33 results about "Deconvolution filter" patented technology

A technique for controlling a loudspeaker system with an artificial neural network includes filtering, with a deconvolution filter, a measured system response of a loudspeaker and a reverberant environment in which the loudspeaker is disposed to generate a filtered response, wherein the measured system response corresponds to an audio input signal applied to the loudspeaker while the loudspeaker is disposed in the reverberant environment. The techniques further include generating, via a neural network model, an initial neural network output based on the audio input signal, comparing the initial neural network output to the filtered response to determine an error value, and generating, via the neural network model, an updated neural network output based on the audio input signal and the error value.

The present invention is a method of processing seismic data in which one or more seismic vibrators are activated with one or more pilot signals and vibrator motions are recorded along with seismic data. Vibrator signatures are computed from measured vibrator motions, such as the ground force signal. A desired impulse response is specified from either a measured vibrator motion or from test data or field data from a location near the location from which the seismic data was acquired. A deconvolution filter is computed from the impulse response and the vibrator signature. Alternatively, a single separation and deconvolution filter is derived from the impulse response and from vibrator signatures from multiple vibrators and sweeps. The deconvolution or deconvolution and separation filter is used to process the seismic data. The vibrators are then moved to a new location, and the activation is repeated.

The present invention is a method of processing seismic data in which one or more seismic vibrators are activated with one or more pilot signals and vibrator motions are recorded along with seismic data. Vibrator signatures are computed from measured vibrator motions, such as the ground force signal. A desired impulse response is specified from either a measured vibrator motion or from test data or field data from a location near the location from which the seismic data was acquired. A deconvolution filter is computed from the impulse response and the vibrator signature. Alternatively, a single separation and deconvolution filter is derived from the impulse response and from vibrator signatures from multiple vibrators and sweeps. The deconvolution or deconvolution and separation filter is used to process the seismic data. The vibrators are then moved to a new location, and the activation is repeated.

A method is provided for averaging a sequence of image frames. A noise-reducing filter is applied to the image frames to generate filtered frames. A deconvolution filter is applied to the filtered frames to generate corresponding deconvolved frames. The filtered frames are transformed by an affine transformation to align them, generating aligned frames. The aligned frames are motion corrected by non-linear transformation based on intensity rank matching, generating a sequence of motion-corrected frames. The motion-corrected frames are averaged to generate a resultant frame.

Provided is an image restoration method and apparatus, in which an image is restored by estimating the blur level of the image. The image restoration apparatus includes a block generation module which generates a plurality of overlapping blocks for an input image; a blur estimation module which generates a plurality of unit step responses for a current overlapping block from the overlapping blocks; a feature-map generation module which generates a feature map for the current overlapping block based on the unit step responses of the current overlapping block; and a filter application module which deblurs the current overlapping block by using a deconvolution filter to filter the feature map of the current overlapping block.

The present disclosure relates to iterative reconstruction of images. In certain embodiments, a deconvolution filter is used to approximate the inversion of a Hessian matrix associated with the reconstruction. In one such embodiment, the desired image is not reconstructed directly in the iterative process. Instead, an image is reconstructed that yields the desired image when filtered by the deconvolution filter.

The invention discloses a deconvolution method of multi-path double spectral domain mixed phase wavelets, comprising the following steps: selecting n earthquake records from the collected seismic records; determining the amplitude spectrum of the wavelet according to the selected n earthquake records; determining the phase spectrum of the wavelet according to the selected n earthquake records; implementing the inverse Fourier transform according to the amplitude spectrum and the phase spectrum of the determined wavelet to obtain the mixed phase wavelet of the time domain; determining the zero phase wavelet which is expected to output according to the amplitude spectrum of the expected output wavelet input by users based on the amplitude spectrum of the determined wavelet; adopting the mixed phase wavelet as input and adopting the zero phase wavelet as output to determine the deconvolution filter through the filter wave matching; and obtaining the earthquake records of which the resolution ratio is improved based on the deconvolution filter and the earthquake records.

Methods and systems for extracting formation properties from formation logging data is disclosed. A method includes obtaining a deconvolution filter; and processing the formation logging data using the deconvolution filter to produce estimates of the formation properties. The estimates of the formation properties may be further processed with a diffusion filter, which may be an anisotropic diffusion filter. A system for extracting formation properties from formation logging data includes a central processing unit and a memory, wherein the memory stores a program having instructions for performing a method that includes designing a deconvolution filter; and processing the formation logging data using the deconvolution filter to produce estimates of the formation properties.

