|
LAND-USE/LAND COVER CLASSIFICATION ANALYSIS USING PIXEL BASED METHODS: CASE OF TAROM CITY, IRAN |
|---|---|
| รหัสดีโอไอ | |
| Creator | Seyyed Behrouz Hosseini, Ali Saremi , Mohammad Hossein Noori Gheydari , Hossein Sedghi, Alireza Firoozfar |
| Title | LAND-USE/LAND COVER CLASSIFICATION ANALYSIS USING PIXEL BASED METHODS: CASE OF TAROM CITY, IRAN |
| Contributor | - |
| Publisher | TuEngr Group |
| Publication Year | 2562 |
| Journal Title | International Transaction Journal of Engineering, Management, & Applied Sciences & Technologies |
| Journal Vol. | 10 |
| Journal No. | 12 |
| Page no. | 10A12G: 1-13 |
| Keyword | Atmospheric Correction, Tarom basin, Maximum Likelihood Classification, Supervised Classification Methods, Land-use map, Landsat-8. |
| URL Website | http://tuengr.com/Vol10_12.html |
| Website title | ITJEMAST V10(12) 2019 @ TuEngr.com |
| ISSN | 2228-9860 |
| Abstract | This research inspects the convenience of Landsat-8 imagery in generating Land-use Land Cover (LULC) maps based on RGB and NIR bands dates back to August 8th, 2017, and at the same time to reveal which type of LULC in Tarom basin can be utilized with maximum accuracy considering the comparison of results with ground samples. Besides necessary preprocessing, land-use classification was done after atmospheric corrections (via FLAASH Algorithm). LULC maps were generated using three pixel-based supervised classification methods, Maximum Likelihood (ML), Support Vector Machine (SVM), and Artificial Neural Networks (ANNs). Results proved that imagery precision based on Kappa statistics and overall accuracy for ML classification method were 0.88 and 91.55, respectively. The acquired outcome indicated that Landsat-8 OLI data, present satisfying LULC classification in the waterbody, mountain and rock, bare land, vegetation, and forest classes. In addition, as the results indicated, it can be stated that all three methods of classification in a region of considerable heterogeneity in terms of elevation (between 280-3000 m), land-use and vegetation such as Tarom, can have significant results. In comparison with the other two methods, classification with the ML method had higher speed and lower complexity for execution in achieving the required maps. |
