Description:TECHNICAL FIELD
[0001] The present disclosure relates to the field of gunshot surveillance. More particularly to classifying a muzzle blast signal from a shockwave signal.

BACKGROUND
[0002] Background description includes information that may be useful in understanding the present invention. It is not an admission that any of the information provided herein is prior art or relevant to the presently claimed invention, or that any publication specifically or implicitly referenced is prior art.
[0003] Shooting incidents are on a rise in the whole world creating threat to public. During such incidents most police personnel are not able to detect where the shooter is exactly located at any given time during the pursuit, especially when the shooter is present in a hostile urban area. Also, it is difficult for humans to rely strictly on their hearing to locate where the sound of gunfire is coming from. Police and military personnel also have trouble relying on witness accounts during such events because people will be in a state of panic and often give inaccurate information related to the place of origin of gunshots. Analyzing acoustics from gunshot might help in determining the origin of the gunshot.
[0004] Guns, including firearms and artillery, generate two types of acoustic signals: a muzzle blast signal and a shockwave signal, differentiating the two signals and analyzing them to determine the location of the shooter.
[0005] Therefore, there is a need for improved techniques for differentiating the muzzle blast signal from the shockwave signal.

OBJECTS OF THE PRESENT DISCLOSURE
[0006] Some of the objects of the present disclosure, which at least one embodiment herein satisfies, are as listed below.
[0007] It is an object of the present disclosure to classify muzzle blast and shockwave signals from a gunshot.
[0008] It is an object to the present disclosure to provide two different modules to process the gunshot signal.
[0009] It is an object of the present disclosure to provide muzzle blast detection module with direct current (DC) offset removal module and a wavelet decomposer.
[0010] It is an object of the present disclosure to provide shockwave signal detection module with DC offset removal module and a moving root mean square (RMS) module.

SUMMARY
[0011] This section is provided to introduce certain objects and aspects of the present disclosure in a simplified form that are further described below in the detailed description. This summary is not intended to identify the key features or the scope of the claimed subject matter.
In one aspect, the present disclosure relates to a system for differentiating signals from a gunshot. The system including an audio sensor, such as, microphones, for obtaining a sound generated by the gunshot, wherein the sound generated by the gunshot includes a first type of signal and a second type of signal and an acquisition unit for receiving signals from the audio sensor and converting to a form suitable for processing by a classification unit, wherein the classification unit comprises a first type of classification unit for classifying the first type of signal and a second type of classification unit for classifying the second type of signal and the acquisition unit comprises an analog to digital converter (ADC) and a field programmable gate array (FPGA). The first type of classification unit includes a muzzle blast classification unit and the first type of signal includes a muzzle blast signal and the second type of classification unit includes a shockwave classification unit and the second type of signal includes a shockwave signal.
[0012] In another aspect, the present disclosure relates to a method for differentiating signals generated from a gunshot. The method includes receiving gunshot signals through one or more audio sensors, converting the received gunshot signals to digital samples by an acquisition unit, and classifying, by a classification unit, the gunshot signals into a muzzle blast signal and a shockwave signal, wherein the classification unit includes a muzzle blast classification unit and a shockwave classification unit.
[0013] The method further includes, the muzzle blast classification unit, removing a direct current (DC) offset from digital samples, applying wavelet decomposition to the DC offset removed digital samples to obtain a set of approximation coefficients, calculating a set of variances for the set of approximation coefficients to detect a presence of a valid acoustic signal by segmenting the set of approximation coefficients, identifying a segment with valid signal and storing the corresponding start and end indices, extracting the valid signal based on start and end indices, calculating a time and frequency duration of the valid signal and comparing the calculated time and frequency duration with a predefined value to determine the presence of a muzzle blast signal.
[0014] The method further includes, the shockwave classification unit, removing a DC offset from digital samples, applying a moving root mean square (RMS) function to the DC offset removed digital samples, segmenting the samples into small windows with a frame length, calculating an average energy content for each window frame length of digital samples subjected to moving RMS function and filtering the calculated energy content, selecting peak signal levels in the filtered energy content having a value greater than a threshold value, wherein the threshold comprises an amplitude threshold, detecting the presence of a valid signal, calculating a half cycle duration, frequency, and time duration of the valid signal and comparing the calculated half cycle duration, frequency, and time duration with a predefined value to determine the presence of a shockwave signal.

