Analysis and visualization of data obtained from camera mounted on unmanned aerial vehicle used in areas of urban transport

Highlights

• Automatic algorithm of counting vehicles driving in opposite directions using unmanned aerial vehicle.

• Data acquired from cameras mounted on UAV for analysis of urban transportation systems and its sustainability.

• Analysis of the road traffic concerning local road restoration, causing road congestions.

• Result of this paper may be used in the future together with the data from existing intelligent transportation systems.

• Case study. Quantitative measurements and the simulation result that justify the use of UAV within smart sustainable cities.

Abstract

Nowadays, it is practically impossible to find a public place, where a person or object would not be under constant camera observation (mainly used to monitor traffic and to increase safety). Interpretation of such repositories, due to the huge amount of data, is not an easy task. Thus there is a need to introduce an intelligent processing systems that, through algorithmic image analysis, are able to detect e.g. situations worth of attention. The article presents methods of data acquisition from cameras mounted on unmanned aerial vehicles (UAV) and their further analysis, which may be used to improve urban transportation systems and its sustainability. The worked out data concern the situation of urban transport in points of intersection of national and local roads. The analysis of the road traffic concerning local road restoration, causing road closure was achieved using image processing and analysis algorithms. The data was obtained from unmanned aerial vehicle flights in critical city locations. Result of this paper may be used in the future together with the data from existing intelligent transportation systems (a fusion of such data will be needed). The application of used methods will allow to extend and support the existing approaches to manage public and freight transport in cities.

Introduction

Urban areas and number of people in the cities are growing at an alarming rate. Such state and the increase in the number of vehicles in cities, makes the safety ensuring processes and optimal traffic flow (public, passenger and freight) more challenging every year. What's more, traffic congestions became a major factor of urban areas pollution (Gately, Hutyra, Peterson, & Sue Wing, 2017; Tang, McNabola, & Misstear, 2020; Zhang & Batterman, 2013). Therefore, it is important to find an intelligent concept of reducing it, by using smart transportation and mobility recommendations, which are key parts in the development process of smart, sustainable cities (Majumdar, Subhani, Roullier, Anjum, & Zhu, 2021). Analysis of issues such as pollution, incorrect driving behaviour, congestion, etc. no matter how complexed, provide a large number of decision useful data. Such analysis generate measurably effects, with many applicable optimization guidelines to reduce energy consumption, improve infrastructure faults, change driving characteristics, estimate the common decision determinants, etc. (Corcoba Magana & Munoz-Organero, 2015; De Penning, D’Avila Garcez, Lamb, Stuiver, & Meyer, 2014). Understanding those guidelines is the key to define low-carbon policies, assess the behavioural tendencies, dynamical feedbacks and more sustainable city systems.

Concept of smart sustainable cities combines economic development, environmental, social, transportation and technological means (Kin, Verlinde, & Macharis, 2017; Pekin, Macharis, Meers, & Rietveld, 2013) to ensure, safety and welfare at reduced costs. Optimal path to achieve that goal is to use an information and communication technologies (ICT) as an integrator, a special utility that combines the rest of smart cities elements (Bamwesigye & Hlavackova, 2019; Haidine, El Hassani, Aqqal, & El Hannani, 2016).

Study explores such issue and a way to utilize UAVs as a part of the traffic safety ensuring mechanism. UAVs’ high mobility and data generation reliability may be the key solution. We aim to highlight the values of UAVs to the smart cities components (e.g. infrastructure, sensors, transportation systems) and to demonstrate their possible application and potential.

Concept itself is closely linked to the sustainable development, which is adopted in urban development and transport strategies. Within that scope, specialized activities and resource are used to reduce the mobility problems (safety, ecology, congestion) experienced by most urban areas in Europe (Cisowski & Szymanek, 2006; Hossain, Hossain, & Sunny, 2019). Road traffic levels and congestion, air pollution, noise, exhaust emissions, consumption of non-renewable resources and road accidents are important challenges requiring sustainable thinking, a different transportation behaviour, such as: rational travel time management, reduction of travel needs, pedestrian, bicycle and public transport participation increase, pollution and noise levels reduction, transport energy efficiency improvement, vehicle loading capacity optimization, infrastructure capacity increasing, urban space quality improvement (Caggiani, Camporeale, & Ottomanelli, 2017; Mercier, Carrier, Duarte, & Tremblay-Racicot, 2016; Wyszomirski, 2011). Those efforts should focus on improving transport accessibility, ensuring the sustainable development of individual modes of transport and improving the conditions for the provision of services related to the carriage of goods and passengers. Flows of goods and passengers should be optimized through synergies between transport, logistics and system activities using all possible tools (Bibri, 2018; Bibri & Krogstie, 2017; Ministerstwo Infrastruktury, 2019). Transportation system designed this way fosters development and innovation processes in cities. Solution proposed in the article will support many transportation analysis and make a positive impact on the results. It may improve rout selection, vehicle accessibility within supply chains and congestion reduction, thanks to: measurement errors reduction, redundancy of data sources, flexible approach to real-time measurements, data range analysis. Wide potential of such solution in the transport sector may enhance road safety and management supporting systems, as well as comprehensive, universal information repositories and economical systems (Ministerstwo Infrastruktury, 2019). However, it is crucial to integrate it with commonly used analysis tools within an intelligent transport systems.

Intelligent Transport Systems (ITS) are emerging worldwide to make transport more efficient, reliable and safe (all of that, thanks to data collection from various sources). ITS make traffic analysis more efficient and the process of pattern and models acquisition easier. Combined with public policy, intelligent transportation systems may deliver substantial system-wide efficiency means and results (Milojevic-Dupont & Creutzig, 2021). To withstand many unwanted issues, ITS needs to be properly optimized. Detection, tracking and counting of moving vehicles are becoming very important for monitoring, planning and controlling traffic flows (Iwan, Kijewska, & Małecki, 2015; Iwan, Małecki, & Korczak, 2013). Lack of such knowledge may have negative impact on the transport itself (Kijewska, Małecki, & Iwan, 2016). Detecting traffic incidents and traffic obstructions is also the most important tasks of large-scale intelligent transport systems. That task is possible only with constant traffic and event detection. Each of such systems requires adequate source information (properly obtained, prepared and classified) for optimal and efficient traffic management, among other things to ensure the safety of road users (Zhou, Ng, Yang, & Xu, 2021). However, classic monitoring and surveillance systems, mostly based on road-mounted sensors, require significant infrastructure related to the financial resources (a large number of cameras or sensors in a small area) and do not always give a complete overview of the situation. Unmanned aerial vehicles that easily move over roads and pedestrian paths have no such limitations.

UAV surveillance may lead to a new class of spatially precise urban planning and evaluation solutions, if certain conditions are fulfilled. Building an analysis of the subject and specialized literature may point many advantages, as well as challenges but one should notice that specialise solutions and issues may support decision making systems and foster systematic knowledge sharing. Such matters should refer to methods of transport analysis and the idea of sustainability.

The next section introduce the overall architecture and main features of traffic analysis methods and tools. Section 3 describes the methodology, the experiment concerning the traffic flow rate measurement and its optimization algorithm outlined in the chosen scenario. The aerial mappings were used (with their advantages and disadvantages) as a key parts of the analysis. Section 4 contains a case study, quantitative measurements and the simulation result that justify the use of UAV within smart sustainable cities system. Future research, potential impact of the study, and research limitations are discussed as a conclusion.