With thousands of kilometers of roads to cover, keeping track of the live situation across the highway network can only be effectively achieved with video surveillance. For traffic authorities, real-time intelligence provided by video surveillance enables a fast, effective response to the now, to achieve a smoother, safer flow of traffic.
But for the longer term, collecting traffic data through video surveillance is important to reveal trends, producing insights for future traffic management. The comprehensive data that network video cameras provide can help authorities improve road planning for less congested, safer journeys, as well as optimize highways maintenance. To meet net zero targets, using this data is also critical to enhance environmental sustainability. As a result, this data can enable smarter, safer traffic.
These capabilities depend on the ability to collect high quality video data, in all weather conditions, day or night, and cameras can be used as a sensor for efficient data collection. But data collection alone is not enough. Crucially, the surveillance cameras must be able to rapidly interrogate the data to achieve actionable insights with effective and reliable analytics. Deep learning and artificial intelligence can uncover patterns buried in surveillance data to enhance improvements and efficient management for urban planners and engineers.
Collecting and analyzing the required data
To optimize the real-time situation on the roads, as well as improve future road planning and management, traffic authorities need to be able to collate and interrogate key data, using it in combination to achieve actionable insights. These data groups typically include:
- Traffic volume along a highway or section is fundamental to understanding congestion. Volume also has an impact on safety, and it’s critical in highways maintenance planning.
- Average speed for the highway section, at given time periods, helps identify congestion in combination with traffic volume. These factors are also important to understand safety and can help identify particular sections for more detailed investigation.
- Road utilization based on vehicle type, from passenger vehicle to heavy truck, is an integral factor in highways maintenance planning, in combination with volume. Vehicle size and weight has an impact on safety, and the ability to classify vehicle types, such as a bus, can also help public transport planning.
While an effective network camera system can provide traffic data across these areas, it can also generate more granular information, drilling down into specific vehicles. Gathering data on vehicle make, model, and color (typically abbreviated to MMC) is possible, and specific vehicle identification can also be achieved through license plate recognition camera technology.
To gather data for actionable insights, collating high quality video under all light conditions and weather situations, is essential. Light sensitivity and Wide Dynamic Range (WDR) are crucial for capturing moving vehicles day or night, generating high clarity with minimal motion blur even in near darkness, combined with high-resolution, full color video. On exposed highways, resistance to high winds that impact image stability is also essential. Electronic Image Stabilization (EIS) provides clear and precise images to capture high speed motion when the camera is subject to vibration.
Identifying specific details or situations can be used to set automated alerts in real time, as well as inform trends for future planning. This is achieved through the processing power of the camera, in combination with analytics based on effective deep learning AI, enabling accurate scene intelligence with minimal false alarms. Edge analytics, performed on the camera itself, optimizes this process by minimizing the response time and network bandwidth requirement compared to server-based video processing.
Real-time traffic data in use
To help manage and ease congestion in real time, traffic authorities in regions such as Hong Kong are using network camera data to direct drivers to faster routes based on live information. This data drives alerts, integrated with mobile apps and highway gantry signs, enabling drivers to navigate the fastest route and easing congestion across the network as a whole. Hong Kong is also implementing a free-flow tolling system with automated payment, removing the need for drivers to stop. A free-flow tolling system is based on RFID (radio frequency identification) technology, using wireless communication between tags and readers. This is supported by cameras equipped with license plate recognition for cars without RFID tags, or as a backup if an RFID tag cannot be detected.
Cameras are also important to enhance road safety by identifying potential hazards before they create an incident, as well as ensuring emergency lanes are kept free from hazards or stationary vehicles. If an incident does occur, automatic incident detection analytics generate an alarm, enabling operators to take rapid action, including guiding emergency services.
Analytics can also help enforce driving regulations to maintain safer highways, from identification of speed, to using prohibited lanes. Similarly, license plate recognition and MMC identification can also optimize sustainability by monitoring and enforcing car use in low-emission zones, as well as lanes dedicated to public transport or multiple occupancy car sharing. Meanwhile, in parking areas, surveillance cameras can monitor the use of charging points for electric vehicles.
Generating statistics to optimize highways planning
Gathering data for analysis is essential to plan the future highways network, generating statistics that identify the most congested road sections, and whether changes to the existing road layout are required. Surveillance cameras, with the capability to capture essential insights, optimize the efficiency and effectiveness of this process.
When planning new highways, or making changes to existing roads, statistics on speed and utilization by vehicle type, combined with incident data, is important to identify safety risk areas. Actionable insights from surveillance camera data can then inform measures to improve safety, including setting speed limits and changing the layout of specific highway sections, through to better understanding of driver behavior to guide driver education.
Ensuring highways maintenance is also vital for safety. Road utilization statistics help plan scheduled maintenance, improving management efficiency as well as reducing the potential for unplanned maintenance. Surveillance cameras can monitor traffic volume combined with vehicle type to identify actual versus expected utilization, and cross checking this data with the road maintenance schedule gives insights into actual maintenance requirements.
Monitoring volume according to vehicle type, and even MMC recognition, can also help inform localized air quality readings as well as CO2 output. Sustainability can also be improved by using surveillance camera data to optimize public transport, where monitoring traffic speed and volume can provide insights to enable measures that improve bus service journey reliability and speed.
Open technology enables statistics generation
Data generated by camera surveillance is crucial to improve the highways network of today and tomorrow. To achieve this, network cameras providing high image quality combined with analytics that can generate actionable insights, are essential.
While insights to improve highways depend on fundamental data sets, building on the capabilities of the camera as a sensor can expand the information and insights achievable from an existing camera network. An open camera platform enables widespread device integration and enhances opportunities for partners to create new developments to improve highways operations and planning.
