336results about "Direction/deviation determination systems" patented technology

The invention discloses an underwater sound environment sensing method based on multi-array-element sparse channel estimation. An underwater sound receiving end receives signals transmitted through an underwater sound multipath channel through a multi-array-element array; performing Hilbert transform on the receiving signal of each array element to obtain an analysis signal, and calculating a cross-correlation function of the analysis signal and the transmitting signal; based on the cross-correlation function, sparse channel parameters, including path amplitude and time delay, of each array element are estimated by adopting an orthogonal matching pursuit algorithm combined with a constant false alarm detection dynamic threshold value; the method comprises the following steps: constructing an array response vector by using sparse channel parameters of a multi-array element array, and searching and estimating angles of arrival, including a direction angle and a pitch angle, of a multipath signal through a spatial spectrum peak value; and based on the arrival angles, amplitudes and time delays of the direct path and the reflection path, inverting an underwater environment structure through a ray acoustic theory, including a reflection point distance and a reflection surface normal vector, so as to realize three-dimensional perception of the underwater reflector. According to the invention, multipath resolution can be improved, and false alarm and missing detection can be effectively reduced.

A method, computer program product, and computing system for generating a plurality of acoustic relative transfer functions associated with a plurality of audio acquisition devices of an audio recording system deployed in an acoustic environment. Acoustic relative transfer functions of at least a pair of audio acquisition devices of the plurality of audio acquisition devices may be compared. Location information associated with an acoustic source within the acoustic environment may be determined based upon, at least in part, the comparison of the acoustic relative transfer functions of the at least a pair of audio acquisition devices of the plurality of audio acquisition devices.

The invention discloses a room geometry inference method based on positioning of a direct sound source and a first-order reflection sound source, which comprises the following steps of: 1) acquiring reverberation signals of a target room by using a microphone array, converting the reverberation signals into FOA signals, and calculating a time-frequency covariance matrix of the FOA signals, estimating the direction of arrival DOAs of the direct sound source s and the direction of arrival DOAs'of the n first-order reflection sound sources s' according to the time-frequency covariance matrix; 2) estimating the distance ds of the direct sound source s according to the DOAs, the DOAs'and the height dh of the microphone array, and determining the position Ps of the direct sound source s; 3) estimating the distance ds'of the first-order reflection sound source s' based on the FOA signal, the DOAs, the DOAs' and ds, and determining the position Ps' of the first-order reflection sound source s'; and 4) obtaining the geometric space structure of the target room according to the position Ps of the direct sound source s and the position Ps'of each first-order reflection sound source s'.

An autonomous underwater vehicle (1) has a housing (10) having a measurement system (13) with three acceleration or speed sensors on different axes from one another, and a processing unit configured to generate seismic data from the received seismic waves (S1) and to determine the direction (D2) of the acoustic waves (S2) transmitted by an acoustic transmitter (20) from a base (2) which are received by at least two sensors of the measurement system (13). The processing unit generates and transmits, to the navigation system (12), guidance data relating to the determined direction (D2) of the received acoustic waves (S2). The navigation system (12) is configured to control the propulsion and steering system (11) of the vehicle according to the guidance data in order to guide the movement of the vehicle towards the base.

The invention provides a DOA estimation and error correction method and device based on joint sparse recovery, and the method comprises the steps: receiving a target signal through a linear array, building an array signal model fusing a phase error, and constructing an optimization model based on multiple measurement vectors; and compensating the phase error in each step of iteration by using an iterative optimization solution algorithm to enable the DOA to gradually approach a true value, thereby obtaining estimation results of the spatial spectrum and the phase error. The method provided by the invention has the advantages that random phase errors caused by array element position errors, system frequency errors, sound wave propagation errors and the like in DOA estimation are corrected. Under the condition that phase errors exist, a signal model and an optimization problem are constructed, and the phase errors and a DOA estimation value are solved through an iterative algorithm, so that a high-precision and low-sidelobe DOA estimation result is obtained. The method is suitable for DOA estimation under the conditions of small snapshots, strong noise and random phase errors.

