KR101561425B1

KR101561425B1 - Method for acoustic signal detection using signal model and apparatus therefor

Info

Publication number: KR101561425B1
Application number: KR1020140049412A
Authority: KR
Inventors: 윤원중; 박규식
Original assignee: 단국대학교 산학협력단; 자인테크놀로지(주)
Priority date: 2013-06-24
Filing date: 2014-04-24
Publication date: 2015-10-20
Also published as: KR20150000397A

Abstract

A method and apparatus for acoustic signal detection using a signal model are disclosed. The acoustic signal detecting apparatus according to the present invention is an acoustic signal detecting method performed in an apparatus for detecting an acoustic signal generated in a firearm, comprising the steps of extracting a feature point having a fluctuation of a bullet shock wave amplitude from a sample of a bullet shock wave, Generating a signal model of the torpedo shockwave based on the extracted feature points, receiving the acoustic signal including the torpedo shock wave generated in the firearm, and comparing the feature point between the generated signal model and the acoustic signal generated in the firearm, And the step According to the present invention, a desired sound signal can be accurately detected in an environment where noise exists.

Description

TECHNICAL FIELD [0001] The present invention relates to a method and apparatus for detecting an acoustic signal using a signal model,

The present invention relates to a method for detecting an acoustic signal generated in a firearm, and more particularly, to a method and apparatus for detecting an acoustic signal generated in a firearm using a signal model.

As the technology for detecting acoustic signals has developed in recent years, there have been many attempts to detect snipers that can be applied in combat. Especially, sniper detection is important to accurately analyze acoustic signals.

The sound signal from a small rifle used by a sniper in a battle is called a gun signal. Gunshot signals from rifle firearms can be divided into gun muzzle blasts when they are shot from a rifle firearm and bullet shockwaves from a bullet flying supersonic. In order to analyze the acoustic signal, the performance is determined according to a spatial arrangement of a microphone as a hardware and a signal analysis technique as software.

Specifically, a method of arranging a plurality of microphones in a 3D space and using them as an array, a method of disposing a microphone at each corner of a vehicle, and a method of disposing a single microphone array composed of a plurality of microphones in a plurality of places Research is being conducted.

In addition to the hardware detection method, a method of detecting a signal in a frequency domain using a fourier transform, a method of measuring an incident signal using filtering, and a method of detecting using a distributed network Approach is also being attempted at the same time.

Such a method for detecting acoustic signals for sniper detection generally detects ammunition shock waves through an N-wave detection algorithm. The N-wave detection algorithm used for detecting the ammunition shock wave has a complex condition for determining the N-wave, and there is a problem that frequent undetected or false detection occurs even in a device over a microphone due to ambient noise.

In addition, since the gunfire generated from a rifle of a sniper rifle does not generate a precise N-wave such as a bullet shock wave, a bandpass filter of 300 Hz to 1000 Hz, which is a range where gunfire exists, is used and a method of detecting and detecting a threshold value is mainly used do. However, such a frequency band is similar to that of most of the electric field noises, so that the detection rate is very low.

In order to overcome the above-described problems, an object of the present invention is to provide a method for detecting an acoustic signal generated in a sniper rifle using a signal model.

Another object of the present invention is to provide an apparatus for detecting an acoustic signal generated in a sniper rifle using a signal model.

According to another aspect of the present invention, there is provided an acoustic signal detection method using a signal model, the acoustic signal detection method being performed in an apparatus for detecting an acoustic signal generated in a firearm, Extracting a point having a variation in the size of the bullet shock wave from a sample of a shockwave that is greater than a preset reference point, generating a signal model of the bullet shock wave based on the extracted feature points, Receiving the acoustic signal including the torpedo shock wave generated in the firearm, and detecting the torpedo shock wave based on the generated signal model and the feature point comparison between the acoustic signals generated in the firearm.

Here, the feature point may include a start point, a positive peak, a negative peak, and an end point of the ammunition shock wave in the sample of the ammunition shock wave.

Here, the step of generating the signal model of the ammunition shock wave may be generated by connecting the start point, the positive peak, the negative peak, and the end point of the ammunition shock wave.

Here, the step of detecting the ammunition shock wave may include comparing a correlation between the signal model of the generated ammunition shock wave and the minutiae points of the received sound signal, and comparing the cross- And determining that the ammunition shock wave is included in the acoustic signal.

