Bridging AI and Ethnobotany: A Deep Learning Approach for Medicinal Plant Identification and Real-World Deployment
Year: 2026
Type: book-chapter
Abstract: Precise identification of medicinal plants is relevant to pharmacological studies and proper use of species, but most currently used image-based methods are tested on small-scale data and do not provide much information on the extrapolation of models.The paper examines how deep learning can be used to classify ten medicinal plant species, which are commonly used in rural and semi-urban areas in Bangladesh.An augmented subset of 5,000 images was augmented to 10,000 and divided into training, validation and test subsets.A variety of convolutional neural networks models, such as EfficientNetB3, InceptionV3, MobileNetV2, and VGG19 were trained and compared.The highest accuracy (99.00%) was attained by Efficient-NetB3 as compared to the other models.Nevertheless, such high accuracy shows that it is necessary to further validate it, including cross-validation and external dataset testing, to determine how well it works in the real world.A web-based prototype that was lightweight was created as well to illustrate that it can be used in practice.Comprehensively, the paper gives a comparative review of the current CNN models and explains the capabilities and their constraints to automated recognition of medicinal plants.
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7163673485