Limits...
Binaural sound localizer for azimuthal movement detection based on diffraction.

Kim K, Choi A - Sensors (Basel) (2012)

Bottom Line: The gradient analysis of the ILD between the structured and unstructured microphone demonstrates the rotation directions as clockwise, counter clockwise, and no rotation of the sound source.Acoustic experiments with different types of sound source over a wide range of target movements show that the average true positive and false positive rates are 67% and 16%, respectively.Spectral analysis demonstrates that the low frequency delivers decreased true and false positive rates and the high frequency presents increases of both rates, overall.

View Article: PubMed Central - PubMed

Affiliation: Division of Electronics & Electrical Engineering, Dongguk University-Seoul, Seoul 100-715, Korea. kwkim@dongguk.edu

ABSTRACT
Sound localization can be realized by utilizing the physics of acoustics in various methods. This paper investigates a novel detection architecture for the azimuthal movement of sound source based on the interaural level difference (ILD) between two receivers. One of the microphones in the system is surrounded by barriers of various heights in order to cast the direction dependent diffraction of the incoming signal. The gradient analysis of the ILD between the structured and unstructured microphone demonstrates the rotation directions as clockwise, counter clockwise, and no rotation of the sound source. Acoustic experiments with different types of sound source over a wide range of target movements show that the average true positive and false positive rates are 67% and 16%, respectively. Spectral analysis demonstrates that the low frequency delivers decreased true and false positive rates and the high frequency presents increases of both rates, overall.

No MeSH data available.


ROC curve for individual rotation direction, time window, and processing frequency.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3472845&req=5

f13-sensors-12-10584: ROC curve for individual rotation direction, time window, and processing frequency.

Mentions: The ROC curve for the given experimental set is expressed in Figure 13. The column and row of the figure in the cluster stand for rotation direction and time window length respectively. Each plot within the figure is generated from the distinctive range of the processing frequency which is whole or selectively cut frequency. The span of the time window is from a half second to four seconds without overlaping for estimating the output gradient that provides the decision criterion. The filter configuration for frequency selection is designed based on the finite impulse response (FIR) filter by placing the zeros onto the specific location on the unit circle. The order of the FIR filter is determined by the number of zeros which is associated with suppressing frequency; therefore, the number of the removed frequency is proportional to the filter order in the given configuration. Accordingly, the devised filter is the high pass filter, which removes the low frequencies around the several hundred hertz. Note that further selection of eliminated frequency indicates the increased range of removal for low frequency.


Binaural sound localizer for azimuthal movement detection based on diffraction.

Kim K, Choi A - Sensors (Basel) (2012)

ROC curve for individual rotation direction, time window, and processing frequency.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3472845&req=5

f13-sensors-12-10584: ROC curve for individual rotation direction, time window, and processing frequency.
Mentions: The ROC curve for the given experimental set is expressed in Figure 13. The column and row of the figure in the cluster stand for rotation direction and time window length respectively. Each plot within the figure is generated from the distinctive range of the processing frequency which is whole or selectively cut frequency. The span of the time window is from a half second to four seconds without overlaping for estimating the output gradient that provides the decision criterion. The filter configuration for frequency selection is designed based on the finite impulse response (FIR) filter by placing the zeros onto the specific location on the unit circle. The order of the FIR filter is determined by the number of zeros which is associated with suppressing frequency; therefore, the number of the removed frequency is proportional to the filter order in the given configuration. Accordingly, the devised filter is the high pass filter, which removes the low frequencies around the several hundred hertz. Note that further selection of eliminated frequency indicates the increased range of removal for low frequency.

Bottom Line: The gradient analysis of the ILD between the structured and unstructured microphone demonstrates the rotation directions as clockwise, counter clockwise, and no rotation of the sound source.Acoustic experiments with different types of sound source over a wide range of target movements show that the average true positive and false positive rates are 67% and 16%, respectively.Spectral analysis demonstrates that the low frequency delivers decreased true and false positive rates and the high frequency presents increases of both rates, overall.

View Article: PubMed Central - PubMed

Affiliation: Division of Electronics & Electrical Engineering, Dongguk University-Seoul, Seoul 100-715, Korea. kwkim@dongguk.edu

ABSTRACT
Sound localization can be realized by utilizing the physics of acoustics in various methods. This paper investigates a novel detection architecture for the azimuthal movement of sound source based on the interaural level difference (ILD) between two receivers. One of the microphones in the system is surrounded by barriers of various heights in order to cast the direction dependent diffraction of the incoming signal. The gradient analysis of the ILD between the structured and unstructured microphone demonstrates the rotation directions as clockwise, counter clockwise, and no rotation of the sound source. Acoustic experiments with different types of sound source over a wide range of target movements show that the average true positive and false positive rates are 67% and 16%, respectively. Spectral analysis demonstrates that the low frequency delivers decreased true and false positive rates and the high frequency presents increases of both rates, overall.

No MeSH data available.