A device for effective weed removal for smart agriculture using convolutional neural network

India loses agricultural produce worth over $11 billion more than the Centers budgetary allocation for agriculture for 2017–18 annually to weeds, according to a study by researchers associated with the Indian Council for Agricultural Research (ICAR). The primary problem is to identify the type of weed in a given agricultural field through real-time monitoring of field by drone. The proposed device will have a drone that will capture real-time images and identify the weed associated with the given crop seedling using multi-class classifier based on Convolutional Neural Network with transfer learning. The weeds are removed by spraying herbicides by drone suggested for the detected weed. The drone will spray herbicides in an effective way so that to bring down weed management costs to enhance profit for farmers while protecting the environment.