scholarly journals Deep-Framework: A Distributed, Scalable, and Edge-Oriented Framework for Real-Time Analysis of Video Streams

Sensors ◽  
2021 ◽  
Vol 21 (12) ◽  
pp. 4045
Author(s):  
Alessandro Sassu ◽  
Jose Francisco Saenz-Cogollo ◽  
Maurizio Agelli
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Video Data ◽  
Video Analytics ◽  
Web Based ◽  
Real Time Analysis ◽  
Open Source Framework ◽  
Cluster Configuration ◽  
Time Requirements ◽  
High Level

Edge computing is the best approach for meeting the exponential demand and the real-time requirements of many video analytics applications. Since most of the recent advances regarding the extraction of information from images and video rely on computation heavy deep learning algorithms, there is a growing need for solutions that allow the deployment and use of new models on scalable and flexible edge architectures. In this work, we present Deep-Framework, a novel open source framework for developing edge-oriented real-time video analytics applications based on deep learning. Deep-Framework has a scalable multi-stream architecture based on Docker and abstracts away from the user the complexity of cluster configuration, orchestration of services, and GPU resources allocation. It provides Python interfaces for integrating deep learning models developed with the most popular frameworks and also provides high-level APIs based on standard HTTP and WebRTC interfaces for consuming the extracted video data on clients running on browsers or any other web-based platform.

scholarly journals On the Performance of One-Stage and Two-Stage Object Detectors in Autonomous Vehicles Using Camera Data

2020 ◽  
Vol 13 (1) ◽  
pp. 89
Author(s):  
Manuel Carranza-García ◽  
Jesús Torres-Mateo ◽  
Pedro Lara-Benítez ◽  
Jorge García-Gutiérrez
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Autonomous Vehicles ◽  
Remote Sensing Data ◽  
Autonomous Driving ◽  
Two Stage ◽  
Detection Systems ◽  
One Stage ◽  
Time Requirements ◽  
Speed Accuracy

Object detection using remote sensing data is a key task of the perception systems of self-driving vehicles. While many generic deep learning architectures have been proposed for this problem, there is little guidance on their suitability when using them in a particular scenario such as autonomous driving. In this work, we aim to assess the performance of existing 2D detection systems on a multi-class problem (vehicles, pedestrians, and cyclists) with images obtained from the on-board camera sensors of a car. We evaluate several one-stage (RetinaNet, FCOS, and YOLOv3) and two-stage (Faster R-CNN) deep learning meta-architectures under different image resolutions and feature extractors (ResNet, ResNeXt, Res2Net, DarkNet, and MobileNet). These models are trained using transfer learning and compared in terms of both precision and efficiency, with special attention to the real-time requirements of this context. For the experimental study, we use the Waymo Open Dataset, which is the largest existing benchmark. Despite the rising popularity of one-stage detectors, our findings show that two-stage detectors still provide the most robust performance. Faster R-CNN models outperform one-stage detectors in accuracy, being also more reliable in the detection of minority classes. Faster R-CNN Res2Net-101 achieves the best speed/accuracy tradeoff but needs lower resolution images to reach real-time speed. Furthermore, the anchor-free FCOS detector is a slightly faster alternative to RetinaNet, with similar precision and lower memory usage.

scholarly journals Deep Learning-Based Cattle Vocal Classification Model and Real-Time Livestock Monitoring System with Noise Filtering

Animals ◽  
2021 ◽  
Vol 11 (2) ◽  
pp. 357
Author(s):  
Dae-Hyun Jung ◽  
Na Yeon Kim ◽  
Sang Ho Moon ◽  
Changho Jhin ◽  
Hak-Jin Kim ◽  
...  
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Model Development ◽  
Classification Model ◽  
Intelligence Analysis ◽  
Noise Filtering ◽  
Meat Industry ◽  
Mel Frequency Cepstral Coefficients ◽  
Web Based ◽  
The Status

