Early-Season Mapping of Johnsongrass (Sorghum halepense), Common Cocklebur (Xanthium strumarium) and Velvetleaf (Abutilon theophrasti) in Corn Fields Using Airborne Hyperspectral Imagery

Accurate information on the spatial distribution of weeds is key to effective site-specific weed management and efficient sustainable use control measures. This work focuses early detection johnsongrass, common cocklebur velvetleaf present in a corn field using high resolution airborne hyperspectral imagery acquired when plants were four six leaf growth stage. Following appropriate radiometric geometric corrections, two supervised classification techniques, such as spectral angle mapper (SAM) mixture analysis (SMA) applied. Two different procedures compared for endmember selections: measurements automatic methods identify pure pixels image. Maps both, overall each three species, obtained with sources. The best results achieved by defining endmembers through collected spectroradiometer. Overall accuracies ranged between 60% 80% SAM maps that do not differentiate species while it decreased 52% individually classified. In this case, SMA technique clearly improved results. proposed methodology shows be promising prospect applicable low cost images new generation sensors onboard unmanned aerial vehicles (UAVs).