Acoustic signals from an acoustic event are captured via sensing nodes of sensor group(s) that comprise a group of sensing nodes at a location comprising spatial boundaries. Each of the sensing nodes comprise a sensor area. Each of the sensor group(s) is based on: range limits of each of the sensing nodes; shared sensing areas of the sensing nodes; and intersections between the sensor area for each of the sensing nodes and the spatial boundaries. Solutions(s) are generated by processing the acoustic signals. The solution(s) indicate the location or trajectory of the acoustic event. A strength of solution compliance value for at least one of the solution(s) is determined. A refined solution is generated employing: sensor contributions of sensing nodes; and the strength of solution compliance value with the spatial boundaries and at least one of the solution(s). A report is created comprising the location or trajectory of the acoustic event.

The invention provides a vector array sparse Bayesian learning direction of arrival estimation method. The method comprises the following steps: constructing a far-field vector sparse signal model; and establishing a noise covariance model under the vector sound field. And estimating a signal power hyper-parameter through a vector array sparse Bayesian learning process. And through a maximum likelihood estimation technology, noise power hyper-parameter estimation is realized. And finally, carrying out peak searching on the converged signal power hyper-parameter to obtain a direction of arrival estimation result. The method has the advantages that the vector array signal processing performance advantage is obtained, higher signal processing gain is obtained, and meanwhile the method has the capability of restraining the azimuth ambiguity problem; and by using the difference between the noise covariance matrix and the signal covariance matrix under the vector noise, the estimation precision of hyper-parameters such as the signal power and the noise power is remarkably improved. Compared with a traditional vector array DOA estimation method, the vector array DOA estimation method has a lower spectrum background and a sharper spatial spectrum peak, and the resolution and DOA estimation precision are remarkably superior to those of other methods.

The invention relates to the field of underwater sound signal processing, in particular to an underwater DOA (direction of arrival) estimation method based on graph nerves and a convolutional neural network, which comprises the following steps: 1, establishing a linear array, and enabling narrow-band signals to simultaneously reach an underwater sound array; 2, performing signal preprocessing to obtain a signal covariance matrix, and performing normalization processing; 3, extracting correlation between array elements and spatial features of array signals, and performing data supplementation on sparse linear array information; 4, forming a double-branch structure, enhancing the information aggregation capability, and extracting features from a space path and a time domain path; and 5, constructing an adjacent matrix, filling node features of damaged array elements, adopting a double-branch structure, extracting spatial features and time domain features, carrying out feature integration, and outputting a DOA estimation result. The spatial correlation between array elements is extracted and the array sparsity problem is processed by using the graph neural network, and the time domain features of the signals are extracted in combination with the convolutional neural network, so that more accurate and more robust DOA estimation can be realized under the conditions of low signal-to-noise ratio and array sparsity.

A half-spectrum search DOA estimation method based on characteristic value gradient jump belongs to the field of array signal processing, and comprises the following steps: modeling a received signal to obtain an array output vector; calculating a received signal covariance matrix; introducing a complex conjugate covariance matrix and a scanning source to construct a new covariance matrix, and constructing a spatial spectrum function; constructing a characteristic value gradient discrimination function, and adaptively estimating a signal source number by using the characteristic value gradient discrimination function; scanning angles are traversed by using a half-spectrum search method, and a signal direction of arrival is verified and determined by combining an MVDR algorithm according to a sharp spectrum peak formed by a spatial spectrum function. According to the method, source number priori knowledge is not needed, the source number can be identified autonomously, the problem that estimation is inaccurate under complex conditions in traditional half-spectrum search is solved, and the method has good robustness under the scene of unknown source number. According to the method, the limitation that few traditional half-spectrum search angle estimation is inaccurate is overcome, and a new thought is provided for half-spectrum search DOA estimation under the condition that the number of information sources is unknown.

