Improving mobility, safety, and the efficient use of roadways
The new Traffic Detector software is founded on deep neural network-based video analytics that support predictive solutions for traffic monitoring and people counting applications. This advanced analytics technology relies on another type of machine learning — deep learning, which uses artificial neural networks that attempt to mimic the human brain allowing it to learn from large amounts of data and recognize patterns to tackle more complex tasks faster, easier, and more accurately. Traffic Detector precisely distinguishes and classifies people and types of vehicles, including cars, trucks, buses, motorcycles, and bicycles, in congested scenes with precise detail.

Once integrated with Intelligent Video Analytics, Traffic Detector data is augmented with, object speed, GPS location, and color. The deep learning algorithm within Traffic Detector is so robust that it ignores possible disturbances caused by vehicle headlights or shadows, to deliver highly accurate data. Data collected by Traffic Detector can be analyzed, consolidated, and visualized using software like Intelligent Insights to convert data into actionable insights. Operators can use these insights to make informed decisions that can help avoid safety issues, ease traffic congestion, and improve the environment. Data provided from Traffic Detector offers an additional advantage for person detection, which is ideal for perimeter protection, people tracking or retail applications like queuing solutions at cashier desks.


Advantages of using Traffic Detector with INTEOX cameras
Traffic Detector is available as a special Object Classification (OC) version in the INTEOX camera line-up. Depending on the application, customers can choose from either the moving or fixed form factors. Later, Traffic Detector will be made available as software licenses that can be complemented with dedicated software maintenance agreements.

Improved intersection monitoring. Detect the presence and volume of vehicles. And accurately separate different vehicles in crowded scenes for robust counting results to help control traffic lights intelligently and keep traffic flowing. Combined with Intelligent Video Analytics, Traffic Detector data can be elevated with vehicle speed, GPS position, and color.

Improved situational awareness to enhance Automatic Incident Detection solutions in tunnels and on highways. Accurately classify vehicle type and detect presence and volume of people and vehicles. Collect data and statistics to enhance Automatic Incident Detection solutions to better understand traffic patterns, congestion points, and more. Improve the efficient use of roadways and safety of people and vehicles in tunnels and on highways.

Manage parking to reduce emissions and traffic jams. Detect occupancy of every parking space to help drivers find open parking quickly. Easily detect trucks with Traffic Detector to help determine duration of (truck) parking to help reduce congestion on city streets.

Unmistakable object detection and tracking. Combined with Intelligent Video Analytics (IVA), Traffic Detector enables the unambiguous tracking of objects of interest using IVA’s Intelligent Tracking feature.

Improved accuracy of people counting in retail, railway and metro, and other crowded spaces. Detect and count people in complex and busy scenes using a wide-angle view.

Seamless integration
Traffic Detector is fully compatible with our Intelligent Video Analytics (IVA) framework, allowing alarm scripting using the same methodology as IVA. It seamlessly integrates with our AIoT software — BVMS and Intelligent Insights, which consolidates, augments, and converts data from multiple sensors (cameras) into actionable insights. These insights are visualized in easy-to-understand dashboards to support informed decision-making and help to predict potential incidents or business opportunities.

Traffic Detector integrates with third-party video management systems from Genetec, Milestone, and others. It is supported by ONVIF profile M (the Metadata profile), which supports analytics configuration and information query for metadata, as well as filtering and streaming of metadata. It has interfaces for generic object classification, and specified metadata for geolocation, vehicle, license plate, human face and human body.

Source link