Weed Detection Using Machine Learning Techniques - A Review

Abstract

Weeds are one of the biggest barriers to crop productivity in recent years. Many efforts are being made to enhance the quality of agricultural crops which is a challenging task for farmers. Weeds are also affecting the quality of the crops along with the productivity of the crops. Weeds are the reason for the lack of water in the crops, because they absorb the moisture of the land, due to which the crops do not get proper water. Weeds also create hindrance during harvesting of crops. When farmers harvest crops, often toxic weeds also get mixed, due to which farmers do not get fair price for their crops. To prevent weeds, farmers spray herbicides uniformly throughout the field without knowing what type of weed it is. Due to which there is a lot of damage to the crops. Spraying herbicides also affects the environment, so it is essential to have a thorough knowledge of weeds so that specific weeds can be controlled. The main objective of this paper is to improve the accuracy of weed detection using machine learning techniques.