A Locally Constrained Linear Coding Method and System for Hyperspectral Images

Abstract

The present invention is applicable to the technical field of image processing, and provides a locally constrained linear coding method of a hyperspectral image, including: extracting features to be encoded of the hyperspectral image to obtain a feature set; determining the hyperspectral image according to the feature points in the feature set Dictionary; obtain the band information of the feature set and the band information of the word in the dictionary respectively; use the band information of the feature point and the band information of the word as the discriminant constraint item, calculate the Euclidean between the feature point of the hyperspectral image and the word The characteristic encoding coefficient is obtained; according to the characteristic encoding coefficient, the feature of the hyperspectral image is encoded. The embodiment of the present invention introduces the band information of the hyperspectral image while utilizing the local features of the hyperspectral image and the local linear constraints of the visual words, and uses it as a discriminant constraint for mapping feature points to dictionary words, thereby reducing the number of feature points. The ambiguity of the mapping to and from dictionary words enhances the ability to represent hyperspectral images.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a local constrained linear coding method and system for a hyperspectral image. Background technique [0002] In terms of feature coding for general image classification, the existing coding algorithms mainly include hard allocation, sparse coding, local feature aggregation descriptor VLAD (vector of locally aggregated descriptors), Fisher vector and locally constrained linear coding algorithm LLC (Locality- constrained linear coding), etc. [0003] Compared with traditional grayscale images, hyperspectral images contain rich spatial, radiometric and spectral triple information. Extending the traditional LLC applied to ordinary images to 3-dimensional LLC and applying it to hyperspectral image classification can achieve better results, but because images are easily affected by factors such as illumination changes and complex backgrounds in the scene, differen...

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved by the present invention is to provide a locally constrained linear encoding method and system for hyperspectral images, aiming at solving the problem of blurred mapping from feature points to dictionary words in the encoding process of existing encoding methods and weak expressive ability of hyperspectral images The problem

Images