New method for cotton fractional vegetation cover extraction based on UAV RGB images

As the key principle of precision farming, distribution fractional vegetation cover is basis crop management within field serves. To estimate FVC rapidly at farm scale, high temporal-spatial resolution imagery obtained by unmanned aerial vehicle (UAV) was adopted. verify application potential consumer-grade UAV RGB in estimated FVC, blue-green characteristic index (TBVI) and red-green (TRVI) were proposed this study according to differences gray value among cotton vegetation, soil shadow field. First, two new constructed indices several published used extract visible light images generate greyscale for each indices. Then, thresholds non-vegetation pixels established based on threshold method which combines support vector machine classification index. Finally, accuracy difference information extraction between newly compared. The results show that extracted TRVI higher than subdivision other (FVC first bud stage cotton: R2=0.832, RMSE=2.307, nRMSE=4.405%; R2=0.981, RMSE=1.393, nRMSE=1.984%; flowering R2=0.893, RMSE=2.101, nRMSE=2.422%; boll R2=0.958, RMSE=1.850, nRMSE=2.050%). Keywords: cotton, UAV, images, cover, method, TRVI, TBVI DOI: 10.25165/j.ijabe.20221504. Citation: Yang H B, Lan Y Lu L Q, Gong D C, Miao J Zhao J. New images. Int Agric & Biol Eng, 2022; 15(4): 172–180.