Convolutional neural network for maize leaf disease image classification
[...]the percentage of accuracy increased while compared to the previous study. 2.RESEARCH METHOD The steps for classification process using CNN are shown in Figure 1. The next process is classification using k-Nearest Neighbor, support machine and decision tree. 2.1.Maize leaf image Image data used...
Saved in:
Published in | Telkomnika Vol. 18; no. 3; pp. 1376 - 1381 |
---|---|
Main Authors | , |
Format | Journal Article |
Language | English |
Published |
Yogyakarta
Ahmad Dahlan University
01.06.2020
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | [...]the percentage of accuracy increased while compared to the previous study. 2.RESEARCH METHOD The steps for classification process using CNN are shown in Figure 1. The next process is classification using k-Nearest Neighbor, support machine and decision tree. 2.1.Maize leaf image Image data used maize leaves that size of 256x256 pixels. First and second convolutional layers have filter size of 64 pixels, third and fourth have filter size of 128 pixels, fifth to seventh have filter size of 256 pixels and eight to thirteenth have filter size of 512. First and second convolutional layers have filter size of 64 pixels, third and fourth have filter size of 128 pixels, 5 to 8 have filter size of 256 pixels and the 9 to 16 have filter size of 512 pixels, 18 ReLU, 5 max-pooling, 3 fully connected layers, 2 dropouts of 0.5 Softmax and output layer. |
---|---|
ISSN: | 1693-6930 2302-9293 |
DOI: | 10.12928/telkomnika.v18i3.14840 |