Noisy data reduction by using tensor and fuzzy c-means algorithm
Mongkol Hunkrajok, Wanrudee Skulpakdee · International Conference on Signal Processing · 2007
Classification of image (both 2D and 3D) and noisy data using eigenvalues of tensor as features is found to be simple, but effective method for reducing noise. The features constitute a systematic structure that can be segmented one from another. We propose the segmentation of class clustering by fuzzy c-mean algorithm which can be applied to classify image and noisy data; thus, unnecessary data from the systems can be removed.