20200512_Lagoon water quality monitoring based on digital image analysis and machine learning estimators

 

1. Title, Journal and Authors

Title: Lagoon water quality monitoring based on digital image analysis and machine learning estimators

Journal: Water research

Authors: Yuanhong Lia,b,c, Xiao Wangc, Zuoxi Zhaoa,b, Sunghwa Hanc, Zong Liuc,*

aDepartment of Engineering, South China Agricultural University

bSouthern Key Laboratory of Agricultural Equipment Machinery, South China Agricultural University

cDepartment of Biological and Agricultural Engineering, Texas A&M University

 

2. Summary

This paper was written to analyze the relationship between the water qualities of the lagoon which treats animal wastewater and their effluent images. Images were augmented by centrifuging each water sample to make a wide range of water qualities. Machine learning methods such as Ridge regression were introduced to investigate the relation between the RGB channel analyses of the water images and their water qualities.

 

The increase in centrifugation time was equivalent to reducing the chemical elements in the sample. Some parameters had insignificant relations, but total solids and total coliform showed significant relations. With this method, water images would be able to predict the water quality.

 

3. Application to research

Most of computer vision analysis were conducted using CNN algorithms. But in this paper, RGB channel analysis were introduced. I can consider this way as one of the image analysis methods.

 

4. Contact

Dae Seong Jeong / Integrated Ph.D. program

 

Environmental Systems Engineering Lab.

School of Environmental Science & Engineering

Gwangju Institute of Science and Technology

1 Oryong-dong Buk-gu Gwangju, 500-712, Korea

 

Phone : +82-10-2003-7860

E-mail : jeongds92@gist.ac.kr