Abstract: Systems and methods that yield highly-accurate classification of acoustic and other non-image events, involving pre-processing data from one or more transducers and generating a visual representation of the source as well as associated features and processing, are disclosed. According to certain exemplary implementations herein, such pre-processing steps may be utilized in situations where 1) all impulsive acoustic events have many features in common due to their point source origin and impulsive nature, and/or 2) the error rates that are considered acceptable in general purpose image classification are much higher than the acceptable levels in automatic impulsive incident classification. Further, according to some aspects, the data may be pre-processed in various ways, such as to remove extraneous or irrelevant details and/or perform any required rotation, alignment, scaling, etc. tasks, such that these tasks do not need to be “learned” in a less direct and more expensive manner in the neural network.

Abstract: Systems and methods that yield highly-accurate classification of acoustic and other non-image events, involving pre-processing data from one or more transducers and generating a visual representation of the source as well as associated features and processing, are disclosed. According to certain exemplary implementations herein, such pre-processing steps may be utilized in situations where 1) all impulsive acoustic events have many features in common due to their point source origin and impulsive nature, and/or 2) the error rates that are considered acceptable in general purpose image classification are much higher than the acceptable levels in automatic impulsive incident classification. Further, according to some aspects, the data may be pre-processed in various ways, such as to remove extraneous or irrelevant details and/or perform any required rotation, alignment, scaling, etc. tasks, such that these tasks do not need to be “learned” in a less direct and more expensive manner in the neural network.

Abstract: Systems and methods that yield highly-accurate classification of acoustic and other non-image events, involving pre-processing data from one or more transducers and generating a visual representation of the source as well as associated features and processing, are disclosed. According to certain exemplary implementations herein, such pre-processing steps may be utilized in situations where 1) all impulsive acoustic events have many features in common due to their point source origin and impulsive nature, and/or 2) the error rates that are considered acceptable in general purpose image classification are much higher than the acceptable levels in automatic impulsive incident classification. Further, according to some aspects, the data may be pre-processed in various ways, such as to remove extraneous or irrelevant details and/or perform any required rotation, alignment, scaling, etc. tasks, such that these tasks do not need to be “learned” in a less direct and more expensive manner in the neural network.

Abstract: Systems and methods for processing information present in a digital audio stream to obtain a measure of gain of an analog-to-digital converter (ADC) preamplifier are disclosed. In one implementation, a method of processing information present in a digitally sampled stream to obtain a measure of ADC preamplifier gain used to digitize the output of a known transducer comprises transforming time-domain digital samples into the frequency domain through use of a discrete Fourier transform (DFT), and using knowledge of the maximum effective frequency associated with the frequency response of the transducer to process frequency-domain data to obtain a measure of the gain of the ADC preamplifier.

Abstract: Systems and methods that yield highly-accurate classification of acoustic and other non-image events, involving pre-processing data from one or more transducers and generating a visual representation of the source as well as associated features and processing, are disclosed. According to certain exemplary implementations herein, such pre-processing steps may be utilized in situations where 1) all impulsive acoustic events have many features in common due to their point source origin and impulsive nature, and/or 2) the error rates that are considered acceptable in general purpose image classification are much higher than the acceptable levels in automatic impulsive incident classification. Further, according to some aspects, the data may be pre-processed in various ways, such as to remove extraneous or irrelevant details and/or perform any required rotation, alignment, scaling, etc. tasks, such that these tasks do not need to be “learned” in a less direct and more expensive manner in the neural network.

Abstract: Systems and methods are disclosed associated with processing origin/location information of a source or event. In one exemplary implementation, there is provided a method of performing improved three-dimensional source location processing including constraint of location solutions to a two-dimensional plane. Moreover, the method includes obtaining a plane of constraint characterized as a plane in which the source is likely to occur, providing one or more virtual sensing elements each characterized as being located on a first side of the plane of constraint in a mirror image/symmetrical position across from a corresponding actual sensing element on an opposite side of the plane, and constraining possible origin locations to be located in the plane of constraint. Other exemplary implementations may include determining the origin location as a function of positions of the sensing elements and the virtual sensing elements as well as time-of-arrival and/or angle-of-arrival information.

