用于信號(hào)特征提取和重建的脊提取算法
Characterization extraction and reconstruction of signals by the ridges of continuous wavelet transform
-
摘要: 該文論述了以信號(hào)在連續(xù)小波變換域上形成的脊作為信號(hào)特征進(jìn)行信號(hào)的特征提取和信號(hào)重建的方法,并且將該理論應(yīng)用到話音信號(hào)的處理中。仿真實(shí)驗(yàn)表明,對(duì)于最高頻率為4kHz的語(yǔ)音信號(hào)來(lái)說(shuō),提取3～5條脊即可以很好地刻劃信號(hào)的全部信息,重建的信號(hào)在主觀感覺(jué)上達(dá)到了較好的效果。
-
關(guān)鍵詞:
- 連續(xù)小波變換; 脊提取; 信號(hào)重建
Abstract: The theory of detecting ridges in the modulus of the continuous wavelet transform is presented as well as reconstructing signal by using information on ridges. Application of the theory on speech signal is discussed in detail. Experimental result shows that as for signals with highest frequency at 4kHz, 3~5 ridges is enough to describe all of the information contained in the signal. The reconstructed signal reserves most of the necessary information of the original signal. -
S. Mallat, S. Zhang, Matching pursuit with time-frequency dictionaries, IEEE Trans. on Signal Processing, 1993, SP-41(12), 3397-3415.[2]N. Delprat, et al., Asymptotic wavelet and Gabor analysis: Extraction of instantaneous frequencies, IEEE Trans. on Info. Theory, 1992, IT-38(2), 644-664.[3]R. Carmona, W. L. Hwang, B. Torresani, Characterization of signals by the ridges of their wavelet transforms, IEEE Trans. on Signal Processing, 1997, SP-45(10), 2586-2590.[4]R. Carmona, W. L. Hwang, B. Torresani, Multiridge detection and time-frequency reconstruction,IEEE Trans. on Signal Processing, 1999, SP-47(2), 480-492.[5]R. Carmona, W. L. Hwang, B. Torresani, Practical Time-Frequency Analysis, San Diego, Academic Press, 1998, Chapter 8. -
計(jì)量
- 文章訪問(wèn)數(shù): 2273
- HTML全文瀏覽量: 108
- PDF下載量: 670
- 被引次數(shù): 0
下載:
