Multi-sound-source localization method for removing outliers by using activity intensity vector weighting
A technology of activity intensity and positioning method, applied in positioning, measurement devices, radio wave measurement systems, etc., can solve the problems of decreased positioning accuracy and mixed outliers, and achieve the effect of improving robustness and good positioning performance
Pending Publication Date: 2021-07-30
BEIJING UNIV OF TECH
View PDF7 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The present invention aims at the problem that in the detection results of the traditional multi-sound source localization method based on a single sound source area, there are always wild points mixed in in the reverberant environment, resulting in a decrease in positioning accuracy. Sound source localization method
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0056]This embodiment is used to detect the positions of multiple sound sources mixed in 150 ms, and the sound sources are located in a silent room environment of 6.0 m×4.0 m×3.0 m. The height of the sound field microphone from the ground is 1.5m, the sound source is located on the horizontal plane 1m away from the sound field microphone, the angular interval of adjacent sound sources is 60°, and the number of sound sources is set to 5. The signal processing software is Matlab2014a.
[0057] During implementation, the present invention is that the algorithm is embedded in the software to realize the automatic operation of each flow process. The present invention will be further explained below with specific implementation steps in conjunction with the accompanying drawings: the concrete work process is as follows:
[0058] Step 1: Time-frequency transform of the signal received by the sound field microphone, divide the time-frequency region, extract the single sound source reg...
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
Login to View More Abstract
The invention provides a multi-sound-source localization method for removing outliers by using activity intensity vector weighting. More accurate orientation estimation is obtained under a reverberation condition by removing time-frequency points with low contribution in a localization process. The multi-sound-source localization method comprises the steps of: firstly, extracting a single sound source region; secondly, calculating an activity intensity vector of each point in the single sound source region and a region average activity intensity vector, solving a point-region activity intensity weight, and weighting time-frequency points; thirdly, acquiring an activity intensity vector of each sound source in a DOA coarse estimation direction, and calculating a point-sound source activity intensity weight; and finally, constructing composite weights, and clustering the composite weights to remove outliers. A histogram is constructed by using composite weights of the residual points, the sound source positions are estimated through post-processing and peak search, the influence of outliers on the positioning precision is minimized, and multi-sound-source accurate positioning in a high-reverberant sound scene is realized.
Description
technical field [0001] The invention belongs to the field of sound source localization in the field of acoustics, and in particular relates to the problem of multi-sound source localization in high reverberation sound scenes Background technique [0002] In recent years, sound source localization technology has attracted the attention of various research teams because of its wide application in the field of audio processing. The spatial position information of all sound sources is obtained from the listening signal recorded by the microphone in the system. The application of sound source localization technology includes but not limited to ambient communication, virtual reality, augmented reality, etc. [0003] In the actual application process, many factors such as the aliasing of recorded signals caused by simultaneous sound sources, the limitation of the microphone array structure, the non-stationary noise in the environment and the existence of reverberation will have a ...
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
Login to View More IPC IPC(8): G01S5/22
CPCG01S5/22
Inventor 贾懋珅高尚李禄张宇
Owner BEIJING UNIV OF TECH