Abstract: Systems and methods are disclosed for locating a weapon fire incident such as an acoustic transient from a gunshot, explosion, weapons launch, etc. In one exemplary implementation, there is provided a method of locating the incident from a combination of propagation phenomena including a discharge time of the weapon fire incident. Moreover, the method may include obtaining a first propagation parameter of the incident from one or more first sensors, obtaining the discharge time from another sensor, and processing the data to determine a location using a common time basis among sensor measurements. According to further exemplary implementations, the discharge time may include a transient event that has a different propagation velocity than that of sound in the atmosphere.

Abstract: Systems and methods are disclosed for locating a weapon fire incident such as an acoustic transient from a gunshot, explosion, weapons launch, etc. In one exemplary implementation, there is provided a method of locating the incident from a combination of propagation phenomena including a discharge time of the weapon fire incident. Moreover, the method may include obtaining a first propagation parameter of the incident from one or more first sensors, obtaining the discharge time from another sensor, and processing the data to determine a location using a common time basis among sensor measurements. According to further exemplary implementations, the discharge time may include a transient event that has a different propagation velocity than that of sound in the atmosphere.

Abstract: Systems and methods are disclosed that detect weapon firing/noise incidents in a region and/or include other related features. According to one or more embodiments, an exemplary method may include detecting acoustic signals from the region by one or more sensors, processing the detected acoustic signals to generate a processed signal, storing the detected acoustic signals with each sensor, and processing the processed signal associated with each sensor to determine if a weapon firing incident occurred. Moreover, exemplary methods may include, if unable to determine whether a weapon firing incident occurred, performing further processing of the acoustic signals and/or determining if a weapon firing incident occurred based upon the stored detected acoustic signals.

Abstract: Systems and methods are disclosed for processing weapon fire information such as gunfire. In one exemplary implementation, there is provided a method of processing gunshot information to determine source location information involving echo/reflection processing features. Moreover, the method may include processing gunshot information received from a source at a sensor having a reflecting surface at a given distance, processing direct arrival time and echo arrival time information, and determining source location information as a function of a virtual sensor calculated behind the surface.

Abstract: Systems and method are disclosed for processing signals. In one exemplary implementation, a method may include transforming initial bullet data associated with one or more sensors into a set of discrete pulses, dividing the discrete pulses into pulse subsets, generating, for the subsets, time domain representations of the pulses, wherein the time domain representations include waveforms having pulse features, and processing the time domain representations to determine alignment between one or more of pulse features, pulses, pairs of channels, and/or pairs of sensors. One or more further implementations may include determining identity of pulses in association with a matching process performed as a function of the alignment, as well as, optionally, other pulse processing features/functionality.

Abstract: Systems and methods are disclosed associated with processing origin/location information of a source or event. In one exemplary implementation, there is provided a method of performing improved three-dimensional source location processing including constraint of location solutions to a two-dimensional plane. Moreover, the method includes obtaining a plane of constraint characterized as a plane in which the source is likely to occur, providing one or more virtual sensing elements each characterized as being located on a first side of the plane of constraint in a mirror image/symmetrical position across from a corresponding actual sensing element on an opposite side of the plane, and constraining possible origin locations to be located in the plane of constraint. Other exemplary implementations may include determining the origin location as a function of positions of the sensing elements and the virtual sensing elements as well as time-of-arrival and/or angle-of-arrival information.

Abstract: A gunshot location system computes candidate gunshot locations from angle-of-arrival information and time-of-arrival information provided by acoustic sensors. In addition to an angle, each sensor calculates an angular uncertainty from impulses received at four or more microphones having rotational symmetry. An intersection of one or more time-of-arrival hyperbolas with one or more angle-of-arrival beams is used to determine a candidate gunshot location. In simple environments, a location can be confirmed with just two sensors allowing sensor density to be significantly reduced, while in complex environments including reflections, blocking, and interfering acoustic events, the additional angle-of-arrival information improves location accuracy and confidence, allowing elimination of candidate locations inconsistent with the combined time-of-arrival and angle-of-arrival information.

