A High-Precision Crop Classification Method Based on Time-Series UAV Images

Timely and accurate information on crop planting structures is crucial for ensuring national food security formulating economic policies. This study presents a method high-precision classification using time-series UAV (unmanned aerial vehicle) images. Before constructing the images, Euclidian distance (ED) was utilized to calculate separability of samples under various vegetation indices. Second, co-occurrence measures gray-level matrix (GLCM) were employed derive texture characteristics, spectral features crops successfully fused. Finally, random forest (RF) other algorithms classify crops, confusion applied assess accuracy. The experimental results indicate following: (1) Time-series remote sensing images considerably increased accuracy classification. Compared single-period image, overall kappa coefficient by 26.65% 0.3496, respectively. (2) object-oriented better suited precise crops. 3.13% 0.0419, respectively, as compared pixel-based results. (3) RF obtained highest in both RF’s producer user cotton, spring wheat, cocozelle, corn area more than 92%. These provide reference statistics agricultural precision management.