223results about How to "Suppress background noise" patented technology

A mobile device that is capable of automatically starting and ending the recording of an audio signal captured by at least one microphone is presented. The mobile device is capable of adjusting a number of parameters related with audio logging based on the context information of the audio input signal.

The invention discloses a depth extraction method of merging motion information and geometric information, which comprises the following steps: (1) carrying out scene segmentation on each frame of two-dimensional video image, and separating the static background and the dynamic foreground; (2) processing a scene segmentation chart by binaryzation and filtering; (3) generating a geometric depth chart of the static background based on the geometric information; (4) calculating the motion vector of the foreground object, and converting the motion vector into the motion amplitude; (5) linearly transforming the motion amplitude of the foreground object according to the position of the foreground object, and obtaining a motion depth chart; and (6) merging the motion depth chart and the geometricdepth chart, and filtering to obtain a final depth chart. The method only calculates the motion vector of the separated dynamic foreground object, thereby eliminating the mismatching points of the background and reducing the amount of calculation. Meanwhile, the motion amplitude of the foreground object is linearly transformed according to the position of the foreground object, the motion amplitude is merged into the background depth, thereby integrally improving the quality of the depth chart.

A speech coding system is provided with time-domain noise attenuation. The speech coding system has an encoder operatively connected to a decoder via a communication medium. A preprocessor processes a digitized speech signal from an analog-to-digital converter. Speech coding systems are used to encode and decode a bitstream. Gains from the speech coding are adjusted by a gain factor Gf that provides time-domain background noise attenuation.

The invention relates to a gas concentration on-line measurement method based on laser absorption spectrum, belonging to the technical field of the tunable laser diode absorption spectrum. The Beer-Lambert principle is subjected to second-order Taylor approximation, the gas concentration is determined based on the ratio of absorption signal second harmonic to background signal first harmonic without calibration experiment. During the measurement process, a laser is subjected to high-frequency modulation on the spectrum line center frequency, and the acquired signal is subjected to discrete Fourier transform (DFT) to obtain each harmonic signal. The gas measurement precision is increased, the wavelength modulation technology is applied at the strong absorption condition, and the application range of the tunable diode laser absorption spectroscopy (TDLAS) is expanded. In the experiment, a calibration and application phase-lock technique applied on the laser is omitted, the measurement process and the measurement system are simplified and the wide application of the TDLAS technique is facilitated.

In estimating propagation path in an OFDM receiver in an OFDM communication system, a CIR estimation unit estimates a group of impulse responses of a propagation path, a valid-impulse discriminator selects impulse responses (CIR), which are greater than a predetermined threshold value, from the impulse-response group, and a propagation path estimation unit generates a matrix expression using a CIR estimation vector that includes the selected CIRs as elements, a matrix S, which is decided based upon number N of points of an IFFT used in OFDM modulation and number Nc of subcarriers used in actual transmission, and a propagation-path response vector, obtains the propagation-path response vector by solving this matrix expression and estimating the propagation path.

The invention provides a voice enhancement method. The method comprises the following steps of: judging whether the current frame has pure noise by using a judgment device; if the current frame has pure noise and a plurality of previous frames of the current frame have pure noise, improving frequency domain signals by using a voice enhancement algorithm of an improved spectral subtraction method, otherwise, improving the frequency domain signals by using an enhancement algorithm of a voice generating model; and transforming the processed frequency domain signals to a time domain, performing de-emphasis processing and acquiring output signals. The invention also provides a device using the method. The voice enhancement method greatly improves the attenuation of residual noise, and ensures the voice intelligibility.

An approach for efficiently reducing background noise from speech signal in real-time applications is presented. A noisy input speech signal is processed through an inverse filter when the spectrum tilt of the input signal is not that of a pure background noise model the noisy input signal is also filtered in order to reduce the spectrum valley areas of the noisy input signal when the background noise is present.

