Cluster-based segmentation for tobacco plant detection and classification

Tobacco is one of the major economical crops in agriculture sector. It essential to detect tobacco plants using unmanned aerial vehicle (UAV) images for improved crop yield and plays an important role early treatment plants. The proposed research work carried out three phases: In first phase, we collect from UAV’s apply French Commision Internationale de l'eclairage (CIE) L*a*b colour space model as pre-processing operations segmentation. And then two prominent motion descriptors namely histogram flow (HOF) boundary (MBH) are combined with optimal oriented gradients (HOG) descriptor exploring trajectory spatial measurements. finally, variations respect scale illumination changes incorporated HOG descriptor. Here both dense patterns refined hierarchical feature selection principal component analysis (PCA). trained evaluated on different UAV image datasets done a comparative machine learning (ML) algorithms. achieves good performance 95% accuracy 92% sensitivity.