Computer vision based robotic weed control system for precision agriculture

India is primarily an agriculture-based country and its economy largely depends upon the agriculture. But, most of the crops grown by the farmer are affected by weeds. Weed identification and control remains one of the most challenging tasks in agriculture. The most widely used methods for weed control is manual spraying of herbicides. But, this method has several negative impacts. Since hand labor is costly, an automated weed control system may be economically feasible. Although there have been many efforts to develop a system to control in-row weeds autonomously, no system is currently available for real-time field use. Further, the Onion is slow-growing, shallow-rooted crop that can suffer severe yield loss from weed competition. In order to overcome the above mentioned problems, the proposed system aims to develop a computer vision based robotic weed control system (WCS) for real-time control of weeds in onion fields. This system will be able to identify weeds and selectively spray right amount of the herbicide. The proposed WCS is an inexpensive and portable wireless system of handheld equipments which can be controlled remotely through a user friendly web interface. It is designed to automate the control of weeds and thus reduces the difficulties of farmers in maintaining the field. The proposed system is based on a combination of image processing, machine learning and internet of things (IoT).

I. Introduction

Agriculture plays a significant role in the economic growth of India. Hence, reducing cost and improving crop yield is highly needed. Precision farming technologies have been developed during the last two decades to refine the agricultural management practices. Every field has weeds growing in it throughout the year and in a country like India, the number of such fields are in the hundreds. The weed is a plant that competes with crops for water, food, nutrients and space, therefore reduces crop yield. Along with the increase in food-grain per capita, the need for weed removal also has increased. Identification and removal of weeds for small scale farms in India comes with the difficulty of hiring labor. Acquiring labors for regular crop maintenance has become one of the greatest challenges. The use of IoT, machine learning & image processing to implement human-free interaction with the crops is highly desirable.