|
The application of unmanned aerial vehicle images for the classification of coral reef area at Man Nai Island, Rayong Province |
|---|---|
| รหัสดีโอไอ | |
| Creator | Peranun Musikarat |
| Title | The application of unmanned aerial vehicle images for the classification of coral reef area at Man Nai Island, Rayong Province |
| Contributor | Anukul Buranapratheprat, Prasarn Intacharoen |
| Publisher | School of Information and Communication Technology, University of Phayao |
| Publication Year | 2567 |
| Journal Title | The Journal of Spatial Innovation Development |
| Journal Vol. | 5 |
| Journal No. | 3 |
| Page no. | 80-92 |
| Keyword | Unmanned aerial vehicle, Aerial photography, Water column correction, Man Nai Island |
| URL Website | https://ph01.tci-thaijo.org/index.php/jsid/index |
| Website title | The Journal of Spatial Innovation Development |
| ISSN | 2730-1494 |
| Abstract | The application of unmanned aerial vehicles (UAVs) to study the classification of coral reef areas around Koh Man Nai in Rayong Province involved the use of Pixel-based classification. This method utilized three models: Maximum Likelihood (ML), Random Trees (RT), and Support Vector Machine (SVM) at 100-meters and 200-meters flight altitudes, categorizing the reef into three types: coral, rocks, and sand. The assessment of coral reef classification accuracy at 100-meters flight altitude using SVM revealed the highest overall accuracy at 72.73%. The Kappa coefficient was 0.748. Comparatively, ML achieved an overall accuracy of 70.91% with a Kappa coefficient of 0.719, while RT showed the least accuracy at 62.27% with a Kappa coefficient of 0.675. Similarly, at 200-meters flight altitude, SVM demonstrated the highest overall accuracy of 67.27%, with a Kappa coefficient of 0.689. ML and RT achieved lower overall accuracies at 65.45% (Kappa coefficient: 0.651) and 63.3% (Kappa coefficient: 0.633), respectively. The results show that the drone's altitude affects the accuracy of land cover classification. Flying at lower altitudes leads to more accurate land cover classification results. SVM stands out as the model that provides the highest accuracy in classification analysis. |
