A Classification Method for High Spatial Resolution Remotely Sensed Image Based on Human Visual Perception Features

High spatial resolution remotely sensed images have made it possible for humans to observe the earth in detail; however, there are new challenges for processing such kind of images. Some different categories in the images are always classified into one category using spectral features based classification method, because they have similar color or intensity. In recent years, the image processing approach based on human visual perception has been a hot research. A classification approach based on human visual perception features is proposed in this paper. A pair of data sets, one is of Quick bird satellite image, the other is of aerial image, are used to evaluate the proposed classification algorithm. The experimental results show that the feature extraction method based on human visual perception is active and effective for the high spatial resolution remotely sensed image, which results in reasonable image classification.