International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 09 | Sep 2024
p-ISSN: 2395-0072
www.irjet.net
Assessing Water Quality through Image Recognition and Machine Learning Mr. Anirudh Vaidison1, Mentor: Dr. Liam Yasin2 1Student, Oberoi International School, Maharashtra, India 2Department of Design Engineering at Imperial College London, London, England
---------------------------------------------------------------------***--------------------------------------------------------------------1.1 PROBLEM STATEMENT
Abstract - This research project explores the application
of machine learning in assessing water quality through image recognition. The study leverages a diverse dataset of water samples collected from various sources across Mumbai, encompassing ponds, lakes, water outlets, and sewage points. Multiple parameters, including pH, conductivity, turbidity, and dissolved oxygen content, are examined in relation to the resulting HEX colour code, which serves as a visual representation of water quality. The central hypothesis posits that employing machine learning algorithms can reliably predict water safety based on these environmental parameters through image recognition of the colour gradient of the water samples. A systematic approach to data collection, standardisation, and logistic regression modeling has been employed. Results demonstrate the effectiveness of the logistic regression model in predicting water safety with an 85% accuracy rate, highlighting its potential for real-time water quality monitoring and risk assessment. Nevertheless, this study recognises its limitations and the need for further research to refine the model's predictive accuracy and address variations across different geographical regions and water sources. This research aims to contribute to the development of innovative approaches for water quality assessment and therefore, environmental preservation.
While water quality using the aforementioned parameters (pH, conductivity, turbidity, and dissolved oxygen) has been tested and analysed in the past, this research paper looks into the feasibility of using image recognition and machine learning as a way of predicting water quality - in essence using technology to look at a very widely prevalent, yet fundamental community problem. The fusion of machine learning with water quality assessment has immense potential to make water safety evaluations more widely accessible.
2. BACKGROUND INFORMATION The World Wildlife Fund (WWF) defines water pollution as “toxic substances entering water bodies such as lakes, rivers, oceans and so on, getting dissolved in them” (“Water Pollution”).
2.1 FACTORS THAT AFFECT WATER POLLUTION The most common reason of poor water quality is a consequence of human activity and consumption. In addition to irresponsible disposal of industrial effluence, there are numerous other factors that contribute to widespread contamination of water.
Key Words: Machine learning, water sources, pH, conductivity, turbidity, dissolved oxygen, HEX colour code, logistic regression
1. Global Warming: One of the most predominant factors that results in water pollution is global warming. Rising global temperatures caused by increased cardon dioxide emissions heat the water, reducing its oxygen content.
1.INTRODUCTION Clean water and adequate sanitation are vital for human existence, environmental sustainability, and economic development. However, the irresponsible disposal of industrial effluence has emerged as a significant threat to water quality in major world cities including Mumbai, a particularly urbanised region. This research paper aims to investigate how pollutants—measured through various water quality parameters such as pH, turbidity, dissolved oxygen and conductivity—affect the colour gradient of various water bodies and therefore, the overall water quality.
© 2024, IRJET
|
Impact Factor value: 8.315
2. Deforestation Felling forests can exhaust water resources and generate organic detritus which allows harmful microorganisms to develop and enter water ways. This leads to increased sediment and nutrient runoff into water bodies, further degrading water quality and allowing toxic substances to dissolve in the water.
|
ISO 9001:2008 Certified Journal
|
Page 292