The invention discloses an indoor multi-sound-source positioning method based on DOA estimation and DOA association, and the method comprises the steps: judging single-source time-frequency points through the correlation of time delay inequality vectors of adjacent time-frequency points between microphones, and constructing a variable-size elastic single-source region; applying the FSSZ distribution condition of the sound source in a reference array to all arrays by using the high correlation characteristic of the FSSZ time-frequency distribution of the same sound source at different microphone arrays in a frame; performing middle DOA estimation on the FSSZs of all the arrays in the whole observation time period by using a circular integral cross spectrum method, and constructing a sound source position histogram in combination with FSSZ aggregation degree weighting; 2D positions of the plurality of sound sources are estimated from the sound source position histogram using 2D-CFAR. According to the invention, under the conditions of indoor reverberation and noise, position estimation can be carried out on a plurality of sound sources with an unknown number, the detection performance under the conditions of positioning precision and missed detection is obviously superior to that of an existing method, and the defects of the prior art are overcome.

For training a perception algorithm to detect an emergency vehicle, respective audio datasets are received from two microphones and respective spectrograms are generated. At least one interaural difference map is generated based on the spectrograms, audio source localization data is generated, which specifies a number of audio sources in respective grid cells of a spatial grid, by applying a CRNN to first input data containing the spectrograms and the least one interaural difference map. An image is received from a camera and output data comprising a bounding box for the emergency vehicle is predicted by applying at least one further ANN to second input data containing the image and the spectrograms. Network parameters are adapted depending on the output data and the audio source localization data.

Method for locating a sound source, particularly in a vehicle environment, comprising the following steps: - Providing (S1) at least two microphones, in particular vehicle microphones, which are attached to different vehicle components in order to obtain raw data; - Providing (S2) position data of the microphones; - transforming (S3) the raw data from a time domain to a frequency domain by means of a spectral transformation in order to obtain spectral data comprising amplitude data and phase data; - preprocessing (S4) the amplitude data and / or the phase data and / or data derived therefrom, using artificial intelligence, to obtain filtered data; - Calling (S5) a beamforming algorithm with the filtered data and the position data of the microphones to obtain directional data for the raw data.

The invention provides an MVDR beam forming method and device, and the method comprises the steps: obtaining an observation acoustic matrix formed after a plurality of acoustic sensors receive a plurality of target signal sources, and enabling the plurality of acoustic sensors to be arranged in an array manner; drawing a potential function numerical diagram based on the observation acoustic matrix to determine the number of target signal sources and the direction of arrival corresponding to each target signal source; calculating a covariance matrix based on the observation acoustic matrix; revising the covariance matrix based on the number of the target signal sources to generate an optimized covariance matrix; and obtaining a beam forming result based on the observation acoustic matrix, the optimization covariance matrix and the array response matrix. The method and the device are used for improving signal extraction quality and accuracy of MVDR beam forming in a multi-signal-source and noise interference environment.

The invention discloses a training method of a multi-sound-source direction-of-arrival estimation model, a multi-sound-source direction-of-arrival estimation method, equipment, a medium and a product, and relates to the technical field of signal processing and artificial intelligence crossing, and the training method comprises the steps: calculating the time-frequency characteristics and cross-correlation characteristics of original signals of a multi-channel array; respectively carrying out position coding on the time-frequency characteristics and the microphone position information; fusing the features into an input matrix, and inputting the input matrix into a backbone network of an improved Transform model; calculating the attention score of the input matrix and generating head output by using the first multi-head self-attention layer, inputting the head output into the second multi-head self-attention layer, calculating the attention score and generating head output, and inputting the output into a multi-task output module to obtain a direction estimation result. The multi-head attention mechanism is introduced, long-time and multi-band complex dependence is captured, the sound source distinguishing capacity is improved, multiple heads capture direction information from different view angles, and it is guaranteed that high accuracy can still be guaranteed in the complex environment.