According to another aspect of the present invention, there is provided an acoustic signal detection method using a signal model, the acoustic signal detection method being performed in an apparatus for detecting an acoustic signal generated in a firearm Extracting a point having a variation of the gun size from a sample of a muzzle blast that is greater than a preset reference point, generating a signal model of the gun based on the extracted feature points, Receiving the acoustic signal including the gassing generated in the firearm, and detecting the gassing based on the comparison of the feature points between the generated signal model and the acoustic signal generated in the firearm.

Herein, the feature point may include a start point of the gunshot, a positive peak, a negative peak, and an end point of the gunshot in the sample of the gunshot.

The step of detecting the shots may include comparing cross-correlations between the received acoustic signals and the minutiae points of the signal model of the generated shots, comparing the cross- A step of determining that the sound signal includes the shot, and a step of normalizing the determined shot to detect a final shot.

Here, the step of detecting the final shot may normalize the detected shot using the cross-correlation between the received sound signal and the signal model of the shot, the signal model of the shot, and the received sound signal .

According to another aspect of the present invention, there is provided an apparatus for detecting an acoustic signal using a signal model, the apparatus comprising: a shockwave detector for detecting an acoustic signal generated in a firearm, And a signal model of a bullet shock wave is generated based on the extracted feature points, and a signal model of the bullet shock wave generated by the firearm is generated based on the extracted feature points, A processor for receiving the acoustic signal including the shockwave and detecting the ammunition shock wave based on the comparison between the generated signal model and the acoustic signal generated by the firearm, And a storage unit.

Here, the feature point may include a start point of the torpedo shock wave, a positive peak, a negative peak, and an end point of the ammunition shock wave.

Here, the processing unit may generate a signal model of the bullet shock wave by connecting the starting point of the bullet shock wave, the positive peak, the negative minimum point, and the end point of the bullet shock wave.

In this case, when the bullet shock wave is detected, the processor compares the correlation between the received acoustic signal and the minutiae points of the signal model of the generated ammunition shock wave, and if the comparison result of the cross- It can be determined that the ammunition shock wave is included in the acoustic signal.

Here, the processing unit extracts a point having a variation in the magnitude of the gun from a sample of muzzle blast that is greater than a preset reference point, and generates a signal model of the gun based on the extracted feature points And receives the acoustic signal including the gassing generated in the firearm, and detects the gassing based on the comparison of the feature point between the generated signal model and the acoustic signal generated in the firearm.

Here, when detecting the totality, the processor compares the correlation between the received acoustic signal and the minutiae points of the signal model of the generated totality, and if the comparison result of the cross-correlation degree between the minutiae points is equal to or higher than a preset reference , It may be determined that the sound signal includes the totality, and the final totality may be detected by normalizing the determined totality.

Here, the processing unit may normalize the determined shots using the cross-correlation degree between the received acoustic signal and the signal model of the shots, the signal model of the shots, and the received acoustic signal.

According to the acoustic signal detection method using the above-described signal model, it is possible to accurately detect a desired acoustic signal in an environment where noise exists, and to prevent a desired acoustic signal from being undetected or mis-detected due to noise .

Also, through the method of detecting an acoustic signal according to the present invention, it is possible to improve the detection rate for detection of a desired acoustic signal in an environment where noises exist.

1 is a conceptual diagram showing a general procedure for detecting bullet shock waves and gunshots in an acoustic signal.
2 is a flowchart illustrating a method of detecting an acoustic signal using a signal model according to an embodiment of the present invention.
3 is a flowchart illustrating a method of detecting an acoustic signal using a signal model according to another embodiment of the present invention.
4 is a graph showing recorded signals for generating a signal model of a bullet shock wave according to the present invention.
FIG. 5 is a graph showing a signal model of the ammunition shock wave generated by modeling the recorded signal shown in FIG.
6 is a graph showing recorded signals to generate a signal model of a gun according to the present invention.
7 is a graph showing a signal model of a shot generated by modeling the recorded signal shown in FIG.
8 is a graph showing an acoustic signal including a torpedo shock wave.
FIG. 9 is a graph showing a result of detection of bullet shock waves from the acoustic signal shown in FIG. 8 using the signal model of the bullet shock wave shown in FIG.
10 is a graph showing an acoustic signal including a bullet shock wave and a gun.
FIG. 11 is a graph showing the first result of detecting the gunshots from the acoustic signal shown in FIG. 10 using the signal model of gunshots shown in FIG.
12 is a graph showing the result of detecting the final gunshot through the normalization process in the first result shown in FIG.
13 is a block diagram illustrating an apparatus for detecting an acoustic signal using a signal model according to an embodiment of the present invention.
FIG. 14 is a conceptual diagram illustrating an experimental environment for checking the performance of a method of detecting an acoustic signal according to the present invention.
FIG. 15 is a table showing the results of detection of ammunition shock waves and gunshots by progressive shooting in the experimental environment shown in FIG. 14. FIG.
FIG. 16 is a table showing the experimental results of detecting the bullet shock wave and the gunfire at the same time in the experimental environment shown in FIG. 14. FIG.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail.

