Application of Unmanned Aerial Vehicle Image To Crop Identification
碩士 === 國立中興大學 === 土木工程學系所 === 103 === This research applied remote sensing technologies to obtain the information of vegetation. Previous automatic identification method was limited by the time and spatial resolution of satellite images. In this study, multi-spectral and high spatial resolution of a...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2015
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/99191971894011446706 |
| Summary: | 碩士 === 國立中興大學 === 土木工程學系所 === 103 === This research applied remote sensing technologies to obtain the information of vegetation. Previous automatic identification method was limited by the time and spatial resolution of satellite images. In this study, multi-spectral and high spatial resolution of aerial images was obtained by using the unmanned aerial vehicle (UAV).
The classification and acreage analysis of vegetation areas could provide the production estimate and periodic environmental detection. The spectral signatures among different species were very similar. In order to improve the classification accuracy, the 3D information was added into the aerial images to aid image interpretation. The result also shows that adding 3D information could improve the classification accuracy. Object-Based classifier is able to consider the relation between pixels, the pixels with higher correlation was considered as one object. Each object will be classified into one class. Object-Based classifier is suitable for agricultural cluster interpretation. The result also shows that the crop interpretation process established by this study has better classification accuracy than pixel-Based classifier in agricultural cluster interpretation.
|
|---|