Cloud, on-premise, or on-the-edge processing

Parquery's Artificial Intelligence (AI) algorithms can analyze the camera footage on your premises or in the cloud, depending on what works best for you.

Parquery's cloud service - the carefree package

With Parquery's cloud service, you only need a camera and an internet connection by ethernet cable, Wifi, or 3G/4G SIM card. The images are transmitted to Parquery's server, then processed, analyzed, and deleted remotely. An Application Programming Interface (API) or our interactive web-based Dashboard provides the results and analytics.

This option works best for customers who want an all-round carefree package, re-use installed cameras, or wish to use any camera manufacturer, not be locked in with a single provider. No need to purchase, run, monitor, and maintain their own servers, but they can focus on what is vital to their business.

Parquery cloud service - the all-round carefree package

On-premise or on your cloud server - data custody

Parquery can install its software on a client's dedicated on-premise or cloud server. This option is best if company policies do not authorize provider software on cloud servers or the data must not leave the premises. Also, for a diverse range of existing and new cameras from any manufacturer, this is a good option.

For on-premise deployment, you will need to install a dedicated server. The camera footage is then transmitted to this server over a local network, processed, and deleted. The results are provided to the facility manager(s) via API or our Dashboard.

Find out about the on-premise server requirements below.

Parquery on-premise or on your cloud server - keep all data in your custody

On the edge - camera footage remains on site

Edge computing allows data to be handled very close to where it is produced rather than transmitting it to the cloud or data server for processing. In practice, this means connecting a camera to a small powerful computer installed closeby, on which Parquery's solution is installed and processes the data on-site.

The data is kept close to where it is recorded and it, thus, also allows for greater processing speeds. This means, this option is best for video analysis, e.g. traffic or people flow monitoring, and if privacy is a major concern, i.e. image or video must not leave the place where it is being recorded. This option is also good if the customer would like to be independent of a specific camera manufacturer and use existing cameras, e.g. surveillance cameras already installed for security purposes.

Anonymous computation results and analytics can be provided via API or the Dashboard.

Parquery edge computing - all footage remains on site.
Extracted anonymous data from several devices can be aggregated on a server and displayed on the dashboard, or used directly with the API

On the camera - all-in-one device

Parquery developed an on-camera application for Axis Communications cameras with ARTPEC-8 system-on-chip (SoC). With the ARTPEC-8 chip, cameras have the processing power to analyze data within themselves: Parquery runs on the camera. On-camera processing means less bandwidth, storage, and server requirements, thus improving scalability compared to on-premise servers.

This option is best if the images must not leave the camera, only anonymous data, and for customers who want to use Axis cameras and easy deployment.

Anonymous computation results and analytics can be provided via API or our Dashboard.

Parquery on-camera processing with Axis ARTPEC-8 - all-in-one device
Extracted anonymous data from several cameras can be aggregated on a server and displayed on the dashboard, or used directly with the API

Puzzled by the choice?

Our Smart Parking experts are happy to advise you.

Get advice

On-premise server requirements

Parquery’s software does not need major infrastructure. Our Operations Team will help you identify the best solution based on your existing servers or advise on purchasing new hardware.

Here is an example of server requirements for an off-street parking area with 200 spots:

Linux Virtual Machine (or specific Linux Machine);
Minimum 2 CPUs;
Minimum 4 Gb RAM;
Approximately 100 Gb of storage space on your hard drive (including space for Linux).

Please contact Parquery’s Operations Team for additional details Get in touch

Privacy and data protection

Both, Parquery’s on-premise and cloud analysis comply with Swiss and European privacy laws (GDPR):

We do not store any images.
We automatically delete images after processing.
We guarantee that no personal data is present in the images. Our solution automatically blurs pedestrians and vehicle number plates before image analysis to ensure data remains anonymous.
If analyzed on-premise, the images do not leave the customer's premises but are processed and analyzed on a server on-site.