Herbal Leaves Classification Based on Leaf Image Using CNN Architecture Model VGG16

Abstract

Herbal leaves are a type that is often used by people in the health sector. The problem faced is the lack of knowledge about the types of herbal leaves and the difficulty of distinguishing the types of herbal leaves for ordinary people who do not understand plants. If any type of plant is used, it will have a negative impact on health. Automatic classification with the help of technology will reduce the risk of misidentification of herbal leaf types. To make identification, a precise and accurate herbal leaf detection process is needed. This research aims to facilitate the classification model of herbal leaf images with a higher accuracy value than previous research. Therefore, the proposed method in this classification process is one of the Transfer Learning methods, namely Convolutional Neural Network (CNN) with a pretrained VGG16 model. This research uses a dataset of herbal leaves with a total of 10 classes: Belimbing Wuluh, Jambu Biji, Jeruk Nipis, Kemangi, Lidah Buaya, Nangka, Pandan, Pepaya, Seledri and Sirih. The performance of the results of the proposed classification method on the test dataset using Classification Report shows an increase in the results of the previous research accuracy value from 82% to 97%. This research also applies Image Data Generator in the augmentation process which aims to improve the image of herbal leaves, reduce overfitting, and improve accuracy.

Herbal Leaf Classification Using Deep Learning Model Efficientnetv2b0

“…The evaluation results of this research on test data using the EfficientNetV2B0 architecture, obtained an accuracy value of 99.14% and a loss value of 1.95%, where these results were able to outperform the evaluation results obtained by previous research using the VGG16 + Data Augmentation architecture [11]. Apart from that, the loss comparison resulting from this research is quite low due to the addition of the MaxPooling2D, Batch Normalization, Flatten layers.…”

“…In other research, the use of the Visual Geometry Group (VGG)16 method which was integrated with the Image Data Generator augmentation technique obtained accuracy results of 96.73%, and training loss of 0.097 at the 100th epoch [11], and if the augmentation process was not used then reducing the accuracy value to 96%.…”

Betel Leaf Disease Classification Using Data Augmentation and Convolutional Neural Network

“…There are different layers, suchas "input,""hidden," and "output,"in thearchitecture of aneural network. A group of hidden layers, including Convo,pooling, and fully connected layers with the normalizing layer,are synchronized and complete the task of securing the dataand images spatially [3]. The author made a comparative studyof the models and the result of the existing work and soybeandiseases using AlexNet and GoogleNett [14].…”

Klasifikasi Jenis Daun Tumbuhan Herbal Berdasarkan Lontar Usada Taru Pramana Menggunakan CNN

“…Adapaun terdapat sekitar 80% masyarakat masih bergantung pada tumbuhan herbal untuk kesehatan [4]. Lontar Usada Taru Pramana merupakan sebuah contoh nyata kearifan budaya lokal Bali yang didalamnya memuat catatan mengenai jenis-jenis tumbuhan herbal dan penggunaan tumbuhan herbal sebagai pengobatan tradisional dimana sebelumnya telah dikaji secara ilmiah dan masih digunakan oleh sebagian besar masyarakat Bali hingga saat ini sebagai pedoman penggunaan tumbuhan herbal [5] [6].…”

“…Selain keterbatasan pengetahuan botani yang dimiliki, tumbuhan-tumbuhan herbal juga sulit diidentifikasi secara akurat, mengingat terdapat kesamaan dalam morfologi di antara beberapa spesies, sehingga sulit untuk dibedakan secara kasat mata. Kesalahan klasifikasi terkadang berdampak buruk apabila berhubungan dengan komposisi racikan tumbuhan herbal dengan tujuan pengobatan [4]. Berkenaan dengan permasalahan tersebut, diperlukan adanya pengembangan sebuah sistem klasifikasi tumbuhan herbal secara otomatis.…”