A mass spectrometer and a method of mass spectrometry are disclosed wherein periodic background noise is effectively filtered out from the mass spectral data. An overall mass window is superimposed upon the mass spectral data. The overall mass window preferably comprises 21 nominal mass windows each preferably having a width of 1.0005 amu. Each nominal mass window preferably comprises 20 channels. An intensity distribution relating to all the first channels of the 21 nominal mass windows is determined. An intensity quantile is determined from the intensity distribution. The intensity quantile is taken to represent the background intensity in the first channel of the central nominal mass window. This process is repeated for the other channels so that the background intensity across the whole of the central nominal mass window is estimated and then subtracted from the raw mass spectral data comprising the central nominal mass window. The overall mass window is then preferably advanced approximately 1 amu and the process is repeated multiple times.

The invention relates to a gas parameter online measurement method based on wavelength modulation spectroscopy and belongs to the field of tunable diode laser absorption spectroscopy. According to the method, a gas absorptivity function is fitted by using an odd number of harmonic signals of an X axis and a Y axis output by a phase-locked amplifier based on the wavelength modulation spectroscopy, normalization processing is carried out on the harmonic signals of the X axis and the Y axis by using first harmonic background signals so as to eliminate the influence of factors, such as background signal, laser intensity and modulation factor, and then, the temperature, concentration, pressure and spectroscopic constant of gas are directly measured by using the gas absorptivity function. The gas parameter online measurement method based on the wavelength modulation spectroscopy has the advantages that the problem that the wavelength modulation spectroscopy needs to calibrate experimental measuring temperature and concentration and cannot measure the pressure and spectroscopic constant of gas nowadays is solved, and the application range of the wavelength modulation spectroscopy is widened.

A system and method for use in filtering of an acoustic signal are provided for producing an output signal of attenuated amount of diffuse sound in accordance with predetermined parameters of desired output directional response and required attenuation of diffuse sound. The system includes a filtration module and a filter generation module including a directional analysis module and filter construction module.

The invention relates to a gas temperature and concentration online measuring method based on a first harmonic signal, which belongs to the technical field of tunable diode laser absorption spectrum (TDLAS). Online measurement of the gas temperature and concentration is realized through the relation between a first harmonic X-axis signal and a first harmonic Y-axis signal and a gas absorption rate functional integral value. The advantages of a direct absorption method in the TDLAS technology are combined with the wavelength modulation method, so that not only is the difficulty that the direct absorption method cannot be applied in a worse environment and a weak absorption condition solved, but also the difficulty that the gas temperature and the gas concentration are determined through a calibration experiment in the measurement of the wavelength modulation method can be solved. Due to the adoption of the method, the measurement precision of the TDLAS technology in the industrial field can be effectively improved, and the application range can be enlarged.

The invention discloses a regional average value kernel density estimation-based moving target detecting method in a dynamic scene. The method comprises the following steps of: firstly, initializing a background model; secondly, building a time and space background model for describing the dynamic complex scene by using a training sample in a background modelling process and considering the time sequence characteristics of pixel points in a video frame and the space characteristics in the adjacent regions of the pixel points; thirdly, continuously updating the background model by using the new video frame sample in a moving target detecting process; fourthly, adapting to the instantaneous background change by the regional kernel density estimating method and adapting to the continuous background change by using single Gauss background model, wherein the combination of the two models can fast and accurately adapt to the continuous change of the background and increases the executing efficiency of the method at the same time; and finally performing a foreground detecting method by providing an adjacent region information amount-based method so as to further remove noise points and inanition of a moving target in the background region in the detecting process and more completely extract the moving object in the foreground. The method can be widely applied to alarming the suspicious moving target in an intelligent monitoring system in an outdoor scene or a prohibited military zone and has wide market prospect and application value.

A sound input device includes a housing, a microphone housing, a vibrator, and first and second sound guide components. The housing has first and second sound collection holes. The microphone housing has first and second microphone sound holes. The first and second sound guide components communicate with the first and second sound collection holes and the first and second microphone sound holes, respectively. The microphone housing has a first microphone inner space that communicates with the first microphone sound hole and one side of the diaphragm of the vibrator, and a second microphone inner space that communicates with the second microphone sound hole and the other side of the diaphragm. A spacing between the first sound collection hole and the second sound collection hole is wider than a spacing between the first microphone sound hole and the second microphone sound hole.

