| Summary: | 碩士 === 國立臺北科技大學 === 電機工程研究所 === 103 === Forests in Taiwan distribute vertically along the central region and can be categorized into broadleaved, mixed, and conifer forests. Terrain features make manual inspection of forests nearly impossible. By utilizing remote sensing data, the amount of field sampling could be significantly reduced. However, the visual interpretation is labor-intensive and heavily dependent on the interpreter’s experience. An automatic algorithm, called multi-level morphological active contour algorithm (MMAC), has been proposed to address these issues in 2011.
The MMAC could effectively increase recognition rate of individual tree in mountainous areas, which is the common case in Taiwan. However, the design of algorithm comes with two drawbacks. The first one is false alarm of tree branches since MMAC was detecting tree tops by local maxima and circular shape. The second one is huge computational complexity for delineation of tree crowns, which prevents it from being implemented practically in medium- or large-scale remote sensing data.
This thesis introduced a model based MMAC to address the above issues. Due to the bell curve property of the Gaussian distribution, the distribution could be modeled as the physical shape of a tree. The proposed algorithm takes advantage of the Gaussian model to improve the original MMAC algorithm. The model was utilized to increase the detection rate of tree tops and decrease the computing time in crown delineation. The improved MMAC will be more suitable for practical purpose since it could provide better detection rate with much less computational complexity.
|
|---|
