This research describes a method for classifying images of medicinal flowers using deep learning techniques. Since their therapeutic qualities have been used for generations, it is essential to correctly identify medicinal flowers before using them. The suggested technique starts with pre-processing the floral photos, then uses convolutional neural networks (CNNs) to extract features. A dataset of 10 distinct medicinal flowers was gathered and annotated with their respective classes in order to assess the performance of the suggested strategy. The study also looked into how various hyper-parameters, such the batch size, learning rate, and number of layers in the CNN model, affected classification accuracy. According to the findings, classification accuracy was increased by reducing the learning rate and adding more layers to the CNN model. The topic of plant identification and classification, particularly that of medicinal plants, has tremendous promise for applications of the suggested approach. This could help pharmacologists, researchers, and botanists identify and analyse various types of medicinal plants, ultimately resulting in the creation of more efficient cures.
Bhavyatha B. , Gagana R. , Ghowli Amrutha , Sapna R. "Image Classification of Medicinal Flowers" Iconic Research And Engineering Journals Volume 6 Issue 12 2023 Page 77-83
Bhavyatha B. , Gagana R. , Ghowli Amrutha , Sapna R. "Image Classification of Medicinal Flowers" Iconic Research And Engineering Journals, 6(12)