The invention relates to a time frequency mask-based single acoustic vector sensor (AVS) target voice enhancement method. According to the method, the arrival angle of the target voice is known, a method of combining a fixed beam former and a post-positioned Wiener filter is adopted for realizing target voice enhancement, and calculation of the weight value of the post-positioned Wiener filter involves self-power spectrum estimation of the target voice. Time frequency sparse characteristics of a voice signal are used, the time frequency point correlation arrival angle for receiving audio signals is estimated through calculating the ISDR (Inter-sensor data ratio) of component signals outputted by two gradient sensors in the AVS, time frequency mask is designed through calculating errors between the time frequency point correlation arrival angle and a target arrival angle, and thus self-power spectrum estimation of the target voice is acquired. According to the method of the invention, any noise prior knowledge does not needed, the target voice can be effectively enhanced in a complicated environment where multiple speakers exist, and interference voice can background noise can be suppressed. In addition, the operation complexity is low, the adopted microphone array size is small (about 1cm<3>), and application on a portable device is excessively facilitated.

Hearing aid for improving the hearing ability of the hard of hearing, comprising an array of microphones, the electrical output signals of which are fed to at least one transmission path belonging to an ear. Means are provided for deriving two array output signals from the output signals of the microphones, the array having two main sensitivity directions running at an angle with respect to one another and each of which is associated to an array output signal. Each array output signal is fed to its own transmission path belonging to one ear of a person who is hard of hearing.

The invention discloses a paper sheet quantitative detection device which is based on a terahertz time-domain spectroscopy technology and a method thereof. The detection device includes: a femtosecond laser, a beam splitter, an optical delay device, a reflective mirror, a focusing lens, a photoconductive antenna type terahertz transmitter, a polyethylene lens, a polarizer, a zinc telluride electro-optical crystal, a phase-locking amplifier, a high-power amplifier and a data collection and processing system. The invention makes use of the polyethylene lens for aligning the terahertz waves to the parallel light beams, and the polyethylene lens is used for focusing the terahertz waves on a terahertz detector after penetrating the paper sheet. The paper sheet quantitative detection method which is disclosed by the invention firstly establishes a paper sheet quantitative detection mathematical model, then respectively measures the time-domain signals of the terahertz waves penetrating the air and penetrating the paper sheet, extracts the delay time and applies the established mathematical model to calculate the paper sheet quantitative value. The invention has high precision of the measurement, strong anti-interference capability, small influence of multiple reflections and safe usage, so the invention can be used in the precise paper sheet quantitative detection of the field of paper making.

To provide a scintillator for neutron detection which has high sensitivity to neutron rays, and is reduced in a background noise attributed to γ rays.[Means to Solve the Problems] A scintillator for neutron detection, comprising a metal fluoride crystal containing, as constituent elements, a metal having a valence of 2 or higher, such as calcium, aluminum or yttrium; lithium; and fluorine, the metal fluoride crystal containing 1.1 to 20 atoms per unit volume (atoms / nm3) of 6Li, having an effective atomic number of 10 to 40, containing a lanthanoid such as cerium, praseodymium or europium, and being represented by LiCaAlF6, LiSrAlF6, LiYF4 etc.

The invention discloses a small target detection method based on regional nomination. The method comprises: step one, carrying out input image preprocessing and converting the input image into one with a unified size; step two, extracting an image base feature map and extracting five-layer feature maps; step three, carrying out feature fusion and carrying out fusion of the fourth-layer feature mapand the fifth-layer feature map; step four, carrying out small target area nomination and generating small target area nomination by using a regional nomination network; and step five, carrying out small target bounding frame refinement and small target area classification. The method has advantages of fast calculation speed, high small target recognition precision, good generalization performance, and good detection performance of common minimum target areas.