BRIEF DESCRIPTION OF DRAWINGS
[0015] The accompanying drawings are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this specification. The drawings illustrate exemplary embodiments of the present disclosure and, together with the description, serve to explain the principles of the present disclosure. The diagrams are for illustration only, which thus is not a limitation of the present disclosure.
[0016] FIG. 1 illustrates an exemplary block diagram of a system (100) for detecting and classifying muzzle blast and shockwave acoustics generated from a gunshot firing, in accordance with some embodiments of the present disclosure.
[0017] FIG. 2 illustrates an exemplary flow chart representing a method (200) for detecting and classifying muzzle blast and shockwave acoustics generated from a gunshot firing, in accordance with some embodiments of the present disclosure.
[0018] FIG. 3 illustrates an exemplary flow chart representing a method (300) for extracting features from acoustics generated from a gunshot firing and classifying it is a muzzle blast signal or not, in accordance with some embodiments of the present disclosure.
[0019] FIG. 4 illustrates an exemplary flow chart representing a method (400) for extracting features from acoustics generated from a gunshot firing and classifying it is a shockwave signal or not, in accordance with some embodiments of the present disclosure.
[0020] FIG. 5 illustrates an exemplary computer system (500) in which or with which embodiments of the present disclosure may be implemented.
[0021] The foregoing shall be more apparent from the following more detailed description of the invention.

