Rummaging Free Parking Spaces using Mask R-CNN

Computer Engineering

Nowadays, finding a vacant parking lot in urban areas is time-consuming. The rise of car ownership has created an imbalance between parking demand and supply. Parking spaces are expensive to build and the parking fee is hiking up day by day. It is hard to enforce and drivers spend an excessive amount of money and time in search of vacant lots. We are proposing a project where our main goal is to detect occupied and empty parking spaces. This project uses deep learning techniques to build a smart parking system through video data analysis that automatically evaluates the availability of vacant parking spaces. For automatic recognition of parking spaces, we are using the Mask R-CNN architecture, a deep learning object detection model, on data obtained from video footage. To mark the parking spot on the input image of the whole parking lot, we use a preprocessed region-based convolutional neural network. And all this can be done with something as simple as a smartphone camera and can notify the user when a parking space is available, via a text message. At the end of the day, we intend to give the best solution for all the parking problems faced in our local area at the least.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.