Use of UAS Multispectral Imagery at Different Physiological Stages for Yield Prediction and Input Resource Optimization in Corn

Abstract

Changes in spatial and temporal variability in yield estimation are detectable through plant biophysical characteristics observed at different phenological development stages of corn. A multispectral red-edge sensor mounted on an Unmanned Aerial Systems (UAS) can provide spatial and temporal information with high resolution. Spectral analysis of UAS acquired spatiotemporal images can be used to develop a statistical model to predict yield based on different phenological stages. Identifying critical vegetation indices (VIs) and significant spectral information could lead to increased yield prediction accuracy. The objective of this study was to develop a yield prediction model at specific phenological stages using spectral data obtained from a corn field. The available spectral bands (red, blue, green, near infrared (NIR), and red-edge) were used to analyze 26 different VIs. The spectral information was collected from a cornfield at Mississippi State University using a MicaSense multispectral red-edge sensor, mounted on a UAS. In this research, a new empirical method used to reduce the effects of bare soil pixels in acquired images was introduced. The experimental design was a randomized complete block that consisted of 16 blocks with 12 rows of corn planted in each block. Four treatments of nitrogen (N) including 0, 90, 180, and 270 kg/ha were applied randomly. Random forest was utilized as a feature selection method to choose the best combination of variables for different stages. Multiple linear regression and gradient boosting decision trees were used to develop yield prediction models for each specific phenological stage by utilizing the most effective variables at each stage. At the V3 (3 leaves with visible leaf collar) and V4-5 (4-5 leaves with visible leaf collar) stages, the Optimized Soil Adjusted Vegetation Index (OSAVI) and Simplified Canopy Chlorophyll Content Index (SCCCI) were the single dominant variables in the yield predicting models, respectively. A combination of the Green Atmospherically Resistant Index (GARI), Normalized Difference Red-Edge (NDRE), and green Normalized Difference Vegetation Index (GNDVI) at V6-7, SCCCI, and Soil-Adjusted Vegetation Index (SAVI) at V10,11, and SCCCI, Green Leaf Index (GLI), and Visible Atmospherically Resistant Index (VARIgreen) at tasseling stage (VT) were the best indices for predicting grain yield of corn. The prediction models at V10 and VT had the greatest accuracy with a coefficient of determination of 0.90 and 0.93, respectively. Moreover, the SCCCI as a combined index seemed to be the most proper index for predicting yield at most of the phenological stages. As corn development progressed, the models predicted final grain yield more accurately.

Ground versus aerial canopy reflectance of corn: Red‐edge and non‐red edge vegetation indices

“…This has led to the discussion whether these inconsistencies should be considered as strengths (Lu et al., 2016; Saberioon et al., 2014) or limitations (Augera et al., 2011) of these technologies. Regardless of the platform, it remains important to differentiate the adequacy of using different vegetation indices (Vis) such as structural indices and indices with and without red‐edge to test their validity of use at different growth stages (Barzin et al., 2020).…”

Predicting Table Beet Root Yield with Multispectral UAS Imagery

“…For example, Zhou et al [18] found that indices from multispectral imagery were correlated (R 2 > 0.7) with rice grain yield at various growth stages. Models using data from different epochs may rely on different vegetation indices and reflectance bands as primary inputs, as found in similar modeling studies of corn [16] and sugar beet [19] yield. A multitemporal approach can also allow for the determination of an optimal timing and growth stage for imagery collection [18,20].…”

“…We thus opted for a relatively simple VI threshold to identify vegetation pixels in each plot. Similar to Barzin et al [16], we determined that while a Normalized Difference Vegetation Index (NDVI) threshold is commonly used for soil masking [37,38], it is not capable of producing a consistent result amongst the imagery from varying growth stages using a single threshold value. We used the triangular vegetation index (TVI), because the TVI of the canopy exhibited a limited variation over the cropping season compared to NDVI.…”

“…UAS and airborne multispectral imagery and vegetation indices derived from combinations of two or more select bands, to enhance the contribution of vegetation properties, have been used for yield prediction in a variety of crop scenarios, including rice [13,14], maize/corn [15,16] and wheat [17]. For example, Zhou et al [18] found that indices from multispectral imagery were correlated (R 2 > 0.7) with rice grain yield at various growth stages.…”

Comparative Analysis of Detectors and Feature Descriptors for Multispectral Image Matching in Rice Crops

“…Smart and precision farming can improve crop yield and quality with the ability to predict and prevent diseases, providing flexible and efficient solutions, using unmanned vehicles, drones, and sensors that allow constant monitoring of the crop [ 1 ]. Thus, spatio-temporal images acquired with drones can be used to develop statistical models to predict crop yields at different phenological stages [ 2 ]. The implementation of sensors to acquire information about the state of crops for monitoring is common in agriculture.…”