Hyperspectral image target prior optimization method based on multitask sparse learning

Abstract

The invention provides a hyperspectral image target prior optimization method based on multitask sparse learning. The method includes performing pre-detection and set operation; obtaining an over-complete target dictionary and a to-be-reconstructed target pixel; building a multi-task sparse representation model, performing complementary learning on sparse representation models of different targetpixels by utilizing spectral similarity of the target pixels, and using an average sparse coefficient of target atoms as a reconstruction weight of the target atoms for an optimal target prior spectrum, so that a target prior optimization effect is improved. The invention has the advantages that background pixels can be prevented from participating in the target priori optimization process, the spectral similarity of the target pixels is fully utilized, and the optimization effect of the target priori is improved.

Description

technical field [0001] The invention relates to the field of hyperspectral images, in particular to the technical field of hyperspectral image processing, and in particular to a hyperspectral image target prior optimization method based on multi-task sparse learning. Background technique [0002] Object detection refers to the process of separating objects of interest from non-target background objects. Hyperspectral imaging has the characteristics of high spectral resolution and map integration, and can provide diagnostic spectral feature information for distinguishing different substances. Therefore, hyperspectral imaging has unique advantages in target detection. At present, hyperspectral image target detection has been applied in topographic survey, natural resource detection, maritime search and rescue, camouflage target recognition and other fields. [0003] For hyperspectral remote sensing image target detection, scholars at home and abroad have proposed many algorit...

AI Technical Summary

Problems solved by technology

This method takes the background pixels into account, and the obtained prior spectrum of the target can be regarded as a linear mixture of the target spectrum and the background spectrum, not the pure target spectrum of the target object. In addition, this method does not take into account the target The spectral similarity between different pixels of the ground object cannot make full use of the target information in the hyperspectral image

Images