| Autor: |
Elfri, Muhammad Afiq Amirul, Rahman, Fatin Hamadah, Newaz, S. H. Shah, Suhaili, Wida Susanty, Au, Thein Wan |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2023, Vol. 2643 Issue 1, p1-8, 8p |
| Abstrakt: |
This study aims to determine paddy crop health from Unmanned Aerial Vehicle (UAV) multispectral images for the whole season and produce useful information to aid in the management for yield optimisation. Agriculture is important to a country because it generates income and provides food security. Crop production in Brunei Darussalam needs to produce sufficient yield in order for the country to be self-sufficient and rely less on oil and gas industries. Factors such as pest and disease have been obstacles and challenges for many farmers, affecting the production yield. In the era of IR4.0, technologies are being used for precision smart agriculture for better management of the ecosystem of production life cycle, sustainability and profitability. The use of an UAV for agriculture in paddy production in Brunei is not only limited to crop spraying, but can also be used in acquiring aerial field image for the whole season. Many of the existing research on image-based disease detection involves close-up image and it is very challenging to detect diseases from a large-scale image. In this study, aerial image used is based on Normalised Difference Vegetation Index (NDVI), which is used to indicate the crop health. With image classification, the plot can be classified into three different field crop health levels: healthy, moderately healthy and non-healthy. The age of the plot is also taken into consideration when classifying the image to produce the true output. To differentiate and distinguish useful objects from undesired objects, semantic segmentation is used in this study. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