The priority placed on animal welfare in the meat industry is increasing the importance of understanding livestock behavior. In this study, we developed a web-based monitoring and recording system based on artificial intelligence analysis for the classification of cattle sounds. The deep learning classification model of the system is a convolutional neural network (CNN) model that takes voice information converted to Mel-frequency cepstral coefficients (MFCCs) as input. The CNN model first achieved an accuracy of 91.38% in recognizing cattle sounds. Further, short-time Fourier transform-based noise filtering was applied to remove background noise, improving the classification model accuracy to 94.18%. Categorized cattle voices were then classified into four classes, and a total of 897 classification records were acquired for the classification model development. A final accuracy of 81.96% was obtained for the model. Our proposed web-based platform that provides information obtained from a total of 12 sound sensors provides cattle vocalization monitoring in real time, enabling farm owners to determine the status of their cattle.

scholarly journals SIAT: A Distributed Video Analytics Framework for Intelligent Video Surveillance

Symmetry ◽  
2019 ◽  
Vol 11 (7) ◽  
pp. 911 ◽  
Author(s):  
Md Azher Uddin ◽  
Aftab Alam ◽  
Nguyen Anh Tu ◽  
Md Siyamul Islam ◽  
Young-Koo Lee
Keyword(s):  
Distributed Computing ◽  
Real Time ◽  
Video Surveillance ◽  
Video Processing ◽  
Large Scale ◽  
Distributed Processing ◽  
Content Management ◽  
Video Data ◽  
Video Analytics ◽  
Intelligent Video Surveillance

In recent years, the amount of intelligent CCTV cameras installed in public places for surveillance has increased enormously and as a result, a large amount of video data is produced every moment. Due to this situation, there is an increasing request for the distributed processing of large-scale video data. In an intelligent video analytics platform, a submitted unstructured video undergoes through several multidisciplinary algorithms with the aim of extracting insights and making them searchable and understandable for both human and machine. Video analytics have applications ranging from surveillance to video content management. In this context, various industrial and scholarly solutions exist. However, most of the existing solutions rely on a traditional client/server framework to perform face and object recognition while lacking the support for more complex application scenarios. Furthermore, these frameworks are rarely handled in a scalable manner using distributed computing. Besides, existing works do not provide any support for low-level distributed video processing APIs (Application Programming Interfaces). They also failed to address a complete service-oriented ecosystem to meet the growing demands of consumers, researchers and developers. In order to overcome these issues, in this paper, we propose a distributed video analytics framework for intelligent video surveillance known as SIAT. The proposed framework is able to process both the real-time video streams and batch video analytics. Each real-time stream also corresponds to batch processing data. Hence, this work correlates with the symmetry concept. Furthermore, we introduce a distributed video processing library on top of Spark. SIAT exploits state-of-the-art distributed computing technologies with the aim to ensure scalability, effectiveness and fault-tolerance. Lastly, we implant and evaluate our proposed framework with the goal to authenticate our claims.

scholarly journals An LSTM-Based Deep Learning Approach for Classifying Malicious Traffic at the Packet Level

2019 ◽  
Vol 9 (16) ◽  
pp. 3414 ◽  
Author(s):  
Ren-Hung Hwang ◽  
Min-Chun Peng ◽  
Van-Linh Nguyen ◽  
Yu-Lun Chang
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Short Term Memory ◽  
Learning Technologies ◽  
Word Embedding ◽  
Detection Systems ◽  
Real Time Analysis ◽  
Detection Delay ◽  
Novel Approach ◽  
Prior Literature

Recently, deep learning has been successfully applied to network security assessments and intrusion detection systems (IDSs) with various breakthroughs such as using Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) to classify malicious traffic. However, these state-of-the-art systems also face tremendous challenges to satisfy real-time analysis requirements due to the major delay of the flow-based data preprocessing, i.e., requiring time for accumulating the packets into particular flows and then extracting features. If detecting malicious traffic can be done at the packet level, detecting time will be significantly reduced, which makes the online real-time malicious traffic detection based on deep learning technologies become very promising. With the goal of accelerating the whole detection process by considering a packet level classification, which has not been studied in the literature, in this research, we propose a novel approach in building the malicious classification system with the primary support of word embedding and the LSTM model. Specifically, we propose a novel word embedding mechanism to extract packet semantic meanings and adopt LSTM to learn the temporal relation among fields in the packet header and for further classifying whether an incoming packet is normal or a part of malicious traffic. The evaluation results on ISCX2012, USTC-TFC2016, IoT dataset from Robert Gordon University and IoT dataset collected on our Mirai Botnet show that our approach is competitive to the prior literature which detects malicious traffic at the flow level. While the network traffic is booming year by year, our first attempt can inspire the research community to exploit the advantages of deep learning to build effective IDSs without suffering significant detection delay.

