scholarly journals A Deep Learning Streaming Methodology for Trajectory Classification

2021 ◽  
Vol 10 (4) ◽  
pp. 250
Author(s):  
Ioannis Kontopoulos ◽  
Antonios Makris ◽  
Konstantinos Tserpes
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
High Frequency ◽  
Tracking System ◽  
High Volume ◽  
Classification Performance ◽  
Automatic Identification ◽  
Mobility Patterns ◽  
Real World Data ◽  
Vessel Tracking

Due to the vast amount of available tracking sensors in recent years, high-frequency and high-volume streams of data are generated every day. The maritime domain is no different as all larger vessels are obliged to be equipped with a vessel tracking system that transmits their location periodically. Consequently, automated methodologies able to extract meaningful information from high-frequency, large volumes of vessel tracking data need to be developed. The automatic identification of vessel mobility patterns from such data in real time is of utmost importance since it can reveal abnormal or illegal vessel activities in due time. Therefore, in this work, we present a novel approach that transforms streaming vessel trajectory patterns into images and employs deep learning algorithms to accurately classify vessel activities in near real time tackling the Big Data challenges of volume and velocity. Two real-world data sets collected from terrestrial, vessel-tracking receivers were used to evaluate the proposed methodology in terms of both classification and streaming execution performance. Experimental results demonstrated that the vessel activity classification performance can reach an accuracy of over 96% while achieving sub-second latencies in streaming execution performance.

scholarly journals Deep Learning-Based Portable Device for Audio Distress Signal Recognition in Urban Areas

2020 ◽  
Vol 10 (21) ◽  
pp. 7448
Author(s):  
Jorge Felipe Gaviria ◽  
Alejandra Escalante-Perez ◽  
Juan Camilo Castiblanco ◽  
Nicolas Vergara ◽  
Valentina Parra-Garces ◽  
...  
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Real Time ◽  
Urban Areas ◽  
Low Cost ◽  
Raspberry Pi ◽  
Automatic Identification ◽  
Signal Recognition ◽  
Main Challenge

Real-time automatic identification of audio distress signals in urban areas is a task that in a smart city can improve response times in emergency alert systems. The main challenge in this problem lies in finding a model that is able to accurately recognize these type of signals in the presence of background noise and allows for real-time processing. In this paper, we present the design of a portable and low-cost device for accurate audio distress signal recognition in real urban scenarios based on deep learning models. As real audio distress recordings in urban areas have not been collected and made publicly available so far, we first constructed a database where audios were recorded in urban areas using a low-cost microphone. Using this database, we trained a deep multi-headed 2D convolutional neural network that processed temporal and frequency features to accurately recognize audio distress signals in noisy environments with a significant performance improvement to other methods from the literature. Then, we deployed and assessed the trained convolutional neural network model on a Raspberry Pi that, along with the low-cost microphone, constituted a device for accurate real-time audio recognition. Source code and database are publicly available.

scholarly journals AIS Politics: The Contested Use of Vessel Tracking at the EU’s Maritime Frontier

2019 ◽  
Vol 44 (5) ◽  
pp. 881-899 ◽  
Author(s):  
Lorenzo Pezzani ◽  
Charles Heller
Keyword(s):  
Mediterranean Sea ◽  
Collision Avoidance ◽  
Tracking System ◽  
Automatic Identification ◽  
Identification System ◽  
Automatic Identification System ◽  
Vessel Tracking ◽  
Collision Avoidance System ◽  
The Mediterranean ◽  
Do So

Automatic identification system (AIS) is a vessel tracking system, which since 2004 has become a global tool for the detection and analysis of seagoing traffic. In this article, we look at how this technology, initially designed as a collision avoidance system, has recently become involved in debates concerning migration across the Mediterranean Sea. In particular, after having briefly discussed its emergence and characteristics, we examine how through different practices of (re)appropriation AIS, and the data it generate, have been seized upon, both to contest and to sustain the exclusionary nature of borders, and the mass dying of migrants at sea to which it leads. We do so by referring to forms of data activism we have contributed to in the frame of our Forensic Oceanography project as well as to situations in which AIS has been mobilized by xenophobic groups to demand even stronger exclusionary measures. At the same time, we point to the multiplicity of actors who participate in the politics of migration through AIS in unexpected ways. We conclude by highlighting the irreducible ambivalence of practices of appropriation and call for persistent attention to one’s own positioning within the global datascape constituted by AIS and other data.

