SpotDash

Description

SpotDash is a web application built using Flask that simplifies the frustration of finding parking lots near you. It consists of three main modules: User, Managers, and Admin.

• User Module: Users can create an account and log in to access the user page, where they can find a list of parking lots within a specific radius of their location.

• Managers Module: Parking area owners can create accounts as managers. They can upload images of their parking areas and use Mask R-CNN to detect available parking spaces. Additionally, managers can detect space availability from video using OpenCV.

• Admin Module: The admin has control over verifying manager accounts and manages user accounts.

Table of Contents

• Installation
• Usage
• Features
• License
• Contact
• Acknowledgments

Installation

To run SpotDash locally, make sure you have the following dependencies installed:

• Python 3.7
• TensorFlow
• Keras
• Mask R-CNN (Refer to Mask R-CNN GitHub Repo)
• OpenCV
• Streamlit
• Flask
• TailwindCSS
• Sqlite3 (Database)(Sqlite3 Database not included)

For enhanced performance (GPU support), use tensorflow-gpu if you have a compatible GPU.

Download Mask R-CNN Model

Download the Mask R-CNN model from this link and place it in the appropriate directory.

Usage

" flask run " (to run flask application[users and managers]) " streamlit run app.py " (to run streamlit application[Admin only])

Features

SpotDash offers the following key features:

• User Module:

  • Create an account and log in.
  • Get a list of parking lots within a specific radius.
  • View parking areas on the main page map.
  • Filter the map to scale up to 10km radius.
  • Select a parking area to view available and occupied spaces.
  • Obtain Google Maps directions to selected parking areas.

• Managers Module:

  • Create an account as a parking area manager.
  • Account activation after admin verification and approval.
  • Log in and add parking details.
  • Upload images of parking areas to detect available parking spaces.
  • Detect occupied spaces (car, motorcycle, or truck) from images.
  • Update parking space status for users to see.
  • Manually mark spaces in videos for detection and update.

• Admin Module:

  • Use Streamlit to manage admin and manager accounts.
  • Approve manager accounts for activation.
  • Get a visualized view of user data.

License

SpotDash is open-source and under the MIT License.

Contact

For any questions, suggestions, or feedback, feel free to contact on LinkedIn: Manukrishna T M.

Acknowledgments

We would like to thank the following individuals/organizations for their contributions and inspiration:

• Mask R-CNN for their amazing work on the Mask R-CNN model.
• This article by Ageitgey for valuable insights on parking space detection.