| Author | Witchawat Sinchai |
| Call Number | AIT Thesis no.RS-09-18 |
| Subject(s) | Trees--Identification--Remote sensing
|
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Remote Sensing and Geographic Information System |
| Publisher | Asian Institute of Technology |
| Series Statement | Thesis ; no. RS-09-18 |
| Abstract | Tree species identification is essential for tropical rain forest management. Remote sensing is one of the most prospective techniques suitably for this purpose. Image captured from satellites or airborne will be useful for remote sensing techniques. In the near future, the satellite may provide higher spatial resolution than the existing resolution. The procedure of analyzing this higher spatial resolution image shall undertake to pursue full advantage of the emerging technology. Hence, the main objective of this study is to establish a method for identifying tree species from higher spatial resolution. This study would be applied optical image captured by digital camera as data source to imitate the image from a satellite. The Benjasiri Park is the study area where is located western of the Emporium Suit Hotel. The resolution image captured from the top of this building is 0.7 cm/pixel. This study applied texture analysis techniques to identify tree species from super high resolution (less than 1 cm/pixel). The texture analysis techniques would be excellently performed and suitably used for this study. The GLCM is another technique which would be applied in this study. The resolution would be diminished in order to define the limitation of applicability of texture analysis in identifying tree species. The result of this study explained that the accomplishment of GLCM with appropriate parameter can identify tree species at 0.7 cm/pixel. The proper parameters depend on the perception of each texture. Thus, each tree species well perform with different parameter. Nevertheless, the parameters should be large enough to perceive the texture of tree species. Although five statistic features of GLCM were applied, only three of them can identify tree species with all parameters; i.e. contrast, entropy, and energy. This study diminished the resolution image into two resolution; i.e. 1.4 and 2.1 cm/pixel. As the result of analyzing this diminished resolution image by GLCM, both resolutions can identify tree into species. However, lower resolution image (2.1cm/pixel) was worse capability than higher resolution image (1.4 cm/pixel). The limitation of this process was likely occurred in larger resolution image. In conclusion, the requirement of resolution image should be equal or more than 2.1 cm/pixel. |
| Year | 2009 |
| Corresponding Series Added Entry | Asian Institute of Technology. Thesis ; no. RS-09-18 |
| Type | Thesis |
| School | School of Engineering and Technology |
| Department | Department of Information and Communications Technologies (DICT) |
| Academic Program/FoS | Remote Sensing (RS) |
| Chairperson(s) | Kibe, Seishiro |
| Examination Committee(s) | Kusanagi, Michiro;Taravudh Tipdecho |
| Scholarship Donor(s) | Royal Thai Government |
| Degree | Thesis (M.Eng.) - Asian Institute of Technology, 2009 |