Classification of rice growth stage based on convolutional neural network. (April 2020)

• Record Type: Journal Article
• Authors: Kusumaningrum, R; Satriaji, W; Endah, S N; Prasetyo, Y; Sukmono, A
• Is Part Of: Journal of physics. Volume 1524(2020)
• Publication Date: 2020-04
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1524/1/012114
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: The proposed method of rice growth classification model based on Convolutional Neural Network (CNN) which had implemented towards LANDSAT images gives the highest accuracy value of 83.4% with the following parameters including batch size 32, drop out 0.5 and band 432. The batch size value is inversely proportional to the level of accuracy obtained, which means the greater the batch size value, the smaller the average level of accuracy obtained, whereas there is no correlation between the change in the drop out value and the accuracy value and in general the best accuracy value in the drop out value is 0.5.