AQI Predictor

Abstract

Our project, AQI Predictor, is an application to predict the quality of Air aka Air Quality Index (AQI) Level using images. The user can input the images by capturing directly from the smartphone’s camera or already captured ones from the gallery. The app uses a trained model that’s hosted on local server at the backend. The trained model is a Convolutional Neural Network that uses machine learning techniques to help us in predicting the Air Quality level (AQL) correctly. The model is trained on a relevant dataset in order to achieve the accuracy. Along with the app, a website will be created that will be synchronized with the app and the data will be collected. The data then will be used to generate a map of the areas the Air quality level has been calculated for. This will help people in deciding which areas have what kind of air quality. The quality of air will be predicted by the application providing us Air Quality Level which will further tell us more details such as if the air is harmful or good and how necessary precautions can be taken. These factors and results will help us and the authorities including the Government and relevant governmental and other bodies decide certain measures to make the air quality better.