A Band Selection Method Based on Chaotic Binary Gravity Search Algorithm

Abstract

The invention relates to a band selection method based on a chaotic binary gravitational search algorithm. The chaotic binary gravitational search algorithm is used to optimize and solve the band selection problem of a hyperspectral image, thereby quickly obtaining a better band subset in the hyperspectral image bands. It can be used in the relevant technical fields of image processing and pattern recognition. The present invention can find a high-quality feasible solution to the band selection problem within an acceptable time cost, it does not need to manually specify the dimension of the band to be selected, and can intelligently achieve a good balance between the correct recognition rate and the dimension of the band , to find a better subset of bands. The invention uses the chaotic binary gravity search algorithm to select the bands of the original band data set of the hyperspectral image, eliminates the irrelevant or redundant bands, takes out the bands that have a greater impact on the correct rate of classification, and reduces the need for irrelevant or redundant bands The calculation time of classification is reduced, and the accuracy and efficiency of image classification are further improved.

Description

Technical field [0001] The invention belongs to the cross application field of hyperspectral image processing and intelligent computing, and relates to the application of swarm intelligence optimization algorithm in image processing, in particular to a solution to the problem of hyperspectral image band selection, and in particular to a chaotic binary gravitation search algorithm The band selection method. Background technique [0002] Hyperspectral remote sensing is one of the major technological breakthroughs made by mankind in earth observation. It uses the nanometer-level spectral resolution of an imaging spectrometer to obtain narrow and spectrally continuous image data. Different from multispectral remote sensing, hyperspectral remote sensing has more abundant ground object spectrum information, which can reflect the subtle spectral properties of the object to be measured in detail, and provides more ground object original data for hyperspectral data processing. At present...

AI Technical Summary

Problems solved by technology

The proposal of these algorithms has greatly helped the dimension reduction of hyperspectral images, but the problem of how to balance the band dimension and the accuracy of data processing has not yet been solved.

In essence, the band selection problem is an NP-hard problem with an exponential computational time complexity.

Images