Estimating Rapeseed Biomass Using UAV Imagery
Author Information
Author(s): Sun Chuanliang, Zhang Weixin, Zhao Genping, Wu Qian, Liang Wanjie, Ren Ni, Cao Hongxin, Zou Lidong
Primary Institution: Jiangsu Academy of Agricultural Sciences
Hypothesis
This study aims to explore the potential of using optical and phenotypic metrics to estimate rapeseed aboveground biomass (AGB).
Conclusion
The study found that integrating hyperspectral and RGB data with advanced artificial intelligence models, particularly deep neural networks, provides an effective approach for estimating rapeseed AGB.
Supporting Evidence
- The DNN model achieved the highest R² values of 0.878 and 0.864 for the training and test datasets, respectively.
- The optimal model combined DNN with VI, TF, and SF features.
- AGB estimation accuracy peaked during the bolting stage.
- Integrating multiple feature types—VI, TF, and SF—offers a more comprehensive approach to estimating AGB.
- Texture and structural parameters significantly improve crop monitoring accuracy.
Takeaway
Researchers used drones to take pictures of rapeseed plants to figure out how much they weigh above the ground, which helps farmers know how healthy their crops are.
Methodology
The study used UAV hyperspectral and RGB imagery to extract vegetation indices, texture features, and structural features, and employed deep neural networks, random forest, and support vector regression to estimate AGB.
Limitations
The study's findings may be limited by the accuracy of UAV data acquisition and the potential mixed pixel problem due to resolution constraints.
Statistical Information
P-Value
p<0.01
Statistical Significance
p<0.01
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website