It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component. And / or < / RTI > includes any combination of a plurality of related listed items or any of a plurality of related listed items.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprises" or "having" and the like are used to specify that there is a feature, a number, a step, an operation, an element, a component or a combination thereof described in the specification, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the relevant art and are to be interpreted in an ideal or overly formal sense unless explicitly defined in the present application Do not.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Hereinafter, the same reference numerals will be used for the same constituent elements in the drawings, and redundant explanations for the same constituent elements will be omitted.

1 is a conceptual diagram showing a general procedure for detecting bullet shock waves and gunshots in an acoustic signal.

Referring to FIG. 1, a general acoustic signal detecting apparatus includes a first half period 110, a first half period and a second half period between a lowest point and a first half period of the input acoustic signal 100, A shockwave is generated based on the similarity degree 130 between the highest point of the remaining half period and the lowest point of the remaining half period, the duration 140 of the remaining half period and the duration 110 of the first half period and the duration 140 of the remaining half period. .

Specifically, the range of the duration 110 of the first half period of the acoustic signal 100 input to the acoustic signal detection apparatus is 70 to 300 占 퐏, the duration 120 between the highest point of the first half period and the lowest period of the remaining half period, Should be in the range of 200 to 400 microseconds.

The error rate of the comparison of the similarity 130 between the highest point of the first half period and the lowest point of the remaining half period of the acoustic signal 100 input to the acoustic signal detection device is 10% or less, and the range of the remaining half period 140 70 to 300 占 퐏 and the error rate of the similarity comparison between the duration 110 of the first half period and the duration 140 of the remaining half period is less than 35%.

In addition, the acoustic signal detection apparatus can detect a muzzle blast by using a band pass filter having a frequency range of 300 to 1000 Hz in which an incident sound signal 100 exists, and then checking a threshold value.

2 is a flowchart illustrating a method of detecting an acoustic signal using a signal model according to an embodiment of the present invention.

Referring to FIG. 2, a method of detecting an acoustic signal using a signal model can be performed in an apparatus for detecting an acoustic signal. The acoustic signal detecting apparatus detects a shock wave from an acoustic signal generated in a firearm using a signal model Can be detected.

First, the acoustic signal detecting apparatus can extract a point having a fluctuation of a bullet shock wave amplitude from a sample of a bullet shock wave of a predetermined reference or more (S210).

For example, the feature points extracted by the acoustic signal detection device include a start zero crossing point, a positive peak, a negative peak, an end zero crossing point, . &Lt; / RTI >

Here, the acoustic signal detecting apparatus can determine that the inclination of the ammunition shock wave starts to be equal to or higher than a preset reference point, and the point at which the size starts to increase is determined as the starting point of the bullet shock wave. After the point at which the size starts to increase, Can be determined as the positive peak.

In addition, the sound signal detection apparatus starts to decrease the size of the bullet shock wave after the point at which the positive peak is determined, and can determine that the point having the slope of 0 and the minimum point is the lowest point of the sound, Can be determined as the end point of the ammunition shock wave.

Thereafter, the acoustic signal detection device may generate a signal model of the torpedo shock wave based on the extracted feature points (S220).

For example, the acoustic signal detection device can generate a signal model by connecting the start point, the positive peak, the negative peak, and the end point of the ammunition shock wave extracted from the feature points.

Thereafter, the acoustic signal detecting apparatus may receive an acoustic signal including a torpedo shock wave generated in a rifle (S230).

Here, the firearm means a weapon using gunpowder and can mean, for example, a pistol, a rifle, a cannon, and the like. In addition, the acoustic signal generated from the fire can be a signal including gunshots generated when a bullet is fired, and a bullet shock including a shot bullet flying at supersonic flight.