A system can operate a speech-controlled device to perform user engagement detection (UED) processing to detect when speech represented in audio data is directed to the device. For example, the device may extract audio features from the audio data and process these audio features using a classifier to estimate an orientation of the user's head, which may be used as a proxy for user engagement. Thus, if the head orientation is within an engagement zone (which varies based on distance to the user), the device may determine that the user is engaged with the device and perform language processing on input speech. In contrast, if the head orientation is outside of the engagement zone, the device may determine that the user is not engaged and ignore the input speech. To enable additional functionality, the classifier may optionally output a coarse estimate of the head orientation along with the UED determination.

The invention discloses a method for detecting radiation characteristics of a to-be-detected target in a shallow sea waveguide environment, which comprises the following steps of: 1, extracting real normal wave modal parameters of a sound source radiation signal in the shallow sea waveguide environment through an orthogonal constraint modal search method according to data received by a vertical array; 2, according to the real normal wave modal parameters, through a normal wave expression in a waveguide environment, performing inversion to obtain a sound field of the whole shallow sea waveguide; and step 3, acquiring sound field information required in the sound field of the shallow sea waveguide, mapping the sound source information into the free space by using the reciprocity theorem, realizing sound field migration from the shallow sea waveguide to the free space, and further acquiring the radiation characteristics of the target to be measured in the free space. The method is used for obtaining the radiation characteristics of the to-be-measured target in the free space according to the data received by the vertical array.

Systems can include an acoustic sensor array configured to receive acoustic signals, an illuminator configured to emit electromagnetic radiation, an electromagnetic imaging tool configured to receive electromagnetic radiation, a distance measuring tool, and a processor. The processor can illuminate the target scene via the illuminator, receive electromagnetic image data from the electromagnetic imaging tool representative of the illuminated scene, receive acoustic data from the acoustic sensor array, and receive distance information from the distance measuring tool. The processor can be further configured to generate acoustic image data of the scene based on the received acoustic data and received distance information and generate a display image comprising combined acoustic image data and electromagnetic image data. The processor can determine depths of various acoustic signals within a scene and generate a representation of the scene the shows the determined depths, including floorplan and volumetric representations.

The application provides a method and an electronic device for guiding a user. The second electronic device can send a first ultrasonic signal, the first electronic device can determine first angle information with the second electronic device according to the first ultrasonic signal, the first electronic device outputs a first prompt signal to prompt the user to rotate the first electronic device. After the user rotates the first electronic device, the second electronic device can send a second ultrasonic signal again, and the first electronic device can determine second angle information with the second electronic device according to the second ultrasonic signal. The first electronic device can output a first guiding signal by using the first angle information, the second angle information and the rotation angle of the first electronic device, so as to guide the user to find the second electronic device. The user can find the second electronic device according to the first guiding signal, and the first electronic device and the second electronic device do not need to be installed with UWB chips, so that the cost can be reduced.

A mobile robot includes a microphone array with a set of microphones. The microphone array is at least partially disposed on the mobile robot. The mobile robot receives audio signals from the microphone array. Audio feature data of acoustic activity is extracted from the audio signals. Direction of arrival (DOA) data of the acoustic activity is generated based on the audio signals. A machine learning model is configured to generate audio event data using the audio feature data. The audio event data identifies at least one sound source of the audio feature data. A knowledge graph is queried using the audio event data to obtain entity data. The entity data has a predetermined relation with the audio event data. Semantic audio scene data is generated using the audio event data, the DOA data, and the entity data. The mobile robot performs an action based on the semantic audio scene data.

