Method for detecting anomaly of hyperspectral image

Abstract

The method discloses a method for detecting anomaly of a hyperspectral image, which can give consideration to advantages of an anomaly detecting method by global and local background modeling. The method comprises the following steps of: projecting a hyperspectral image to orthogonal subspaces of the background based on singular value decomposition so as to obtain a residual image including noise and anomaly; on the basis, introducing a space rank depth for local background modeling and dividing the residual image into two sample sets of noise background and potential anomaly; and finally, carrying out global background modeling by virtue of a multielement gaussian model, calculating the mahalanobis distance of various samples in the potential anomaly set, and comparing the distance with a threshold value to achieve anomaly detection, thus finally generating a binary mapping image capable of reflecting anomaly position coordinates and distribution regulation.

Description

technical field [0001] The invention relates to a spectral image processing method, in particular to a hyperspectral image problem detection method. Background technique [0002] Precise spectral diagnostic information of hyperspectral images is very beneficial to detect objects with subtle differences from the background. Among them, anomaly detection is to detect objects with spectral differences from the surrounding environment under the condition of lack of spectral prior knowledge. It is one of the key technologies for hyperspectral image object detection and has strong practicability. [0003] There are many methods for hyperspectral image anomaly detection. According to the different background modeling methods, they can be divided into two anomaly detection methods: local background modeling and global background modeling, but the two methods can only be used to a certain extent. To adapt to the complexity of hyperspectral image background and abnormal targets, the ...

AI Technical Summary

Problems solved by technology

[0003] There are many methods for hyperspectral image anomaly detection. According to the different background modeling methods, they can be divided into two anomaly detection methods: local background modeling and global background modeling, but the two methods can only be used to a certain extent. To adapt to the complexity of hyperspectral image background and abnormal targets, the anomaly detection method based on local background modeling usually needs to estimate the size of abnormal targets in advance, and the detection performance for larger abnormal targets is poor, while the global background modeling based Although the model anomaly detection method can better detect large anomalies, the ability of the global background model to describe the subtle changes of the entire image background is relatively lacking, and it has a low probability of detecting abnormal objects.

Images