Skip to main content
ARS Home » Plains Area » Fort Collins, Colorado » Center for Agricultural Resources Research » Water Management and Systems Research » Research » Publications at this Location » Publication #371115

Research Project: Improving the Sustainability of Irrigated Farming Systems in Semi-Arid Regions

Location: Water Management and Systems Research

Title: Estimating above-ground biomass of maize using features derived from UAV-based RGB imagery

Author
item NIU, YAXIAO - Northwest A&f University
item ZHANG, LIYUAN - Northwest A&f University
item Zhang, Huihui
item HAN, WENTING - Northwest A&f University
item PENG, XINGSHUO - Northwest A&f University

Submitted to: Remote Sensing Reviews
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/24/2019
Publication Date: 5/28/2019
Citation: Niu, Y., Zhang, L., Zhang, H., Han, W., Peng, X. 2019. Estimating above-ground biomass of maize using features derived from UAV-based RGB imagery. Remote Sensing Reviews. 11(11):1261. https://doi.org/10.3390/rs11111261.
DOI: https://doi.org/10.3390/rs11111261

Interpretive Summary: In the study, we used a consumer-grade UAV to acquire RGB imagery to estimate maize above-ground biomass (AGB). A few vegetation indices (combination of R, G, B band) and plant height were derived from these RGB imagery and used to developed models for AGB estimation. The results showed that plant height directly derived from RGB point clouds had a high correlation with ground-truth plant height measurements. The AGB exponential regression models based on plant height alone had higher correlations for both fresh and dry AGB than the linear regression model. The results confirm that vegetation index derived from RGB imagery had great potential to estimate maize AGB. And when using multivariable linear regression based on vegetation indices, a higher correlation was obtained. There was no significant improvement of the estimation performance when plant height was added into the multivariable linear regression model.

Technical Abstract: The rapid, accurate, and economical estimation of crop above-ground biomass at the farm scale is crucial for precision agriculture management. An unmanned aerial vehicle (UAV) remote sensing system has a great application potential with the ability of obtaining remote-sensing imagery with high temporal-spatial resolution. To verify the application potential of consumer-grade UAV RGB imagery in estimating maize above-ground biomass, in this study, vegetation indices and plant height derived from consumer-grade UAV RGB imagery were adopted. To obtain a more accurate observation, plant height was directly derived from UAV RGB point clouds. To search the optimal estimation method, the estimation performance of models based on vegetation indices alone, based on plant height alone, and based on both vegetation indices and canopy height were compared. The results showed that plant height directly derived from UAV RGB point clouds had a high correlation with ground-truth data with a R2 value of 0.90 and a RMSE value of 0.12 m.The above-ground biomass exponential regression models based on plant height alone had higher correlations for both fresh and dry above-ground biomass with R2 values of 0.77 and 0.76, respectively, compared to the linear regression model (both R2 values were 0.59). The vegetation index derived from UAV RGB imagery had great potential to estimate maize above-ground biomass with R2 values ranged from 0.63 to 0.73. And when estimating above-ground biomass of maize by using multivariable linear regression based on vegetation indices, a higher correlation was obtained with a R2 value of 0.82. There was no significant improvement of the estimation performance when plant height derived from UAV RGB imagery was added into the multivariable linear regression model based on vegetation indices. When estimating crop above-ground biomass based on UAV RGB remote-sensing system alone, looking for optimized vegetation indices and establishing estimation models with high performance based on advanced algorithms (e.g. machine learning technology) may be the correct way.