High Spatial Resolution Remote Sensing Image Classification Based on Pixel Shape Index Method

Then the pixel shape index method is used to extract features of high spatial resolution remote sensing images and supplement spectral features.The experimental results show that feature can effectively distinguish ground objects with similar but different geometric shapes, superior classification in accuracy.Compared small ripple multi-scale region method, also achieves better results.On other hand, it found experiment easily affected by detailed information image, effect rich not ideal.In this paper, object targets high-resolution images. Compared features, have great advantages accuracy, conducive features.