Towards Real-Time Moment Tensor Inversions in a Data Rich Micro-Seismic Environment using Deep Learning

2021 ◽  
Author(s):  
Nima Nooshiri ◽  
Christopher J. Bean ◽  
Francesco Grigoli ◽  
Torsten Dahm
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Real Time ◽  
High Frequency ◽  
Moment Tensor ◽  
Source Parameters ◽  
Data Set ◽  
Real Time Information ◽  
Time Information ◽  
Current Inversion

<p>Despite advanced seismological methods, source characterization for micro-seismic events remains challenging since current inversion and modelling of high-frequency waveforms are complex and time consuming. For a real-time application like induced-seismicity monitoring, these methods are slow for true real-time information because they require repeated evaluation of the often computationally expensive forward operation. Moreover, because of the low amplitude and high-frequency content of the recorded micro-seismic signals, routine inversion procedure can become unstable and manual parameter tuning is often required. Therefore, real-time and automatic source inversion procedures are difficult and not standard. A more promising alternative to the current inversion methods for rapid source parameter inversion is to use a deep-learning neural network model that is calibrated on a data set of past and/or possible future observations. Such data-driven model, once trained, offers the potential for rapid real-time information on seismic sources in a monitoring context.</p><p>In this study, we investigate how a supervised deep-learning model trained on a data set of synthetic seismograms can be used to rapidly invert for source parameters. The inversion is represented in compact form by a convolutional neural network which yields seismic moment tensor. In other words, a neural-network algorithm is trained to encapsulate the information about the relationship between observations and underlying point-source models. The learning-based model allows rapid inversion once seismic waveforms are available. Moreover, we find that the method is robust with respect to perturbations such as observational noise and missing data. In this study, we seek to demonstrate that this approach is viable for micro-seismicity real-time estimation of source parameters. As a demonstration test, we plan to apply the new approach to data collected at the geothermal field system in the Hengill area, Iceland, within the framework of the COSEISMIQ project funded through the EU GEOTHERMICA programme.</p>

Smart Surveillance System Using Computer Vision and Internet of Things

2020 ◽  
Vol 17 (1) ◽  
pp. 68-73
Author(s):  
M. Hemaanand ◽  
V. Sanjay Kumar ◽  
R. Karthika
Keyword(s):  
Deep Learning ◽  
Internet Of Things ◽  
Real Time ◽  
Tracking System ◽  
Single Shot ◽  
Stream Data ◽  
Advanced Technique ◽  
The Real ◽  
Smart Surveillance ◽  
Bounding Boxes

With the evolution of technology ensuring people for their safety and security all around the time constantly is a big challenge. We propose an advanced technique based on deep learning and artificial intelligence platform that can monitor the people, their homes and their surroundings providing them a quantifiable increase in security. We have surveillance cameras in our homes for video capture as well as security purposes. Our proposed technique is to detect and classify as well as inform the user if there is any breach in security of the classified object using the cameras by implementing deep learning techniques and the technology of internet of things. It can serve as a perimeter monitoring and intruder alert system in smart surveillance environment. This paper provides a well-defined structure for live stream data analysis. It overcomes the challenge of static closed circuit cameras television as it serves as a motion based tracking system and monitors events in real time to ensure activities are limited to specific persons within authorized areas. It has the advantage of creating multiple bounding boxes to track down the objects which could be any living or non-living thing based on the trained modules. The trespasser or intruder can be efficiently detected using the CCTV camera surveillance which is being supported by the real-time object classifier algorithm at the intermediate module. The proposed method is mainly supported by the real time object detection and classification which is implemented using Mobile Net and Single shot detector.

