Accurate as well as automated medicinal leaf categorization is a critical chore in medicinal plant species identification. However, manual categorization is time compelling, error prone and mostly reliant on expert skills due to high inter as well as intra class variability among medicinal plant categories. Recent advancement in image processing and artificial intelligence has enabled automated plant species identification, providing reliable as well as scalable alternative to traditional procedures. This research represents a comparative analysis of Recurrent Neural Network (RNN) and Quantum Convolutional Neural Network (QCNN) for automated medicinal leaf categorization in terms of performance evaluation including accuracy, precision, recall and F1 score. Experimental outcome shows that QCNN significantly outperforms the RNN. RNN exhibits 68% accuracy, macro avg. precision 67%, recall 67%, f1 score 66% and weighted avg. precision 70%, recall 68%, f1 score 68% which is less as compared to QCNN which shows 96% accuracy, macro avg. precision 96%, recall 96%, f1 score 96% and weighted avg. precision 96%, recall 96%, f1 score 96%. Owing to its superior performance QCNN further integrated into a user interface framework to enable real time medicinal leaf categorization. The developed interface offers a user friendly, efficient and scalable platform for medicinal leaf identification application. The suggested system establishes the effectiveness of quantum motivated deep learning model in medicinal leaf image categorization as well as its usage and highlights the potential of QCNN trusted systems for intelligent medical applications.
Garnayak et al. (Tue,) studied this question.