The invention provides a multi-sound-source arrival direction estimation method and device based on a frequency focusing spatial spectrum. The method comprises the following steps: firstly, generating microphone array data and constructing a training data set by using a preset simulation environment and a microphone array signal simulator; secondly, constructing a mask estimation network model through sound source masking training; then, sub-bandwidth division is carried out on the enhanced multi-sound-source signals in a frequency band splicing mode, and a weighted focusing covariance matrix is obtained; thirdly, broadband focusing spatial spectrum features are constructed through a sub-bandwidth frequency focusing covariance matrix obtained through beam forming, and a convolutional neural network is constructed based on the features; and finally, according to the mask estimation network model and the multi-sound-source DOA spatial spectrum estimation network model, decoding to obtain a multi-sound-source DOA spatial spectrum, and completing multi-sound-source DOA estimation. According to the invention, the intermediate representation based on the broadband frequency focusing spatial spectrum reduces the influence of nonlinear distortion of the neural network on DOA estimation; and the stability of algorithm performance and the fault tolerance of DOA estimation are improved.

Some systems include an acoustic sensor array configured to receive acoustic signals, an electromagnetic imaging tool configured to receive electromagnetic radiation, a user interface, a display, and a processor. The processor can receive electromagnetic data from the electromagnetic imaging tool and acoustic data from the acoustic sensor array. The processor can generate acoustic image data of the scene based on the received acoustic data, generate a display image comprising combined acoustic image data and electromagnetic image data, and present the display image on the display. The processor can receive an annotation input from the user interface and update the display image based on the received annotation input. The processor can be configured to determine one or more acoustic parameters associated with the received acoustic signal and determine a criticality associated with the acoustic signal. A user can annotated the display image with determined criticality information or other determined information.

The invention discloses an automatic classroom director method and electronic equipment. Audio data and video data of a current classroom scene are acquired; based on the audio data, sound source direction positioning is carried out through a microphone array, and the current sound source direction is determined; face key points are extracted according to the video data, lip movement recognition is carried out according to the face key points, and a lip movement recognition result is obtained; extracting human body key points according to the video data, and performing action recognition according to the human body key points to obtain a target action recognition result; based on the sound source direction, the lip movement recognition result and the target action recognition result, a current shooting picture of the camera device is controlled, and the current shooting picture of the camera device is a director picture; and outputting and displaying the director picture. According to the application, audio and video multi-mode information is fused, high-precision identification and natural and accurate automatic picture switching of the speaker and the interaction object in the classroom scene are realized, and the automation level of director and the overall classroom recording and broadcasting effect are improved.

Systems and methods directed toward acoustic analysis can include a plurality of acoustic sensor arrays, each including a plurality of acoustic sensor elements, and a processor in communication with the plurality of acoustic sensor arrays. The processor can be configured to select one or more of the plurality of acoustic sensor arrays based on one or more input parameters, and generate acoustic image data representative of an acoustic scene based on received acoustic data from the selected one or more acoustic sensor arrays. Such input parameters can include distance information and / or frequency information. Different acoustic sensor arrays can share acoustic sensor elements in common or can be entirely separate from one another. Acoustic image data can be combined with electromagnetic image data from an electromagnetic imaging tool to generate a display image.

Approaches presented herein provide for identification of sound from a sound source relative to an array of microphones of a potentially unknown configuration using, in part, differences in the audio signals received by the microphones. In at least one embodiment, audio signals are captured using an array of microphones and audio features are extracted from those signals. The audio features can be processed using a first neural network to generate a feature vector representing a spatial location of an audio source with respect to the plurality of microphones, where the spatial location is inferred based on audio differences and independent of an availability of information indicating a physical configuration of the plurality of microphones. The feature vector can be provided to a task-specific model to perform at least one audio-related task based in part on the spatial location.

