KLASIFIKASI KEBUTUHAN NUTRISI MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK PADA DAUN TANAMAN PADI
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
WARDANI, SYUKRON ADITIYA
ruli, riki
indrianto, indrianto
ruli, riki
indrianto, indrianto
Subject
Teknik Informatika
Datestamp
2023-05-29 08:01:55
Abstract :
Indonesia is an agricultural country with a large agricultural area. Today, growing rice is mainly done indoors hydroponically with a multilevel farming system or indoor vertical farming because of its effectiveness in reducing the size of agricultural land and still producing large amounts of rice. Nutrients in plants grown hydroponically need attention. This study aimed to classify the nutritional needs of rice plant leaves and determine the accuracy of the convolutional neural network (CNN) algorithm in determining nutrition in rice plants. Detecting plant nutrients can be done automatically using the CNN algorithm. CNN is a development and modification of the Multi-Layer Perceptron (MLP), which is intended to perform two-dimensional image data processing. This study has 1156 datasets that are processed using 2 types of architecture, MobileNetV2, and VGG16, with 3 different pixels, including 128, 192, and 224. The limited hyperparameters in this study are split data, 80% training data, 10% valid data, and 10 % test data, batch size 20, and epoch 15, 25, and 35. The best results in this study were obtained on the MobileNetV2 architecture with 224x224 pixels, and epoch 25, which resulted in an accuracy of 87.07% means that the CNN algorithm can classify the nutritional needs of rice plant leaves.
Indonesia is an agricultural country with a large agricultural area. Today, growing rice is mainly done indoors hydroponically with a multilevel farming system or indoor vertical farming because of its effectiveness in reducing the size of agricultural land and still producing large amounts of rice. Nutrients in plants grown hydroponically need attention. This study aimed to classify the nutritional needs of rice plant leaves and determine the accuracy of the convolutional neural network (CNN) algorithm in determining nutrition in rice plants. Detecting plant nutrients can be done automatically using the CNN algorithm. CNN is a development and modification of the Multi-Layer Perceptron (MLP), which is intended to perform two-dimensional image data processing. This study has 1156 datasets that are processed using 2 types of architecture, MobileNetV2, and VGG16, with 3 different pixels, including 128, 192, and 224. The limited hyperparameters in this study are split data, 80% training data, 10% valid data, and 10 % test data, batch size 20, and epoch 15, 25, and 35. The best results in this study were obtained on the MobileNetV2 architecture with 224x224 pixels, and epoch 25, which resulted in an accuracy of 87.07% means that the CNN algorithm can classify the nutritional needs of rice plant leaves.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
