| Autor: |
Kelvianto, Kevin, Kusnadi, Adhi |
| Předmět: |
|
| Zdroj: |
Engineering & Technology Quarterly Reviews; Dec2023, Vol. 6 Issue 2, p33-38, 6p |
| Abstrakt: |
This study investigates the use of artificial neural networks (ANN) employing the backpropagation method to identify flower types based on petal shapes. Android devices were utilized to capture flower images and transmit them directly to a server. Once a sufficient number of images were gathered, the training of the artificial neural network commenced. The flower images were processed to extract shape features using Sobel edge detection followed by thresholding. Subsequently, the data were normalized and fed into the ANN for training. Once the training was complete, the Android devices were capable of capturing new flower images and using the ANN to identify them. The findings of this research indicate that by using a single hidden layer with 35 hidden nodes, the system achieved a flower detection accuracy of 80%. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
