A GENERIC FRAMEWORK FOR EFFICIENT LEAF DISEASE DETECTION USING DEEP LEARNING MODELS AND TRANSFER LEARNING

Fecha de publicación: 29/01/2021
Fuente: Wipo "precision agriculture"
This invention is pertaining to leaf disease detection which is part of Precision Agriculture (PI). It has novel architecture based on deep learning models and transfer learning to realize a generic framework for efficient leaf disease detection. It takes leaf images live through satellite or remote sensing technology and follows a supervised learning approach for the training and prediction of diseases. It helps in leveraging technology based agriculture with the ability to detect leaf diseases for given crop. It incorporates different modules in order to support detection of leaf diseases of various crops with minor changes in the configurations. The invention can be integrated with any existing PA applications used in agricultural domain. The framework is extensible and support future algorithms to improve the state of the art. The invention is useful to different stakeholders such as farmers, agriculture department professionals, governments, academia and researchers. It provides an out of the box solution for leaf disease prediction and take measures in near real time basis.