The embodiment of the invention relates to the electric power inspection technical field, and discloses a power transmission line detection method and device; the method comprises the following steps: carrying out frame difference operation between two adjacent frame images of a collected visible light video sequence so as to obtain a difference image, and binarizing same to obtain a binary difference image; using a preset threshold edge detection algorithm to carry out edge detection for the binary difference image so as to obtain an edge binary difference image; carrying out Hough conversion so as to detect linear sections; carrying out cycle iterations and screening merging for detected linear sections according to the transmission line parallel characteristics until the obtained target linear section number is smaller than a preset value, and marking and displaying same; calculating the central position of a target zone, and sending the central position to a hang cabin so as to adjust the posture of the hang cabin. The method and device can effectively improve the power transmission line detection efficiency and accuracy, thus reducing work load of inspection personnel.

The invention discloses a weak signal detection method based on joint denoising and frequency modulation, and belongs to the technical field of signal processing. The method comprises the steps that 1) partial noise suppressing is carried out on low signal to noise ratio signals which are to be detected and are mixed with noise through an autocorrelator; 2) wavelet threshold transforming is carried out on the signals which are processed in the step of 1) and are to be detected, so as to remove residual noise after signal autocorrelation; 3) a modulatable standard signal is used to modulate the frequency of the signals output in the step of 2), so that the modulated signal frequency can meet the detection condition of a dual Duffing oscillator difference system, and high frequency signals in the modulated signals are filtered out through a low pass filter; 4) the signals output by the low pass filter in step of 3) are input into the dual Duffing oscillator difference detection system, and initial state parameters of the system are set; and 5) whether the dual Duffing oscillator difference detection system has the phenomenon of regular intermittence chaos is judged, and if the phenomenon of regular intermittence chaos occurs, a continuous zero method is used to acquire an intermittence chaos period and the frequency of the signals to be detected. Compared with a traditional detection method, the method can be used to detect the weak signals of any frequency in a background with strong noise, and is of great significance in practical application.

The invention discloses an intelligent gear defect analysis method based on fractional wavelet transform and a BP neutral network. The intelligent gear defect analysis method includes: taking transform order as a variable to perform fractional Fourier transform on gear vibration signals to determine optimal order, and performing fractional wavelet transform on the gear vibration signals under the optimal order for denoising to realize separation of useful component and background noise of the gear vibration signals; calculating feature parameters of the signals after being denoised to form a group of feature vectors which are used for representing features of gear vibration after denoising; averagely dividing the feature vectors into two groups which serve as a training sample and a testing sample respectively, and inputting the feature vectors into the BP neutral network for learning and classifying. By the intelligent gear defect analysis method, background noise mixed in the gear meshing vibration signals is inhibited well, useful signal component related to defects is retained, and gear defect features can be extracted effectively; self learning and classifying capability of the BP neutral network is utilized, so that defect mode of gears can be quickly recognized qualitatively with high accuracy.

The present invention discloses a wireless telephone system using microphone arrays together with additional signal processing to suppress the background noise in the surrounding environment. The signal processing resources of a wireless telephone and multi-channel transmission capabilities of the Bluetooth transmission are used to suppress the background noise. The wireless telephone system includes a Bluetooth transceiver communicating to a wireless telephone through a multi-channel Bluetooth transmission, and an array of microphones coupled to the Bluetooth transceiver. The array of microphones includes a first microphone producing a first audio signal output and a second microphone producing a second audio signal output. The first audio signal output and second audio signal output are transmitted to the wireless telephone through the first channel and second channel of multi-channel Bluetooth transmission respectively. The system and method of the present invention allows using low cost Bluetooth transceiver(s) with multiple microphone arrays to provide the background noise suppression.

A multi-beam sounding sonar water body imaging beamforming algorithm relates to the field of signal processing. The multi-beam sounding sonar water body imaging beamforming algorithm mainly comprises:compensating for the propagation loss of the acoustic wave according to a time gain curve in each detection sampling time, and obtaining a background noise level of the current detected water area after time averaging; performing near-field focusing beamforming on the signal and estimating the number of sources under the current snapshot sequence number according to the current background noise level; performing covariance matrix estimation on a signal vector with a snapshot number of 1, and obtaining a new pseudo covariance matrix by reconstructing the data covariance matrix after the forward and backward smoothness; performing singular value decomposition on the pseudo-covariance matrix, using conventional beamforming output results and an array manifold to construct a spatial spectralfunction, and obtaining a multi-beam sounding sonar water body imaging result. The algorithm can be widely applied to the multi-beam sounding sonar water body imaging function, can effectively suppress the background noise of the multi-beam sounding sonar water body imaging, and can improve the sonar imaging quality.

