Traffic Data Types Provided by GoodVision

There are many various types of traffic data and information about traffic behavior which can be collected in the streets or from the roadways. This data provide precious insights to traffic engineers on road usage, traffic patterns or safety. Traditional data collection methods like human surveyors or pressure hoses give very limited set of this data in contrast to video analytic solutions. This article provides the overview of the traffic data that are collected and provided by GoodVision Video Insights product at my.goodvisionlive.com.

Traffic engineers rely on accurate, high-quality traffic data. It is a crucial starting point to complete their projects. While they can choose to collect that data using traditional methods, including loops, pucks, and road tubes, many organizations are moving to video data collection using artificial intelligence as a reliable alternative to gather traffic data under various conditions. GoodVison provides following traffic data (seen details and examples below):

  1. Spot traffic counts
  2. Turning movement counts
  3. Roundabout traffic counts
  4. Origin-destination traffic counts
  5. Vehicle classification counts
  6. Pedestrian and bicycle counts
  7. Exact traffic trajectories and positions
  8. Vehicle delay data
  9. Vehicle transit time
  10. Time-distance between vehicles
  11. Traffic collision points
  12. Peak traffic hours
  13. Traffic volume time series
  14. Traffic frequency heat-maps
  15. Traffic speed heat-maps
  16. Traffic delay/hold-up map

About GoodVision Video Insights

GoodVision Video Insights (GVVI) is a software tool providing fully automated traffic data collection from video recordings. It also provides powerful analytical features for deep traffic analysis and reporting all in a single tool accessible to anyone. It operates online on a pay-as-you-go basis and works with your existing cameras.

1. Most common traffic data

Spot Traffic Counts

You can draw virtual lines or virtual zones directly on the screen and focus on traffic counting in these areas. How many cars are driving the highway? Draw the virtual spots on the desired lanes and count them. The counting works on unlimited number of road lanes, if visible.

Virtual lines provide bi-directional counts, virtual zones are more versatile for vehicle traffic.

Turning Movement Counts (TMCs)

Sometimes called intersection counts are one of the most sought traffic data type for traffic modelling, to determine the intersection capacity or to analyze signal timing. You can isolate any traffic turning movement (e.g. left, right turns, or through)on the intersection by connecting virtual zones (and lines) together. These movements can be exported as an OD matrix.

By combining more virtual areas into complex filters, all traffic movements on the intersection can be covered.

Roundabout Traffic Counts

Similarly you can monitor any specific traffic movement on the roundabouts. In a roundabout count, entering and circulating vehicles are tracked to their exit point. Since roundabouts are not signalized, the traffic flow is less predictable and capturing those movements accurately with traditional methods is very challenging. With GoodVision it is simple as any other counts.

Traffic movements on roundabouts work in the similar way as on intersections.

Origin-destination Traffic Counts

Similarly as TMCs or roundabout counts, any custom defined traffic movement can be isolated in GoodVision and reported the same way as TMC counts.

Vehicle Classification Counts

Each traffic count provided by GoodVision is composed of counts for each vehicle class separately. We recognize 8 classes of objects: car (taxi), van (ogv1), truck (ogv2 + heavy), bus, motorcycle, bicycle, pedestrian, animal.

All counts in GoodVision are by vehicle class.

On visual analytics screen you can distinguish the trajectories of these classes by following color code:

Pedestrian and Bicycle Counts

Non-motorized modes of traffic are traditionally surveyed using the same types of sensors used for motorized vehicles which provides unpleasant results. These types of projects are most commonly provided to help cities understand the current state of facility usage, to help prioritize locations in which to implement pedestrian- and bike-related facilities, and to evaluate the effects of projects, such as bike lanes. GoodVision recognizes and collects pedestrians and bicycles from the video automatically from roads, crosswalks, bicycle paths — from anywhere on the scene.

Pedestrians or bicycles are detected by GoodVision anywhere on the scene.

2. Advanced contextual traffic data

Exact Traffic Trajectories

GVVI visualizes exact object routes in the video, from entering the scene, to the time they left it. Exact trajectories help you to accurately position your counting spots and give you exact overview of the traffic behavior on the scene.

Displaying the trajectories is the good way to start your analysis and get the overall picture.

Exact Traffic Positions in Time

GVVI visualizes exact traffic object positions and the density of movement using position maps.

Vehicle Stay/Delay in Area

GVVI allows you to isolate just the traffic which has spent a certain period of time within the specific area. This helps to count delayed vehicles, e.g. just the vehicles involved in the traffic jam.

Vehicle Transit Time

GVVI calculates exact transit time of vehicles trespassing a specific virtual area or the time of transit from origin to destination.

Time-Distance of Vehicles

Knowing the time-distance between individual vehicles helps to analyze traffic behavior, creation of traffic jams and to solve many traffic safety issues. GVVI provides time distance calculation as a standard output for any traffic movement on the scene.

Traffic Collision Points

These are the places on intersection of various types where different traffic flows are crossing each other simultaneously which leads to increased accident risk. GVVI provides and visualizes real collision points in contrast to those calculated using traditional methodologies.

Advanced traffic reports

Total Volume, Average Volume and Peak Traffic Hours

Knowing the total traffic count, average count per interval and peak hours of heaviest traffic is a must. GVVI reports these data on every reports screen for all traffic movements.

Traffic Volume Time Series

There are several available bar/area/heat-map diagrams available to display various aspects of traffic volume distribution i time by given filters — e.g. traffic in specific direction. These diagrams helps you to identify traffic flow time patterns, traffic variety in time, traffic jamming or deaf time spots. GVVI allows to adjust the time granularity (reported interval) from 1 sec to 24 hours and more.

Traffic Frequency (Heat-map)

GVVI visualizes the frequency of traffic movement in the video. Hotter areas indicate higher traffic volume while cold areas indicate lower or no traffic.

Traffic Acceleration/Deceleration (Heat-map)

GVVI visualizes the relative speed of traffic in the video. Hotter areas indicate decelerating traffic while cold areas indicate accelerating traffic. This report is useful to indicate potentially unsafe areas or areas with huge traffic delays.

Traffic Delay/Hold-up Heat-maps

GVVI visualizes the average delay of the traffic within various sectors of the scene. While speed heat-maps inform you about delaying/decelerating traffic, hold-up heat-map informs about exact traffic delay/waiting time in the area.

Comprehensive Traffic Movements Report

All designed traffic movements on your video scene can be exported into Excel report using your desired time interval granularity, object classes, etc.


Daniel Stofan is a co-founder and CEO of GoodVision — a company devoted to innovate the ways of traffic data collection. Our product GoodVision Video Insights is the autonomous traffic data collection cloud service providing highly reliable traffic data from standard surveillance cameras via Artificial Intelligence. It provides advanced data analytics, data visualization and managerial reporting — all in a single platform covering the whole process of operation from data collection to decision making, with 1-hour delivery of results.