2021 Volume 35 Pages 32-40
Plant culture experiments in space require non-destructive and automated assessment of growth that is efficient and unattended. Currently, plant growth conditions in spacecraft are often captured by ordinary RGB (red, green blue) cameras. In addition to images in the RGB wavelengths, near-infrared (NIR) images that are not affected by plant pigments, such as chlorophylls, could be used to improve the quality of growth-related information. In this study, we examined the possibility of using NIR to assess the growth status of a model moss plant, Physcomitrium (Physcomitrella) patens. Mosses cultured in multi-well plates were imaged with a hyperspectral (HS) sensor, and the length of gametophores, the number of leaves and gametophores, and the amount of photosynthetic pigments in the moss colonies were measured. Then we used the R, G, B, and NIR (850 nm) bands from the obtained HS images to extract the moss colony areas and evaluate the amount of growth. Our results suggested that spectral indices using R and NIR bands were suitable for assessing the morphology of moss colonies and extracting the colony area. The extracted area can be used to estimate the amount of growth.
