Machine-Learning Approach to Non-Destructive Biomass and Relative Growth Rate Estimation in Aeroponic Cultivation
(2023) In Agriculture 13(4).- Abstract
- We train and compare the performance of two machine learning methods, a multi-variate regression network and a ResNet-50-based neural network, to learn and forecast plant biomass as well as the relative growth rate from a short sequence of temporal images from plants in aeroponic cultivation. The training dataset consists of images of 57 plants taken from two different angles every hour during a 5-day period. The results show that images taken from a top-down perspective produce better results for the multi-variate regression network, while images taken from the side are better for the ResNet-50 neural network. In addition, using images from both cameras improves the biomass estimates from the ResNet-50 network, but not those from the... (More)
- We train and compare the performance of two machine learning methods, a multi-variate regression network and a ResNet-50-based neural network, to learn and forecast plant biomass as well as the relative growth rate from a short sequence of temporal images from plants in aeroponic cultivation. The training dataset consists of images of 57 plants taken from two different angles every hour during a 5-day period. The results show that images taken from a top-down perspective produce better results for the multi-variate regression network, while images taken from the side are better for the ResNet-50 neural network. In addition, using images from both cameras improves the biomass estimates from the ResNet-50 network, but not those from the multi-variatemultivariate regression. However, all relative growth rate estimates were improved by using images from both cameras. We found that the best biomass estimates are produced from the multi-variate regression model trained on top camera images using a moving average filter resulting in a root mean square error of 0.0466 g. The best relative growth rate estimates were produced from the ResNet-50 network training on images from both cameras resulting in a root mean square error of 0.1767 g/(g·day). (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/3c5b5932-3d2a-494c-9d52-86e745b1b2e6
- author
- Åström, Oskar ; Hedlund, Henrik and Sopasakis, Alexandros LU
- organization
- publishing date
- 2023-03-29
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- machine learning, aeroponics, hydroculture, neural network, regression, biomass, fresh weigh, relative growth rate, image analysis
- in
- Agriculture
- volume
- 13
- issue
- 4
- article number
- 801
- pages
- 13 pages
- publisher
- MDPI AG
- external identifiers
-
- scopus:85153722794
- scopus:85153722794
- ISSN
- 2077-0472
- DOI
- 10.3390/agriculture13040801
- language
- English
- LU publication?
- yes
- id
- 3c5b5932-3d2a-494c-9d52-86e745b1b2e6
- date added to LUP
- 2023-04-02 14:17:11
- date last changed
- 2023-05-16 09:17:12
@article{3c5b5932-3d2a-494c-9d52-86e745b1b2e6, abstract = {{We train and compare the performance of two machine learning methods, a multi-variate regression network and a ResNet-50-based neural network, to learn and forecast plant biomass as well as the relative growth rate from a short sequence of temporal images from plants in aeroponic cultivation. The training dataset consists of images of 57 plants taken from two different angles every hour during a 5-day period. The results show that images taken from a top-down perspective produce better results for the multi-variate regression network, while images taken from the side are better for the ResNet-50 neural network. In addition, using images from both cameras improves the biomass estimates from the ResNet-50 network, but not those from the multi-variatemultivariate regression. However, all relative growth rate estimates were improved by using images from both cameras. We found that the best biomass estimates are produced from the multi-variate regression model trained on top camera images using a moving average filter resulting in a root mean square error of 0.0466 g. The best relative growth rate estimates were produced from the ResNet-50 network training on images from both cameras resulting in a root mean square error of 0.1767 g/(g·day).}}, author = {{Åström, Oskar and Hedlund, Henrik and Sopasakis, Alexandros}}, issn = {{2077-0472}}, keywords = {{machine learning; aeroponics; hydroculture; neural network; regression; biomass; fresh weigh; relative growth rate; image analysis}}, language = {{eng}}, month = {{03}}, number = {{4}}, publisher = {{MDPI AG}}, series = {{Agriculture}}, title = {{Machine-Learning Approach to Non-Destructive Biomass and Relative Growth Rate Estimation in Aeroponic Cultivation}}, url = {{http://dx.doi.org/10.3390/agriculture13040801}}, doi = {{10.3390/agriculture13040801}}, volume = {{13}}, year = {{2023}}, }