Abstract: A gunshot location system computes candidate gunshot locations [314] from angle-of-arrival information [304, 308] and time-of-arrival information [312] provided by acoustic sensors [300, 302]. In addition to an angle, each sensor calculates an angular uncertainty [306, 310] from impulses received at four or more microphones having rotational symmetry. An intersection of one or more time-of-arrival hyperbolas with one or more angle-of-arrival beams [322] is used to determine a candidate gunshot location. In simple environments, a location can be confirmed with just two sensors allowing sensor density to be significantly reduced, while in complex environments including reflections, blocking, and interfering acoustic events, the additional angle-of-arrival information improves location accuracy and confidence, allowing elimination of candidate locations inconsistent with the combined time-of-arrival and angle-of-arrival information.

Abstract: Systems and method are disclosed for processing signals. In one exemplary implementation, a method may include transforming initial bullet data associated with one or more sensors into a set of discrete pulses, dividing the discrete pulses into pulse subsets, generating, for the subsets, time domain representations of the pulses, wherein the time domain representations include waveforms having pulse features, and processing the time domain representations to determine alignment between one or more of pulse features, pulses, pairs of channels, and/or pairs of sensors. One or more further implementations may include determining identity of pulses in association with a matching process performed as a function of the alignment, as well as, optionally, other pulse processing features/functionality.

Abstract: A gunshot location system computes candidate gunshot locations [314] from angle-of-arrival information [304, 308] and time-of-arrival information [312] provided by acoustic sensors [300, 302]. In addition to an angle, each sensor calculates an angular uncertainty [306, 310] from impulses received at four or more microphones having rotational symmetry. An intersection of one or more time-of-arrival hyperbolas with one or more angle-of-arrival beams [322] is used to determine a candidate gunshot location. In simple environments, a location can be confirmed with just two sensors allowing sensor density to be significantly reduced, while in complex environments including reflections, blocking, and interfering acoustic events, the additional angle-of-arrival information improves location accuracy and confidence, allowing elimination of candidate locations inconsistent with the combined time-of-arrival and angle-of-arrival information.

Abstract: Systems and methods are disclosed for archiving and processing information from an array of remote sensors. Aspects of the innovations may be incorporated, for example, into a gunshot detection and location system to preserve sensor information surrounding a gunshot event for later review or analysis. According to one implementation, an exemplary method may including sampling video signals from the sensors and storing samples in a storage device such that the video signals detected from the sensors may be selected and/or reproduced. Other exemplary implementations may include creating a searchable database wherein the video information is indexed as a function of parameters, which may include data associated with the capture of video signals such as sensor information, time of capture, etc.

Abstract: A gunshot location system computes candidate gunshot locations [314] from angle-of-arrival information [304, 308] and time-of-arrival information [312] provided by acoustic sensors [300, 302]. In addition to an angle, each sensor calculates an angular uncertainty [306, 310] from impulses received at four or more microphones having rotational symmetry. An intersection of one or more time-of-arrival hyperbolas with one or more angle-of-arrival beams [322] is used to determine a candidate gunshot location. In simple environments, a location can be confirmed with just two sensors allowing sensor density to be significantly reduced, while in complex environments including reflections, blocking, and interfering acoustic events, the additional angle-of-arrival information improves location accuracy and confidence, allowing elimination of candidate locations inconsistent with the combined time-of-arrival and angle-of-arrival information.

Abstract: A system and method for archiving and retrieving information from an array of remote sensors. In a preferred embodiment the invention is incorporated in a gunshot detection and location system to preserve audio information surrounding a gunshot event for later review or analysis. In a preferred embodiment the system includes a plurality of acoustic sensors deployed in an array, a computer for processing gunshot information from the sensors, and a mass storage device for temporary archival of audio information. When a gunshot event is detected, the location of the audio information of the data within the spool is stored in an index to facilitate later retrieval of the information.

Abstract: A gunshot location system computes candidate gunshot locations [314] from angle-of-arrival information [304, 308] and time-of-arrival information [312] provided by acoustic sensors [300, 302]. In addition to an angle, each sensor calculates an angular uncertainty [306, 310] from impulses received at four or more microphones having rotational symmetry. An intersection of one or more time-of-arrival hyperbolas with one or more angle-of-arrival beams [322] is used to determine a candidate gunshot location. In simple environments, a location can be confirmed with just two sensors allowing sensor density to be significantly reduced, while in complex environments including reflections, blocking, and interfering acoustic events, the additional angle-of-arrival information improves location accuracy and confidence, allowing elimination of candidate locations inconsistent with the combined time-of-arrival and angle-of-arrival information.