A Comprehensive Literature Review on AI-Driven Herbal Plant Identification: Machine Learning and Deep Learning based Approaches

Deep Learning (DL) and Machine Learning (ML) approaches have been extensively explored in recent times for their efficiency in identifying and classifying plant species and their diseases. These techniques have the potential to increase efficiency and precision in diagnosing diseases, thereby revolutionizing agriculture by improving output and yields. Many industries depend on these plants for their products, making the prospect of their disappearance a critical issue. Therefore, identifying these plants has become increasingly important. Given the vast number of plant species, manually identifying them is impractical due to the experience and time required. To expedite this process, experts suggest using DL and ML techniques. This review provides a comprehensive analysis of various DL and ML methods for identifying plant diseases, highlighting technological approaches, data sources, and metrics of efficacy. Key models discussed include Support Vector Machines (SVMs), Convolutional Neural Networks (CNNs), and hybrid models that combine multiple feature extraction techniques. While these methods have shown promising results, they still face challenges such as inconsistent datasets, high resource demands, difficulties in result interpretation, and practical implementation issues. This study offers an in-depth review of the current research landscape, identifies persistent problems, and suggests future directions for advancing the application of these technologies in the agricultural industry.