Hyperspectral anomaly detection method based on discrimination forest subspace selection

Abstract

The invention provides a hyperspectral anomaly detection method based on discrimination forest subspace selection, and the method comprises the steps: randomly selecting a part of pixels from a hyperspectral image, constructing a subspace selection isolated binary tree, and constructing an isolated discrimination forest according to the subspace selection isolated binary tree; traversing the hyperspectral image through the constructed isolated discrimination forest, and calculating the average path length; and calculating an abnormal score value of each pixel to realize detection of an abnormal target. According to the invention, an isolated discrimination forest model is used to repeatedly learn and estimate the distribution rule of background classes and abnormal classes in the image inthe form of subsets. On the basis, an axis parallel subspace selection method is introduced, a wave band which is more beneficial to abnormal information discrimination is selected, the problem that abnormal information is buried due to wave band redundancy and too high dimension is avoided, abnormal score value information of the image is solved, and a final result of hyperspectral image abnormaltarget detection is obtained.

Description

technical field [0001] The invention belongs to the technical field of computer image processing, and relates to a hyperspectral image target anomaly detection method, in particular to a hyperspectral anomaly target detection method based on discriminant forest subspace selection. Background technique [0002] Hyperspectral remote sensing imagery combines traditional two-dimensional imaging remote sensing technology and spectral technology, and has the characteristics of high spectral resolution and map-spectrum integration. Each pixel on the image has spectral information of dozens or even hundreds of bands, which can provide diagnostic spectral feature information for distinguishing different substances. Therefore, hyperspectral remote sensing images have the ability to distinguish subtle spectral differences between different substances Ability. This characteristic of hyperspectral remote sensing images enables it to effectively use the fine spectral features of ground o...

AI Technical Summary

Problems solved by technology

[0004] However, these abnormal target detection methods based on statistical distribution modeling still have shortcomings in distinguishing background and abnormal object types. Therefore, there is an urgent need in this field for a detection method that breaks the model limitation and fully estimates and learns background information in hyperspectral images.

Images