Unlock instant, AI-driven research and patent intelligence for your innovation.

Self-adaptive background sound elimination method

An adaptive background and background sound technology, applied in voice analysis, instruments, stethoscopes, etc., can solve problems such as signal distortion, excessive output signal distortion, disappearance, etc., and achieve the effect of background sound interference elimination

Pending Publication Date: 2021-11-09
潘嘉晟
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an adaptive background sound elimination method, which effectively solves the problem that the current background sound elimination research has achieved certain results, but the background sound interference is essentially in the time domain and The frequency domain is sparse, that is, it does not exist all the time, nor does it exist in all frequency bands, so it is not necessary to eliminate the background sound in all time frames and frequency bands, otherwise it will lead to excessive elimination of the results, making some The specific components associated with the disease disappear. In addition, the traditional heart sound denoising method does not introduce a mechanism to control signal distortion after cancellation, resulting in the problem that the output signal may be too distorted

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Self-adaptive background sound elimination method
  • Self-adaptive background sound elimination method
  • Self-adaptive background sound elimination method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] Embodiment one, by Figure 1-3 Provide, the present invention provides a kind of self-adaptive background sound elimination method, comprise the following steps, the judgment on the time frame, the judgment on the frequency domain, the background sound elimination and change into time series signal, the judgment on the described time frame: first to The input data is divided into frames to obtain the kth frame data vector x of the mixed heart sound signal k and the kth frame vector d of the background noise interference signal k , each frame has 256 sampling points, and the overlap between two adjacent frames is 50%, and then the judgment of each frame is made. In order to meet the real-time requirements of the background sound cancellation algorithm in the electronic stethoscope, the Raida criterion is selected. To detect the background sound, that is, calculate the mean value μ of each frame of the background sound data k and standard deviation σ k , set a backgrou...

experiment example

[0074] The evaluation indicators of this experimental example mainly include the following three:

[0075] (1) Output signal-to-noise ratio, oSNR is defined as the clean heart sound signal power P s and background sound signal power P d The ratio of , the larger the value, the less background sound remains in the signal:

[0076]

[0077] (2) Mean square error, MSE means that there is unnecessary information in the denoising signal. For the clean heart sound signal after the background sound is eliminated, its value must be lower than that of the background sound signal, and the smaller the value, the better the denoising effect. It is defined as follows :

[0078]

[0079] Among them, x(j) and s(j) represent the original heart sound signal and the clean heart sound signal after the background sound is eliminated, respectively, and L represents the length of the signal.

[0080] (3) Correlation coefficient, CCF is a common index to measure the effect of denoising. It ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

A self-adaptive background sound elimination method comprises the following steps of judgment on a time frame, judgment on a frequency domain, background sound elimination and conversion into a time sequence signal. Compared with a traditional heart sound denoising method, a mechanism for controlling signal distortion after offset is introduced, so that the problem of overlarge distortion of an output signal is avoided. Whether background sound exists or not is judged in a time domain and a frequency domain, background sound elimination is conducted on a time frame and a frequency band where the background sound exists through frequency domain Wiener filtering, and a signal distortion control factor is added during gain calculation and used for balancing background sound elimination and heart sound signal fidelity. The final contrast experiment proves that the background sound elimination method can effectively eliminate the background sound interference while keeping the heart sound signal true.

Description

technical field [0001] The invention relates to background sound elimination, in particular to an adaptive background sound elimination method. Background technique [0002] For a long time, cardiovascular disease has been one of the main diseases that endanger human life and health. With the aggravation of my country's population aging trend, the situation of prevention and treatment of cardiovascular diseases is becoming more and more severe. Heart sound auscultation is one of the convenient methods for diagnosing cardiovascular diseases. It acquires physiological and pathological sounds accompanied by cardiovascular activities in a non-invasive way, thus giving quick diagnostic results. The use of traditional stethoscopes requires physicians to have rich clinical experience, and the results given are relatively subjective. With the development of computer and Internet of Things technology in recent years, it is possible to realize an electronic stethoscope that can auto...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G10L21/0208G10L21/0224G10L21/0232A61B7/04
CPCG10L21/0208G10L21/0224G10L21/0232A61B7/04
Inventor 潘嘉晟
Owner 潘嘉晟