Adulterating medicinal plant leaves with undesirable species poses a challenge to the authenticity and effectiveness of herbal medicines. Modern technologies, such as artificial intelligence, are widely used to differentiate medicinal plant leaves from adulterants to overcome this issue and authenticate medicinal plant leaves. These automated frameworks offer a reliable and efficient solution to deal in real-time with adulteration problems, quality control, species misidentification, etc., especially in the herbal medicine industry. So, this paper presents an automated plant species identification technique, which uses a 15-layer convolutional neural network (CNN) model to correctly detect two distinct plant leaf species. These leaves include Azadirachta Indica, commonly known as ‘Neem,’ which has excellent medicinal significance, and another plant having similar leaf characteristics, i.e., Melia Azedarach, widely known as ‘Baikan.’ Firstly, leaf images are pre-processed and supplied to the given CNN model to classify them precisely. This model also serves as a feature extractor employed with machine learning classifiers, including histogram boosting gradient, light gradient boosting machine, random forest, extremely randomized trees, and support vector machine to recognize plant leaf species. Examining outcomes of the suggested CNN model and its combination with other machine learning classifiers results in an accurate identification of plant species using that model with a maximum accuracy of 97.56%. Hence, the suggested 15-layer CNN model is a super-effective technique for accurately identifying different types of medicinal and other plant species.
Singh et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: