A Preprocessing Method for Hyperspectral Target Detection Based on Tensor Principal Component Analysis

Abstract

Traditional target detection (TD) algorithms for hyperspectral imagery (HSI) typically suffer from background interference. To alleviate this problem, we propose a novel preprocessing method based on tensor principal component analysis (TPCA) to separate the background and target apart. With the use of TPCA, HSI is decomposed into a principal component part and a residual part with the spatial-spectral information of the HSI being fully exploited, and TD is performed on the latter. Moreover, an effective distinction in scheme can be made between a HSI tensor's spatial and spectral domains, which is in line with the physical meanings. Experimental results from both synthetic and real hyperspectral data show that the proposed method outperforms other preprocessing methods in improving the TD accuracies. Further, target detectors that combine the TPCA preprocessing approach with traditional target detection methods can achieve better results than those of state-of-the-art methods aiming at background suppression.

Supervised Distance-Based Feature Selection for Hyperspectral Target Detection

“…The signs of the values in t are not important, as they only denote the direction of autocorrelation. Therefore, the absolute values of t are employed in Equation (5). Since R is a positive matrix, e also has positive elements in Equation (4).…”

“…(1) For target d L×1 , the TD projector k L×1 in Equation (1) is constructed while using the autocorrelation matrix R L×L defined in Equation (2). (2) e L×1 and t L×1 are calculated based on Equations (4) and (5). (3) For feature f i , the a i is obtained, as follows:…”

“…To continue, similar to what is done with AFS, the target signature and the cluster mean vectors are transferred into the detection space while using the TD projector defined in Equation (1). The target vector in the detection space is obtained using Equation (5). Subsequently, the cluster mean vectors in the detection space are obtained, as follows:…”

“…In Equation (11), each c i represents the mean column vector for cluster i. As in Equations (4) and (5), the asterisk in Equation (10), * , represents the element-wise multiplication. Again, similar to Equations (4) and (5), the absolute values in Equation (10) are used in the FS process.…”

“…Hyperspectral imagery (HSI) have properties that provide scientists with various applications, such as crop and mineral identification in agriculture and geology, improved classification map production [1], subpixel target and anomaly detection [2][3][4][5][6][7], spectral unmixing [8], and data fusion. However, huge data volumes are produced due to the high number of spectral bands.…”