A study on the method of transformation using mathematical tool

摘要

Scale is a term with high occurrence frequencies in geography study recent years. Especially, the scale issue in remote sensing is one heated topic in geography field. Although people have pay attention to the issue, there are few studies on the transformation among remote sensing data and the scale transformation is still in the incompetence stage. The author analyzes the issue of scale transformation from the view of mathematics, mainly basing on FOURIER and WAVELET. We can get the approximate information of the image with high resolution by a proper low-pass filter, because two images with the same time, the same area, but the different resolution have the conformability in the spatial frequency spectrum. And we can also get the restructuring graph which is in close proximity to the real image with low resolution by combining the detail information getting from wavelet transformation. The coefficient of correlation is as high as 0.914. And the overall evaluating index suggested in the paper is as high as 98.97%, which verify the method based on mathematics is correct.