IoT Based Weed Detection and Removal in Precision Agriculture

Abstract

Weeds are undesirable plants growing among the good ones. Weeds have the capability to grow rapidly and interrupt the growth of other plants by becoming a competent for nutrients, water, and space. Weed identification is a critical phase in the weeding process. Weed identification was usually done by skilled workers who were hired for this purpose. It is done by scrutinizing each and every spot in the field. Later, herbicides were used to kill the weeds. But this is a labour-intensive process. When herbicides are sprayed manually, it is been sprayed on the crops as well. When these crops are ingested, they cause harmful effects. With the continual improvement in agricultural production levels, it is critical to precisely discriminate crops from weeds and to achieve precision weed-only spraying. Of late, technology is used to control weed growth in farms. Many weed removal solutions have been tried out in the past. But those methods are less reliable. We rely on a system that gives reliable results with low cost and which exercise less human labour. For this purpose, a robot is designed which will capture snapshots of the field and identify the weeds using TFL Classify and spray the herbicides on the weeds. The robot when powered on follows the command given through laptop or mobile like directions and spraying. This robot is very easy to use. This system requires very less labour.