scholarly journals GPU Accelerated real-time Melanoma Detection

2018 ◽  
Vol 7 (3) ◽  
pp. 1208
Author(s):  
Ajai Sunny Joseph ◽  
Elizabeth Isaac
Keyword(s):  
Real Time ◽  
Video Processing ◽  
Processing Technique ◽  
Video Data ◽  
Image Processing Technique ◽  
Processing Unit ◽  
Central Processing ◽  
Melanoma Detection ◽  
Real Time Analysis ◽  
Graphical Processing

Melanoma is recognized as one of the most dangerous type of skin cancer. A novel method to detect melanoma in real time with the help of Graphical Processing Unit (GPU) is proposed. Existing systems can process medical images and perform a diagnosis based on Image Processing technique and Artificial Intelligence. They are also able to perform video processing with the help of large hardware resources at the backend. This incurs significantly higher costs and space and are complex by both software and hardware. Graphical Processing Units have high processing capabilities compared to a Central Processing Unit of a system. Various approaches were used for implementing real time detection of Melanoma. The results and analysis based on various approaches and the best approach based on our study is discussed in this work. A performance analysis for the approaches on the basis of CPU and GPU environment is also discussed. The proposed system will perform real-time analysis of live medical video data and performs diagnosis. The system when implemented yielded an accuracy of 90.133% which is comparable to existing systems.  

scholarly journals Development of processes allowing near real-time refinement and validation of triage tools during the early stage of an outbreak in readiness for surge: the FLU-CATs Study

2015 ◽  
Vol 19 (89) ◽  
pp. 1-132 ◽  
Author(s):  
Sudhir Venkatesan ◽  
Puja R Myles ◽  
Gerard McCann ◽  
Antonis A Kousoulis ◽  
Maimoona Hashmi ◽  
...  
Keyword(s):  
Primary Care ◽  
Real Time ◽  
Patient Outcomes ◽  
Proof Of Concept ◽  
Time Analysis ◽  
Web Based ◽  
Clinical Assessments ◽  
Levels Of Care ◽  
Real Time Analysis ◽  
Future Outbreak

BackgroundDuring pandemics of novel influenza and outbreaks of emerging infections, surge in health-care demand can exceed capacity to provide normal standards of care. In such exceptional circumstances, triage tools may aid decisions in identifying people who are most likely to benefit from higher levels of care. Rapid research during the early phase of an outbreak should allow refinement and validation of triage tools so that in the event of surge a valid tool is available. The overarching study aim is to conduct a prospective near real-time analysis of structured clinical assessments of influenza-like illness (ILI) using primary care electronic health records (EHRs) during a pandemic. This abstract summarises the preparatory work, infrastructure development, user testing and proof-of-concept study.Objectives(1) In preparation for conducting rapid research in the early phase of a future outbreak, to develop processes that allow near real-time analysis of general practitioner (GP) assessments of people presenting with ILI, management decisions and patient outcomes. (2) As proof of concept: conduct a pilot study evaluating the performance of the triage tools ‘Community Assessment Tools’ and ‘Pandemic Medical Early Warning Score’ to predict hospital admission and death in patients presenting with ILI to GPs during inter-pandemic winter seasons.DesignProspective near real-time analysis of structured clinical assessments and anonymised linkage to data from EHRs. User experience was evaluated by semistructured interviews with participating GPs.SettingThirty GPs in England, Wales and Scotland, participating in the Clinical Practice Research Datalink.ParticipantsAll people presenting with ILI.InterventionsNone.Main outcome measuresStudy outcome is proof of concept through demonstration of data capture and near real-time analysis. Primary patient outcomes were hospital admission within 24 hours and death (all causes) within 30 days of GP assessment. Secondary patient outcomes included GP decision to prescribe antibiotics and/or influenza-specific antiviral drugs and/or refer to hospital – if admitted, the need for higher levels of care and length of hospital stay.Data sourcesLinked anonymised data from a web-based structured clinical assessment and primary care EHRs.ResultsIn the 24 months to April 2015, data from 704 adult and 159 child consultations by 30 GPs were captured. GPs referred 11 (1.6%) adults and six (3.8%) children to hospital. There were 13 (1.8%) deaths of adults and two (1.3%) of children. There were too few outcome events to draw any conclusions regarding the performance of the triage tools. GP interviews showed that although there were some difficulties with installation, the web-based data collection tool was quick and easy to use. Some GPs felt that a minimal monetary incentive would promote participation.ConclusionsWe have developed processes that allow capture and near real-time automated analysis of GP’s clinical assessments and management decisions of people presenting with ILI.Future workWe will develop processes to include other EHR systems, attempt linkage to data on influenza surveillance and maintain processes in readiness for a future outbreak.Study registrationThis study is registered as ISRCTN87130712 and UK Clinical Research Network 12827.FundingThe National Institute for Health Research Health Technology Assessment programme. MGS is supported by the UK NIHR Health Protection Research Unit in Emerging and Zoonotic Infections.

