Parquery’s smart parking solution provides real-time results for parking facility managers, as well as operational partners interested in taking the few simple, straightforward steps necessary to integrate Parquery’s results into their existing platforms.
Learn more about the requirements for Parquery's algorithms to work for you below.
Parquery detects objects in images from any camera, even those already installed. All we need are the images they produce. No video stream required. Parquery can also detect vehicles in video camera images at any interval the client requires, for example every 10 seconds, every thirty seconds, once a minute, every five minutes, etc.
If you haven’t installed any cameras yet, Parquery will help you identify the optimal number and most effective locations for installation. All we need from you is a map of your parking areas and the available installation height. Using this data and based on a selection of common off-the-shelf cameras, we create a simulation of the camera views. Our goal is to accomplish total coverage of the parking area with the minimum number of cameras.
This translates directly into saving you money: we monitor as many parking spots as possible with as few cameras as possible.
If you haven’t installed any cameras yet, it’s important to select a reliable power source, eg. cable or solar panels.
Ideally, the cameras will be mounted on buildings, tall posts or atop street lamps. Once connected to a power source and running, the cameras can send images in real time or at any time interval chosen by the customer.
For solar-powered cameras, the frequency for sending images depends on battery capacity.
The camera images are either processed remotely on Parquery’s cloud servers or onsite on a dedicated server. In either case, the cameras will have to be connected to a network:
Regardless of whether you run our software locally or on Parquery’s cloud servers, you have two options:
In either case our Operations Team will help you configure your cameras for the best possible results.
Camera installation can be carried out by the customer, an installation company commissioned by the customer, or by professional installers (Parquery partners).
Image analysis and data processing can either be done on Parquery’s cloud servers or on the customer’s premises on a dedicated server. Our Smart Parking experts can support you in making the decision based on your individual requirements.
Parquery’s software doesn’t need any major infrastructure. Our Operations Team will help you identify the best solution for your parking areas based on your existing servers or advise on purchasing new hardware.
Please contact Parquery’s Operations Team for additional details.
As soon as Parquery receives your camera images, we can deliver results. The accuracy of the initial results will be at 95% and will improve to +99% within the first month.
During this period our Operations Team will collect images from your cameras and train our software to adapt to your specific environment. In order to reach the desired accuracy, we need to cover a range of lighting and weather conditions, and identify and compensate for potential edge cases including extreme shadowing, partial occlusions eg. by trees, adverse weather conditions, etc.
Before we deliver the first results, we define a data format together with you that meets your individual requirements.
Once we have collected sufficient training data and achieve consistent high accuracy, we deliver the results to you. We then arrange a second meeting after about a month to get your feedback and modify the dashboard as needed.
Depending on your requirements and suggestions, we will also use this meeting to define next steps.
After comparing with other providers, one of the reason we chose to work with Parquery was the ease of implementation, deployment time, configuration, and calibration.
We are very satisfied with Parquery for their support and their solution's reliability.
M. Franco, SPIE Switzerland Ltd