An Improved Quantum-Behaved Particle Swarm Optimization for Endmember Extraction
Authors:
- 3 April 2019
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 57 (8) , 6003-6017
Abstract
Endmember extraction (EE) plays an important role in the quantitative analysis of hyperspectral images, as the main step in the decomposition of mixed pixels. At present, scholars have proposed many EE algorithms based on the linear spectral mixture model and the convex geometry principle, such as the pixel purity index (PPI) and the vertex component analysis (VCA). At the same time, many intelligent optimization algorithms, such as the particle swarm optimization (PSO) and the discrete PSO (DPSO), have been applied to EE, which can get promising results for real images. However, PSO and DPSO have theoretical limitations and cannot guarantee the global convergence. The problem of premature convergence will reduce the accuracy of the EE result. The quantum-behaved PSO (QPSO) can theoretically guarantee the convergence of the algorithm by combining the quantum mechanics into the PSO. In order to increase the accuracy of the algorithm, this paper proposes an improved QPSO (IQPSO) algorithm for EE. IQPSO has made innovations in population coding and initialization methods. Besides, the collaborative approach for updating the optimal positions of particles can help to solve the difficulties caused by high dimensions. The experimental results show that IQPSO can extract endmembers efficiently and effectively.Keywords
Funding Information
- National Natural Science Foundation of China (41871243, 41431175, 61822113)
- National Key R&D Program of China (2018YFA0605501)
- Natural Science Foundation of Hubei Province (2018CFA050)
This publication has 45 references indexed in Scilit:
- Efficient, simultaneous detection of multi-class geospatial targets based on visual saliency modeling and discriminative learning of sparse codingISPRS Journal of Photogrammetry and Remote Sensing, 2014
- A Signal Processing Perspective on Hyperspectral Unmixing: Insights from Remote SensingIEEE Signal Processing Magazine, 2013
- Quantum-Behaved Particle Swarm Optimization: Analysis of Individual Particle Behavior and Parameter SelectionEvolutionary Computation, 2012
- Comparative study between a new nonlinear model and common linear model for analysing laboratory simulated‐forest hyperspectral dataInternational Journal of Remote Sensing, 2009
- Hyperspectral remote sensing of plant pigmentsJournal of Experimental Botany, 2006
- A Cooperative Approach to Particle Swarm OptimizationIEEE Transactions on Evolutionary Computation, 2004
- A Quantitative and Comparative Analysis of Endmember Extraction Algorithms From Hyperspectral DataIEEE Transactions on Geoscience and Remote Sensing, 2004
- Spectral unmixingIEEE Signal Processing Magazine, 2002
- Fully constrained least squares linear spectral mixture analysis method for material quantification in hyperspectral imageryIEEE Transactions on Geoscience and Remote Sensing, 2001
- Imaging Spectrometry for Earth Remote SensingScience, 1985