Abstract
This study presents a comparative study of multispectral and RGB (red, green, and blue) sensor-based cotton canopy cover modelling using multi-temporal unmanned aircraft systems (UAS) imagery. Additionally, a canopy cover model using an RGB sensor is proposed that combines an RGB-based vegetation index with morphological closing. The field experiment was established in 2017 and 2018, where the whole study area was divided into approximately 1 x 1 m size grids. Grid-wise percentage canopy cover was computed using both RGB and multispectral sensors over multiple flights during the growing season of the cotton crop. Initially, the normalized difference vegetation index (NDVI)-based canopy cover was estimated, and this was used as a reference for the comparison with RGB-based canopy cover estimations. To test the maximum achievable performance of RGB-based canopy cover estimation, a pixel-wise classification method was implemented. Later, four RGB-based canopy cover estimation methods were implemented using RGB images, namely Canopeo, the excessive greenness index, the modified red green vegetation index and the red green blue vegetation index. The performance of RGB-based canopy cover estimation was evaluated using NDVI-based canopy cover estimation. The multispectral sensor-based canopy cover model was considered to be a more stable and accurately estimating canopy cover model, whereas the RGB-based canopy cover model was very unstable and failed to identify canopy when cotton leaves changed color after canopy maturation. The application of a morphological closing operation after the thresholding significantly improved the RGB-based canopy cover modeling. The red green blue vegetation index turned out to be the most efficient vegetation index to extract canopy cover with very low average root mean square error (2.94% for the 2017 dataset and 2.82% for the 2018 dataset), with respect to multispectral sensor-based canopy cover estimation. The proposed canopy cover model provides an affordable alternate of the multispectral sensors which are more sensitive and expensive.
Highlights
Numerous studies are being conducted on cotton crop growth monitoring for precision agriculture
From both the 2017 and 2018 experiments, it was observed that the normalized difference vegetation index (NDVI)-based and RGB reference-based average canopy cover (CC) per grid followed the same trend throughout the growing season, and there was a one-to-one correspondence between the two when plotted as a straight line at an intercept of one with very high R2 values
With a multi-year CC analysis, MS sensor-based CC estimation was used as a reference, as it is considered a stable and accurate form of estimation
Summary
Numerous studies are being conducted on cotton crop growth monitoring for precision agriculture. There is a growing need for phenotyping to match this high pace breeding process. Plant breeders and agriculture scientists have recognized the need for a high-throughput phenotyping (HTP) system that can efficiently measure phenotypic traits such as crop height, volume, canopy cover, and vegetation indices (VIs) with reasonable accuracy [3]. HTP is an extensively discussed phenomenon; until recently, its implementation has been rather fragmentary [4]. The change in this situation has been mainly attributed to the recent developments in unmanned aircraft systems (UAS). Lightweight platforms combined with consumer grade imaging sensors have provided an affordable system to perform the necessary remote sensing activities for precision agriculture, especially with low altitude flights that provide high temporal and spatial resolution data [5,6,7,8]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
