Image denoising method combining Bayes layered learning with space-spectrum combined priori

Abstract

The invention discloses a high spectral image denoising method combining Bayes layered learning with space-spectrum combined priori; the method comprises the following steps: carrying out three dimensional slide block segmentation for a high spectral image according to a space spectrum correlation and non-local self-similarity thereof, and using relative distance priori based on fused features tonon-locally select a plurality of block data pieces to serve as synergy block data, wherein the selected block data is most similar to to-be-observed block data; using layered priori to build a Bayeslow-rank decomposition model, and realizing synergy block data learning and expression. The mode uses low-rank decomposition to carve statistics characteristics of synergy block data, and combines Dirichlet process mixing Gaussian distribution to express noise statistics characteristics; the method uses a variation Bayes method to solve the model, thus effectively reducing image noises. An existing method cannot simultaneously inhibit a plurality of noises in the high spectral image; the high spectral image denoising method combining Bayes layered learning with space-spectrum combined priori can solve said problems, is accurate in result, and can provide strong analysis basis for follow up applications.

Description

technical field [0001] The invention belongs to the field of information technology, and in particular relates to a hyperspectral image noise reduction method combined with Bayesian layered learning and space-spectrum joint prior. Background technique [0002] Hyperspectral images have strong spatial-spectral correlations. Based on this property, block learning has been widely used in hyperspectral image analysis. However, most noise reduction methods based on block learning place all blocks in the same position for learning, ignoring the differences in edges and local structures between different blocks of hyperspectral image imaging, so they cannot effectively express hyperspectral images. . Especially when the pixels in the image block are seriously polluted by noise, the effective information in the image block is very little at this time, so the image block cannot be directly used to restore noise-free data. On the other hand, existing studies have shown that hyperspe...

AI Technical Summary

Problems solved by technology

However, most noise reduction methods based on block learning place all blocks in the same position for learning, ignoring the differences in edges and local structures between different blocks of hyperspectral image imaging, so they cannot effectively express hyperspectral images.

Especially when the pixels in the image block are seriously polluted by noise, there is little effective information in the image block at this time, so the image block cannot be directly used to restore the noise-free data.

On the other hand, existing studies have shown that hyperspectral images are usually polluted by a variety of noises with distinct statistical properties, such as signal-dependent noise, spatial domain or interspectral domain dependent noise, and mixed noise.

Most of the existing denoising algorithms only analyze part of the noise or model the noise with different statistical characteristics separately, which cannot effectively characterize the inherent noise characteristics of hyperspectral images, and perform poorly in the adequacy of noise reduction

Images