HERBAL PLANT DETECTION BASED ON LEAVES IMAGE USING CONVOLUTIONAL NEURAL NETWORK WITH MOBILE NET ARCHITECTURE
Abstract
Indonesia is a country with a variety of flora/plant diversity. One type of flora wealth is herbal plants. Herbal plants are plants that have uses to treat a disease. The diversity of herbs often makes our mistakes in recognizing the type. Therefore we need a system that can recognize the types of herbs automatically with their use. In this study, the CNN (Convolutional Neural Network) algorithm is used. This algorithm is a deep learning method that can recognize and classify an object. In this study, we use 500 images for 5 types of leaves of herbal plants. Mobilenet architecture is used on an Android-based system so that it has the thickness of the convex filter that matches the image thus saving the size of the learning model. Based on the test results on 30 new images obtained an accuracy rate of 86.7%. So it can be concluded that the use of the CNN algorithm is quite good at detecting herbal plants based on the training data used.
Downloads
References
F. Liantoni and H. Nugroho, “Klasifikasi Daun Herbal Menggunakan Metode Naïve Bayes Classifier Dan Knearest Neighbor,” J. Simantec, vol. 5, no. 1, pp. 9–16, 2015.
M. Damayanti et al., “Pengenalan Daun Tanaman Obat Menggunakan Jaringan Syaraf Tiruan Backpropagation,” vol. 4, no. 2, pp. 98–103, 2019.
F. S. Ni’mah, T. Sutojo, and D. R. I. M. Setiadi, “Identifikasi Tumbuhan Obat Herbal Berdasarkan Citra Daun Menggunakan Algoritma Gray Level Co-occurence Matrix dan K-Nearest Neighbor,” J. Teknol. dan Sist. Komput., vol. 6, no. 2, pp. 51–56, 2018, doi: 10.14710/jtsiskom.6.2.2018.51-56.
Hendriyana and Y. H. Maulana, “Identifikasi Jenis Kayu menggunakan Convolutional Neural Network,” J. RESTI (Rekayasa Sist. dan Teknol. Informasi), vol. 4, no. 1, pp. 70–76, 2020.
Z. Imaduddin and H. A. Tawakal, “Aplikasi Mobile Untuk Deteksi Dan Klasifikasi Daun Secara Real Time,” J. Teknol. Terpadu, vol. 1, no. 1, 2015.
W. S. Eka Putra, “Klasifikasi Citra Menggunakan Convolutional Neural Network (CNN) pada Caltech 101,” J. Tek. ITS, vol. 5, no. 1, 2016, doi: 10.12962/j23373539.v5i1.15696.
F. Liantoni, “Klasifikasi Daun Dengan Perbaikan Fitur Citra Menggunakan Metode K-Nearest Neighbor,” J. Ultim., vol. 7, no. 2, pp. 98–104, 2016, doi: 10.31937/ti.v7i2.356.
D. C. Khrisne, P. Studi, T. Elektro, F. Teknik, and U. Udayana, “Pengenalan Pola Motif Kain Tenun Gringsing Menggunakan Metode Convolutional Neural Network Dengan Model Arsitektur,” vol. 6, no. 3, pp. 159–168, 2019.
A. Michele, V. Colin, and D. D. Santika, “Mobilenet convolutional neural networks and support vector machines for palmprint recognition,” Procedia Comput. Sci., vol. 157, pp. 110–117, 2019, DOI: 10.1016/j.procs.2019.08.147.
N. Purnama, “Deteksi Tanaman Herbal Berdasarkan Citra Daun Menggunakan Convolutional Neural Network Dengan Arsitektur Mobilenet.” Denpasar, 2020.
Abstract viewed = 657 times
PDF downloaded = 588 times
-a.jpg)
-b.jpg)
