DocumentCode :
693850
Title :
Research on Reefs Extraction Method Based on Multi-spectral Remote Sensing
Author :
Gang Li ; Qingzhong Wang
Author_Institution :
Coll. of Autom., Harbin Eng. Univ., Harbin, China
fYear :
2013
fDate :
14-16 Nov. 2013
Firstpage :
14
Lastpage :
17
Abstract :
Reefs Extraction from satellite remote sensing images is used widely in maritime safety, marine surveying and marine military. Based on analyzing and processing IKONOS satellite remote sensing data, a reefs extraction algorithm is proposed in this paper, in which principal components analysis transformation is combined with edge enhancement. The principal component analysis transformation is applied for all the four bands of the IKONOS data, and the edge is enhanced by using the sharpened near-infrared image. Then Gauss Laplace operator is used to extract the edge of the reefs. The results show that the proposed method removes the linear correlated redundant information in the multi-spectral bands, and higher extraction accuracy of reefs edge can be obtained. From the view of computational complexity and computational efficiency, this method is also better than other algorithms.
Keywords :
feature extraction; geophysical image processing; image enhancement; oceanographic techniques; remote sensing; Gauss Laplace operator; IKONOS satellite remote sensing data; edge enhancement; marine military; marine surveying; maritime safety; multispectral remote sensing; principal components analysis transformation; reefs extraction algorithm; reefs extraction method; satellite remote sensing images; sharpened near-infrared image; Algorithm design and analysis; Data mining; Educational institutions; Image edge detection; Principal component analysis; Remote sensing; Satellites; Atmospheric Correction; Edge Enhancement; Edge Stretching; LOG Operater Itroduction; Principle Component Analysis;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Business Intelligence and Financial Engineering (BIFE), 2013 Sixth International Conference on
Conference_Location :
Hangzhou
Print_ISBN :
978-1-4799-4778-2
Type :
conf
DOI :
10.1109/BIFE.2013.4
Filename :
6961081
Link To Document :
بازگشت