• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Sixth International Conference on Electronic Measurement amp; Instruments (ICEMI '2003) Vol.1; Aug 18-21, 2003; Taiyuan, China >A Robust Denoising Algorithm Based on Discrete Wavelet Transform (DWT)
【24h】

A Robust Denoising Algorithm Based on Discrete Wavelet Transform (DWT)

机译：基于离散小波变换（DWT）的鲁棒去噪算法

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Signal denoising is very important in speech transmit and speech coding. In this paper, an effective and simple denoising method by DWT is proposed. The traditional methods usually filter the signal in time or frequency domain. The properties usually become poorer when the energy of the noise are high. Therefore a new robust denoising method is desired. In this paper, base on the theory that human ear has the similar frequency distinguish ability as DWT, we filter the original speech in wavelet domain. Firstly, we transform the raw signal into wavelet domain, and then using the wavelet coefficients of 2, 3 and 4 order to synthesize the signal we need. After experiments in different noising circumstance, we conclude the new denoising method is a simple and robust one.
机译：信号降噪在语音传输和语音编码中非常重要。本文提出了一种有效且简单的DWT去噪方法。传统方法通常在时域或频域对信号进行滤波。当噪声的能量很高时，性能通常会变差。因此，需要一种新的鲁棒去噪方法。本文基于人耳具有与DWT相似的频率区分能力的理论，在小波域对原始语音进行过滤。首先，将原始信号转换到小波域，然后使用2、3和4阶的小波系数合成所需的信号。经过在不同噪声环境下的实验，我们得出结论：这种新的去噪方法是一种简单而强大的方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号