A Survey on Different Methods for Medicinal Plants Identification and Classification System

Abstract

Medicinal herbs are getting popular in the pharma industry because they have minimal side effects and become less expensive than contemporary pharmaceuticals. Several people have indicated a strong interest in the topic of automated medicinal plant identification as a result of these findings. There's several avenues for progress in developing a strong classification that can reliably recognize medicinal herbs in real - time basis. The effectiveness and reliability of many machine learning techniques for plant categorization employing leaf pictures which have been employed in recent years are discussed in this paper. Also included are assessments of their advantages and disadvantages. For certain machine learning methods, the study provides the image processing techniques that are employed to identify leaf and retrieve significant leaf attributes. The effectiveness of these machine learning algorithms while identifying leaf images depending on typical plant properties, such as form, vein, texture, and a mixture of numerous aspects, is categorized. The leaf datasets that are publicly accessible for computerized plant identification are also examined, and the article closes with a summary of existing study and potential for improvement in this field.