ABSTRACT
Medicinal plants are gaining popularity in the pharmaceutical sector because they have fewer side effects and are less expensive than modern pharmaceuticals. Many researchers have expressed a strong interest in the study of automatic medicinal plant recognition as a result of these facts. There are several avenues for progress in developing a strong classifier that can reliably categorise medicinal plants in real time. The effectiveness and reliability of various machine learning techniques for plant classifications using leaf pictures that have been utilised in recent years are discussed in this research. The review discusses the image processing approaches used to detect leaves and extract significant leaf features for several machine learning classifiers. The performance of these machine learning classifiers when classifying leaf images based on standard plant features, such as shape, vein, texture, and a combination of several aspects, is classified. We also go over the leaf databases that are publicly available for automatic plant recognition, and we wrap off with a discussion of some of the most important ongoing studies and potential for improvement in this area.
