Crop and Weed Semantic Segmentation using a Fuzzy Transform based Active Contour Model

Abstract

The idea of semantic segmentation is crucial in many different fields, Including robotic vision, medicine, and manv others. Weeds are one of the main factors that could reduce crop productivity, thus it is essential to understand how crucial semantic segmentation is in the agricultural industry in segmenting crops from weed. Images are usually taken in a variety of climatic environments. often making them with low contrast. Using a fuzzy transform, which improves Image quality reasonably and will able to brinz out shapes or areas of interest within an image. To the resultant image enhancement using fuzzy transform, the study applies an active contour model which with the heln of the level set method identifies the boundaries and objects which were hidden due to low contrast. The propossed method when applied in the sezmentation of crop and weed delivers promising results. The outcomes of this strategy demonstrate its effectiveness.