Thereafter, the acoustic signal detecting apparatus can compare the correlation between the signal model of the generated ammunition shock wave and the minutiae points of the received acoustic signal (S240).

Thereafter, when the comparison result of the cross-correlation between feature points is equal to or higher than a predetermined reference value, the acoustic signal detection apparatus can determine that the acoustic signal includes the ammunition shock wave (S250).

On the other hand, the acoustic signal detecting apparatus can determine that the acoustic signal does not include the ammunition shock wave when the result of the cross-correlation comparison between the minutiae points is less than a preset reference.

The acoustic signal detection method using the signal model according to the present invention can detect the ammunition shock wave from the acoustic signal generated in the firearm as described above with reference to steps S210 to S250.

3 is a flowchart illustrating a method of detecting an acoustic signal using a signal model according to another embodiment of the present invention.

Referring to FIG. 3, a method of detecting an acoustic signal using a signal model can be performed in an apparatus that detects an acoustic signal, and can detect a gun from an acoustic signal generated in a firearm.

First, the acoustic signal detection apparatus can extract feature points whose variation in size is larger than a preset reference from the sample of the gunness (S310). Here, the minutiae points extracted by the acoustic signal detection device may be the same as those described with reference to step S210 of FIG.

Thereafter, the acoustic signal detection device may generate a signal model of the shots based on the extracted feature points (S320). Here, the specific method by which the acoustic signal detection device generates the signal model of the shots may be the same as that described with reference to step S220 of FIG.

Thereafter, the acoustic signal detection device may receive the acoustic signal including the fire generated in the firearm (S330). Here, the acoustic signal received by the acoustic signal detecting device may be the same as that described with reference to step S230 of FIG.

Thereafter, the acoustic signal detection device can compare the cross-correlation between the received acoustic signal and the minutiae points of the signal model of the generated grossness (S340). Here, a concrete method of comparing the cross-correlation between feature points by the acoustic signal detection apparatus may be the same as that described with reference to step S240 of FIG.

Thereafter, the acoustic signal detection apparatus may determine that the acoustic signal includes a gender if the comparison result of the cross-correlation between the feature points is equal to or greater than a preset reference (S350). On the other hand, the acoustic signal detecting apparatus can judge that the acoustic signal does not include the shots when the result of the cross-correlation comparison between the feature points is less than a preset reference.

Thereafter, it can be determined that the acoustic signal, which the acoustic signal detecting device judges that the shot is included, includes not only the shot but also the ammunition shock wave. Accordingly, the acoustic signal detection apparatus can normalize the determined gunshot to detect the final gunshot (S360).

Here, the acoustic signal detecting apparatus can perform the normalization processing on an acoustic signal which simultaneously includes a torpedo shock and a gun, thereby removing the torpedo shock and detecting the final gunshot.

Specifically, the acoustic signal detection apparatus can normalize the detected shots based on the following Equation (1).

In Equation (1), XcorrResult denotes a result of comparing the degree of cross correlation between the signal model of the gun and the received sound signal, MBmodel denotes the signal model of the gun, x denotes the received sound signal generated in the firearm, it means.

The acoustic signal detection method using the signal model according to the present invention can detect the gun firing from the sound signal generated in the firearm as described above with reference to steps S310 to S360.

Hereinafter, with reference to Figs. 4 to 7, a method of extracting a feature point from a sample of a bullet shock wave and a shot, and generating a signal model of a bullet shock wave and a shot based on extracted feature points will be described.

Here, the sample of the bullet shock wave and the shot that the acoustic signal detecting device extracts the characteristic point records the acoustic signal generated by shooting the 5.56 mm bullet in the firearm at a distance of 100 m.

In addition, the sample of the ammunition shock wave and the shot was recorded by setting the separation distance between the acoustic signal detection device and the flying ammunition, that is, the closest point of arrival (CPA) from the acoustic signal to 5 m.

FIG. 4 is a graph showing a recorded signal for generating a signal model of a bullet shock wave according to the present invention, and FIG. 5 is a graph showing a signal model of a bullet shock wave generated by modeling the recorded signal shown in FIG.

Referring to FIG. 4, the acoustic signal detection apparatus can extract feature points whose fluctuation of the ammunition shock wave amplitude in the sample 400 of the ammunition shock wave is equal to or greater than a preset reference.

