Land cover classification with satellite optical and radar image fusion

Synthetic Aperture Radar (SAR) is widely used in remote sensing and landcover objects. Comparing with normal optical images, SAR could produce more detailed images of objects from large distance, without being affected by weather conditions such as clouds. Since optical images can generate high-reso...

Full description

Saved in:
Bibliographic Details
Main Author: Zhu, Di
Other Authors: Lu Yilong
Format: Final Year Project
Language:English
Published: 2019
Subjects:
Online Access:http://hdl.handle.net/10356/78334
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Synthetic Aperture Radar (SAR) is widely used in remote sensing and landcover objects. Comparing with normal optical images, SAR could produce more detailed images of objects from large distance, without being affected by weather conditions such as clouds. Since optical images can generate high-resolution images, a combination of SAR and optical images could provide a better land cover classification result for further study. This report mainly explores and compares several image processing methodologies applied on SAR images. Our main target is to find a combination of image filtering and segmentation method for better landcover classification result. Processing methods discussed in this report including Wiener filtering, morphological filtering, Chan-Vese segmentation and K-means segmentation. The processed SAR image would then be fused with the optical image of the same area to achieve a more appealing classification result.