An optimized K-means Clustering Algorithm Based on BC-QPSO For Remote Sensing Image

The Euclid distance based K-means clustering is among the hard classification algorithms. When dealing with deterministic remote sensing data, it is difficult to gain satisfactory classification results using K-means algorithm. The traditional K-means clustering algorithm is faced with several shortcomings such as locally converged optimization, being sensitive to initial clustering centers, etc. This paper proposes a K-means clustering algorithm based on the Binary Correlation Quantum Behaved Particle Swarm Optimization (BC-QPSO) to relieve the above shortcomings. Convergence is guaranteed in this improved K-means algorithm with probability 1 by means of the powerful global searching ability offered by BC-QPSO. The swarm fitness variance determines the transition between BC-QPSO and K-means. The experiment results on clustering analysis show that the improved K-means clustering algorithm outperforms the traditional algorithm with regard to remote sensing imaging precision.