An Image Analysis Of Sound By Dual Microphone Perception System

At present, the research of sound source localization technology based onmicrophone array has a large development, Sound source localization method basedon time delay estimation becoming gradually systematic. Sound source localizationtechnology based on microphone array can be divided into three categories: one soundsource localization technology is formed based on the maximum output power of thesteerable beam; another sound source localization technology based onhigh-resolution spectral estimation; the third sound source localization technologybased on the time difference of arrival (TDOA), this is time delay estimation.Currently the third method has been most widely used, although the algorithm itselfhas still many inadequacies, but because of its small amount of computation, real-time,low-cost, and after proper improvement this method can realize good positioningaccuracy even in the presence of noise and reverberation, therefore this method issuitable for real-time applications in practice. This paper introduces sound sourcelocalization technology mainly based on this method.The sound source localization technique achieves a sound source positioningmainly by analysis and study of the microphone array, and by series processing theacquisition signal. Sound source localization technique based on time delay estimationmainly estimating time difference between the sound source reach the twomicrophones, because each time difference corresponding to only one hyperboloid, soa plurality of microphones can be corresponding to plurality of hyperboloid theintersection of these hyperboloids is the position of the sound source.Human ear is the main organ that perception outside sounds, its internalcomplexity and special sense of sound mechanisms generate a lot of inspiration forour research on microphone array sound source localization. we can even directlycompare human ear to the microphone, then the binaural will composed a microphonearray. Sound source localization technology based on the mechanism of the human auditory has been widely applied in the field of robotics, and achieved good results.The research object of this paper is a dual-microphone array composed by binaural,firstly, introduce conduction mechanisms of sound in the ear from biological point,then rise the spread of the sound source and positioning to theory height, based on theJeffress model, position the sound source by the time difference and intensitydifference of the interaural. Use the theory of small aperture array positioning methodand the positioning accuracy, through the computer software (matlab) analysis,processing the perception of a sound signal by binaural, and eventually express theresult of the analysis in the form of image, so the position of the sound source by thearray will be more intuitive, so we can realize visual observation of sound sourcelocalization.The main innovation of this study:1. Link the array sound source localization technology with the human earperceives, explore a sound source localization technology based on the auditoryperception of binaural. Research the sound perception of binaural from the view ofbiology, physics and psychology. Let the wonderful acoustic phenomenon becomesscientific, concrete, and more interesting.2. Research a Fine Interpolation of Correlation Peak (FICP) time delayestimation algorithm based on Modified Chirp Z Transform (MCZT), this algorithmimproved the positioning accuracy, apply it to small aperture array for positioning themoving sound source. This algorithm has a high Computing speed, a small amount ofcalculation, and high accuracy on delay estimation.3. Express sound perception of binaural and the position of the sound source inthe form of image, by comparing various sound son graphic, find the spectrogram thedifference between the pleasing and offensive sound. By observing the distribution ofthe energy of the band, we can identify the sound source localization in the mostaccurate angle. We can apply this conclusion to arrange a concert hall or a broadvenue, to make people feel the best sound effects.