A disruptive smart-city machine vision platform.
Cloudparc is a disruptive smart city machine vision platform for parking management. It allows users to see an availability of parking slots in an area with statistics and predictive analytics. We developed the camera, that allows detecting a free parking spot nearby. With an onboard decision making, it transfers this information to the system in a real-time, and guides driver to the nearest available parking spot through navigation tool in their mobile app. We used visual recognition, OpenCV, deep learning, on board decision-making tools.
CloudParc is truly smart and autonomous platform, that is perfect for a big city. It allows eliminating the usual parking automates and sensors as well as ticket issuance and collection – it is able to identify a license plate, to determine how long the particular car has been parked and automatically issue a bill to the user through the mobile application. Machine vision, along with a mobile app, allows drivers to find the nearest parking spot.
Installing high-resolution connected cameras on light poles, the technology is able to monitor the street and spot the vacant places.
Our solution is using 10 different machine learning algorithms to ensure the same stable performance in different environmental conditions (low lights, very bright light, dust, snow, etc).
No need for human touch – all the operations and management are made by cameras, hardware, and software and data is automatically stored for further processing, predictive analytics reports, that ensure more smart decision-making.
Machine vision and machine learning tools
We used OpenCV library to take advantage of multi-core processing and Learn Haar cascade object recognition as an object detection method. In the context of a multiplicity of a camera, parameters, and sources of information, we applied onboard decision-making tools.
Case study is available for this project: How to get fully autonomous parking solution using machine vision
Interested in machine vision? Learn also how machine vision could be used for unmanned aerial vehicles.