Moment Invariant Based Weed/Crop Discrimination For Smart Farming

In order to sustain a growing worldwide population with sufficient farm production, new smart farming methods are required to increase or maintain the crop yield while minimizing the environmental impacts. Selective weed treatment is a critical step in autonomous management of crop. It also reduces the use of herbicides by applying only to weed effected area instead of spraying uniformly in the field. Therefore, reduction in the amount of herbicides used in modern agriculture is a relevant step towards sustainable agriculture. Therefore, in automatic weed control technique, accurate weed detection is an essential task. In this paper image processing based weed detection has been acomplished. As the shape of leaf consists of abandant information, regarding the plant. Therefore, shape feature has an important role in weed detection. Here, moment invariant based shape feature is proposed to detect the weed in crop. The invariants of the central moment of image yields different values for weed and crop. Here we have analysized the moment invariant values for different weed and crop images and obtained success upto some extend.