A Novel Change Detection Method for Unregistered Optical Satellite Images

Wang LUO; Hongliang LI; Guanghui LIU; Guan GUI

doi:10.1587/transcom.E95.B.1890

A Novel Change Detection Method for Unregistered Optical Satellite Images

Wang LUO, Hongliang LI, Guanghui LIU, Guan GUI

Full Text Views

0

Share
Cite this

Summary :

In this letter, we propose a novel method for change detection in multitemporal optical satellite images. Unlike the tradition methods, the proposed method is able to detect changed region even from unregistered images. In order to obtain the change detection map from the unregistered images, we first compute the sum of the color difference (SCD) of a pixel to all pixels in an input image. Then we calculate the SCD of this pixel to all pixels in the other input image. Finally, we use the difference of the two SCDs to represent the change detection map. Experiments on the multitemporal images demonstrates the good performance of the proposed method on the unregistered images.

Publication: IEICE TRANSACTIONS on Communications Vol.E95-B No.5 pp.1890-1893

Publication Date: 2012/05/01

Publicized

Online ISSN: 1745-1345

DOI: 10.1587/transcom.E95.B.1890

Type of Manuscript: LETTER

Category: Multimedia Systems for Communications

Cite this

Copy

Wang LUO, Hongliang LI, Guanghui LIU, Guan GUI, "A Novel Change Detection Method for Unregistered Optical Satellite Images" in IEICE TRANSACTIONS on Communications, vol. E95-B, no. 5, pp. 1890-1893, May 2012, doi: 10.1587/transcom.E95.B.1890.
Abstract: In this letter, we propose a novel method for change detection in multitemporal optical satellite images. Unlike the tradition methods, the proposed method is able to detect changed region even from unregistered images. In order to obtain the change detection map from the unregistered images, we first compute the sum of the color difference (SCD) of a pixel to all pixels in an input image. Then we calculate the SCD of this pixel to all pixels in the other input image. Finally, we use the difference of the two SCDs to represent the change detection map. Experiments on the multitemporal images demonstrates the good performance of the proposed method on the unregistered images.
URL: https://globals.ieice.org/en_transactions/communications/10.1587/transcom.E95.B.1890/_p

Copy

@ARTICLE{e95-b_5_1890,
author={Wang LUO, Hongliang LI, Guanghui LIU, Guan GUI, },
journal={IEICE TRANSACTIONS on Communications},
title={A Novel Change Detection Method for Unregistered Optical Satellite Images},
year={2012},
volume={E95-B},
number={5},
pages={1890-1893},
abstract={In this letter, we propose a novel method for change detection in multitemporal optical satellite images. Unlike the tradition methods, the proposed method is able to detect changed region even from unregistered images. In order to obtain the change detection map from the unregistered images, we first compute the sum of the color difference (SCD) of a pixel to all pixels in an input image. Then we calculate the SCD of this pixel to all pixels in the other input image. Finally, we use the difference of the two SCDs to represent the change detection map. Experiments on the multitemporal images demonstrates the good performance of the proposed method on the unregistered images.},
keywords={},
doi={10.1587/transcom.E95.B.1890},
ISSN={1745-1345},
month={May},}

Copy

TY - JOUR
TI - A Novel Change Detection Method for Unregistered Optical Satellite Images
T2 - IEICE TRANSACTIONS on Communications
SP - 1890
EP - 1893
AU - Wang LUO
AU - Hongliang LI
AU - Guanghui LIU
AU - Guan GUI
PY - 2012
DO - 10.1587/transcom.E95.B.1890
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E95-B
IS - 5
JA - IEICE TRANSACTIONS on Communications
Y1 - May 2012
AB - In this letter, we propose a novel method for change detection in multitemporal optical satellite images. Unlike the tradition methods, the proposed method is able to detect changed region even from unregistered images. In order to obtain the change detection map from the unregistered images, we first compute the sum of the color difference (SCD) of a pixel to all pixels in an input image. Then we calculate the SCD of this pixel to all pixels in the other input image. Finally, we use the difference of the two SCDs to represent the change detection map. Experiments on the multitemporal images demonstrates the good performance of the proposed method on the unregistered images.
ER -