Agriculture is an essential occupation to the people of India. It is considered as the backbone of most of the Indian population. However, one of the biggest concerns of agriculture is the growth of weeds. These weeds have to be removed to get a fruitful harvest. This process of removing weeds is weeding, which must be done with utmost care without affecting the valuable crops. Using agricultural chemicals is one of the most popular ways to manage weeds. However, weed identification is one of the challenging parts of cultivation, as the use of chemicals throughout the plantation is harmful to the environment and the agricultural ecosystem. In addition, manually removing the weed is possible but not entirely practical, considering human error and labor charges that must be paid to them. This identification of weeds leads to the demand for alternatives to weed control and identification techniques. Therefore, industries continue to seek human-free automated mechanisms that are relatively inexpensive. In this regard, machine vision comes into action for agricultural automation. Machine vision technology uses cameras rather than the naked eye to identify. In recent years, machine vision technologies have rapidly developed, and the progress achieved is remarkable. Machine vision technology has been proven to help build automation in agriculture resulting in cost-effective, highly efficient, and high-precision solutions. In addition, the increased computational power of the hardware, decreased costs, and advancements in the accuracy and efficiency of the algorithms have made it possible to construct feasible and practical automatic weeding strategies. This chapter focuses on the exploration of numerous machine vision strategies involved in automated weeding and their applications, use cases, and research challenges.