粉末冶金课程教学改革实践

Abstract:We put forward a quick clustering method for large numbers of multidimensional images, which is based on the sensitive subspace consisting of the image set's sensitive dimensions. We first estimate the probability density of each dimension by the parzen window algorithm, enhance its optional ability by extracting zero and smoothness processing, then gain the sensitive dimensions to compose the sensitive subspace through recognizing the number of the rallying points, and last do the Rival Penalized Competitive Learning (RPCL) clustering in the subspace. Moreover, we detected the red tide of multidimensional images using this method, which proved it could quickly get similar results with one-ninth time.

Volume:13E

Translation or Not:no