In precision agriculture, image segmentation is essential for tasks like yield calculation, weed identification, and crop disease detection. This research study performs an in-depth review of the several image segmentation methods that can be used in the agricultural sector. We explore thresholding techniques, watershed algorithms, superpixel partitioning, clustering, semantic and instance segmentation. This study explains how well deep learning architectures like convolutional neural networks (CNNs) work to solve the problems associated with agricultural image segmentation. This study intends to direct future research and development in the field of precision agriculture by identifying the possible uses and constraints of various segmentation approaches.