Android-based mobile App to Predicting Air Quality using computer vision and machine learning - My final year project

 

this video is a demo and explanation of our project

System Introduction

Our app is android based which predict air quality index using machine learning and computer vision. Portable air quality index meters are very expensive and inconvenient to carry around. Nowadays everyone has a phone and camera in it. We will make it cheaper to check the air quality and this will help to solve this ever-growing world problem.

Background of the System

An air quality index (AQI) is used by government agencies^[1] to communicate to the public how polluted the air currently is or how polluted it is forecast to become. Public health risks increase as the AQI rises. Different countries have their own air quality indices, corresponding to different national air quality standards. Some of these are the Air Quality Health Index (Canada), the Air Pollution Index (Malaysia), and the Pollutant Standards Index (Singapore).

The EPA (Environment protection department) maintains a system of rating the safety of the air in a given area, called the Air Quality Index. Understanding the Air Quality index is important because it gives people vital information about the conditions of the air in their location and how the quality of the air in their city can impact their health.

They are many apps who is providing the air quality index but they only use geotagging. In our project, we are using computer vision and machine learning models for this task and also providing a user-friendly interface, also easily understandable system that can be understandable to the layman.

  Objectives of the System

our system has the following objectives:

·         Predict air quality

·         In Cheapest way

·         Heath precautions

·         Free for all 

·         Simple layout

·         AQI focuses on health effects

Significance of the System

The citizens play a crucial role in improving the air quality of a nation. The citizens should be well-informed about the local and global air pollution problems and the ways to mitigate them.

Think of the AQI as a yardstick that runs from 0 to 500. The higher the AQI value, the greater the level of air pollution and the greater the health concern. For example, an AQI value of 50 or below represents good air quality, while an AQI value over 300 represents hazardous air quality.

Product Scope

This could be done with pollution sensors — although they can be expensive to deploy at scale. Our goal was to design a reliable and inexpensive air quality estimation solution, accessible to everyone with a smartphone. Our goal is to develop an Android-based mobile application to provide local, real-time air quality estimation using smartphone camera images. 

  Product Functionality

Following is the product functionality of our app:

• The Mobile Application. This is used to capture images and predict AQI levels. The application processes images on-device.
• TensorFlow Lite is used to power on-device inference, in a small binary size (which is important for download speed, when bandwidth is limited) for the trained machine learning model.
• Firebase. Parameters extracted from the images (described below) are sent to Firebase. Whenever a new user uses the app, a unique ID is created for them. This can be used later to customize the machine-learning model for different geo-locations.
• We train our models here, using these parameters and the PM values from the geo-location.
• ML Kit. Trained models are hosted on ML Kit, and automatically onto the device, then run with TensorFlow Lite.