The invention provides a total internal reflection fluorescence microscopic imaging method and device based on senior secondary axisymmetric polarized beams. The device comprises a laser device for emitting laser beams, a pinhole filter for performing spatial filtration on the laser beams emitted from the laser device, a collimation lens for collimating the laser beams subjected to spatial filtration into parallel light beams, a polarization conversion system for performing polarization state conversion on the parallel light beams to obtain the senior secondary axisymmetric polarized beams, a pupil filter and an annular diaphragm which are used for modulating the vibration amplitude and the phase of the obtained senior secondary axisymmetric polarized beams, a reflecting and focusing system for focusing the modulated axisymmetric polarized beams on an interface between glass and a sample to produce an evanescent field on the interface between the glass and the sample because of total reflection, a reflecting and filtering system for focusing a fluorescence signal emitted from the evanescent field on a pinhole array plate, a photoelectric detector for detecting the signal, and a signal analysis processing system.

A sound pickup device is provided, the device including (1) a directionality forming unit that forms directionality to output of a microphone array, (2) a target area sound extraction unit that extracts non-target area sound from output of the directionality forming unit, and that suppresses non-target area sound components extracted from output of the directionality forming unit so as to extract target area sound, (3) a determination information computation unit that computes determination information, (4) an area sound determination unit that determines whether or not target area sound is present using the determination information computed by the determination information computation unit, and (5) an output unit that outputs the target area sound extracted only in cases in which the target area sound is determined to be present by the area sound determination unit.

A voice input device includes a first microphone, a second microphone, and a processor. The second microphone has a lower distance decay rate than the first microphone. The processor is configured to acquire noise information of noise by comparing a first signal obtained from the first microphone with a second signal obtained from the second microphone. The processor is further configured to perform noise suppression processing based on the noise information.

A mobile device that is capable of automatically starting and ending the recording of an audio signal captured by at least one microphone is presented. The mobile device is capable of adjusting a number of parameters related with audio logging based on the context information of the audio input signal.

A semantic segmentation method and apparatus for a three-dimensional image, and a storage medium are provided. The method includes: obtaining a three-dimensional image; slicing the three-dimensional image according to three directional planes, to obtain two-dimensional slice images of an x axis, two-dimensional slice images of a y axis, and two-dimensional slice images of a z axis; invoking a first segmentation model, a second segmentation model, and a third segmentation model to respectively perform semantic segmentation on the two-dimensional slice images of the x axis, the y axis, and the z axis, to obtain distribution probability maps of a target object on the three directional planes; and obtaining a three-dimensional distribution binary image of the target object by invoking an adaptive fusion model to perform three-dimensional fusion on the three distribution probability maps respectively corresponding to an x-axis directional plane, a y-axis directional plane, and a z-axis directional plane.

The invention discloses an all-weather active type panoramic sensing device which comprises a mobile object laser light source, a multi-element output all-around visual sensor and a micro processor, wherein a mirror surface with double curved surfaces and two camera units are arranged in the multi-element output all-around visual sensor; a polarization splitting prism is arranged on a refraction and reflection light path of the mirror surface with double curved surfaces is used for splitting the refraction and reflection light of the mirror surface with double curved surfaces into light containing a certain polarized component and light not containing a certain polarized component; the two camera units are respectively positioned on reflection and refraction light paths of the polarization splitting prism, and are used for respectively acquiring a first panoramic video image containing polarized light information and a second panoramic video image containing light intensity information; the micro processor is used for fusing point cloud geometrical information in the first panoramic video image with the color information in the second panoramic video image, and constructing a panoramic 3D model. The invention further discloses a 3D panoramic modeling approach adopting the all-weather active type panoramic sensing device.