Abstract
Abstract. Traffic flow analysis is fundamental for urban planning and management of road traffic infrastructure. Automatic number plate recognition (ANPR) systems are conventional methods for vehicle detection and travel times estimation. However, such systems are specifically focused on car plates, providing a limited extent of road users. The advance of open-source deep learning convolutional neural networks (CNN) in combination with freely-available closed-circuit television (CCTV) datasets have offered the opportunities for detection and classification of various road users. The research, presented here, aims to analyse traffic flow patterns through fine-tuning pre-trained CNN models on domain-specific low quality imagery, as captured in various weather conditions and seasons of the year 2018. Such imagery is collected from the North East Combined Authority (NECA) Travel and Transport Data, Newcastle upon Tyne, UK. Results show that the fine-tuned MobileNet model with 98.2 % precision, 58.5 % recall and 73.4 % harmonic mean could potentially be used for a real time traffic monitoring application with big data, due to its fast performance. Compared to MobileNet, the fine-tuned Faster region proposal R-CNN model, providing a better harmonic mean (80.4 %), recall (68.8 %) and more accurate estimations of car units, could be used for traffic analysis applications that demand higher accuracy than speed. This research ultimately exploits machine learning alogrithms for a wider understanding of traffic congestion and disruption under social events and extreme weather conditions.
Highlights
1.1 BackgroundTraffic monitoring and analysis are crucial for an efficient urban planning and management of road traffic infrastructure
automatic number plate recognition (ANPR) systems do not always offer an overall extent of road users due to specific focus on recognising characters in car plates and discarding candidate detections if car plate is not fully recognised (Buch et al, 2011). Such limitations have been overcome with the use of state-of-the-art open-source deep learning technologies (Shi et al, 2017) alongside the freely-available closed-circuit television (CCTV) datasets, enabling training of models for detection and classification of various road users as well as traffic monitoring and prediction (Lv et al, 2015)
Second analysis was performed on 26-28th of February 2018, that is before and during a snow event, taken on A193 Newbridge Street Roundabout where a CCTV sensor monitors the traffic over a part of the A167 Central Motorway (North East Combined Authority (NECA), 2018a)
Summary
1.1 BackgroundTraffic monitoring and analysis are crucial for an efficient urban planning and management of road traffic infrastructure. ANPR systems do not always offer an overall extent of road users due to specific focus on recognising characters in car plates and discarding candidate detections if car plate is not fully recognised (Buch et al, 2011). Such limitations have been overcome with the use of state-of-the-art open-source deep learning technologies (Shi et al, 2017) alongside the freely-available closed-circuit television (CCTV) datasets, enabling training of models for detection and classification of various road users as well as traffic monitoring and prediction (Lv et al, 2015). Number of cars, constituting a main parameter in traffic analysis, can be estimated at various time scales from spatially heterogeneous CCTV locations This type of multiscale spatiotemporal observations supports understanding of traffic congestion before, during and after a disruptive event
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
