Assessment of Invasive and Weed Species by Hyperspectral Imagery in Agrocenoses Ecosystem

The present study aimed to investigate the possibility of using hyperspectral imaging data identify invasive and weed species in agrocenoses ecosystem. most common weeds grain agrocenoses, i.e., Ambrosia artemisiifolia L., Euphorbia seguieriana Neck., Atriplex tatarica Glycyrrhiza glabra Setaria pumila (Poir.) Roem. Schult, served as objects. population weeds, especially is for selected region study. Therefore, shooting objects was carried out with a camera, Cubert UHD185, values 100 spectral channels were obtained from images. 80 vegetation indices (VIs) calculated. material processed mathematical statistics (analysis variance, t-test) search methods analysis (principal component analysis, decision tree, random forest). Using statistical methods, simultaneous use several VIs differentiated between more deliberately precisely. combination Derivative index (D1), Chlorophyll content (Datt3), Pigment specific normalized difference (PSND) can be used identification. tree method, established good division into groups; (1) perennial rhizomatous (Euphorbia seguieriana, glabra), (2) annual (A. artemisiifolia, A. tatarica, S. pumila); These Vis are (CI), D1, Datt3. forest that have greatest impact on Mean Decrease Accuracy Gini Datt3, PSND, Double Peak Index (DPI). channel identification plant principal showed worse results than when VIs. A great similarity help analysis.