采用不同融合方法对土地利用现状进行自动分类的研究

LandsatTM图像在土地利用上的应用极为广泛,但TM图像的空间分辨率比较低,因此单独采用TM图像进行土地利用分类的精度也较低。通过融合高空间分辨率全色图像的高频纹理信息来提高TM图像的空间分辨率,使融合后的图像既保留TM图像的多光谱特性,又具有高空间分辨率的特性,易于图像的解译,达到提高自动分类的精度,减少自动分类的误差的目的。笔者运用IHS和KL融合方法,并对生成的新图像作分类精度比较研究,探讨融合图像在土地利用中的应用。

Study on classification of the image fusion in land use

The Landsat TM images have very widely uses in the land use and the result of classification is also low because spatial resolution of TM image is low. The high spatial resolution image and the multi-spectral image are fused, and the fusion can enhance the spatial resolution of TM image. It can produce the new image that has the multi-spectral and high spatial resolution. The fusion image can enhance the degree of the classification and decrease the error of classification. The study on the classification of the Fusion of TM and SPOT Image and application is conducted.