Research objectives Use hyperspectral data from UAV platform for crop/weed discrimination Generation of spraying task /application maps for SSWM and validation The research will focus on two kind of weeds i.e. thistles in grassland and bindweeds in maize fields for case study. Field aerial hyperspectral images of thistles will be collected at the different growth stages and different flight height. The aerial images of bindweeds will also be taken in the early growth stage in the maize fields. Next, the effective spectral wavelengths which could be used to discriminate the spectral response of weeds and crops. The textural features and morphological parameters of species will be discussed and explored using Object-based image analysis (OBIA) algorithms and the traditional Otsu’s threshold method. Then the different combined feature sets will be regarded as inputs to different discrimination models e.g. Support Vector Machines (SVM) or Linear Discrimination Analysis (LDA) to classify species, some deep learning models will also be explored. The accuracy of the resulting models will be assessed by overall classification accuracy. To facilitate precision spraying, the generation of relevant task maps, that delineate the spray zones combined with the appropriate doses will be produced. Then field validation experiments comparing the developed technology for SSWM with blanket field spraying will be performed to evaluate and confirm its biological efficacy and usability. An economic analysis in terms of costs of plants protection products and labor costs will be done.