A video-based vehicle counting system using an embedded device in realistic traffic conditions

One of the most important features of smart cities is efficient traffic monitoring. Currently, many monitoring approaches focus on video-processing techniques using traffic surveillance cameras. However, video analytics for traffic monitoring on edge devices like cameras is a difficult task, due to limited computational resources and variety of unknown traffic scenarios. To overcome these difficulties, we designed and evaluated a real-time vehicle counting system using deep neural networks in an embedded device. Experimental results were carried out to determine the best system configuration parameters and to analyze the impact of changing environmental conditions on our system performance. For urban vehicle counting, our approach could achieve a recall and precision values of 99% within a video processing time of 10 frames per second.

[1]  Weisong Shi,et al.  Edge Computing: Vision and Challenges , 2016, IEEE Internet of Things Journal.

[2]  Rituparna Chaki,et al.  Applications of wireless sensor network in Intelligent Traffic System: A review , 2011, 2011 3rd International Conference on Electronics Computer Technology.

[3]  Bing-Fei Wu,et al.  A Real-Time Vision System for Nighttime Vehicle Detection and Traffic Surveillance , 2011, IEEE Transactions on Industrial Electronics.

[4]  Simon Elias Bibri,et al.  Smart sustainable cities of the future: An extensive interdisciplinary literature review , 2017 .

[5]  Wei Liu,et al.  SSD: Single Shot MultiBox Detector , 2015, ECCV.

[6]  Yingjie Xia,et al.  Towards improving quality of video-based vehicle counting method for traffic flow estimation , 2016, Signal Process..

[7]  Reza Safabakhsh,et al.  Vehicle detection, counting and classification in various conditions , 2016 .

[8]  Minsub Kim,et al.  Investigation on performance and energy efficiency of CNN-based object detection on embedded device , 2017, 2017 4th International Conference on Computer Applications and Information Processing Technology (CAIPT).

[9]  Gabriel-Miro Muntean,et al.  A Communications-Oriented Perspective on Traffic Management Systems for Smart Cities: Challenges and Innovative Approaches , 2015, IEEE Communications Surveys & Tutorials.

[10]  Le Thanh Nguyen-Meidine,et al.  A comparison of CNN-based face and head detectors for real-time video surveillance applications , 2017, 2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA).

[11]  Pascal Perez,et al.  Edge-Computing Video Analytics for Real-Time Traffic Monitoring in a Smart City , 2019, Sensors.

[12]  Nikhil Yadav,et al.  Comparative Study of Object Detection Algorithms , 2017 .

[13]  Dave Evans,et al.  How the Next Evolution of the Internet Is Changing Everything , 2011 .

[14]  Gabriel Barros-Gavilanes,et al.  Edge vs. cloud computing: where to do image processing for surveillance? , 2019, Security + Defence.

[15]  Abhirup Khanna,et al.  Intelligent Traffic Management System for Smart Cities , 2018, Communications in Computer and Information Science.

[16]  Noel E. O'Connor,et al.  Performance of video processing at the edge for crowd-monitoring applications , 2018, 2018 IEEE 4th World Forum on Internet of Things (WF-IoT).

[17]  Bo Tang,et al.  Incorporating Intelligence in Fog Computing for Big Data Analysis in Smart Cities , 2017, IEEE Transactions on Industrial Informatics.

[18]  Axel Jantsch,et al.  Fog Computing Fundamentals in the Internet-of-Things , 2018 .

[19]  Louahdi Khoudour,et al.  Automatic vehicle counting system for traffic monitoring , 2016, J. Electronic Imaging.

[20]  Andrés Heredia,et al.  Comparing SVM and SSD for classification of vehicles and pedestrians for edge computing , 2019, 2019 IEEE Colombian Conference on Applications in Computational Intelligence (ColCACI).

[21]  Andrés Heredia,et al.  Video processing inside embedded devices using SSD-Mobilenet to count mobility actors , 2019, 2019 IEEE Colombian Conference on Applications in Computational Intelligence (ColCACI).

[22]  Choong Seon Hong,et al.  Edge-Computing-Enabled Smart Cities: A Comprehensive Survey , 2019, IEEE Internet of Things Journal.

[23]  Alberto Quattrini Li,et al.  Deep Neural Networks: A Comparison on Different Computing Platforms , 2018, 2018 15th Conference on Computer and Robot Vision (CRV).

[24]  Wolfram Burgard,et al.  Probabilistic Robotics (Intelligent Robotics and Autonomous Agents) , 2005 .

[25]  Bo Chen,et al.  MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications , 2017, ArXiv.

[26]  Bhagya Nathali Silva,et al.  Towards sustainable smart cities: A review of trends, architectures, components, and open challenges in smart cities , 2018 .

[27]  Patan Rizwan,et al.  Real-time smart traffic management system for smart cities by using Internet of Things and big data , 2016, 2016 International Conference on Emerging Technological Trends (ICETT).