Hyperspectral target detection based on transform domain adaptive constrained energy minimization

• Published in: International journal of applied earth observation and geoinformation Vol. 103; p. 102461
• Main Authors: Zhao, Xiaobin, Hou, Zengfu, Wu, Xin, Li, Wei, Ma, Pengge, Tao, Ran
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.12.2021; Elsevier
• Subjects: Constrained energy minimization; Fractional Fourier transform; Hyperspectral imagery; Multi-direction double window; Target detection; Constrained energy minimization; Multi-direction double window; Hyperspectral imagery; Target detection; Fractional Fourier transform
• ISSN: 1569-8432; 1872-826X
• DOI: 10.1016/j.jag.2021.102461

•A transform domain-based revised constrained energy minimization detector is proposed.•The fractional Fourier transform is adopted to improve the separability of background and target.•Multi-direction double window summation strategy is used to make full of the local spatial statistical information.•Pearson correlation coefficient is weighted to further adaptively adjust detection accuracy. Traditional hyperspectral target detection methods use spectral domain information for target recognition. Although it can effectively retain intrinsic characteristics of substances, targets in homogeneous regions still cannot be effectively recognized. By projecting the spectral domain features on the transform domain to increase the separability of background and target, fractional domain-based revised constrained energy minimization detector is proposed. Firstly, the fractional Fourier transform is adopted to project the original spectral information into the fractional domain for improving the separability of background and target. Then, a newly revised constrained energy minimization detector is performed, where sliding double window strategy is used to make the best of the local spatial statistical characteristics of testing pixel. In order to make the best of inner window information, the mean value of Pearson correlation coefficient is measured between prior target pixel and testing pixel associated with its four neighborhood pixels. Extensive experiments for four real hyperspectral scenes indicate that the performance of the proposed algorithm is excellent when compared with other related detectors.