Parquery: Smart Parking Application for Smart City
An ETH Zurich based “Spin Off” named Parquery has collaborated with SELISE in order to deliver their product, a computer vision based system for monitoring on-street parking occupancy. This Smart City application helps cities optimize their on-street parking administration and enforcement while at the same time making parking spot availability accessible through a web-based application.
SELISE in Smart City Business
In Locarno, Switzerland, where the first solution has been deployed on street parking occupancy is now monitored in real time. Twenty cost effective outdoor IP cameras analyse chosen parking areas all over the city. The cameras contain a 3G module, that sends images to the Parquery cloud which processes them using highly sophisticated computer vision algorithms. The result of this process is exact real-time parking availability information which is independent from weather conditions, daytime and angle at which the images are recorded.
The Modern City
Most of the cities are suffering from traffic caused by cars cruising around while searching for parking. Such search traffic does not only waste time and fuel for the drivers but is also detrimental to the municipality, as it causes pollution, traffic congestion, degraded pedestrian environments and lowers income of small shops and restaurants in various dimensions.
Today parking occupancy information is mostly available only for parking garages and off-street parking, where parked cars are easily counted. While on-street parking is more preferable to drivers since it is usually more convenient, occupancy is hardly obtained. Thus, parking search assistance is turning out to be one of the major unsolved problems in urban mobility.
Parquery, with its smart computer vision algorithm, figures out the real time parking occupancy in different parking spots and lets a modern Java stack handle application and data concerns. The occupancy information is stored in a NoSQL database (MongoDB) from where data is analytically processed (OLAP) and dumped into a data warehouse using Hive. For real time high velocity data processing and fault tolerance, Apache Storm and Kafka are being used in Parquery.
Parquery provides a responsive web application built on Spring Boot and AngularJS, which uses Google map API to help the users find an empty parking spot. The scope of this application is not only limited to finding a parking spot though. With the help of the smart algorithm, parking inspectors can figure out overtime occupancy in parking spots and take the necessary measures. Parquery furthermore analyzes historical data and generates detailed, configurable parking occupancy reports that help the city authority to better plan future administrative decisions.
Parquery is the next generation solution for city dwellers. It helps reducing traffic by up to 30% in relevant areas. Real time vacancy data and smart routing have shown up to 43% improvement in reducing parking spot searching time in crowded business districts. Smart parking also helps to increase revenues and save costs by applying better payment discipline, make enforcement and collection automated and simple for the city administration.
Parquery was featured in RSI News, a concern of Swiss Broadcasting Corporation. In the report, features and prospect of Parquery was discussed.
Hadoop, Hive, Spring, Hibernate, AngularJS