Provided is a processing method for seismic records The method comprises the following steps: multi-channels of seismic records are selected and each channel of seismic records is converted into a cepstrum domain in order to determine a cepstrum sequence of each channel of earthquake record; an amplitude spectrum of an input wavelet and an input wavelet trend fitting curve are determined through the cepstrum sequence of the input wavelet of seismic records; Fourier inversion is carried out on the amplitude spectrum of the input wavelet to determine the input wavelet; signal-to noise-ratio spectrums of the multi-channels of seismic records are computed and based on the input wavelet trend fitting curve, and with reference to a curved of the signal-to noise-ratio spectrum, an amplitude spectrum of the expected output wavelet is obtained; Fourier inversion is carried out on the amplitude spectrum of the expected output wavelet to determines the expected output wavelet; a deconvolution filter is obtained by means of carrying out matched filtering of the input wavelet and the expected output wavelet; deconvolution is carried out on the seismic records based on a convolution model in order to obtain a reflection coefficient sequence; and convolution is carried out on the deconvolution filter and the reflection coefficient sequence to obtained optimized seimic records.

Methods and systems for extracting formation properties from formation logging data is disclosed. A method includes obtaining a deconvolution filter; and processing the formation logging data using the deconvolution filter to produce estimates of the formation properties. The estimates of the formation properties may be further processed with a diffusion filter, which may be an anisotropic diffusion filter. A system for extracting formation properties from formation logging data includes a central processing unit and a memory, wherein the memory stores a program having instructions for performing a method that includes designing a deconvolution filter; and processing the formation logging data using the deconvolution filter to produce estimates of the formation properties.

The invention discloses a single-frame defogging method based on multi-scale deconvolution filtering. Through a minimum value filtering method and a deconvolution filtering method, a rough transmittance image and a fine transmittance image are acquired respectively. Multi-scale integration is performed on the rough transmittance image and the fine transmittance image so as to acquire a final minimum value filtering transmittance image and a deconvolution filtering transmittance image. And then according to an atmospheric physical model, image defogging calculating is performed so that a degraded input image can be recovered. Through minimum value filtering, a rough image is acquired, and through deconvolution filtering, a fine image is acquired. In the invention, recovered image detail performance and a contrast ratio can be increased.

The invention provides an air gun array wavelet directional deconvolution method and system. The method comprises the following steps: obtaining a directional deconvolution filter through calculationaccording to the spatial position of each air gun and corresponding imaginary wavelets; obtaining angle domain seismic data through [tau]-p transform on the basis of common detection point seismic data; obtaining seismic record data through calculation on the basis of the angle domain seismic data and the directional deconvolution filter; on the basis of the seismic record data, obtaining final seismic data through inverse Fourier transform and inverse [tau]-p transform. In this way, the spatial distribution of an air gun array and the influence of sea surface ghosting are comprehensively taken into consideration, the directional deconvolution filter is constructed to perform directional deconvolution on offshore seismic data, directional effects of wavelets of the air gun array are eliminated, and the processing accuracy of offshore seismic data is improved.

A three-dimensional imaging device (10) according to an aspect of the present invention, which outputs first and second viewpoint images due to a pupil division on a light flux from a single imaging optical system (12, 14), includes a defocus map calculation unit (61) calculating a defocus map representing a defocus amount at each position in the first and second viewpoint images; a restoration filter storage (62) storing restoration filters corresponding to the defocus amount and an image height at each position in the viewpoint images, the restoration filters being deconvolution filters based on a point image intensity distribution representing an imaging performance of the imaging optical system; and a restoration unit (63) selecting a restoration filter, which corresponds to the defocus amount and the image height at each position in the viewpoint images, for each position in the first and second viewpoint images, and restoring the first and second viewpoint images by performing a deconvolution for each position in the first and second viewpoint images based on the selected restoration filter.

A method of filtering seismic signals is described using the steps of obtaining the seismic signals generated by activating a seismic source and recording signals emanating from the source at one or more receivers; defining a source signature deconvolution filter to filter the seismic signal, wherein the filter is scaled by a frequency-dependent term based on an estimate of the signal-to-noise (S / N) based on the spectral power of a signal common to a suite of angle-dependent far-field signatures normalized by the total spectral power of the signatures within the angular suite and performing a source signature deconvolution using the source signature deconvolution filter.

A method is provided for averaging a sequence of image frames. A noise-reducing filter is applied to the image frames to generate filtered frames. A deconvolution filter is applied to the filtered frames to generate corresponding deconvolved frames. The filtered frames are transformed by an affine transformation to align them, generating aligned frames. The aligned frames are motion corrected by non-linear transformation based on intensity rank matching, generating a sequence of motion-corrected frames. The motion-corrected frames are averaged to generate a resultant frame.

