Paper Key : IRJ************885
Author: Shubhangi Bodke,Eshwari Deepak Bhorekar,Chndrakant Sanjay Thakar
Date Published: 03 Apr 2024
Abstract
Air pollution primarily originates fromsubstances that are directly emitted from natural oranthropogenic processes, such as carbon monoxide (CO) gasemitted in vehicle exhaust or sulfur dioxide (SO2) releasedfrom factories. However, a major air pollution problem isparticulate matter (PM), which is an adverse effect ofwildfires and open burning. Application tools for airpollution monitoring in risk areas using real-time monitoringwith drones have emerged. A new air quality index (AQI) formonitoring and display, such as three-dimensional (3D)mapping based on data assessment, is essential for timelyenvironmental surveying. The objective of this paper is topresent a 3D AQI mapping data assessment using a hybridmodel based on a machine-learning method for drone realtime air pollution monitoring (Dr-TAPM). Dr-TAPM wasdesigned by equipping drones with multi-environmentalsensors for carbon monoxide (CO), ozone (O3), nitrogendioxide (NO2), particulate matter (PM2.5,10), and sulfurdioxide (SO2), with data pre- and post-processing with thehybrid model. The hybrid model for data assessment wasproposed using backpropagation neural network (BPNN) andconvolutional neural network (CNN) algorithms.Experimentally, we considered a case study detecting smokeemissions from an open burning scenario. As a result,PM2.5,10 and CO were detected as air pollutants from openburning. 3D AQI map locations were shown and thevalidation learning rates were apparent, as the accuracy ofpredicted AQI data assessment was 98%.Keywords: 3D AQI mapping; BPNNCNN model; smokedetection; open burning; Dr-TAPM; data assessment