© 2019. Samia Sultana. This is a research/review paper, distributed under the terms of the Creative Commons Attribution-
Noncommercial 3.0 Unported License http://creativecommons.org/licenses/by-nc/3.0/), permitting all non commercial use,
distribution, and reproduction in any medium, provided the original work is properly cited.
Volume 19 Issue 1 Version 1.0 Year 2019
Type: Double Blind Peer Reviewed International Research Journal
Publisher: Global Journals
A Comparative Analysis of Air Pollution Detection Technique
using Image Processing, Machine Learning and Deep Learning
Approach
By Samia Sultana
Stamford University Bangladesh
Abstract- Air pollution is one of the principal environmental issues for the industrial emission and
infection of the atmosphere which is caused by the climatic and traffic elements, burning of fossil
fuels, etc. For the past several years, various methods and models have been discovered to
detect the pollution of the air. In this paper, among all of those, three mechanisms have been
focused, which are image processing approach, machine learning, and deep learning technique.
A comparative study has developed among these three methods to detect the pollutant of the air
in the account of time, cost and efficiency so that different scenario and system can choose the
best method according to their need. The objective of this paper is to assimilate the procedure of
these methods in brief and utilize this study to estimate the best solution for the corresponding
requirement of any particular circumstances.
Keywords: air pollution, image processing, machine learning, deep learning.
GJCST-A Classification: I.2.m
AComparativeAnalysisofAirPollutionDetectionTechniqueusingImageProcessingMachineLearningandDeepLearningApproach
Global Journal of Computer Science and Technology: A
Hardware & Computation
Strictly as per the compliance and regulations of:
Online ISSN: 0975-4172 & Print ISSN: 0975-4350