For example, the acoustic signal detection apparatus includes a start point 410, a positive peak 420, a negative point 430 and an end point 440 of a bullet shock wave in a sample 400 of a bullet shock wave, Can be extracted.

Specifically, in the acoustic signal detecting apparatus, the inclination of the bullet shock wave sample 400 starts to become a predetermined reference or more and the point where the size starts to increase can be determined as the starting point 410 of the bullet shock wave, And the point at which the slope becomes 0 after the point where it starts to be the positive peak 420.

In addition, the sound signal detecting apparatus starts to decrease the size of the bullet shock wave after the point at which it is determined that the positive peak 420 is located. Thus, the point at which the slope is 0 and the magnitude is lowest can be determined as the lowest point 430, The point at which the size becomes 0 after the point determined as the lowest point 430 of the bullet shock wave 440 can be determined as the end point 440 of the bullet shock wave.

5, the acoustic signal detecting apparatus connects the start point 510, the positive peak 520, the negative peak 530, and the end point 540 of the ammunition shock wave, The signal model 500 can be generated.

FIG. 6 is a graph showing a recorded signal to generate a signal model of a gun according to the present invention, and FIG. 7 is a graph showing a signal model of a gun generated by modeling the recorded signal shown in FIG.

Referring to FIG. 6, the acoustic signal detecting apparatus can extract feature points having a variation of the magnitude of the magnitude in the sample (600) of the magnitude above a preset reference.

For example, the acoustic signal detection device may extract a starting point 610, a positive peak 620, a negative bottom point 630, and a gun end point 640 of the gun that starts gunshots in the sample 600 of gunshots .

Here, the concrete method of determining the starting point 610 of the gun, the positive peak 620, the negative bottom 630 and the end point 640 of the gunshots is the same as that described with reference to Figs. 4 to 5 can do.

7, the acoustic signal detection apparatus connects the start point 710 of the extracted gunshot, the positive peak 720, the negative ear 730 and the end point 740 of the gun, 700).

Hereinafter, with reference to FIGS. 8 to 12, a method of detecting a bullet shock wave and a gun using a signal model from an acoustic signal generated by a firearm will be described.

Here, the firearm fires a bullet having a diameter of 5.56 × 45 mm at a distance of 200 m, and sets a distance between the acoustic signal detecting device and the flying ammunition, that is, a minimum distance reaching point from the acoustic signal to 10 m.

FIG. 8 is a graph showing an acoustic signal including a torpedo shock wave, and FIG. 9 is a graph showing a result of detecting a torpedo shock wave from the acoustic signal shown in FIG. 8 using the signal model of the torpedo shock wave shown in FIG.

Referring to FIG. 8, the sound signal generated from the firearm may include a gunshield generated when a bullet is fired from a muzzle of a firearm, and a bullet shock wave generated while flying a bullet with a supersonic speed.

Referring to FIG. 9, the acoustic signal detecting apparatus can detect a torpedo shock wave by comparing the acoustic signals generated from the firearm with the signal models of the torpedo shock waves shown in FIG. 5.

It can be seen that, when the acoustic signal detection device detects the ammunition shock wave using the signal model, only the specific portion including the ammunition shock wave has a rapidly changing value and the remaining portion has a value close to zero.

Accordingly, the acoustic signal detecting apparatus can determine that the acoustic signal generated from the firearm contains the ammunition shock wave.

FIG. 10 is a graph showing acoustic signals including a torpedo shock wave and a gun, FIG. 11 is a graph showing a first result of detecting the gunshots from the acoustic signal shown in FIG. 10 using the signal model of gunshots shown in FIG. And FIG. 12 is a graph showing the result of detecting the final gunshot through the normalization process in the first result shown in FIG.

10, an acoustic signal generated from a firearm may include a gun fire 1020 generated when a bullet is fired from a muzzle of a firearm and a bullet shock wave 1010 generated while flying a bullet at a supersonic speed.

Referring to FIG. 11, the acoustic signal detection apparatus can detect the gunness by comparing the cross-correlation between the signal models of the gunshots shown in FIG. 7 and the sound signals generated from the firearms.

When the acoustic signal detecting device detects the gun using the signal model, the magnitude of the specific portion including the gunshot 1120 as well as the specific portion including the gunshot shock wave 1110 is rapidly changed, It can be seen that a result having a value close to 0 appears.

