A study of image classification based on MLC combined with spectral angle
The image information in abundance,reflects the feature of the objects. Spectral angle emphasizes the shape of the spectral and it isn't sensitive to the enhancement of the image. Maximum Like hood Classification MLC) is one of popular methods for remote sensing image classification. Spectral angle expressed as probability is added to the discriminant function. By this way, spectral angle is used to classify image as a feature. The experimental classification is carried out and the results show that the classification based on MLC combined with spectral angle has a higher accuracy compared with traditional MLC and SAM.
image classificationspectral angleMLCdiscriminant function
NETL
NSTL
万方数据
维普
