A Binary Social Spider Optimization algorithm for unsupervised band selection in compressed hyperspectral images

作者：

Highlights：

• Proposed a new Binary social spider algorithm for selecting optimal set of bands.

• Feature selection using entropy and first spectral difference in DWT domain.

• Clustering of spectrally unfolded image with K-means algorithm.

• Lower time complexity as compared to the other existing works.

摘要

•Proposed a new Binary social spider algorithm for selecting optimal set of bands.•Feature selection using entropy and first spectral difference in DWT domain.•Clustering of spectrally unfolded image with K-means algorithm.•Lower time complexity as compared to the other existing works.

论文关键词：Discrete wavelet transform,Temporal redundancy,Binary social spider algorithm,Entropy,First spectral derivative

论文评审过程：Received 27 March 2017, Revised 18 December 2017, Accepted 19 December 2017, Available online 22 December 2017, Version of Record 30 December 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.12.034