ViSense CrowdDynamics v1.6.0

This document provides an overview of the changes in ViSense CrowdDynamics v1.6.0, released on April 11th, 2017.

The following sections describe the most notable changes in this release (aside from bugfixes and minor improvements). For the complete changelog, see the end of this document.

New: Pedestrian trajectories in GPS coordinates

Since version 1.1, ViSense CrowdDynamics supports the conversion from the camera positions of each detected pedestrian to a calibrated distance in meters from the camera,
also known as real-world coordinates. This allows for a top-view of the measured area where each object can be mapped on the scene.
Starting with version 1.6, it is also possible to convert these real-world coordinates to GPS coordinates. These coordinates can be used to track the pedestrians in street views or maps. Such overview is useful for crowd management or the visualization of city dynamics. As with camera calibrations for real-world analysis, GPS calibration is performed by our experts and can be purchased as an additional service. Please note that GPS calibration is currently only provided using the ViSense API, and is only available in the Netherlands. Please contact us for more information and pricing.

New: ViSense tools

ViSense CrowdDynamics is not only for usage as a standalone application, but also as a data provider for system integrators. This enables integrators to enrich their information provision and dashboard using the advanced ViSense analytics. By integrating the ViSense API, real-time data can be retrieved and integrated in your own applications. To further support integration of the ViSense API, we present the ViSense tools which provide ViSense API integration examples. Currently, using ViSense tools you can handle authentication and retrieve system information from any ViSense system.
The ViSense tools are written in NodeJS (available on github and NPM) and are actively maintained and further developed. We encourage system integrators to actively participate in the ViSense tools development by submitting bug reports and pull requests.

Special message for system integrators: GPS API

The data of the GPS calculations is exposed to system integrators with a GPS API. Both the real-time GPS coordinates are available (using /object/gps) as bundled GPS trajectories (using /object/track/gps). If you want to extract the GPS trajectories from the API, you can connect to the following websockets:

ws://xxx.xxx.xxx.xxx/api/v1/data/main/object/gps
ws://xxx.xxx.xxx.xxx/api/v1/data/main/object/track/gps

Please note that the GPS information is only available after calibration of the ViSense system, as mentioned before. Furthermore, GPS information is currently only available in
the Netherlands. Finally, please note that GPS tracks are not stored locally and thus GPS coordinates can only be calculated with the latest GPS calibration. Therefore, if the camera is moved or zoomed and a new calibration has been performed, previous object trajectories can no longer be converted to GPS coordinates.

Additional API calls for system integrators

To further ease the integration process for system integrators, we provide extra API calls that can be used during integration development. The authentication API has a new command: clearUserSessions. With this command, one can clear all open sessions for a single user to free session slots. It can be called by HTTP POST, supplying the username and password. The API can be found at:

http://xxx.xxx.xxx.xxx/api/v1/authenticate

An API was added to retrieve basic device information without an authentication session. Currently the only supported information field is the service tag, which is used to identify the system. This service tag is also shown in the login page of the web interface. The device info API path is:

http://xxx.xxx.xxx.xxx/api/v1/device/info

An API call was added with which a firmware restart can be requested, which is faster than a full system reboot. Rebooting was already available as an API call in the ViSense system, and takes about 20 seconds, a firmware restart takes only a few seconds. A firmware restart can be requested by calling:

http://xxx.xxx.xxx.xxx/api/v1/firmware?command=restart

For further information, please view the documentation on your ViSense box at:

http://xxx.xxx.xxx.xxx/getting-started.html

If you have no access to your ViSense system, click here.

Suggestions?

We are continuously improving our current system and documentation. Therefore, we are always eager to hear your suggestions to improving ViSense.

Complete changelog

New features

  • Pedestrian trajectories available as GPS coordinates in API.
  • ViSense tools: integration examples in Node.js available on github and NPM.
  • New API-related items
    • Authentication API call to clear all login sessions for a single user
    • Device info API (currently supported: service tag)
    • Firmware restart call in device/firmware API

Improvements

  • New software architecture that enables the use of multiple, separate detectors.
  • Show service tag in the login-page of the web interface.

Bug fixes

  • Fixed error when visualizing the top-view perspective while having a stretched camera view.
  • Fixed counting error when requesting many counting intervals from the cumulative count API.
  • Fixed visualization errors when visualizing the counts of the previous hour of the current day and of the last hour.
  • Fixed minor memory bug.

Schrijf u in voor de Nieuwsbrief.






*
Verplichte velden