Compressed Sensing Reconstruction for Hyperspectral Images Using Immune Clone Algorithm Based on Gaussian Dictionary

Li Wang ； Wei Wang ； Boni Liu

《International Core Journal of Engineering》 6卷5期 (2020/05) Pp. 269-275

https://doi.org/10.6919/ICJE.202005_6(5).0039

摘要

An immune clone algorithm (ICA) for compressed sensing (CS) reconstruction for hyperspectral images is proposed. The core idea of compressed sensing theory is that, an image of interest is sparse or compressible in some domain, and then it could be reconstructed accurately through a complex optimization algorithm. ICA is a global optimal search algorithm which could ensure the algorithm converges to the optimal solution with probability 1. The paper takes advantage of the ICA to solve a complex optimization problem to reconstruct the images which are sampled by compressed sensing. The proposed algorithm was evaluated on some hyperspectral images and the results illustrate that the reconstructed images has high peak signal-to-noise ratio (PSNR) owing to the merit of ICA.

關鍵字

Hyperspectral images ； Compressed sensing reconstruction ； Immune clone algorithm ； Reconstructed peak signal-to-noise ratio

參考文獻

REN Z, WU L. Spectral-spatial Classification for Hyperspectral Imagery Based on Intrinsic Image Decomposition [J]. Spacecraft Recovery & Remote Sensing, 2019, 40(3): 111-120. DOI:10.3969/ j.issn. 1009-8518.2019.03.014.

CHEN S, ZHOU Y, QI R. Joint sparse representation of hyperspectral image classification based on kernel function [J]. Systems Engineering and Electronics, 2018, 40(3): 692-698. DOI: 10.3969 /j.issn. 1001- 506X. 2018. 03. 31.

FANG L, HE N, LI S, et al. A New Spatial-Spectral Feature Extraction Method for Hyperspectral Images Using Local Covariance Matrix Representation [J]. IEEE Transactions on Geoscience and Remote Sensing, 2018, 56(6): 3534-3546. DOI: 10.1109/ TGRS.2018.2801387.

Donoho. D. Compressed sensing [J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289-1306.

Candès. E., Romberg. J., and Tao. T. Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information [J]. IEEE Transactions on Information Theory, 2006, 52(2): 489-509.