Robust signal processing compression for clustering of speech waveform and image spectrum
- 1 April 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1801-1805
Abstract
We address the issue regarding big data of high dimensional computation. Various statistical and signal processing techniques are used for compression and dimension reduction in last decade for data mining research. Most prominent technique in recent trend of computation for feature extraction is PCA,ICA, MDS, LDA, DWT, DFT and S-transform. These techniques are being used in different dimension of research arena. In this paper we have tried to dissect wavelet transform in speech processing, image processing with respect to time and frequency domain. After noise free decomposition the observation or samples correlation and similarity computation performed to form cluster. The cluster analysis is based on k-means cluster.Keywords
This publication has 4 references indexed in Scilit:
- Neural networks used for speech recognitionJournal of Automatic Control, 2010
- Hierarchical browsing and search of large image databasesIEEE Transactions on Image Processing, 2000
- A theory for multiresolution signal decomposition: the wavelet representationIEEE Transactions on Pattern Analysis and Machine Intelligence, 1989
- Decomposition of Hardy Functions into Square Integrable Wavelets of Constant ShapeSIAM Journal on Mathematical Analysis, 1984