All Collections
Webinar Archive
Webinar Series
Webinar n. 7. - Advanced traffic data analytics for modellers with GoodVision
Webinar n. 7. - Advanced traffic data analytics for modellers with GoodVision
Valeriya Murzyna avatar
Written by Valeriya Murzyna
Updated over a week ago

Transport modellers rely on traffic data provided by their providers - surveyors. Today, 90% of traffic data is still collected manually. Modellers are receiving spreadsheets of simple volume data that is impossible to analyse deeply. But the process is becoming smarter and effective with the help of GoodVision.

This webinar will teach you how to use digitally collected traffic data for various traffic behavior analyses you need for traffic modelling, especially:

1. Vehicle speed analysis

2. Gap-acceptance analysis

3. Saturation flows analysis

4. Occupancy and dwell time analysis

Here you can download the Gap acceptance analysis tool and all reports that were shown during the webinar.



In the webinar, you are showing the drone scenes that are so clear, but footages from real traffic surveys are very often a way lower quality, from the fixed cameras, sometimes from the low-position cameras. Does the speed calculation or gap acceptance work also for these fixed cameras?


Yes, exactly, it works for fixed cameras. The drone scene is nice to show everything clearly, therefore we tried to show you the low-quality footage during the webinar, so you can see that it is also available for various types of video quality.


Do you need a specific type of video configuration for your analysis?


Yes, you need, but configurations are flexible. We have recommendations and guidelines for how you should prepare your video footage. The GoodVision system from the beginning was designed with a bold vision to be compatible with existing infrastructure without knowing what the quality is. You can provide analysis on high-quality footage and very low-quality footage. We are getting various qualities, resolutions, framerates, ratios from all around the world, and we have to deal with it. The ideal resolution is HD or FullHD, but low-quality cameras are not useless. It depends on the use case. It doesn't matter what the camera's resolution is. What's matter is the resolution of the object that is monitored. If you have a simple two-lane road and you are counting vehicles there, you are free to go with the low-quality, even a VGA or 4ATP camera. If you are looking for more detailed analysis, for example, bicycle or big roundabout analysis with lots of movements in the back, far away from the camera, we recommend to go with a high resolution. Then you will get a guaranteed accuracy on all your movements.


Are you flexible with the vehicle classifications for your analysis since this mostly varies among countries?? Can a client make recommendations on the vehicle classification and does that come with an added cost?


Yes, we are offering this type of service like a client-sponsored development. We providing custom classification schemes for your countries. Currently, the vehicle classification scheme that GoodVision provides is COBA 8 that is accepted all around the world, but still, there are specifics. For example, east Asia has specific types of vehicles like tricycles, rickshaws, jeepneys, etc. When these are the requirements from the client, we work closely with a client, and we are preparing the custom model for him. It is a bundled service with a clear process and clear delivery time, clear expectations, and inputs, so the sure answer is YES, we do that, and it comes with added costs.


What do I need as a modeler to start working with GoodVision? Should I upload videos by myself? Because usually this is done by our supplier.


We are not able to replace your supplier. We are providing support to you and your supplier and bringing you closer. The GoodVision Video Insight platform is a great platform that also works as a traffic project management platform. Instead of manual counts, your supplier can process the video footage through GoodVision Video Insights. After the footage is processed, he can share the data without any SD cards or sending links to OneDrive and so on. This is very secure, so the project results can be shared in one click from you to your customer. GoodVision provides pretraining for you and your suppliers, and it comes free of charge.


Can you touch a little on the security of the traffic data sent to your outfit??


We don't use the video data for our company use. When the data are processed with GoodVision Video Insights, it only stays at the platform for so long as it's needed for processing, which is 1 - 4 hours for the whole survey. After the processing, which is fully automated, with no humans touche, the video content is deleted. However, the user can choose that he wants to save or store the footage in GoodVision Vault. We have a service for this, for example, for further forensic analysis and so on. Because without video, you cannot look for events, but we are very strict here when it comes to the data protection policy. It's the client here who owns the data, not us. We don't keep it, we delete it, and the whole process is auditable so it can be proven when the data arrived when it was processed when it was deleted, and we guarantee this. And in case it is needed, we are ready to sign or perform the non-disclosure agreement on some extra terms.


Are your final results editable by the Client?


Yes, all the traffic reports are in .cvs or .xls format, so you can edit it in a way as you wish. You can put it into your own template. The data are in the standard format, so you can easily change whatever you need. The important thing is that you can always regenerate the report. It is not paid by the number of generated reports. Once you have the data, you can report it as many times as you wish. You can also be able to have a specific format of your report that is also customizable.

Did this answer your question?