DETAILED DESCRIPTION
[0022] The following is a detailed description of embodiments of the disclosure depicted in the accompanying drawings. The embodiments are in such detail as to clearly communicate the disclosure. However, the amount of detail offered is not intended to limit the anticipated variations of embodiments; on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the scope of the present disclosure as defined by the appended claims.
[0023] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the present invention. It will be apparent to one skilled in the art that embodiments of the present invention may be practiced without some of these specific details.
[0024] The present disclosure relates to classification of acoustics from a gunshot firing, more particularly to a system for classifying a first type of signal, for example, a muzzle blast signal and a second type of signal, for example, a shockwave signal generated from a gunshot firing. The system receives input signals i.e, gunshot signals from one or more sources such as, without limitations, microphones. The input signal is converted to digital samples with an analog to digital converter (ADC) at an acquisition unit. The digital samples are given as input signals to a classification unit for differentiating shockwave and muzzle blast signals. The classification unit comprises two separate modules for receiving digital samples of the acoustic signal, wherein a first type of module classifies the first type of signal, and a second type of module classifies the second type of signal, wherein the first type of module includes muzzle blast classification module/unit to extract muzzle blast signal and the second type of module includes shockwave classification module to extract shockwave signals. Each module analyzes time width, frequency range and temporal shape characteristics of the gunshot acoustics. The input digital signal is decomposed into detailed and approximation coefficients of wavelet transformation applied at different number of levels. The approximation coefficients at a certain level number are used for further processing. The approximation coefficients represent the low frequency content of the original signal and is the approximated signal of the original input signal. The level number is calculated based on the sampling frequency of the acquisition system. The presence of valid signal and noisy signal in a given set of input digital samples is determined based on a dynamic threshold, wherein the dynamic threshold value is calculated by calculating variance of entire approximated signal and multiplying by a constant number. The given set of input digital samples is divided into segments of fixed length and variance for each segment is calculated. The calculated variance is then compared with the dynamic threshold value to identify presence of valid signal. Each segment with a valid signal is further processed to classify it as muzzle blast or shockwave by calculating the frequency and time duration of the signal. The time duration and frequency are compared with fixed values which are unique for muzzle blast and shockwave signal. For muzzle blast detection module, pre-processing techniques like wavelet decomposition are applied to enhance the muzzle blast signal and eliminate shockwave signal. On the other hand, for shockwave detection module, the input digital signals are treated with moving root means square (RMS) function.
[0025] Various embodiments of the present disclosure will be explained with reference to FIGs. 1-5.
[0026] FIG. 1 illustrates an exemplary block diagram of a system (100) for detecting and classifying muzzle blast and shockwave acoustics generated from a gunshot firing, in accordance with some embodiments of the present disclosure. In FIG. 1, the system (100) comprising audio sensor, such as, without limitations, microphones (102), an acquisition unit (104), a classification unit (106) comprising a muzzle blast classification unit (108) and a shockwave classification unit (110) are shown. The microphones (102) receive the gunshot firing sound and transmit the data to the acquisition unit (104). The acquisition unit (104) includes an analog to digital converter (ADC) and a field programmable gate array (FPGA). The ADC samples the analog input i.e, the gunshot audio and generates digital signals for processing by the classification unit (106). The classification unit (106) includes the muzzle blast classification unit (108) for separating the muzzle blast signal from the input digital signal and the shockwave classification unit (110) for separating the shockwave signal from the input digital signal.
[0027] Although FIG. 1 shows exemplary components of the system (100), in other embodiments, the system (100) may include fewer components, different components, differently arranged components, or additional functional components than depicted in FIG. 1. Additionally, or alternatively, one or more components of the system (100) may perform functions described as being performed by one or more other components of the system (100).
[0028] FIG. 2 illustrates an exemplary flow chart representing a method (200) for detecting and classifying muzzle blast and shockwave acoustics generated from a gunshot firing, in accordance with some embodiments of the present disclosure. The method (200) includes receiving, at step 210, a gunshot signal by the microphones (102) as shown in FIG. 1, wherein the gunshot signal may be from a true source or may be simulated one for purpose of testing the classification unit (106) as shown in FIG. 1, acquiring, at step 220, the received signals by the acquisition unit (104) as shown in FIG. 1, wherein the acquired signals are converted to digital signals by the ADC coupled to the acquisition unit (104). Referring to FIG. 2, the method (200), further includes, classifying, at step 230, the converted digital signals as either muzzle blast or shockwave signal by the classification unit (106) as shown in FIG. 1.
[0029] FIG. 3 illustrates an exemplary flow chart representing a method (300) for extracting features from acoustics generated from a gunshot firing and classifying whether it is a muzzle blast signal or not, in accordance with some embodiments of the present disclosure. The method (300) includes removing, at step 302, a DC offset present in the digital signal from the ADC and calculating a level number for decomposition, it is important to calculate the precise level number so that no useful data is eliminated and no unwanted data is prominent in the approximate coefficients signal at that particular level; after calculating the level number based on the sampling frequency of the acquisition system (104) of FIG. 1, then applying, at step 304, a wavelet decomposition to obtain the approximation coefficients at required level of decomposition, wherein the wavelet decomposition eliminates higher frequencies including any shockwaves, detecting, at step 306, a presence of signal corresponding to muzzle blast in the original input signal based on an approximation signal. To detect the presence of muzzle blast signal, a variance is calculated for the entire approximation signal by dividing the signal into smaller segments of specific window length. The variance for each segment is compared with a predefined variance threshold, wherein the predefined variance threshold is given as 10*variance(approximation_signal). The segment having a variance above the predefined variance threshold is marked as 1 and the segment having a variance below the predefined variance threshold is marked as 0. Referring to FIG. 3, the method (300) proceeds with storing, at step 308, in a variable, for example, a window flag, based on the variance associated with the segment a ‘1’ or a ‘0’. Once the window flag is obtained start and end indices are calculated and arranged pairwise to determine segment which has signal presence. The start indices have a transition from 0 to 1 and end indices have a transition from 1 to 0. To obtain a proper pair of indices for the approximation signal, the start and end indices pair having a difference of less than 4 samples are eliminated similarly if the difference between one end index and the next immediate start index is less than 6 then they are eliminated, thereby combining two smaller windows with 1's into a single window with a start and end index to enable fine tuning to get a proper pair of indices. The indices thus obtained are for approximation signal. To obtain the indices for original signal, the indices are multiplied with a scaling factor. The scaling factor is calculated as [length(original_input)/length(approximation_signal)].
