Share Email Print

Proceedings Paper

Validation of the morphological compositing method for ZY-3 satellite imagery
Author(s): Shuna Feng; Yindi Zhao
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Each scene of image generated from earth observation satellites can only cover a certain area. When one scene cannot cover a user’s area of interest, two or more scenes are needed to be registered and combined into a single image, and this composition process is referred to as image mosaicking. The key issue in image composition is to decide where to place the seam line in overlapping region. The optimal seam line which joins several scenes of images covered the entire study area, is usually determined by texture and other characteristics of the overlap region for seamless quality. Recently, a morphological image compositing algorithm was proposed which is able to automatically delineate seam lines along salient image structures. And this algorithm uses the ideas of marker-controlled segmentation for image mosaicking and divides the overlap region into a determined number of areas. The resulting seam lines of the morphological image compositing algorithm cut along high gradient regions which are object edges in initial images. However, the morphological compositing method only applied to delineate the invisible seam line to the human eyes based on Landsat ETM+ data which is the representation of medium resolution data. In this paper, we test the validation of the morphological compositing method to generate visually pleasing seam line for image mosaic without changing the image radiometry and feasibility to handle two adjacent scenes simultaneously on high spatial resolution imagery by using ZY-3 multispectral image data. The focus of this paper is developing a quantitative evaluation measure which is usually formulated as the sum of morphological gradient of the image mosaic along the seam line divided by the length of the seam to quantitatively estimate the ‘quality’ of the automatically delineate seam line.

Paper Details

Date Published: 18 November 2014
PDF: 8 pages
Proc. SPIE 9263, Multispectral, Hyperspectral, and Ultraspectral Remote Sensing Technology, Techniques and Applications V, 926322 (18 November 2014); doi: 10.1117/12.2068895
Show Author Affiliations
Shuna Feng, China Univ. of Mining and Technology (China)
Yindi Zhao, China Univ. of Mining and Technology (China)
Jiangsu Key Lab. of Resources and Environmental Information Engineering (China)

Published in SPIE Proceedings Vol. 9263:
Multispectral, Hyperspectral, and Ultraspectral Remote Sensing Technology, Techniques and Applications V
Allen M. Larar; Makoto Suzuki; Jianyu Wang, 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?