nearby vehicles. © 2025 IEEE.

关键词： Motor transportation

来源：评论

学校读者我要写书评

暂无评论

Gesture Recognition of Indian Sign Language Using Hybrid CNN-SVM Classifier

Gesture Recognition of Indian Sign Language Using Hybrid CNN...

引用

International Conference on Systems, Control and Automation, ICSCA 2023

作者： Singh, Utkrisht Singh, Awadhesh Kumar Department of Computer Engineering National Institute of Technology Haryana Kurukshetra India

ISBN: (纸本)9789819773831

Sign language is a type of nonverbal communication method used by people with hearing impairments and the deaf to convey their sentiments and emotions. According to the 2011 Census of India data, there are over 50 lakh people in India who are deaf, and nearly 20 lakh people have speech impairments. In this paper, we have presented a technique to recognise the hand movements of Indian Sign Language efficiently by using a hybrid model of a strong convolutional neural network (CNN) and support vector machine (SVM). In this work, an Indian Sign Language (ISL) dataset is used to train the model. This dataset contains various two handed Indian Sign Language images. In the proposed hybrid model, CNN is used to serve as an automated feature extractor that extracts the most informative features, whereas SVM serves as a classifier. The experimental findings show the correctness of the suggested framework by reaching an identification accuracy of 99.78% for this model on the ISL dataset, and after including Gaussian and random noise to the images, the CNN-SVM model’s accuracy drops to 92.72%. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Generating Time-Series Data Using Generative Adversarial Networks for Mobility Demand Prediction

引用

computers, Materials & Continua 2023年第3期74卷 5507-5525页

作者： Subhajit Chatterjee Yung-Cheol Byun Department of Computer Engineering Jeju National UniversityJeju-si63243Korea Department of Computer Engineering Major of Electronic EngineeringJeju National UniversityInstitute of Information Science&TechnologyJeju63243Korea

The increasing penetration rate of electric kickboard vehicles has been popularized and promoted primarily because of its clean and efficient *** kickboards are gradually growing in popularity in tourist and education-centric *** the upcoming arrival of electric kickboard vehicles,deploying a customer rental service is *** to its freefloating nature,the shared electric kickboard is a common and practical means of *** plans for shared electric kickboards are required to increase the quality of service,and forecasting demand for their use in a specific region is *** demand accurately with small data is *** data is necessary for training machine learning algorithms for effective *** generation is a method for expanding the amount of data that will be further accessible for *** this work,we proposed a model that takes time-series customers’electric kickboard demand data as input,pre-processes it,and generates synthetic data according to the original data distribution using generative adversarial networks(GAN).The electric kickboard mobility demand prediction error was reduced when we combined synthetic data with the original *** proposed Tabular-GAN-Modified-WGAN-GP for generating synthetic data for better prediction *** modified The Wasserstein GAN-gradient penalty(GP)with the RMSprop optimizer and then employed Spectral Normalization(SN)to improve training stability and faster ***,we applied a regression-based blending ensemble technique that can help us to improve performance of demand *** used various evaluation criteria and visual representations to compare our proposed model’s *** data generated by our suggested GAN model is also *** TGAN-Modified-WGAN-GP model mitigates the overfitting and mode collapse problem,and it also converges faster than previous GAN models for synthetic data creation.

关键词： Machine learning generative adversarial networks electric vehicle time-series TGAN WGAN-GP blend model demand prediction regression

来源：评论

学校读者我要写书评

暂无评论

Harnessing Natural Entity Recognition for Intuitive Spreadsheet Control: A Voice-Driven Google Sheets Extension 5

Harnessing Natural Entity Recognition for Intuitive Spreadsh...

引用

5th IEEE International Conference for Emerging technology, INCET 2024

作者： Joglekar, Pushkar Jha, Divyanshu Dhorage, Prathamesh Thakkar, Dhruv Dhumal, Om Department of Computer Engineering Vishwakarma Institute of Technology Pune India

ISBN: (纸本)9798350361155

This paper introduces an innovative Voice-Controlled Google Sheets Extension employing Natural Language Processing (NLP) capabilities. The system seamlessly integrates a Chrome extension and a meticulously trained Named Entity Recognition (NER) model. Trained on custom datasets, the NER model proficiently extracts crucial entities, such as command types, cell references, and values from user speech, enabling accurate interpretation of complex commands. The extension enhances user interaction with spreadsheets, offering a voice-driven interface that simplifies spreadsheet management. The backend, utilizing Flask-based REST APIs, facilitates seamless communication with the Google Sheets API, ensuring precise execution of actions dictated by the interpreted commands. Experimental results underscore the system's accuracy and efficiency, establishing its potential as an intelligent productivity tool. This research contributes to the evolution of smart computing interfaces, providing users with an intuitive and accessible means of interacting with spreadsheet applications. © 2024 IEEE.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Real-time automated fabric defect detection system 5th

