Share Email Print

Proceedings Paper

The selection of optimum scale in coast area classification of high-resolution remote sensing imagery
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Spatial conception exists in remote sensing imagery as well as spectral information. It acts as more importance role in dominant landscape objects detection in high-resolution remote sensing imagery. Multiscale analysis is a new approach to meet the requirement of how to use spatial information in classification. Compared with traditional pixel based classification methods, multiscale analysis is composed of two fundamental components: the generation of a multiscale representation and information extraction. The paper focuses on one segmentation techniques- Fractal Net Evolution Approach (FNEA) and its usage in improvement in coastal remotely sensed image classification. FNEA is considered as one of effectual region-based segmentation and its threshold is a combination of size and homogeneity. We discuss two different segmental strategies which are speed-first and scale-first, and their impacts on image-objects. We can get the optimal segmental scale by analyzing the relationship between average size of each image-object and the different scale.

Paper Details

Date Published: 31 October 2005
PDF: 8 pages
Proc. SPIE 5983, Remote Sensing for Environmental Monitoring, GIS Applications, and Geology V, 59831Q (31 October 2005); doi: 10.1117/12.627375
Show Author Affiliations
Jianyu Chen, Second Institute of Oceanography (China)
Shanghai Institute of Technical Physics, CAS (China)
Delu Pan, Second Institute of Oceanography (China)
Zhihua Mao, Second Institute of Oceanography (China)
Xiaoyu Zhang, Second Institute of Oceanography (China)
Zhejiang Univ. (China)

Published in SPIE Proceedings Vol. 5983:
Remote Sensing for Environmental Monitoring, GIS Applications, and Geology V
Manfred Ehlers; Ulrich Michel, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?