12, the acoustic signal detection apparatus removes the bullet shock wave 1210 from the result of simultaneous detection of the bullet shock wave 1210 and the gun bulb 1220, and outputs the gun bulb 1220 to be detected, Can be accurately detected.

Here, the normalization process performed by the acoustic signal detection device may be the same as that described with reference to step S360 of FIG.

13 is a block diagram illustrating an apparatus for detecting an acoustic signal using a signal model according to an embodiment of the present invention.

13, an acoustic signal detection apparatus 1300 using a signal model according to the present invention can perform an acoustic signal detection method using the signal model described with reference to FIGS. 2 to 3. The acoustic signal detection apparatus 1300 includes a processing unit 1310, And a storage unit 1320.

The processing unit 1310 of the acoustic signal detection apparatus 1300 using the signal model according to the present invention can detect the torpedo shock wave from the acoustic signal generated in the firearm.

Specifically, the processing unit 1310 can extract the minutiae from the sample of the ammunition shock wave whose fluctuation of the ammunition shock wave amplitude is equal to or larger than a preset reference. Here, the minutiae may include the starting point of the ammunition shock wave, the positive peak, the negative minimum point, and the end point of the ammunition shock wave. Here, a specific method of extracting the minutiae from the sample of the bullet shock wave by the processing unit 1310 may be the same as that described with reference to step S210 of Fig.

Thereafter, the processing unit 1310 can generate a signal model of the ammunition shock wave based on the extracted minutiae points. Here, the processing unit 1310 can generate a signal model by connecting the starting point, the positive peak, the negative peak, and the end point of the ammunition shock wave, which are extracted minutiae points. Here, the specific method by which the processing unit 1310 generates the signal model may be the same as that described with reference to step S220 of FIG.

Thereafter, the processing unit 1310 receives an acoustic signal including a torpedo shock wave generated in the firearm, and can detect a torpedo shock wave based on a comparison of feature points between the generated signal model and the acoustic signal generated in the firearm. Here, the processing unit 1310 can compare the degree of cross correlation between the acoustic signals and the minutiae points of the signal model.

Thereafter, the processing unit 1310 can determine that the sound signal contains the ammunition shock wave when the cross-correlation degree comparison result between the minutiae points is equal to or greater than a preset reference. Here, a specific method by which the processing unit 1310 detects the torpedo shock wave may be the same as that described with reference to step S250 of FIG.

In addition, the processing unit 1310 of the acoustic signal detection apparatus 1300 using the signal model according to the present invention can detect the shots from the acoustic signals generated in the firearm.

Specifically, the processing unit 1310 can extract feature points whose fluctuation of the magnitude of the gunshots is larger than a predetermined reference from the sample of the gunshots. Here, the minutiae may include the starting point of gunfire, the positive peak, the negative peak, and the end point of the gun. The specific method by which the processing unit 1310 extracts the minutiae from the samples of gunshot can be the same as that described with reference to step S310 of Fig.

Thereafter, the processing unit 1310 can generate a signal model of the shots based on the extracted feature points. Here, the processing unit 1310 may generate the signal model by connecting the extracted minutiae points, that is, the starting point of the gunshot, the peak of the positive, and the endpoint of the gun. A specific method by which the processing unit 1310 generates the signal model may be the same as that described with reference to step S320 of FIG.

Thereafter, the processing unit 1310 receives the sound signal including the gunshots generated in the firearm, and can detect the gunshots based on the comparison of the feature points between the generated signal model and the sound signals generated in the firearm. Here, the processing unit 1310 can compare the degree of cross correlation between the acoustic signals and the minutiae points of the signal model.

Thereafter, the processing unit 1310 may determine that the sound signal includes the gunshots, and may determine the final gunshots by normalizing the determined gunshots if the cross-correlation degree comparison result between the minutiae points is equal to or greater than a preset reference. Here, a specific method of normalizing the determined severity by the processing unit 1310 and detecting the final severity may be the same as that described with reference to step S360 of FIG.

Here, the processing unit 1310 may include a processor and a memory. The processor may refer to a general purpose processor (e.g., a Central Processing Unit (CPU), etc.) or a dedicated processor for a method of detecting acoustic signals using a signal model. The program code for the acoustic signal detection method using the signal model can be stored in the memory. That is, the processor can read the program code stored in the memory, and can perform each step of the acoustic signal detection method using the signal model based on the read program code.