The invention discloses a camera module, electronic equipment, an optical element and a preparation method of the camera module. The camera module comprises a lens group, the optical element, an imagesensor and a processor, wherein the optical element is arranged in a Fourier transform plane between two lenses; the image sensor receives the diffracted image light reduced by the optical element and converts the image light into a first image signal, and the first image signal is used for being transmitted to a convolution filter in the processor to be subjected to deconvolution filtering processing so as to compensate blurring to obtain a second image signal. Optical elements are arranged in Fourier transformation planes of the two lenses, so that diffraction of light passing through the camera module is reduced. The diffracted image light is received by the image sensor and is converted into a first image signal by the image sensor, so that deconvolution filtering processing is carried out on the first image signal to further compensate blurring to obtain a second image signal, and the diffraction of the light passing through the camera module can be reduced.

An ultrasonic measuring device including: an ultrasonic transducer device; an emission unit for emitting an ultrasonic beam; a reception unit for receiving an ultrasonic echo reflected by a test subject; and a processing unit for processing reception, wherein the processing unit identifies a transfer function with respect to the ultrasonic transducer device and the test subject based on a first reception signal corresponding to an ultrasonic beam radiated to a first area in the test subject, a second reception signal corresponding to an ultrasonic beam radiated to a second area in the test subject, and a third reception signal corresponding to an ultrasonic beam radiated to a third area located between the first area and the second area in the test subject, and performs ultrasonic image generation processing including filter processing using a deconvolution filter including the transfer function performed on the reception signals.

The embodiment of the invention discloses a method and device for imaging an adjoining well by acoustic logging. The method comprises the following steps of collecting dipole longitudinal wave data infour orientations when a dipole sound source is excitated, wherein the dipole sound source is in an orthogonal dipole mode; obtaining dipole longitudinal wave data in a plurality of geographic orientations through vector rotation according to the collected dipole longitudinal wave data in four orientations; carrying out deconvolution filtering processing on the obtained dipole longitudinal wave data in each orientation to obtain filtered sound wave data; carrying out wave field separation on the filtered sound wave data to obtain separated upstream reflective longitudinal wave data and downstream reflective longitudinal wave data; carrying out migration imaging processing on the upstream reflective longitudinal wave data and the downstream reflective longitudinal wave data obtained by separating the wave field in each geographic orientation to obtain imaging data of a borehole of the adjoining well. According to the scheme, the adjoining well can be imaged by means of acoustic logging.

An ultrasonic measuring device including: an ultrasonic transducer device; an emission unit for emitting an ultrasonic beam; a reception unit for receiving an ultrasonic echo reflected by a test subject; and a processing unit for processing reception, wherein the processing unit identifies a transfer function with respect to the ultrasonic transducer device and the test subject based on a first reception signal corresponding to an ultrasonic beam radiated to a first area in the test subject, a second reception signal corresponding to an ultrasonic beam radiated to a second area in the test subject, and a third reception signal corresponding to an ultrasonic beam radiated to a third area located between the first area and the second area in the test subject, and performs ultrasonic image generation processing including filter processing using a deconvolution filter including the transfer function performed on the reception signals.

The present disclosure relates to iterative reconstruction of images. In certain embodiments, a deconvolution filter is used to approximate the inversion of a Hessian matrix associated with the reconstruction. In one such embodiment, the desired image is not reconstructed directly in the iterative process. Instead, an image is reconstructed that yields the desired image when filtered by the deconvolution filter.

The invention relates to a method for keeping signal to noise ratio and enhancing seismic record resolution, which is applied to the field of seismic exploration signal processing. The method mainly comprises the following steps of: manually exciting and recording a seismic signal; carrying out the conventional seismic signal processing, namely noise suppression, deconvolution, velocity analysis, excursion, and the like, on the seismic signal; estimating a seismic wavelet amplitude spectrum; estimating a signal purity spectrum; computing a deconvolution filtering operator by utilizing the seismic wavelet amplitude spectrum and the signal purity spectrum, and carrying out deconvolution processing on a seismic record; and drawing the seismic record into a seismic section image capable of reflecting underground structural features. The seismic record processed by the method disclosed by the invention can better identify the geologic structure and the fracture system which are smaller in size and more concealed.

