TrafficSurvey – Video post-processing platform

Let's meet!
Intertraffic 2024
16 - 19 April 2024
Amsterdam / Netherlands
Traffex
22 - 23 May 2024
Coventry / England

Extending the use of smart video analytics in the USA with the FLOW TrafficCamera

Introduction

As municipalities and public road management organizations embark on a journey to optimize their traffic management strategies, video analytics has emerged as a game-changing solution. The remarkable advancements in hardware, including the latest generation of NVIDIA and INTEL chips, combined with groundbreaking developments in deep learning algorithms and computer vision, have propelled video analytics technology to new heights. Now, a cost-effective traffic sensing solution has arrived – one that outperforms traditional inductive loops while offering a multitude of additional functionalities. This solution is none other than FLOW.

The emergence of FLOW has ignited a paradigm shift in the Intelligent Transportation Systems (ITS) landscape. Its price-competitiveness and unparalleled capabilities have paved the way for a new era of traffic analysis and management. Gone are the days of relying solely on inductive loops; FLOW takes traffic sensing to a whole new level.

The versatility of FLOW is underscored by its ability to tackle a wide array of traffic management scenarios with utmost effectiveness. Through real-world applications, FLOW has proven to be a valuable asset in improving traffic efficiency, promoting safety, and enabling informed decision-making for municipalities and traffic solution integrators.

In this article we showcase the transformative potential of FLOW, highlighting its versatility and effectiveness in revolutionizing traffic management across a multitude of use cases. Experience firsthand how FLOW empowers organizations to embrace smarter, more efficient approaches to traffic operations.

 
FLORIDA CUT OUT1

Videoanalytics setup in Florida with gates gathering Origin-Destionation data

Current state of traffic sensing technology in the US market and its paradigms

In the ever-evolving landscape of traffic management, a seismic shift is reshaping the USA market. For years, the industry relied on traditional sensing technologies, with their inherent limitations and prescribed compatibility requirements. However, the advent of video analytics has ushered in a new era, challenging the status quo and transforming the way data is collected and analyzed.

Gone are the days when legacy systems dictated the boundaries of what was deemed possible. With the rise of video analytics powered by new-generation traffic sensing and data processing technology, a profound shift has occurred. This innovative approach harnesses the power of computer vision and deep learning algorithms, leveraging the capabilities of cameras to capture and analyze real-time traffic data.

One notable legacy requirement in data collection was the need for extensive onboard data storage on specific traffic monitoring edge devices. However, the advancements in IoT connectivity have paved the way for cities to rapidly deploy devices and gather smart city data on updated aggregation platforms. Real-time connections with reduced latency have revolutionized data download and availability, minimizing the need for prolonged periods between login and retrieval. Moreover, these advancements have improved the aggregation, display, and storage of information on server-side and user PC platforms, reducing reliance on local storage.

The evolution of smart sensors has brought about a dual-purpose capability. With the highly accurate FLOW analytics, not only are object detection capabilities enhanced in all weather conditions, but there is also the ability to classify a wide range of road users, including motor vehicles and various categories of active mobility. This expanded data set plays a vital role in meeting traffic operations requirements, providing accurate detections to traffic controllers while simultaneously capturing a rich array of object characteristics within the camera’s field of view.

 

Multi counter multipurpose traffic monitoring

Example video analytics setup for gathering multimodal traffic data on cars, bicycles and pedestrians at the same time

With FLOW video analytics edge devices, real-time data is seamlessly transmitted to traffic controllers via NTCIP or open API network connections, complementing traditional SDLC or I/O methods. Simultaneously, long-term traffic counts and other essential statistics are continuously updated on servers, smart city platforms, traffic control center devices, and other systems that rely on this valuable data. Thanks to direct connections via Ethernet, Wi-Fi, or LTE, FLOW edge devices become an integral part of the city’s network, enabling efficient data transmission and analysis.

As we embrace these advancements in video analytics technology, a new era unfolds, where real-time insights and comprehensive data management drive smarter and more efficient traffic operations. Join us as we explore the transformative power of FLOW and its role in shaping the future of video analytics for municipalities and integrators alike.

The “trusted technology bias” no longer holds back the adoption of video analytics solutions over traditional counterparts like inductive loops, radar detection, or analog cameras. Video analytics’ superior detection capability and enriched data set have overwhelmingly favored their implementation. Successful deployments have proven the cost-effective benefits, including improved accuracy, reliability, ease of installation, verification, connection, and remote monitoring. This new generation of video analytics technology brings forth a wealth of insights from roadside cabinets, revolutionizing traffic management. Intersection performance measures are now readily available, eliminating the need for manual reporting and enabling proactive decision-making through smart sensors and smart city applications.

