DeepHerb: A Vision Based System for Medicinal Plants Using Xception Features

The conservation of biodiversity is crucial as many plant species are critically under extinction. traditional medicinal system, an alternative to synthetic drugs, promote healthy living and mainly depends on the wide repository plants. A vision-based automatic identification system proposed using different neural network techniques in computer vision deep learning. challenge lies unavailability herb dataset. paper showcases a novel leaf dataset entitled DeepHerb comprising 2515 images from 40 varied Indian herbs. efficacy revealed by comparing pre-trained convolution architectures such VGG16, VGG19, InceptionV3 Xception. work concentrates adopting transfer learning technique models extract features classify Artificial Neural Network (ANN) Support Vector Machine (SVM). SVM hyperparameters tuned further Bayesian optimization achieve better performance model. model learned Xception ANN outperformed 97.5% accuracy. cross-platform mobile application HerbSnap developed integrating identifies image with prediction time 1 second per reveals pertinent details herbs database. This research will focus expanding benefit stakeholders thus, enriches society knowledge their properties.