scholarly journals Novel Feature-Extraction Methods for the Estimation of Above-Ground Biomass in Rice Crops

Sensors ◽  
2021 ◽  
Vol 21 (13) ◽  
pp. 4369
Author(s):  
David Alejandro Jimenez-Sierra ◽  
Edgar Steven Correa ◽  
Hernán Darío Benítez-Restrepo ◽  
Francisco Carlos Calderon ◽  
Ivan Fernando Mondragon ◽  
...  
Keyword(s):  
Feature Extraction ◽  
Vegetation Index ◽  
Ground Truth ◽  
Gaussian Mixture ◽  
Extraction Methods ◽  
Temporal Variations ◽  
Biomass Estimation ◽  
Growth Stages ◽  
Above Ground Biomass ◽  
Ground Biomass

Traditional methods to measure spatio-temporal variations in above-ground biomass dynamics (AGBD) predominantly rely on the extraction of several vegetation-index features highly associated with AGBD variations through the phenological crop cycle. This work presents a comprehensive comparison between two different approaches for feature extraction for non-destructive biomass estimation using aerial multispectral imagery. The first method is called GFKuts, an approach that optimally labels the plot canopy based on a Gaussian mixture model, a Montecarlo-based K-means, and a guided image filtering for the extraction of canopy vegetation indices associated with biomass yield. The second method is based on a Graph-Based Data Fusion (GBF) approach that does not depend on calculating vegetation-index image reflectances. Both methods are experimentally tested and compared through rice growth stages: vegetative, reproductive, and ripening. Biomass estimation correlations are calculated and compared against an assembled ground-truth biomass measurements taken by destructive sampling. The proposed GBF-Sm-Bs approach outperformed competing methods by obtaining biomass estimation correlation of 0.995 with R2=0.991 and RMSE=45.358 g. This result increases the precision in the biomass estimation by around 62.43% compared to previous works.

scholarly journals Estimating Above-Ground Biomass of Potato Using Random Forest and Optimized Hyperspectral Indices

2021 ◽  
Vol 13 (12) ◽  
pp. 2339
Author(s):  
Haibo Yang ◽  
Fei Li ◽  
Wei Wang ◽  
Kang Yu
Keyword(s):  
Random Forest ◽  
Vegetation Index ◽  
Tuber Formation ◽  
Growth Stages ◽  
Spectral Indices ◽  
Above Ground Biomass ◽  
Full Spectrum ◽  
Ground Biomass ◽  
Tuber Bulking ◽  
Input Variables

Spectral indices rarely show consistency in estimating crop traits across growth stages; thus, it is critical to simultaneously evaluate a group of spectral variables and select the most informative spectral indices for retrieving crop traits. The objective of this study was to explore the optimal spectral predictors for above-ground biomass (AGB) by applying Random Forest (RF) on three types of spectral predictors: the full spectrum, published spectral indices (Pub-SIs), and optimized spectral indices (Opt-SIs). Canopy hyperspectral reflectance of potato plants, treated with seven nitrogen (N) rates, was obtained during the tuber formation and tuber bulking from 2015 to 2016. Twelve Pub-SIs were selected, and their spectral bands were optimized using band optimization algorithms. Results showed that the Opt-SIs were the best input variables of RF models. Compared to the best empirical model based on Opt-SIs, the Opt-SIs based RF model improved the prediction of AGB, with R2 increased by 6%, 10%, and 16% at the tuber formation, tuber bulking, and for across the two growth stages, respectively. The Opt-SIs can significantly reduce the number of input variables. The optimized Blue nitrogen index (Opt-BNI) and Modified red-edge normalized difference vegetation index (Opt-mND705) combined with an RF model showed the best performance in estimating potato AGB at the tuber formation stage (R2 = 0.88). In the tuber bulking stage, only using optimized Nitrogen planar domain index (Opt-NPDI) as the input variable of the RF model produced satisfactory accuracy in training and testing datasets, with the R2, RMSE, and RE being 0.92, 208.6 kg/ha, and 10.3%, respectively. The Opt-BNI and Double-peak nitrogen index (Opt-NDDA) coupling with an RF model explained 86% of the variations in potato AGB, with the lowest RMSE (262.9 kg/ha) and RE (14.8%) across two growth stages. This study shows that combining the Opt-SIs and RF can greatly enhance the prediction accuracy for crop AGB while significantly reduces collinearity and redundancies of spectral data.