The storage unit 1320 may store the processed information and the processed information in the processing unit 1310. For example, the storage unit 1320 may store samples of ammunition shock waves and gunshots, preset criteria for determining minutiae points, generated torpedo shock waves, and signal models of gunshots.

Hereinafter, experiments and experimental results for confirming the performance of the acoustic signal detecting method according to the present invention will be described with reference to FIG. 14 to FIG.

FIG. 14 is a conceptual diagram illustrating an experimental environment for checking the performance of a method of detecting an acoustic signal according to the present invention.

FIG. 15 is a table showing experimental results of detecting bullet shock waves and gunfire in sequential shooting in the experimental environment shown in FIG. 14, and FIG. 16 is a table showing the result of shooting in the experimental environment shown in FIG. Shock wave, and shot.

Referring to FIG. 14, in order to confirm the performance of the method of detecting an acoustic signal according to the present invention, an experiment for detecting a bullet shock wave and a gun was performed using an acoustic signal detection method during a shooting operation in a shooting range.

A firearm 1410 used in this experiment used a bullet having a diameter of 5.56 mm and a diameter of 45 mm. A first sound signal detector 1420 was installed at a range of 100 meters from the firearm 1410, A signal override device 1430 is installed. In addition, the distances between the bullet fired through the fire and each sound signal detection device, that is, the minimum distance from the sound signal, were divided into 5, 10, 15, 20, 25, 30, 35 and 40m.

The method of proceeding the shooting proceeded in a sequential shooting method in which each shot was fired at a sufficient time difference, and a simultaneous shooting method in which shooting was performed simultaneously from the prepared shooter regardless of the order.

Referring to FIG. 15, when sequential shooting is performed in the shooting environment as shown in FIG. 14, the results of the torpedo shock and gunshots detected in the sound signal detecting apparatus and the general sound signal detecting apparatus according to the present invention .

The acoustic signal detection apparatus according to the present invention detects 56 ammunition shock waves generated from a firearm at a distance of 100 meters, 56 ammo shock waves out of 56 guns, and 56 guns.

On the other hand, the conventional acoustic signal detection device detected only 56 bullet shock waves generated from a firearm having a range of 100 meters, 51 bullet shock waves out of 56 guns, and 54 guns.

In addition, the acoustic signal detecting apparatus according to the present invention detects 56 ammunition shock waves generated from a firearm with a range of 200 m, 56 ammunition shock waves and 56 shots out of 56 guns.

On the other hand, a typical acoustic signal detection device detected 56 ammunition shock waves generated from a firearm having a range of 200 meters, 55 ammunition shock waves and 56 shots out of 56 gunshots.

As described above, it can be seen that the acoustic signal detecting apparatus according to the present invention exhibits an accurate and high detection rate as compared with a general acoustic signal detecting apparatus.

Referring to FIG. 16, in the shooting environment as shown in FIG. 14, when the shooting is performed simultaneously, the results of the torpedo shock wave and the gunshot detected in the acoustic signal detecting apparatus and the general sound signal detecting apparatus according to the present invention are shown have.

The acoustic signal detection apparatus according to the present invention detects 56 torpedo shock waves generated from a firearm at a range of 100 m, 42 torpedo shock waves out of 56 guns, and 56 guns.

On the other hand, a typical acoustic signal detector detected 56 ammunition shock waves generated from a firearm at a distance of 100 meters, 48 ammunition shock waves and 48 guns out of 56 gunshots.

In addition, the acoustic signal detecting apparatus according to the present invention detects 49 torpedo shock waves generated from a firearm at a distance of 200 m, 46 torpedo shock waves out of 56 gunshots, and 48 gunshots.

On the other hand, a typical acoustic signal detection device detected 49 bullet shock waves generated from a firearm having a range of 200 m, 46 bullet shock waves out of 56 gunshots, and 36 guns.

As described above, it can be seen that the acoustic signal detecting apparatus according to the present invention exhibits an accurate and high detection rate in a noisy environment to the surroundings as compared with a general acoustic signal detecting apparatus.

It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined in the appended claims. It will be possible.

