Hello! Our project, Parking Analysis via Image Processing, leverages a convolutional neural network to scan still shots of parking lots and determine how many vehicles are present. The data is then pushed to our server to be used to model a forecast for future occupancy levels.
The project overview linked above contains information about the goals and initial motivations for the project, as well as some of the methodologies used.
Above is a video that details the full workflow starting from image to predictions of our system.
Above is a link directed at our live website, where the most recent scan is displayed for users.