An Advanced Calibration Method for Image Analysis in Laboratory-Scale Seawater Intrusion Problems

G. Robinson, S. Moutari, A. A. Ahmed, G. A. Hamill

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Image analysis is a useful tool for visualising flow through laboratory-scale aquifers but existing methods of converting image light intensity to concentration can be labour intensive and time consuming. The new approach proposed in this study utilises the Random Forest machine learning technique to build a calibration model to replace the requirement for unique calibrations of each test aquifer. Calibration images from a previous experimental study were used to train the Random Forest model and the output was compared to the results from ahigh resolution pixel-wise methodology. The Random Forest model provided a trade-off inaccuracy with increased efficiency and reduced sensitivity to image desynchronisation when compared to the pixel-wise method. The reduced accuracy was attributed in part to non-linear lighting distribution across the sandbox, which could be corrected by orientating the backlights effectively. Time savings of around 35% were achieved for this experimental study and this is expected to increase for larger scale studies. The new calibration approach exhibits some promising features in terms of its robustness to experiment
LanguageEnglish
Number of pages16
JournalWater Resources Management
Early online date11 Apr 2018
DOIs
Publication statusEarly online date - 11 Apr 2018

Fingerprint

Salt water intrusion
Calibration
seawater
calibration
Aquifers
pixel
experimental study
Pixels
aquifer
light intensity
image analysis
Image analysis
trade-off
train
Learning systems
savings
labor
Lighting
Personnel
methodology

Cite this

@article{3b6325887f714819a035458782ac22ef,
title = "An Advanced Calibration Method for Image Analysis in Laboratory-Scale Seawater Intrusion Problems",
abstract = "Image analysis is a useful tool for visualising flow through laboratory-scale aquifers but existing methods of converting image light intensity to concentration can be labour intensive and time consuming. The new approach proposed in this study utilises the Random Forest machine learning technique to build a calibration model to replace the requirement for unique calibrations of each test aquifer. Calibration images from a previous experimental study were used to train the Random Forest model and the output was compared to the results from ahigh resolution pixel-wise methodology. The Random Forest model provided a trade-off inaccuracy with increased efficiency and reduced sensitivity to image desynchronisation when compared to the pixel-wise method. The reduced accuracy was attributed in part to non-linear lighting distribution across the sandbox, which could be corrected by orientating the backlights effectively. Time savings of around 35{\%} were achieved for this experimental study and this is expected to increase for larger scale studies. The new calibration approach exhibits some promising features in terms of its robustness to experiment",
author = "G. Robinson and S. Moutari and Ahmed, {A. A.} and Hamill, {G. A.}",
year = "2018",
month = "4",
day = "11",
doi = "10.1007/s11269-018-1977-6",
language = "English",
journal = "Water Resources Management",
issn = "0920-4741",
publisher = "Springer Netherlands",

}

TY - JOUR

T1 - An Advanced Calibration Method for Image Analysis in Laboratory-Scale Seawater Intrusion Problems

AU - Robinson, G.

AU - Moutari, S.

AU - Ahmed, A. A.

AU - Hamill, G. A.

PY - 2018/4/11

Y1 - 2018/4/11

N2 - Image analysis is a useful tool for visualising flow through laboratory-scale aquifers but existing methods of converting image light intensity to concentration can be labour intensive and time consuming. The new approach proposed in this study utilises the Random Forest machine learning technique to build a calibration model to replace the requirement for unique calibrations of each test aquifer. Calibration images from a previous experimental study were used to train the Random Forest model and the output was compared to the results from ahigh resolution pixel-wise methodology. The Random Forest model provided a trade-off inaccuracy with increased efficiency and reduced sensitivity to image desynchronisation when compared to the pixel-wise method. The reduced accuracy was attributed in part to non-linear lighting distribution across the sandbox, which could be corrected by orientating the backlights effectively. Time savings of around 35% were achieved for this experimental study and this is expected to increase for larger scale studies. The new calibration approach exhibits some promising features in terms of its robustness to experiment

AB - Image analysis is a useful tool for visualising flow through laboratory-scale aquifers but existing methods of converting image light intensity to concentration can be labour intensive and time consuming. The new approach proposed in this study utilises the Random Forest machine learning technique to build a calibration model to replace the requirement for unique calibrations of each test aquifer. Calibration images from a previous experimental study were used to train the Random Forest model and the output was compared to the results from ahigh resolution pixel-wise methodology. The Random Forest model provided a trade-off inaccuracy with increased efficiency and reduced sensitivity to image desynchronisation when compared to the pixel-wise method. The reduced accuracy was attributed in part to non-linear lighting distribution across the sandbox, which could be corrected by orientating the backlights effectively. Time savings of around 35% were achieved for this experimental study and this is expected to increase for larger scale studies. The new calibration approach exhibits some promising features in terms of its robustness to experiment

U2 - 10.1007/s11269-018-1977-6

DO - 10.1007/s11269-018-1977-6

M3 - Article

JO - Water Resources Management

T2 - Water Resources Management

JF - Water Resources Management

SN - 0920-4741

ER -