A Review and Taxonomy on Crop and Weed Classification Based on Machine Learning Techniques

. Agriculture plays a crucial role in global food security, yet crop productivity is significantly affected by the presence of weeds. Weeds compete with crops for essential resources such as nutrients, water, sunlight, and space, leading to reduced yields and economic losses. Traditional weed management practices, including manual weeding and blanket application of herbicides, are often labor-intensive, costly, time-consuming, and environmentally harmful. In this context, crop–weed classification using machine learning and deep learning techniques has emerged as a promising solution for enabling precise, efficient, and sustainable weed management. Conventional weed control methods largely rely on human expertise or uniform herbicide spraying across entire fields. Manual weed identification is highly subjective, prone to human error, and impractical for large-scale farms. On the other hand, indiscriminate use of herbicides increases production costs, causes herbicide resistance in weeds, degrades soil health, and contaminates water resources. These limitations highlight the need for intelligent automated systems capable of accurately distinguishing crops from weeds in real time. The basics of machine learning based classifiers applied to image classification have also been discussed. Salient features of existing techniques along with the research gap have been clearly highlighted. The research gaps identified in existing work allows future researchers in leveraging the limitations of existing approaches and devising novel methods. Finally the evaluation metrics to evaluate the performance of existing work have been presented for comparative performance evaluation.