Researchers Propose a Novel Spectral-Spatial Anomaly Detector for Hyperspectral Data Based on Improved Isolation Forest

Editor: SONG Xiangyu | Oct 9, 2021

 

Recently, a team led by Prof. HE Bin from the Changchun Institute of Optics, Fine Mechanics and Physics (CIOMP) of the Chinese Academy of Sciences proposed a new spectral-spatial anomaly detector for hyperspectral images. Their up-to-date result was published on IEEE Transactions on Geoscience and Remote Sensing.

 

Anomaly detection in hyperspectral images (HSIs) does not need any kind of target information. In other words, anomaly detection aims to locate and search for targets which are generally unknown, but relatively small with low probabilities in an image scene. Anomalous targets are pixels that cannot be identified by prior knowledge, but deviate from the expected behavior and can hold interesting information.

 

Anomaly detection in hyperspectral image is affected by redundant bands and the limited utilization capacity of spectral-spatial information. Thus, it is difficult to distinguish outliers in an unsupervised way. Most methods have concentrated on examination of the role of HSI spectral signatures in anomaly detection, employing exclusively the spectrum of a given pixel to determine its outlier status.

 

Prof. HE and his team proposed a spectral-spatial anomaly detector based on improved isolation forest (SSIIFD) to enhance the detection accuracy by making full use of global and local information, as well as spectral and spatial information of HSIs.

 

The advantages of the proposed SSIIFD method are threefold: first, the method fully utilizes spectral and spatial information in HSIs; second, this method fully employs global and local information in HSIs; third, this method detects anomaly pixels more clearly and accurately at a lower false alarm rate (FAR). The experiments on four real hyperspectral data sets reveal that SSIIFD is stable and superior to other state-of-the-art methods in terms of both objective and subjective evaluations.

 

 

 

Contact:
Author: Dr.SONG Xiangyu 
Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences (CIOMP), Changchun 130033, China, and also with the University of Chinese Academy of Sciences, Beijing 100049, China.
E-mail: songxiangyu17@ mails.ucas.edu.cn
Article Links:
https://ieeexplore.ieee.org/document/9521674

 

 

Fig. 1. Flowchart of the proposed SSIIFD method. (Image by SONG)

 

Fig. 2. San Diego-I data set. (a) Pseudocolor image, (b) ground truth map, and detection maps of (c) RXD, (d) CRD, (e) PTA, (f) KIFD, (g) MFIFD, and (h) proposed SSIIFD.

 

 

 

Join Us
Links
Contact
中文
Copyright 2003-2015 All Rights Reserved Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences
No.3888 Dong Nanhu Road,Changchun,Jilin,China 130033  Email:english@ciomp.ac.cn   TEL:+86-431-86708197  FAX:+86-431-86708198