Herbal plant leaves classification for traditional medicine using convolutional neural network
Open Access
- 1 September 2024
- journal article
- research article
- Published by Institute of Advanced Engineering and Science in IAES International Journal of Artificial Intelligence (IJ-AI)
- Vol. 13 (3) , 3322-3329
Abstract
The classification of herbal plant leaves can be implemented in agriculture and traditional medicine. Primarily, sorting leaves was done before it was processed into medicinal ingredients. Currently, the sorting was still done manually by writing it on notes. Sometimes there were errors in the selection of leaves for medicinal ingredients. Herbal plants had various forms and are very greatly. Artificial intelligence technology was needed to have fast-paced time efficiency in sorting leaves. In the field of artificial intelligence, there was a specific or detailed learning process known as deep learning. The objective of this research was to classify herbal plant leaves images by applying and combining the convolutional neural network (CNN) deep learning method with data augmentation methods without the pre-trained architecture such as MobileNet and LeNet. This technique consisted of 4 main stages such as collecting data, preprocessing or normalizing data, building a model, and evaluating. The dataset used in this research were 4 types of herbal plants that do not flower and do not bear fruit including gulma siam, piduh, sirih, and tobacco. Each class had 250 images with total dataset used in this research was 1,000 images of herbal plant leaves and divided into 2 data, namely 80% data training 20% data testing, and validation. The data was trained with the epoch of 100 for the best training. It had an accuracy score of 98.74%. Without the data augmentation process it had an accuracy score of 91.43%.Keywords
This publication has 24 references indexed in Scilit:
- Medicinal plants from the 14th edition of the Russian Pharmacopoeia, recent updatesJournal of Ethnopharmacology, 2021
- Toward Robust Pedestrian Detection With Data AugmentationIEEE Access, 2020
- A Shallow Convolutional Neural Network for Apple ClassificationIEEE Access, 2020
- Centralized/decentralized event-triggered pinning synchronization of stochastic coupled networks with noise and incomplete transitional rateNeural Networks, 2020
- A robust system for road sign detection and classification using LeNet architecture based on convolutional neural networkSoft Computing, 2019
- Quality Classification of Enoki Mushroom Caps Based on CNN2019
- A Study on Growth Stage Classification of Paddy Rice by CNN using NDVI Images2019
- Fruit Classification Based on Six Layer Convolutional Neural Network2018
- Automatic Music Genre Classification using Convolution Neural Network2018
- Performance effect analysis for insect classification using convolutional neural network2017