Real-time automated fabric defect detection system

引用

5th International Conference on Advances in AI for Biomedical Instrumentation, Electronics and Computing, ICABEC 2023

作者： Mandawkar, Umakant Shahade, Makarand Wadekar, Samruddhi Kachhava, Chetan Patil, Yash Mandwekar, Sakshi Department of Computer Engineering SVKM Institute of Technology Dhule India

ISBN: (纸本)9781032642987

Even today the Textile Industries are using manpower to detect and classify the fabric defect, the process of depending on human inspectors for fabric defect detection can be considered as not an optimized way since it is time-consuming as well as an error prone process. To tackle this issue, here we propose a groundbreaking development- A real-time, automated fabric defect detection system. Making the most use of all the available cutting-edge advanced image processing techniques and machine learning algorithms, the proposed system will identify, locate and categorize a wide range of fabric defect where precision and accuracy will be the hallmarks of the proposed methodology The system will operate in real time, capturing highresolution fabric surface images, leading to precise and swift defect identification. Its automated and real-time capabilities render it to be an essential instrument in textile manufacturing as it will enhance its quality by diminishing the requirement of human inspectors, and evolving overall inspection process efficiency. Comprehensive and thorough testing across various fabric materials will prove its high accuracy rates in defect detection. It will possess the power to revolutionize the textile industry by enhancing product quality, cutting costs and boosting productivity. © 2024 The Author(s).

关键词： Textile industry

来源：评论

学校读者我要写书评

暂无评论

Feature Extraction From Single-Channel EEG Using Tsfresh and Stacked Ensemble Approach for Sleep Stage Classification

引用

International Journal of e-Collaboration 2023年第3期19卷 1-20页

作者： Radhakrishnan, B.L. Ezra, Kirubakaran Jebadurai, Immanuel Johnraja Department of Computer Science and Engineering Karunya Institute of Technology and Sciences Coimbatore India Department of Computer Science and Engineering GRACE College of Engineering Thoothukudi India

The smart world under Industry 4.0 is witnessing a notable spurt in sleep disorders and sleep-related issues in patients. Artificial intelligence and IoT are taking a giant leap in connecting sleep patients remotely with healthcare providers. The contemporary single-channel-based monitoring devices play a tremendous role in predicting sleep quality and related issues. Handcrafted feature extraction is a time-consuming job in machine learning-based automatic sleep classification. The proposed single-channel work uses Tsfresh to extract features from both the EEG channels (Pz-oz and Fpz-Cz) of the SEDFEx database individually to realise a single-channel EEG. The adopted mRMR feature selection approach selected 55 features from the extracted 787 features. A stacking ensemble classifier achieved 95%, 94%, 91%, and 88% accuracy using stratified 5-fold validation in 2, 3, 4, and 5 class classification employing healthy subjects data. The outcome of the experiments indicates that Tsfresh is an excellent tool to extract standard features from EEG signals. © 2023 IGI Global. All rights reserved.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

Multiple Disease Prediction using Machine Learning and Doctor Appointment Scheduling System

Multiple Disease Prediction using Machine Learning and Docto...

引用

2024 OPJU International technology Conference on Smart Computing for Innovation and Advancement in Industry 4.0, OTCON 2024

作者： Magar, Anand Desai, Rutuja Deshmukh, Siddhi Deshpande, Samarth Dhamne, Sakshi Vishwakarma Institute of Technology Department of Computer Engineering Pune India

ISBN: (纸本)9798350373783

This paper introduces a pioneering approach aimed at elevating healthcare services through the integration of machine learning (ML) techniques for predicting Heart (Cardiovascular) disease, Parkinson's disease, Diabetes, and Breast cancer, alongside an intelligent doctor appointment scheduling system. In this paper, the prediction of all diseases is conducted using machine learning algorithms including RF (Random Forest), LR (Logistic Regression), KNN (K Nearest Neighbor), and Naive Bayes. Data preprocessing and oversampling are carried out. Confusion matrix is utilized to calculate performance parameters. The performance parameters of all four algorithms on training and testing data were evaluated and compared. Random Forest is the best-performing ML algorithm, which yields the highest values, greater than 95%, for all parameters compared to the other three algorithms. These results are saved as SAV files and utilized in the web interface formation of disease prediction. These results indicate that machine learning algorithms are effective in predicting multiple diseases. © 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

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

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

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：