Search

Ratoon rice production has been an emerging cropping system to increase food quality and productivity worldwide. Efficient monitoring of ratoon rice aboveground biomass (AGB) over large areas is valuable for precision agriculture, as AGB is closely related to crop grain yield and quality. Unmanned aerial vehicle (UAV) remote sensing has opened an unprecedented opportunity to efficiently monitor crop AGB instead of labor-intensive ground measurements. Vegetation indices (VIs)-based approach for estimating AGB is easily affected by background materials and often suffers saturation problems at high AGB levels. Although the combined use of UAV-collected structural and spectral features can alleviate these problems to some extent, uncertainties on the AGB estimation still exist for crop with significant difference in canopy architecture and AGB composition throughout the growing season (e.g., ratoon rice). There is a hypothesis that the combination of spectral, textural and structural features can improve ratoon rice AGB estimations across different developmental stages. Therefore, the utility of UAV-based spectral, textural and structural features were quantified in estimating ratoon rice AGB at field level with contrasting agronomic treatments (i.e., nitrogen fertilizer and stubble height), in which multiple linear regression (MLR) and gaussian process regression (GPR) methods were applied and compared. Results showed that (1) each feature had its own respective limitation: specifically, spectral and textural features exhibited insufficient sensitivity to AGB variability of remaining stubbles or stem at early stage and suffered saturation problem at grain filling stage; structural features were difficult to detect the emergence of panicles from panicle initiation to heading stages; (2) the combination of three types of features can complement each other and achieved the highest accuracy using GPR method: the combination of spectral, structural and textural features achieved the best estimation accuracy for estimating ratoon rice AGB with an R2 of 0.94 and RMSE of 81.4 g m−2 across different developmental stages, which significantly improved the model performance compared to the combination of spectral and textural features (R2 = 0.56, RMSE = 170.2 g m−2) and the combination of spectral and structural features (R2 = 0.86, RMSE = 138.8 g m−2). In summary, this study provides a novel approach for efficiently estimating ratoon rice AGB at field level, which is critical for timely decision making (e.g., determine when and where to apply fertilizer or pesticide) in precision agriculture. [ABSTRACT FROM AUTHOR]

Southern corn leaf blight (SCLB) seriously threatens corn production. The timely and accurate monitoring of SCLB conditions (e.g., detection during the asymptomatic stage and severity classification during the symptomatic stage) is valuable for precision agriculture, because the application of pesticides depends on disease conditions. Compared with time-consuming and laborious field surveys, spectroscopy is a promising tool for plant disease monitoring. The unique advantages of combining multiple spectral enhancement features for monitoring rice and wheat diseases have been recognized. However, physiological and biochemical differences between maize leaves and rice and wheat leaves, along with the specific spectral response of SCLB, are likely to affect the performance of combining multiple spectral enhancement features. In addition, similar previous studies have not combined spectral slope features, i.e., first-order spectral derivatives (FSDs), with spectral bands (SBs) and spectral indices (SIs) and wavelet features (WFs) to improve plant disease detection. Thus, the performance of a method that combines FSDs, WFs, SBs, and SIs for SCLB asymptomatic detection, symptomatic detection, and symptomatic severity classification should be evaluated further. Here, the utility of combining SBs, SIs, WFs, and FSDs was quantified and evaluated in the asymptomatic detection, symptomatic detection, and symptomatic severity classification of SCLB. Various forms of spectral enhancement features that were sensitive to SCLB infection from the asymptomatic stage to the severe stage were first identified and combined using the RELIEF-F and sequential floating forward selection algorithms on the basis of two independent inoculation experiments. Finally, SCLB asymptomatic detection, symptomatic detection, and symptomatic severity classification models were developed and evaluated using the support vector machine algorithm. Results showed that combining FSDs with SBs, SIs, and WFs achieved the best performance in SCLB spectroscopic monitoring. (1) SCLB asymptomatic detection and symptomatic detection were moderately improved, i.e., overall accuracy (OA) and macro F1 (MF1) improved by ~ 1% to 2%. The OA of SCLB asymptomatic detection was 87.1% with an MF1 of 0.87, and that of symptomatic detection was 93.1% with an MF1 of 0.93. (2) SCLB symptomatic severity classification was significantly improved, i.e., OA and MF1 improved by ~ 7%. The OA of severity classification was 81.8% with am MF1 of 0.82. This study demonstrated that the complementary relationships among SBs, SIs, WFs, and FSDs could effectively improve SCLB spectroscopic monitoring. The proposed method provides a novel tool for large-scale SCLB spectroscopic monitoring. It has broad implications for assisting management decisions (i.e., when and where to apply pesticides and how much to apply) in precision agriculture. [ABSTRACT FROM AUTHOR]