scholarly journals Real-Time and Deep Learning Based Vehicle Detection and Classification Using Pixel-Wise Code Exposure Measurements

Electronics ◽  
2020 ◽  
Vol 9 (6) ◽  
pp. 1014 ◽  
Author(s):  
Chiman Kwan ◽  
David Gribben ◽  
Bryan Chou ◽  
Bence Budavari ◽  
Jude Larkin ◽  
...  
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Compressive Sensing ◽  
Real Time ◽  
Classification Performance ◽  
Video Data ◽  
Real Time System ◽  
Processing Power ◽  
Compressed Data ◽  
Processing Device

One key advantage of compressive sensing is that only a small amount of the raw video data is transmitted or saved. This is extremely important in bandwidth constrained applications. Moreover, in some scenarios, the local processing device may not have enough processing power to handle object detection and classification and hence the heavy duty processing tasks need to be done at a remote location. Conventional compressive sensing schemes require the compressed data to be reconstructed first before any subsequent processing can begin. This is not only time consuming but also may lose important information in the process. In this paper, we present a real-time framework for processing compressive measurements directly without any image reconstruction. A special type of compressive measurement known as pixel-wise coded exposure (PCE) is adopted in our framework. PCE condenses multiple frames into a single frame. Individual pixels can also have different exposure times to allow high dynamic ranges. A deep learning tool known as You Only Look Once (YOLO) has been used in our real-time system for object detection and classification. Extensive experiments showed that the proposed real-time framework is feasible and can achieve decent detection and classification performance.

scholarly journals Challenges of an Autonomous Wildfire Geolocation System Based on Synthetic Vision Technology

Sensors ◽  
2018 ◽  
Vol 18 (11) ◽  
pp. 3631 ◽  
Author(s):  
Victor Arana-Pulido ◽  
Francisco Cabrera-Almeida ◽  
Javier Perez-Mato ◽  
B. Dorta-Naranjo ◽  
Silvia Hernandez-Rodriguez ◽  
...  
Keyword(s):  
Real Time ◽  
High Speed ◽  
Tracking System ◽  
High Volume ◽  
Human Interaction ◽  
Detection Methods ◽  
Synthetic Vision ◽  
Technological Advantage ◽  
Wildfire Suppression ◽  
Dry Conditions

Thermographic imaging has been the preferred technology for the detection and tracking of wildfires for many years. Thermographic cameras provide some very important advantages, such as the ability to remotely detect hotspots which could potentially turn into wildfires if the appropriate conditions are met. Also, they can serve as a key preventive method, especially when the 30-30-30 rule is met, which describes a situation where the ambient temperature is higher than 30 ∘ C, the relative humidity is lower than 30%, and the wind speed is higher than 30 km/h. Under these circumstances, the likelihood of a wildfire outburst is quite high, and its effects can be catastrophic due to the high-speed winds and dry conditions. If this sort of scenario actually occurs, every possible technological advantage shall be used by firefighting teams to enable the rapid and efficient coordination of their response teams and to control the wildfire following a safe and well-planned strategy. However, most of the early detection methods for wildfires, such as the aforementioned thermographic cameras, lack a sufficient level of automation and usually rely on human interaction, imposing high degrees of subjectivity and latency. This is especially critical when a high volume of data is required in real time to correctly support decision-making scenarios during the wildfire suppression tasks. The present paper addresses this situation by analyzing the challenges faced by a fully autonomous wildfire detection and a tracking system containing a fully automated wildfire georeferencing system based on synthetic vision technology. Such a tool would provide firefighting teams with a solution capable of continuously surveilling a particular area and completely autonomously identifying and providing georeferenced information on current or potential wildfires in real time.

Real-time Aircraft Tracking System: A Survey and A Deep Learning Based Model

2021 ◽  
Author(s):  
Muhammed Emir cakici ◽  
Feyza Yildirim Okay ◽  
Suat Ozdemir
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Tracking System ◽  
System A ◽  
Aircraft Tracking
Sign in / Sign up

Export Citation Format

Share Document