scholarly journals Assessment of Forest above Ground Biomass Estimation Using Multi-Temporal C-band Sentinel-1 and Polarimetric L-band PALSAR-2 Data

2018 ◽  
Vol 10 (9) ◽  
pp. 1424 ◽  
Author(s):  
Xiaodong Huang ◽  
Beth Ziniti ◽  
Nathan Torbick ◽  
Mark Ducey
Keyword(s):  
Optical Sensors ◽  
Forest Canopy ◽  
Alpha Angle ◽  
Ground Truth ◽  
Biomass Estimation ◽  
Above Ground Biomass ◽  
Ground Truth Data ◽  
Ground Biomass ◽  
Multi Temporal ◽  
L Band

Synthetic Aperture Radar (SAR), as an active sensor transmitting long wavelengths, has the advantages of working day and night and without rain or cloud disturbance. It is further able to sense the geometric structure of forests more than passive optical sensors, making it a valuable tool for mapping forest Above Ground Biomass (AGB). This paper studies the ability of the single- and multi-temporal C-band Sentinel-1 and polarimetric L-band PALSAR-2 data to estimate live AGB based on ground truth data collected in New England, USA in 2017. Comparisons of results using the Simple Water Cloud Model (SWCM) on both VH and VV polarizations show that C-band reaches saturation much faster than the L-band due to its limited forest canopy penetration. The exhaustive search multiple linear regression model over the many polarimetric parameters from PALSAR-2 data shows that the combination of polarimetric parameters could slightly improve the AGB estimation, with an adjusted R2 as high as 0.43 and RMSE of around 70 Mg/ha when decomposed Pv component and Alpha angle are used. Additionally, the single- and multi-temporal C-band Sentinel-1 data are compared, which demonstrates that the multi-temporal Sentinel-1 significantly improves the AGB estimation, but still has a much lower adjusted R2 due to the limitations of the short wavelength. Finally, a site-level comparison between paired control and treatment sites shows that the L-band aligns better with the ground truth than the C-band, showing the high potential of the models to be applied to relative biomass change detection.

Above-ground biomass estimation of Indian tropical forests using X band Pol-InSAR and Random Forest

2021 ◽  
Vol 21 ◽  
pp. 100462
Author(s):  
Sadhana Yadav ◽  
Hitendra Padalia ◽  
Sanjiv K. Sinha ◽  
Ritika Srinet ◽  
Prakash Chauhan
Keyword(s):  
Random Forest ◽  
Tropical Forests ◽  
Biomass Estimation ◽  
Above Ground Biomass ◽  
Ground Biomass ◽  
X Band

Using photography to estimate above-ground biomass of small trees

2020 ◽  
pp. 1-7
Author(s):  
Brandon R. Hays ◽  
Corinna Riginos ◽  
Todd M. Palmer ◽  
Benard C. Gituku ◽  
Jacob R. Goheen
Keyword(s):  
Cost Effective ◽  
Allometric Equation ◽  
Biomass Estimation ◽  
Above Ground Biomass ◽  
Tree Biomass ◽  
East African ◽  
Ground Biomass ◽  
Herbivore Exclusion ◽  
Significant Difference ◽  
Allometric Relation

Abstract Quantifying tree biomass is an important research and management goal across many disciplines. For species that exhibit predictable relationships between structural metrics (e.g. diameter, height, crown breadth) and total weight, allometric calculations produce accurate estimates of above-ground biomass. However, such methods may be insufficient where inter-individual variation is large relative to individual biomass and is itself of interest (for example, variation due to herbivory). In an East African savanna bushland, we analysed photographs of small (<5 m) trees from perpendicular angles and fixed distances to estimate above-ground biomass. Pixel area of trees in photos and diameter were more strongly related to measured, above-ground biomass of destructively sampled trees than biomass estimated using a published allometric relation based on diameter alone (R2 = 0.86 versus R2 = 0.68). When tested on trees in herbivore-exclusion plots versus unfenced (open) plots, our predictive equation based on photos confirmed higher above-ground biomass in the exclusion plots than in unfenced (open) plots (P < 0.001), in contrast to no significant difference based on the allometric equation (P = 0.43). As such, our new technique based on photographs offers an accurate and cost-effective complement to existing methods for tree biomass estimation at small scales with potential application across a wide variety of settings.

Sign in / Sign up

Export Citation Format

Share Document