Apparatus and method for recognizing horn using sound signal processor

• US 9,615,186 B2
• Filed: 12/13/2014
• Issued: 04/04/2017
• Est. Priority Date: 10/15/2014
• Status: Active Grant

• Current Assignee: Hyundai Motor Company (Hyundai Motor Group)
• Original Assignee: Hyundai Motor Company (Hyundai Motor Group)
• Inventors: Kim, Jae Kwang, Jang, Yoon Ho
• Primary Examiner(s): Sniezek, Andrew L
• Application Number:
  US14/569,687
• Publication Number:
  US 20160112816A1
• Time in Patent Office: 843 Days
• CPC Class Codes: B60Q 9/008 for anti-collision purposes
• CPC Class Codes: B60W 30/12 Lane keeping
• CPC Class Codes: G08G 1/0965 responding to signals from another vehicle, e.g. emergency vehicle
• CPC Class Codes: G08G 1/166 for active traffic, e.g. moving vehicles, pedestrians, bikes
• CPC Class Codes: G08G 1/167 Driving aids for lane monitoring, lane changing, e.g. blind spot detection
• CPC Class Codes: H04R 29/00 Monitoring arrangements; Testing arrangements for hearing aids H04R25/30; detection of loudspeaker connection H04R5/04; sound-field adaptation dependent on speaker detection H04S7/308

Abstract

An apparatus and a method for recognizing a horn using a sound signal processor are provided. The apparatus includes a data storage unit configured to acquire and store surrounding sound data and a sound signal processor configured to sample a sound signal within the sound data, perform signal conversion, and define and extract the frequency signals. In addition, a horn recognizer is configured to combine the extracted frequency signals, accumulate the frequency signals, and digitize a change rate of the accumulated frequency signals. Further, an output data determiner is configured to determine whether the digitized change rate of the frequency signals is equal to or greater than a threshold value to determine whether and when the horn is generated.

Claims

1. An apparatus for recognizing a horn using a sound signal process, comprising:

a memory configured to store program instructions;

a sound signal processor configured to execute program instructions, the program instructions when executed configured to;

acquire and store surrounding sound data from a microphone;

sample a sound signal within the sound data;

perform signal conversion on frequencies of the sampled sound signal;

define and extract frequency signals that exhibit horn sound characteristics from the converted frequency signals;

combine the extracted frequency signals;

accumulate the combined frequency signals over time;

calculate a change of the accumulated frequency signals; and

determine whether the calculated change of the frequency signals is equal to or greater than a threshold value to determine whether and when the horn is generated.

2. The apparatus according to claim 1, wherein the frequency signals which exhibit the horn sound characteristics include:

a fundamental frequency signal of a frequency band of about 300 Hz to about 500 Hz; and

a harmonics set from about 2 to about 10 multiples of the frequency band of the fundamental frequency signal.

3. The apparatus according to claim 1, wherein in the extraction of the frequency signals that exhibit the horn sound characteristics, the sound signal processor is further configured to use a total power spectrum and a spectrum flux to extract the frequency signals.

4. The apparatus according to claim 3, wherein the total power spectrum is a sum of total power within a frame section of the sampled sound signal.

5. The apparatus according to claim 3, wherein the spectrum flux is a change rate of power between frames of the sampled sound signal.

6. A method for recognizing a horn using a sound signal process, comprising:

acquiring, by a processor, and storing surrounding sound data using a microphone;

sampling, by the processor, a sound signal within the sound data;

performing, by the processor, signal conversion on frequencies of the sampled sound signal;

defining and extracting, by the processor, frequency signals which exhibit horn sound characteristics from the converted frequency signals;

combining, by the processor, the extracted frequency signals;

accumulating, by the processor, the combined frequency signals over time;

calculating, by the processor, a change of the accumulated frequency signals; and

determining, by the processor, whether the calculated change of the frequency signals is equal to or greater than a threshold value to determine whether and when a horn is generated.

7. The method according to claim 6, wherein the frequency signals which exhibit the horn sound characteristics include:

a fundamental frequency signal of a frequency band of about 300 Hz to about 500 Hz; and

a harmonics set from about 2 to about 10 multiples of the frequency band of the fundamental frequency signal.

8. The method according to claim 6, wherein in the extraction of features of the frequency signals includes:

using a total power spectrum and a spectrum flux to extract the feature of the frequency signals which exhibit the horn characteristics.

9. The method according to claim 8, wherein the total power spectrum is a sum of total power within a frame section of the sampled sound signal.

10. The method according to claim 8, wherein the spectrum flux is a change rate of power between frames of the sampled sound signal.

11. A non-transitory computer readable medium containing program instructions executed by a processor, the computer readable medium comprising:

program instructions that acquire and store surrounding sound data from a microphone;

program instructions that sample a sound signal within the sound data;

program instructions that perform signal conversion on frequencies of the sampled sound signal;

program instructions that define and extract frequency signals that exhibit horn sound characteristics from the converted frequency signals;

program instructions that combine the extracted frequency signals;

program instructions that accumulate the combined frequency signals over time;

program instructions that calculate a change of the accumulated frequency signals; and

program instructions that determine whether the calculated change of the frequency signals is equal to or greater than a threshold value to determine whether and when the horn is generated.

12. The non-transitory computer readable medium of claim 11, wherein the frequency signals which exhibit the horn sound characteristics include:

a fundamental frequency signal of a frequency band of about 300 Hz to about 500 Hz; and

a harmonics set from about 2 to about 10 multiples of the frequency band of the fundamental frequency signal.

13. The non-transitory computer readable medium according to claim 11, wherein in program instructions that extract of features of the frequency signals includes:

program instructions that use a total power spectrum and a spectrum flux to extract the feature of the frequency signals which exhibit the horn characteristics.

14. The non-transitory computer readable medium according to claim 13, wherein the total power spectrum is a sum of total power within a frame section of the sampled sound signal.

15. The non-transitory computer readable medium according to claim 13, wherein the spectrum flux is a change rate of power between frames of the sampled sound signal.