|
Rice Variety Identification Using Morphological Characteristics of Paddy Seed via Decision Tree and Fuzzy Logic |
|---|---|
| รหัสดีโอไอ | |
| Creator | Athasart Narkthewan |
| Title | Rice Variety Identification Using Morphological Characteristics of Paddy Seed via Decision Tree and Fuzzy Logic |
| Contributor | Patcharaporn Narkthewan, Noppadol Maneerat |
| Publisher | Thai Socities of Agricultural Engineering |
| Publication Year | 2565 |
| Journal Title | Thai Socities of Agricultural Engineering Journal |
| Journal Vol. | 28 |
| Journal No. | 2 |
| Page no. | 49-59 |
| Keyword | Rice variety identification, Morphology, Decision tree, Fuzzy logic |
| URL Website | https://li01.tci-thaijo.org/index.php/TSAEJ/index |
| Website title | Thai Socities of Agricultural Engineering Journal |
| ISSN | 2651-222X |
| Abstract | The collection of various rice varieties is a way of rice seed conservation. The conservation of rice varieties solves the problem of shortage of pure and good quality rice seeds as well as helps to conserve rice varieties from being lost. Moreover, the selected rice varieties can be used as good breeders for future breeding. The morphological characteristics of rice varieties were used as the components for rice variety identification. Therefore, rice specialists and tools are an important role in rice variety identification. The aim of the study was to present the idea of using computers to build an artificial intelligence method for rice variety identification. The fuzzy logic and decision tree techniques were used to perform the proposed system for problem-solving rice variety identification. The morphological analysis of paddy seed is composed of two steps. In the first step, the decision tree technique was applied to filter the quality data of the paddy seed morphology. Subsequently, the other quantitative data of the paddy seed morphology were processed with fuzzy logic techniques to identify the rice varieties. A total of twenty-six rice varieties were used in the experiment as the proposed method and twelve characteristics of paddy seed morphological data were evaluated to identify rice varieties. The results summarized that the computer system and artificial intelligence technique in the proposed method successfully identified rice variety identification. The eighteen rice varieties of twenty-six rice varieties demonstrated a proportion of true prediction of more than 70% from a total prediction. |