The invention provides a subspace smooth sparse reconstruction passive direction of arrival estimation method under strong interference. The method comprises the following steps: firstly, constructing a far-field airspace sparse signal model received by an array; and obtaining a signal subspace covariance matrix through subspace projection and enhanced space smoothing. And estimating the distribution of the signal power in the spatial domain by using a covariance fitting criterion. And a far-field grid point evolution criterion is formulated, so that the grid points are gradually split along with iteration according to the rule. And outputting the signal power hyper-parameter as a spatial spectrum estimation result, and performing peak searching on the spatial spectrum to obtain a direction-of-arrival estimation result. The method has the advantages that the subspace projection technology is introduced, and the adverse effect of strong interference signals on weak target detection is greatly weakened; by enhancing the spatial smoothing technology, the robustness of coherent signals is improved, and meanwhile, the spatial spectrum reconstruction precision is improved. A grid evolution method is utilized, so that grid points non-uniformly cover an interested airspace with emphasis, and the calculation efficiency of the method is remarkably improved.

The invention discloses a multipath chirp signal arrival time estimation method based on instantaneous frequency fitting. The method comprises the following steps: S1, transmitting a reference signal to the surrounding environment by a loudspeaker of the smart phone, and receiving the reference signal reflected by the surrounding environment in a multi-path manner through a microphone of the smart phone as an echo signal; s2, performing Hilbert transform, derivation and discretization on the echo signal in sequence to obtain a discrete instantaneous frequency, and processing by using a difference method to obtain an instantaneous frequency increment of the discrete instantaneous frequency; and S3, according to the discrete instantaneous frequency, the instantaneous frequency increment and the reference signal, performing variance function construction, linear fitting and calculation processing to obtain the arrival time estimation of the echo signal. The method has the advantages of low calculated amount consumption, robust anti-multipath capability and high-precision estimation result, and can more accurately estimate the arrival time of the signal in a complex indoor acoustic multipath space.

The invention belongs to the field of signal processing, and relates to a dynamic direction of arrival estimation method based on a fusion neural network and a beam forming technology. The method comprises the following steps: S1, receiving an observation signal, and extracting complementary features from three dimensions of a time-frequency domain, a spatial domain and a perception domain by using a multi-modal feature extraction method to form a basis of multi-modal feature representation; s2, based on the multi-modal features extracted in the S1, performing adaptive fusion among the features by adopting a gated cross attention mechanism; and S3, realizing DOA estimation in a complex scene through a synergistic effect of neural network parameter learning and beam forming theory constraint by adopting a multi-stage optimization framework fusing physical prior and data driving based on the features subjected to adaptive fusion in the S2. According to the method, the DOA of the signal can be efficiently, accurately and dynamically estimated, the signal receiving quality is improved, and a new thought with both theoretical preciseness and engineering practicability is provided for underwater acoustic monitoring.

A method performed by an audio system comprising a headset. The method sends a playback signal containing user-desired audio content to drive a speaker of the headset that is being worn by a user, receives a microphone signal from a microphone that is arranged to capture sounds within an ambient environment in which the user is located, performs a speech detection algorithm upon the microphone signal to detect speech contained therein, in response to a detection of speech, determines that the user intends to engage in a conversation with a person who is located within the ambient environment, and, in response to determining that the user intends to engage in the conversation, adjusts the playback signal based on the user-desired audio content.

The invention relates to the technical field of deep learning, and discloses a direction of arrival estimation method, system and device, and a medium. The method comprises the following steps: acquiring a multi-sound-source signal in a target environment; the multi-sound-source signals are converted into a GCC-PHAT matrix, then the GCC-PHAT matrix is converted into a GCC-PHAT matrix image, and the GCC-PHAT matrix image comprises time delay information of the signals obtained by the microphones; the multi-sound-source signals are converted into a covariance matrix, then the covariance matrix is converted into a covariance matrix image, and the covariance matrix image comprises spatial distribution information of the signals acquired by the microphones; and fusing the GCC-PHAT matrix image and the covariance matrix image, inputting the fused image into a Vision Mama network, extracting a time delay feature and a spatial distribution feature of the multi-sound-source signal, and performing direction of arrival estimation on the multi-sound-source signal according to the time delay feature and the spatial distribution feature to improve the direction of arrival estimation precision.