Fecha de publicación: 21/08/2020 Fuente: Wipo "precision agriculture" This invention is related to technology driven and automated leaf disease detection towards Precision Agriculture (PI). The invention has provision to collect leaf images through satellite imagery. Since satellite imagery throw challenges on quality of images, there is provision for noise removal that leverages quality of input images. The Afterwards the images are subjected to segmentation process in order to identify Regions of Interest (ROI). Once the interested regions are known, those regions are used for feature extraction and feature optimization to improve quality of training. A deep learning based CNN is employed for learning features. With the learned knowledge and knowledge gained from pre-trained leaf images, the deep CNN learning results in a leaf disease prediction model that is capable of discriminating healthy leave from infected ones. For extracting features a descriptor known as daisy descriptor is used. Daisy descriptor represents features of given leaf image and is best used for image matching in the supervised learning and prediction process. The proposed system is part of precision agriculture and can be integrated with any application that deals with PI.