Skip to content

Kotlin Android mobile application to facilitate garbage collection by connecting individuals and NGOs

License

Notifications You must be signed in to change notification settings

pk-218/EcoConnect

 
 

Repository files navigation

EcoConnect - Google Developers Solution Challenge 2021

Team Coders.ktx

Description

We aim to solve the targets:

  1. Solid Waste Management under the Goal 11 - Sustainable Cities.
  2. Improving recycling rates under the Goal 12 - Responsible Consumption & Production.
  • EcoConnect connects individual households to NGOs near them that provide the means of garbage collection(recyclable waste) through scheduling pickups from households where they have been notified of and as well as processing the garbage for recycling or reuse. This is facilitated with the help of Google Maps. Along with the ability to connect and donate waste, users can detect whether the waste they have can be recycled or not. With the help of AutoML Vision Edge, a ML model is used to classify images of garbage. Thus, the user becomes aware of what can be donated and is given a means of whom/where to donate.

    You can find the APK for our app here

    Demonstration of our app on YouTube

    IMAGE ALT TEXT HERE

Tech Stack

  • We built an Android mobile application using the programming language Kotlin. MVVM (Model-View-ViewModel) architecture was followed throughout for ease of scaling and smooth communication with the UI.

  • For the backend, Firebase was used extensively.

    • For authentication of individuals and organizations, Firebase Google Authentication was used.
    • For storing app data efficiently, Cloud Firestore was used.
    • For creating the image classification machine learning model, AutoML Vision Edge of Firebase and Google Cloud Platform were used. On GCP, Google Cloud Storage and Cloud Vision API were used. Using Cloud Vision API, a ML TFLite model was made for image classification using our custom images and labels.
  • XML was used with Kotlin as the frontend. Navigation drawer was used and to make efficient use of XML, fragments were used extensively.

  • Android Studio 4.2 Beta 6 was the IDE used to develop the Android mobile application. The app was thoroughly tested on the virtual device, Google Pixel 3a and other physical Android devices.

  • Git and GitHub were used for version control and effective collaboration.

Installation

Clone the app in your directory using:

git clone https://github.com/azeez-72/EcoConnect.git

Once cloned, open Android Studio in that directory where you have cloned the project and run the app.

About

Kotlin Android mobile application to facilitate garbage collection by connecting individuals and NGOs

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Kotlin 100.0%