Satellite polSAR microwave imaging of Singapore and its surrounding area

Nowadays, one major application of remotely sensed data is land cover classification. This report introduces the procedure of land cover classification of polSAR image. One step of land cover classification is image segmentation, it is an important pre-process for final land cover classification....

Full description

Saved in:
Bibliographic Details
Main Author: Dai, Ruozhen
Other Authors: Lu Yilong
Format: Final Year Project
Language:English
Published: 2016
Subjects:
Online Access:http://hdl.handle.net/10356/68323
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Nowadays, one major application of remotely sensed data is land cover classification. This report introduces the procedure of land cover classification of polSAR image. One step of land cover classification is image segmentation, it is an important pre-process for final land cover classification. Two image segmentation methods are introduced, one is ChanVese active contours without edges and the other is watershed segmentation. The matlab implementation of the two methods on a polSAR image provided was discussed. Once segmented result was produced from original polSAR image, image fusion between segmented polSAR image and optical image is conducted to get a more comprehensive image of the area. Through compare between the two different segmentation methods, the limitations, advantages, disadvantages of them are found out.