An Automated Approach to Weed Out Prosopis Juliflora(Karuvelam) Trees

Computing technology is not only crucial but also related to environmentalists and other related authorities. Agriculture is the backbone of India, which is greatly affected due to the insufficient water supply. Prosopis juliflora (karuvelam trees) which is used only as a firewood has greater environmental impacts such as water absorption from the soil and moisture from atmosphere, emits low oxygen and more carbon dioxide comparatively. This project has been designed to give a complete solution for the eradication of karuvelam trees, as the manual removal and using heavy machines may increase the number of labours and cost for removing these trees with long penetrating roots. Much more number of trees can be eradicated using technical methods.In this project, we propose a cloud based approach for identification of karuvelam trees affecting agricultural lands using Deep learning. The system consists of a setup which can identify karuvelam leaves from images and retrieve the GPS location and update the cloud. The Data is processed in the cloud and the instruction is delivered to the chemical spraying setup through IOT board to kill the weeds. Our project aims at designing a prototype to prove that the code that we have developed for identification of karuvelam trees works well in real system. The concept can be incorporated in real world using OPTiM agri drone(automatic weedicide sprayer on selected weeds) designed by Japan.