[0030] Referring to FIG. 3, the method (300) further includes searching, at step 310, a positive and negative peak i.e, a local maximum for each segment and calculating a start and end index for a required acoustic signal based on the predefined variance threshold and maximum peaks obtained. For example, the part of the data from the start of the data segment to the index where maximum positive peak is present is extracted and the corresponding indices with the value less than 0.001*(maximum positive peak value) is considered. These indices are stored in a variable and the starting time of arrival index is taken as highest value among the indices-1. On the other hand, a part of the data from the maximum negative peak to the end of the data segment is extracted and those indices are found where the value is greater than 3*0.001*(maximum positive peak value). These indices are stored in a variable and the end of signal index is taken as lowest value among the indices+1.
[0031] Upon obtaining the indices, the method (300) includes calculating, at step 312, a time duration and a frequency of the obtained acoustic signal, and determining, at step 314, if the
Frequency F is greater than Fll (frequency lower limit) and less than Ful (frequency upper limit) and if the acoustic time duration T is greater than Tll (time lower limit) and less than Tul,(time upper limit) classifying, at step 318, the acoustic signal as muzzle blast signal, otherwise, classifying, at step 316, the signal is classified as muzzle blast acoustics.
[0032] FIG. 4 illustrates an exemplary flow chart representing a method (400) for extracting features from acoustics generated from a gunshot firing and classifying whether it is a shockwave signal or not, in accordance with some embodiments of the present disclosure. The method (400) includes, removing, at step 402, the DC offset from the input digital signal (i.e, the signal from the microphone passed through ADC converter) and calculating a moving root means square (RMS) signal for the input digital signal, calculating, at step 404, an average energy content for each window frame length of the calculated moving RMS signal and filtering the average energy content to obtain a smoothened signal. The method (400), further includes detecting, at step 406, local peaks in the smoothened signal and comparing with a peak threshold and selecting the peaks having a value more than the peak threshold, wherein the peak threshold is calculated based on (( maximum value of the peaks)/3), creating a signal presence flag by assigning ‘1’ to signal peaks more than the peak threshold and assigning ‘0’ to signal peaks less than the peak threshold and calculating start and end indices from the signal presence flag and arranging it pairwise to determine segment having the signal presence. The start indices are the indices with transition from 0 to 1 and end indices are the indices having transition from 1 to 0. The start and end indices pair having a difference of less than 20 samples are eliminated to fine tune to obtain proper data segments indicating signal presence, wherein the length of each segment is taken based on start and end index of a certain length. The method (400) further includes finding, at step 408, all the local positive peaks and negative peaks above the peak threshold in each segment, wherein a threshold for positive peaks is calculated as (maximum of local positive peak)/4 and the threshold for negative peaks is calculated as (maximum of local negative peak)/4 and sorting the peaks according to the position in which they appear in the data segment, determining, at step 410, for any two peaks, whether a first peak >0 and second peak<0 and if yes calculating, at step 412, a half cycle duration, for example, a time difference between peaks and calculating a start and end index of required acoustic signal. The start and end indices are calculated by collecting a certain number of samples before and after the indices of first and second peak and finding absolute difference between peaks and the other indices, the indices at which maximum absolute differences occur at both the first and second peaks are selected as start and end index.
[0033] Referring to FIG. 4, the method (400) further includes calculating, at step 414, a time duration and frequency of the required acoustic signal and determining, at step 416, whether the half cycle duration is greater than Hll (Half cycle lower limit) and less than Hul (half cycle upper limts), the time duration of the acoustic signal is greater than Tll (time lower limit) and less than Tul (time upper limit) and if the frequency of the signal is greater than Ful (frequency upper limit). If the condition is satisfied, classifying, at step 420, the acoustic signal as a shockwave signal. Otherwise, i.e, if the condition is not satisfied, classifying, at step 418, the acoustic signal as not a shockwave signal.
[0034] In one embodiment of the present invention, if the half cycle time condition is satisfied then the entire time duration of the signal is calculated by finding its accurate start and end indices. For example, the start index is calculated by taking a certain range of indices before the maximum positive peak and finding the index within the range which has maximum absolute difference value between peak value and value at that index. The end index is calculated by taking a certain range of indices after the maximum negative peak and finding the index within the range which has maximum absolute difference value between negative peak value and the value at that index.
[0035] FIG. 5 illustrates an exemplary computer system (500) in which or with which embodiments of the present disclosure may be implemented.
[0036] As shown in FIG. 5, the computer system (500) may include an external storage device (510), a bus (520), a main memory (530), a read only memory (540), a mass storage device (550), communication port(s) (560), and a processor (570). A person skilled in the art will appreciate that the computer system (500) may include more than one processor (570) and communication port(s) (560). The processor (570) may include various modules associated with embodiments of the present disclosure. The communication port(s) (560) may be any of an RS-242 port for use with a modem based dialup connection, a 10/100 Ethernet port, a Gigabit or 10 Gigabit port using copper or fiber, a serial port, a parallel port, or other existing or future ports. The communication port(s) (560) may be chosen depending on a network, such a Local Area Network (LAN), Wide Area Network (WAN), or any network to which computer system connects. The memory (530) may be Random Access Memory (RAM), or any other dynamic storage device commonly known in the art. The read-only memory (530) may be any static storage device(s) e.g., but not limited to, a Programmable Read Only Memory (PROM) chips for storing static information e.g., start-up or basic input/output system (BIOS) instructions for the processor (570). The mass storage device (550) may be any current or future mass storage solution, which can be used to store information and/or instructions. Exemplary mass storage solutions include, but are not limited to, Parallel Advanced Technology Attachment (PATA) or Serial Advanced Technology Attachment (SATA) hard disk drives or solid-state drives (internal or external, e.g., having Universal Serial Bus (USB) and/or Firewire interfaces), one or more optical discs, Redundant Array of Independent Disks (RAID) storage.
[0037] The bus (520) communicatively couples the processor (570) with the other memory, storage, and communication blocks. The bus (520) may be, e.g. a Peripheral Component Interconnect (PCI) / PCI Extended (PCI-X) bus, Small Computer System Interface (SCSI), Universal Serial Bus (USB) or the like, for connecting expansion cards, drives and other subsystems as well as other buses, such a front side bus (FSB), which connects the processor (570) to the computer system (500).
[0038] Optionally, operator and administrative interfaces, e.g. a display, keyboard, and a cursor control device, may also be coupled to the bus (520) to support direct operator interaction with the computer system (500). Other operator and administrative interfaces may be provided through network connections connected through the communication port(s) (560). Components described above are meant only to exemplify various possibilities. In no way should the aforementioned exemplary computer system (500) limit the scope of the present disclosure.
[0039] While the foregoing describes various embodiments of the invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof. The scope of the invention is determined by the claims that follow. The invention is not limited to the described embodiments, versions or examples, which are included to enable a person having ordinary skill in the art to make and use the invention when combined with information and knowledge available to the person having ordinary skill in the art.

