A new method for integration feature extraction of respiratory signals is presented in this paper. According to the characters of the non-stationary sounds, we subdivided the wavelet space in L2(R) based on the wavelet multiscale analysis theory. Advantages of this approach are made both in algorithm and recognition capability. The experimental results demonstrate that the proposed feature extraction method is more powerful and more efficient.
Citation:
Liu Yi, Zhang Caiming, "A New Feature Extraction Method Based on Feature Integration," icicic, vol. 3, pp.170-173, First International Conference on Innovative Computing, Information and Control - Volume III (ICICIC'06), 2006