Rival penalization controlled competitive learning for aerial remote sensing hyper-spectral data clustering with optimizing seed point selection and data dimensionality reduction

Key Words:

Abstract:RPCCL (Rival Penalization Controlled Competitive Learning) algorithm has been extensively used in data clustering problems but its performance is sensitive to the selection of the initial cluster center. This paper, therefore, further investigates the RPCCL and proposes an optimizing seed point selection which chooses non-neighbor data points of greatest local density as seed points. Compared to the existing RPCCL with random seed points, the clustering by this RPCCL with selecting seed points can converge more stably and effectively. Moreover, it is applicable to deal with the data in red tide ARSHD (Aerial Remote Sensing Hyper-spectral Data) image. Additionally, because of the redundancy of high dimensions in the red tide hyper-spectral data, a dimensionality reduction method is also described. The experiments show the promising results of this improved RPCCL approach.

Volume:14

Issue:SUPPL. 2

Translation or Not:no