Remote sensing image amplification method based on PDE and wavelet

Abstract

The invention discloses a remote sensing image amplification method based on PDE and wavelet, comprising the following steps: (1) a remote sensing image is amplified for the first time with a bicubic interpolating method according to the required amplification factor; (2) wavelet decomposition is carried out on a result amplified with the bicubic interpolating method, and low-frequency components obtained after the wavelet decomposition are replaced by original images, and then wavelet synthesis is carried out to obtain a wavelet amplification result; (3) a self-snake model in a PDE method is adopted to carry out post-treatment on the images amplified by wavelet to obtain an amplification image with strengthened and smooth edges. Images amplified with the remote sensing image amplification method have obvious edges. The remote sensing image amplification method eliminates the edge faintness caused by common amplification methods and can make edges smooth. On the basis of the self-snake model method, sawtooth is eliminated, and the edges are smoothened.

Description

technical field [0001] The invention relates to a remote sensing image amplification method based on PDE and wavelet, and belongs to the technical field of remote sensing image processing. Background technique [0002] Image upscaling is an image processing technique that obtains a high-resolution version of a low-resolution image. With the wide application of image processing technology, as the basic operation of image processing, image magnification has important application value. At present, remote sensing is applied in a wide range of fields, such as terrestrial water resource survey, land resource survey, vegetation resource survey, geological survey, urban remote sensing survey, marine resource survey, surveying and mapping, archaeological survey, environmental monitoring and planning management, etc. Remote sensing images are also developing towards high resolution, hyperspectral / hyperspectral and other directions. Although the improvement of remote sensing image r...

AI Technical Summary

Problems solved by technology

However, the existing various image enlargement methods all have their own defects.

Such as the nearest interpolation method, it is easy to cause aliasing effect in the image; the more commonly used bilinear interpolation method still inevitably has blurring at the boundary of the image area; the cubic B-spline interpolation method and Bezier surface interpolation method are better than the above The two methods are better, and can obtain smoother edges, but there will be spots and offsets in light and dark areas; wavelet interpolation effectively enhances the high-frequency information of the image, but the edges are not smooth enough; fractal interpolation maintains the texture of the image , but there is still blurring at the edges

Images