A High Speed Mixel Estimation Method by Using the Genetic Algorithm

Author

Ito, Seiji ; Oguro, Yoshinari

Author_Institution

Dept. of Global Environ. Studies, Hiroshima Inst. of Technol., Hiroshima, Japan

fYear

2009

fDate

12-14 Sept. 2009

Firstpage

742

Lastpage

745

Abstract

The mixel (mixed pixel) which includes the multiple categories is one of the error factor in the land cover classification. Also to solve the mixel problem it requires a lot of calculations. A high speed mixel estimation method by using the Genetic Algorithm (GA) is proposed in this paper. In the result of the simulation for the actual remotely sensed data it shows the computation time reduces approximately 75% than the conventional method. For the quantitative evaluation by the simulated mixel data and actual image data, moreover, it is found that the accuracies of the mixel analysis of the proposed method is slightly better than those in some cases.

Keywords

genetic algorithms; geophysical techniques; geophysics computing; image classification; remote sensing; Genetic Algorithm; Hiroshima city; Japan; computation time; conventional method; error factor; high speed mixel estimation method; image data; land cover classification; remotely sensed data; simulated mixel data; Analytical models; Computational modeling; Genetic algorithms; Helium; Image analysis; Image resolution; Indium tin oxide; Signal processing algorithms; Smart pixels; Vegetation mapping; GA; mixel; remotely sensed image;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Information Hiding and Multimedia Signal Processing, 2009. IIH-MSP '09. Fifth International Conference on

Conference_Location

Kyoto

Print_ISBN

978-1-4244-4717-6

Electronic_ISBN

978-0-7695-3762-7

Type

conf

DOI

10.1109/IIH-MSP.2009.87

Filename

5337273