AQE-Net: A Deep Learning Model for Estimating Air Quality of Karachi City from Mobile Images

Air quality has a significant influence on the environment and health. Instruments that efficiently inexpensively detect air could be extremely valuable in detecting indices. This study presents robust deep learning model named AQE-Net, for estimating from mobile images. The algorithm extracts features patterns scene photographs collected by camera device then classifies images according to index (AQI) levels. Additionally, an dataset (KARACHI-AQI) of high-quality outdoor was constructed enable model’s training assessment performance. sample data were monitoring station Karachi City, Pakistan, comprising 1001 hourly datasets, including photographs, PM2.5 levels, AQI. compares examines traditional machine algorithms, e.g., support vector (SVM), models, such as VGG16, InceptionV3, AQE-Net KHI-AQI dataset. experimental findings demonstrate that, compared other achieved more accurate categorization quality. 70.1% accuracy, while SVM, InceptionV3 56.2% 59.2% respectively. In addition, MSE, MAE, MAPE values calculated our (1.278, 0.542, 0.310), which indicates remarkable efficacy approach. suggested method shows promise fast way estimate classify pollutants only captured photographs. flexible scalable potential fill gaps gathered costly devices around world.