Estimating Absorption Coefficient from a Single Image via Entropy Minimization
Junya Katahira (Kyushu Institute of Technology), Ryo Kawahara (Kyushu Institute of Technology), Takahiro Okabe (Kyushu Institute of Technology)*
The 34th British Machine Vision Conference
Abstract
When light passes through a liquid, its energy is attenuated due to absorption. The attenuation depends on both the spectral absorption coefficient of a liquid and the optical path length of light, and is described by the Lambert-Beer law. The spectral absorption coefficients of liquids are often unknown in real-world applications and to be measured/estimated in advance, because they depend not only on liquid media themselves but also on dissolved materials. In this paper, we propose a method for estimating the three-band (RGB) absorption coefficient of a liquid only from a single color image of an under-liquid scene taken from the outside of the liquid in a passive and non-contact manner. Specifically, our proposed method investigates the observed colors in the log of chromaticity band-ratio space, and estimates the absorption coefficient up to a certain ambiguity via entropy minimization. Moreover, we reveal the effects of the ambiguity on the applications to under-liquid image/scene analysis. We conducted a number of experiments using real images, and confirmed that our method works well and is useful for under-liquid shape recovery, absorption removal, and reflectance recovery.Video
Citation
@inproceedings{Katahira_2023_BMVC,
author = {Junya Katahira and Ryo Kawahara and Takahiro Okabe},
title = {Estimating Absorption Coefficient from a Single Image via Entropy Minimization},
booktitle = {34th British Machine Vision Conference 2023, {BMVC} 2023, Aberdeen, UK, November 20-24, 2023},
publisher = {BMVA},
year = {2023},
url = {https://papers.bmvc2023.org/0674.pdf}
}
Copyright © 2023
The British Machine Vision Conference is organised by