100: acoustic signal
400: Sample of ammunition shock wave
500: Signal model of ammunition shock wave
600: Sample of gunshot
700: signal model of gunshot
1300: Sound signal processing device using signal model
1310:
1320:
1410: Firearms
1420: First sound signal detection device
1430: Second acoustic signal detection device

Claims

A method for detecting an acoustic signal in an apparatus for detecting an acoustic signal generated in a firearm,
Extracting a point having a variation in the size of the bullet shock wave from a sample of a bullet shockwave having a predetermined reference value or more;
Generating a signal model of the bullet shock wave based on the extracted feature points;
Receiving an acoustic signal including a torpedo shock wave generated in the firearm; And
And detecting the ammunition shock wave based on a correlation comparison between the generated signal model and the minutiae of the acoustic signal generated in the firearm.

The method according to claim 1,
The feature point may be,
A positive peak, a negative peak, and an end point of the ammunition shock wave in the sample of the ammunition shock wave.

The method of claim 2,
Wherein the step of generating the signal model of the ammunition shock wave comprises:
A positive peak, a negative peak, and an end point of the ammunition shock wave are connected to each other to generate an acoustic signal.

The method according to claim 1,
The step of detecting the bullet shock wave includes:
Comparing the degree of cross correlation between the generated signal of the torpedo shock wave and the feature point of the received acoustic signal; And
And determining that the ammunition shock wave is included in the acoustic signal when the comparison result of the cross-correlation between the feature points is equal to or greater than a preset reference.

A method for detecting an acoustic signal in an apparatus for detecting an acoustic signal generated in a firearm,
Extracting a point having a variation of the magnitude of the magnitude from a sample of muzzle blast that is higher than a preset reference;
Generating a signal model of the gun based on the extracted feature points;
Receiving an acoustic signal including a gunshot generated in the firearm; And
Detecting the shots based on a correlation comparison between the generated signal model and the feature points of the acoustic signals generated in the firearm.

The method of claim 5,
The feature point may be,
Wherein the signal includes a starting point of the gunshot, a peak of a positive, a bottom of a negative and an endpoint of the gunshot in the sample of the gunshot.

The method of claim 5,
The method of claim 1,
Comparing cross-correlations between the received acoustic signal and the minutiae points of the signal model of the generated gross;
Determining that the sound signal includes the totality if the comparison result of the cross-correlation between the minutiae points is equal to or greater than a preset reference; And
And detecting a final gunshot by normalizing the determined gunshot to detect the final gunshot.

The method of claim 7,
Wherein detecting the final gunshot comprises:
Wherein the detected signal is normalized using a cross-correlation between the received acoustic signal and the signal model of the signal, the signal model of the signal, and the received acoustic signal.

An apparatus for detecting an acoustic signal generated in a firearm,
Extracting a point having a variation in the size of the bullet shock wave from a sample of a bullet shockwave of a predetermined reference or more and generating a signal model of the bullet shock wave based on the extracted feature point, A processor for receiving the acoustic signal including the torpedo shock wave generated in the firearm and detecting the torpedo shock wave based on a correlation between the generated signal model and the correlation between the minutiae points of the acoustic signal generated in the firearm; And
And a storage unit for storing information processed by the processing unit and processed information.

The method of claim 9,
The feature point may be,
A positive peak, a negative peak, and an end point of the ammunition shock wave.

The method of claim 10,
Wherein,
Wherein the signal model of the bullet shock wave is generated by connecting the starting point of the bullet shock wave, the positive peak, the negative peak, and the end point of the bullet shock wave.

The method of claim 9,
Wherein,
And comparing the cross-correlation degree between the received acoustic signal and the minutiae points of the signal model of the generated ammunition shock wave when the ammunition shock wave is detected, and when the comparison result of the cross- And determines that the ammunition shock wave is included.

The method of claim 9,
Wherein,
Extracting a point having a variation in the magnitude of the gun from a sample of a muzzle blast equal to or greater than a preset reference point and generating a signal model of the gun based on the extracted feature point, Wherein the acoustic signal detecting unit receives the acoustic signal including the generated gait and detects the gait based on the comparison of the feature point between the generated signal model and the acoustic signal generated by the firearm.

14. The method of claim 13,
Wherein,
Comparing the cross-correlation between the received acoustic signal and the minutiae points of the signal model of the generated grossness, and if the cross-correlation degree comparison result between the minutia points is equal to or greater than a preset reference, Is included, and a final gunshot is detected by normalizing the determined gunshot, and detecting the final gunshot.

15. The method of claim 14,
Wherein,
Wherein the signal processing unit normalizes the determined grossness using a cross-correlation between the received acoustic signal and the signal model of the grossness, a signal model of the grossness, and a received acoustic signal.