ADVANTAGES OF THE DISCLOSURE
[0040] The present disclosure provides for classifying muzzle blast from shockwave signal generated during a gunshot firing.
[0041] The present disclosure provides for detecting the place from where the gunshot was fired.
  , Claims:1. A system (100) for differentiating signals from a gunshot, said system (100) comprising:
an audio sensor (102) for obtaining a sound generated by the gunshot, wherein the sound generated by the gunshot includes a first type of signal and a second type of signal; and
an acquisition unit (104) for receiving signals from the audio sensor and converting to a form suitable for processing by a classification unit (106),
wherein the classification unit (106) comprises a first type of classification unit for classifying the first type of signal and a second type of classification unit for classifying the second type of signal.

2. The system (100) as claimed in claim 1, wherein the acquisition unit (104) comprises an analog to digital converter (ADC) and a field programmable gate array (FPGA).

3. The system (100) as claimed in claim 1, wherein the audio sensor (102) includes one or more microphones.

4. The system (100) as claimed in claim 1, wherein the first type of signal comprises a muzzle blast signal and the second type of signal comprises a shockwave signal.

5. The system (100) as claimed in claim 1, wherein the first type of classification unit includes a muzzle blast classification unit (108) and the second type of classification unit includes a shockwave classification unit (110).

6. A method for differentiating signals generated from a gunshot comprising:
receiving gunshot signals through one or more audio sensors;
converting the received gunshot signals to digital samples by an acquisition unit; and
classifying, by a classification unit, the gunshot signals into a muzzle blast signal and a shockwave signal.

7. The method as claimed in claim 6, wherein the classification unit comprises a muzzle blast classification unit and a shockwave classification unit.

8. The method as claimed in claim 7, further comprising:
removing, by the muzzle blast classification unit, a direct current (DC) offset from digital samples;
applying, by the muzzle blast classification unit, wavelet decomposition to the DC offset removed digital samples to obtain a set of approximation coefficients;
calculating, by the muzzle blast classification unit, a set of variances for the set of approximation coefficients to detect a presence of a valid acoustic signal by segmenting the set of approximation coefficients;
identifying, by the muzzle blast classification unit, a segment with valid signal and storing the corresponding start and end indices;
extracting, by the muzzle blast classification unit, the valid signal based on start and end indices and calculating a time and frequency duration of the valid signal; and
comparing, by the muzzle blast classification unit, the calculated time and frequency duration with a predefined value to determine the presence of a muzzle blast signal.

9. The method as claimed in claim 7, further comprising:
removing, by the shockwave classification unit, a DC offset from digital samples; applying, by the shockwave classification unit, a moving root mean square (RMS) function to the DC offset removed digital samples and segmenting the samples into small windows with a frame length;
calculating, by the shockwave classification unit, an average energy content for each window frame length of digital samples subjected to moving RMS function and filtering the calculated energy content;
selecting, by the shockwave classification unit, peak signal levels in the filtered energy content having a value greater than a threshold value;
detecting, by the shockwave classification unit, the presence of a valid signal;
calculating, by the shockwave classification unit, a half cycle duration, frequency, and time duration of the valid signal; and
comparing, by the shockwave classification unit, the calculated half cycle duration, frequency, and time duration with a predefined value to determine the presence of a shockwave signal.

10. The method as claimed in claim 9, wherein the threshold comprises an amplitude threshold.