Hyperspectral image denoising method based on non-convex low-rank matrix approximation and total variation regularization

Abstract

The invention discloses a hyperspectral image denoising method based on non-convex low-rank matrix approximation and total variation regularization. The hyperspectral image denoising method comprisesthe following steps: 1) acquiring hyperspectral image data to be denoised, wherein M, N and K represent the number of width and height spectral bands of a hyperspectral image respectively; 2) constructing a denoising model based on non-convex low-rank matrix decomposition and TV regularization; (3) optimizing the model by adopting an Augmented Lagrange function (ALM) algorithm, and optimizing themodel by adopting an ALM algorithm; 4) optimizing and solving the model by adopting an augmented Lagrange function algorithm; and 5) outputting the hyperspectral image after the mixed noise is removed.

Description

technical field [0001] The invention relates to the field of hyperspectral image processing, in particular to a hyperspectral image denoising method based on non-convex low-rank matrix approximation and full variation regularization. Background technique [0002] Remote sensing technology has undergone major changes and innovations in terms of theory, technology and application. Among them, the emergence and development of hyperspectral image (HSI) technology is more prominent. Hyperspectral image data is acquired by a hyperspectral resolution sensor and is composed of hundreds of adjacent narrow spectral band images, which can provide spectral information of hundreds of continuous bands of the same scene. Therefore, hyperspectral images are widely used in food safety, biomedical imaging, military surveillance and other fields. However, during the imaging process of the sensor, hyperspectral images are inevitably polluted by various noises, including Gaussian noise, impuls...

AI Technical Summary

Problems solved by technology

The above algorithms all construct the target model through prior conditions such as low rank, convex function, and neighborhood similarity to achieve the denoising effect. Among them, the most classic regular constraint item of the low rank model is the nuclear norm, which combines all singular values Addition does not treat different singular values ​​equally like the rank function, which means that larger singular values ​​will be penalized more

Images