Artificial Intelligence-Aided Automated Detection of Railroad Trespassing

2019 ◽  
Vol 2673 (7) ◽  
pp. 25-37 ◽  
Author(s):  
Asim Zaman ◽  
Baozhang Ren ◽  
Xiang Liu
Keyword(s):  
Artificial Intelligence ◽  
Real Time ◽  
Video Clip ◽  
Text Message ◽  
The United States ◽  
Video Data ◽  
Surveillance Video ◽  
Railroad Industry ◽  
Real Time Analysis ◽  
Safety Research

Trespassing is the leading cause of rail-related deaths and has been on the rise for the past 10 years. Detection of unsafe trespassing of railroad tracks is critical for understanding and preventing fatalities. Witnessing these events has become possible with the widespread deployment of large volumes of surveillance video data in the railroad industry. This potential source of information requires immense labor to monitor in real time. To address this challenge this paper describes an artificial intelligence (AI) framework for the automatic detection of trespassing events in real time. This framework was implemented on three railroad video live streams, a grade crossing and two right-of-ways, in the United States. The AI algorithm automatically detects trespassing events, differentiates between the type of violator (car, motorcycle, truck, pedestrian, etc.) and sends an alert text message to a designated destination with important information including a video clip of the trespassing event. In this study, the AI has analyzed hours of live footage with no false positives or missed detections yet. This paper and its subsequent studies aim to provide the railroad industry with state-of-the-art AI tools to harness the untapped potential of an existing closed-circuit television infrastructure through the real-time analysis of their data feeds. The data generated from these studies will potentially help researchers understand human factors in railroad safety research and give them a real-time edge on tackling the critical challenges of trespassing in the railroad industry.

Asset Management Using a Web-Based Accounting Online System To Maintain Value of Company Assets

2017 ◽  
Vol 1 (1) ◽  
pp. 31-37
Author(s):  
Ruli Supriati ◽  
Diah Aryani ◽  
Siti Maesaroh
Keyword(s):  
Real Time ◽  
Asset Management ◽  
Economic Value ◽  
Selling Price ◽  
Accounting System ◽  
Web Based ◽  
Asset Value ◽  
High Level ◽  
Asset Management System ◽  
Asset Data

Assets are entities that are tangible or intangible and have economic value. In a company, assets are important, because assets are assets that must be managed properly to provide good benefits also to the company. Companies must be able to keep the value of each company's assets to always be at a high level, avoiding damage so that the value of the asset does not fall in the selling price. With asset management, companies are increasingly eager in planning outgoing expenses to maintain the value of existing assets, monitoring assets that are bought, sold, or otherwise depreciated. Asset management based on this online accounting system, companies get asset data reports, accumulated depreciation of each asset until the asset value is updated or real-time. Any assets that have been added in the system can be archived, edited or deleted. Import feature on this system also facilitate the company in doing penginputan asset data in large quantity. Thus, with the asset management system directly synchronized with the company's financial data, will provide work efficiency in monitoring each company's assets.   Keywords​: Asset Management, Real-time, Entities, and Asset Values

Real time analysis of Intelligent placing system for vehicles using IOT with Deep learning

2021 ◽  
Author(s):  
Sushma Jaiswal ◽  
Dilip Kumar Sharma ◽  
Tarun Jaiswal ◽  
Bhimraj Basumatary ◽  
Mohit Tiwari ◽  
...  
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Time Analysis ◽  
Real Time Analysis
Sign in / Sign up

Export Citation Format

Share Document