Adaptive autoregressive model-based hyper-spectral imagery super-resolution method

Abstract

The invention belongs to the technical field of computer vision, and provides a hyper-spectral imagery super-resolution acquisition method which has wide application and can obtain high-quality and hyper-spectral images. For the purpose, the adopted technical scheme is that the adaptive autoregressive model-based hyper-spectral imagery super-resolution method comprises the following steps: amplifying the image of each waveband from the hyper-spectral image of the first wave band by means of high-resolution color images in sequence; searching an image with the highest similarity to an input ith waveband image from three closest wave band images for the ith waveband image, and then projecting the image having exceeded the super-resolution corresponding to the most similar image to the current waveband to obtain a projection image; realizing the super-resolution of the image of the current ith waveband through an adaptive autoregressive model based on the projection image and the high-resolution color image to finally realize the image super-resolution of all wavebands. The method is mainly applied to image processing.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and relates to an autoregressive theory, a signal projection transformation space theory and a hyperspectral image super-resolution method, specifically, to a hyperspectral image superresolution method based on an adaptive autoregressive model. technical background [0002] For a long time, the restoration of high-resolution hyperspectral images from low-resolution hyperspectral images has made important breakthroughs in key technologies, and has become mature and widely used in remote sensing geography, medical images, agricultural science, and climate science. application. However, the traditional method of super-resolution hyperspectral images has certain limitations on the spectrum, and the super-resolution results are not satisfactory. The hyperspectral image super-resolution method based on an adaptive autoregressive model can achieve higher resolution with the help of a high-resol...

AI Technical Summary

Problems solved by technology

At present, one method of common hyperspectral super-resolution technology is to reconstruct a high-resolution hyperspectral image from a single or multiple low-resolution hyperspectral images, for example (Hyperspectral imagery super-resolution by sparse representation and spectral regularization, Yongqiang Zhao*, Jinxiang Yang, Qingyong Zhang, Lin Song, Yongmei Cheng and Quan Pan, Journal on Advances in Signal Processing, 2011), however, such methods can achieve The resolution magnification is very limited, and the result is not very good; another method is to obtain a high-resolution hyperspectral image by means of a fusion of a high-resolution color image and a low-resolution hyperspectral image, for example (High -resolution Hyperspectral Imaging via Matrix Factorization, Rei Kawakami et al.CVPR2011), but this method has great limitations on the band of hyperspectral images, requiring the input hyperspectral images to be images in the visible light band, so this method cannot To achieve a wide range of practical applications

Images