 

SMARTEK bridge 4 BEFORE

Boston bridge traffic lane distribution before changes (4x vehicle traffic lanes 2 for each way)

Boston bridge project

In this project, the objective was to conduct comprehensive traffic monitoring and data collection for construction zones transitioning into permanent traffic management systems. The Massachusetts Department of Transportation (DOT) sought to monitor traffic on a bridge and evaluate the impact of converting two out of four car lanes into bicycle lanes. However, due to the bridge’s structure, embedding sensors without compromising its longevity was not feasible. After an industry survey conducted by the Department, Smartek ITS won the screening process with DataFromSky’s FLOW, namely the TrafficCamera with onboard video analytics.

The FLOW video analytics system proved invaluable during the bridge monitoring program, enabling real-time reporting and data collection while the traffic configurations evolved. With FLOW’s advanced object classification capabilities encompassing 17 categories, the sensor technology seamlessly adapted as the vehicle lanes were transformed into dedicated bicycle lanes. The four general traffic lanes were reconfigured into two lanes for bicycles and two for cars. Remarkably, these lane adjustments did not necessitate any modifications to the FLOW traffic monitoring system, thanks to the availability of AI classification and the flexible placement of “virtual loops” for continuous and precise classification and directional analysis of all bridge users.

The Maryland DOT appreciated the seamless transition from using the same sensor infrastructure for both work zone and project monitoring to permanent usage for smart city applications. The significant benefits of installation and maintenance, such as avoiding relocation or recalibration, played a decisive role in their selection of FLOW video analytics.

 

SMARTEK bridge 3

Boston bridge video analytics with highlighted pedestrian and bicycle trajectories with the new traffic lane layout – 2x car lanes in the center and 2x bicycle lanes on the outside

FLOW Dashboard

FLOW dashboard example with real-time traffic statistics – data aggregared on the edge device

FLOW video analytics is very versatile and easy to use thanks to its implementation via an ever-improving visual programming language providing the user the ability to filter or trigger based on detected objects’ position or actions within the FOV. This enables everything from simple loop emulation at a stop line to more complicated scenarios such as U-turn detection, pedestrian cross-walk encroachment, blocking vehicle detection, queue detection, red light or stop sign violations, or average speed measuring in addition to simple count studies. The highly optimized and fully configurable trajectory processing engine evaluates dozens of detection tasks in each camera view in parallel, 24/7/365 accurately and reliably. There are even features such as expression and string widget that allows custom coding so you are free to set up the detection and conditioning to track exactly what you need in the way you need it. The ability to add new AI object classes or performance upgrades if required, once identified, and a proven history of performance enhancements at a rapid pace from the DataFromSky’s FLOW development team are also a huge plus!

Conclusion

DataFromSky, a pioneer in computer vision with over 10 years of experience in traffic analysis, is at the forefront of revolutionizing traffic sensing technology worldwide. With a robust portfolio of FLOW video analytics solutions, DataFromSky has already made significant strides in enhancing traffic management across the globe. From North and South America to Europe, Asia, and Australia, FLOW products have been successfully deployed, with over 900 server camera streams and more than 80 TrafficXROADS units in operation. This widespread adoption reflects DataFromSky’s commitment to improving traffic flow, reducing emissions, and ensuring the safety of all road users, be it at intersections, highways, roads, or tunnels. Join us on this transformative journey!

Read about our successful projects.

Read more about DataFromSky.

You can explore different use cases in depth on the individual use-case pages or learn in detail about the different FLOW products from the menu on top of the page.

 

Want to read more? Go back to the BLOG.

Camera closeup fixing ratio EXPORT v.1

Deployments info

  • Use case – Traffic monitoring and data gathering for construction zones transitioning to permanent data collection and traffic management 
  • Product: FLOW – TrafficCamera
  • Vertical: Municipality, Massachusetts Deparment of transportation
  • Location: Boston
  • Local Integrators: Smartek ITS

Results and Highlights

  • Vehicle detection range of 55 meters (60 yards) on Boston bridge
  • Multimodal detection (16 classes) – including vehicles, pedestrians, and bicycles (From FLOW v 1.16.0.)
  • Price competitive solution with high-quality data
  • Dynamic Anonymization
  • 24/7 real-time traffic monitoring
  • Non-intrusive nature and fast deployment of the solution
  • Remote and on-demand addition of new analytics features / traffic metrics thanks to FLOW framework and DataFromSky’s flexibility
  • Ease of installation, configuration, calibration and operation
  •  Ability to integrate to the existing traffic systems such as Highway Traffic Management System
  • Ability to export detection data results, even as raw trajectories
  • Ongoing operation and maintenance services available

Like the story? Share it!