The application discloses a camera module, electronic equipment, optical elements and a method for preparing the camera module. The camera module includes a lens group, an optical element, an image sensor and a processor. In the conversion plane, the image sensor receives the image light after the diffraction is reduced by the optical element and converts the image light into a first image signal, and the first image signal is sent to the convolution filter in the processor for deconvolution filter processing to Blur is compensated to obtain a second image signal. The diffraction of light passing through the camera module is reduced by arranging optical elements in the Fourier transform plane of the two lenses. The image light after the reduced diffraction is received by the image sensor and converted into the first image signal by the image sensor, which facilitates the deconvolution filtering process on the first image signal to further compensate for the blur to obtain the second image signal, so that the camera module can Reduces the diffraction of light passing through its interior.

The invention discloses a seismic exploration multi-frequency data fusion method based on weight deconvolution. The method comprises the following steps that: an objective function based on a plurality of single-frequency seismic data is constructed; a weight deconvolution filtering operator corresponding to the single-frequency seismic data is obtained based on the objective function by using the signal-to-noise ratio difference between different data, and the operator is applied to the corresponding single-frequency seismic data; and finally, superposition is performed to obtain a fusion result. According to the method, direct association between the objective function and the single-frequency seismic data is established, and weight filtering processing is performed by using the signal-to-noise ratio differences of the different data, so that the problem of possible excessive noise amplification in a subsequent deconvolution process caused by introduction of low-signal-to-noise-ratio data is avoided.

The invention discloses a construction method for a short-time remote photoplethysmography signal detection model. The method comprises the following steps: acquiring a face video image sequence, and preprocessing the face video image sequence as an initial data set; processing the collected photoplethysmography signals to serve as a target set; and training a short-time remote photoplethysmography signal detection model. An encoder, a decoder, a branch loss module and a residual constant block in feature extraction constructed based on a 3D space-time convolution filter and a deconvolution filter, and a significant feature extraction module based on a CBAM attention mechanism are designed. The design of the encoder and the decoder is used for carrying out scale transformation under time-space domain features and time domain features, it is guaranteed that effective features highly related to short-time remote photoplethysmography signal time sequence information are reserved in the feature extraction process, and the performance of the model is improved. On the basis of a CBAM attention mechanism-based significant feature extraction module, the perceptual feature extraction capability is improved, and the problem of low robustness in the prior art is solved.

Provided is a processing method for seismic records The method comprises the following steps: multi-channels of seismic records are selected and each channel of seismic records is converted into a cepstrum domain in order to determine a cepstrum sequence of each channel of earthquake record; an amplitude spectrum of an input wavelet and an input wavelet trend fitting curve are determined through the cepstrum sequence of the input wavelet of seismic records; Fourier inversion is carried out on the amplitude spectrum of the input wavelet to determine the input wavelet; signal-to noise-ratio spectrums of the multi-channels of seismic records are computed and based on the input wavelet trend fitting curve, and with reference to a curved of the signal-to noise-ratio spectrum, an amplitude spectrum of the expected output wavelet is obtained; Fourier inversion is carried out on the amplitude spectrum of the expected output wavelet to determines the expected output wavelet; a deconvolution filter is obtained by means of carrying out matched filtering of the input wavelet and the expected output wavelet; deconvolution is carried out on the seismic records based on a convolution model in order to obtain a reflection coefficient sequence; and convolution is carried out on the deconvolution filter and the reflection coefficient sequence to obtained optimized seimic records.

The invention discloses an on-line measurement method for underwater installation of a jacket based on image restoration, and the method comprises the following steps: firstly, carrying out the imagecollection of the underwater jacket through a camera, and obtaining an original sequence image; secondly, carrying out time domain mean filtering on the obtained original sequence image through a high-pass filter; thirdly, carrying out deconvolution filtering through a Gaussian Wiener space; and finally, solving the part pitch through the calibrated underwater camera parameter model. The method has the advantages that the method is suitable for underwater three-dimensional measurement, the algorithm execution speed is lower than 0.5s, the average error of pitch detection of three-dimensional measurement is within 0.1 mm, the execution speed is reduced, the detection precision is improved, and the working efficiency is improved for underwater installation of the jacket.

The invention discloses a resolution enhancement method for an ultrasonic microscope. The method comprises the following steps: firstly, acquiring ultrasonic tomography data of a test piece; carryingout multi-scale analysis on the ultrasonic signal of each point by adopting discrete wavelet decomposition to generate a low-frequency component and a high-frequency component; processing the low-frequency components by using a Wiener deconvolution filter, and separating the superposed signals; conducting discrete wavelet inverse transformation by using the separated signals and high-frequency components to reconstruct ultrasonic tomography data; and generating vertical and horizontal profile scan images of higher resolution in the test piece by using the reconstructed ultrasound tomography scan data. The problem that the imaging resolution of an existing ultrasonic microscope is difficult to improve is solved.