27 results about "Curvelet transform" patented technology

Curvelet-based low level fusion of camera and RADAR sensor information is disclosed and includes processing RADAR data corresponding to a scene to generate a RADAR point cloud and processing camera image data corresponding to the scene using a curvelet transform to identify a target of interest in the scene and generate for the target of interest a target type, (x,y) coordinate values, and a curvelet magnitude per decomposition level. If discrepancies exist between (x,y) coordinate values of the RADAR point cloud and the target type, (x,y) coordinate values, and curvelet magnitude per composition level of the target of interest, a portion of the RADAR data processing is repeated to regenerate the RADAR point cloud; otherwise, the RADAR point cloud to a perception stack of a vehicle.

The invention relates to a vector seismic data random noise suppression method and device based on fast quaternion curvelet transformation, a medium and equipment, and the method comprises the steps: representing X, Y and Z as a quaternion array form Q = Xi + Yj + Zk in a time-space domain, and forming a two-dimensional quaternion signal f; according to the data features, fast quaternion curvelet forward transformation is used for decomposing f in Ncsalt scales and Ncoarse directions, and a corresponding quaternion curvelet transformation coefficient c is obtained; estimating thresholds under various scales and directions, performing threshold operation on c of different scales and different directions to obtain a quaternion curvelet transform coefficient after threshold processing, and performing fast quaternion curvelet inverse transformation to represent fast quaternion curvelet inverse transformation; reconstructing the denoised quaternion signal, taking out three corresponding components from three imaginary part positions of the denoised quaternion signal, completing vector denoising, and obtaining a denoising result sum;

The invention discloses a three-dimensional seismic data reconstruction method based on a weighted fast iterative shrinkage threshold algorithm, and the method aims at solving a problem that a conventional three-dimensional seismic data reconstruction method neglects the potential correlation between adjacent slices, and comprises the steps: firstly dividing three-dimensional seismic data into a series of time slices; and carrying out multi-scale multi-direction two-dimensional curvelet transformation to obtain a curvelet domain coefficient, then introducing a fast iterative shrinkage threshold algorithm, and carrying out reconstruction by adopting an index threshold parameter formula. In the process, according to the characteristic that curvelet coefficient support sets of two adjacent time slices have a large intersection, a weighting operator is constructed in a curvelet domain by using a prior support set provided by a previous reconstructed time slice, and then the weighting operator is fused into a fast iterative shrinkage threshold algorithm. According to the method, the signal-to-noise ratio of the reconstructed signal is increased, the calculation speed is increased, the requirement for a computer memory is reduced, weak effective wave signals are protected, and the reflected wave event is more continuous and clearer.

The present invention discloses a method for denoising distributed optical sensing signals based on curvelet transform, comprising the following steps: continuously collecting original Rayleigh backscattered signals generated by multiple optical pulses; performing orthogonal demodulation on the original signals to obtain demodulated data; stacking the demodulated data in the order of the optical pulse time to form a two-dimensional matrix; performing fast discrete curvelet transform and scale decomposition on the two-dimensional matrix to obtain a curvelet coefficient matrix; performing threshold processing on the submatrix blocks corresponding to each direction of each scale to obtain a processed curvelet coefficient matrix; performing fast discrete inverse curvelet transform on the processed curvelet coefficient matrix to obtain a transformed matrix; performing moving difference on the transformed matrix to determine the disturbance position on the optical fiber using the amplitude difference between the backscattered signals generated by optical pulses at different times. The present invention has the beneficial effects of being simple and direct, having strong adaptability, and being able to perform fast and effective noise reduction.

The application relates to the technical field of petroleum geophysical exploration processing, and provides a method, device, equipment, medium and product for improving the signal-to-noise ratio of reconstructed data, which is characterized in that the signal-to-noise ratio of each trace of a target layer in an irregular observation system is calculated as the signal-to-noise ratio of the target layer, a three-dimensional array is constructed for data of a regular observation system designed based on the irregular observation system, the three-dimensional array is valued based on the irregular observation system, the three-dimensional array after the valuation is calculated based on three-dimensional curvelet transformation, and re-valuation is performed according to the signal-to-noise ratio of the data of the target layer of adjacent traces; the use proportion of high signal-to-noise ratio data in the reconstruction process is improved; after iterative calculation, the reconstructed data is output; the continuity constraint in each direction and the signal-to-noise ratio information of the target layer in the three-dimensional data can be fully utilized; the use proportion of high signal-to-noise ratio data in the reconstruction process is improved; and the signal-to-noise ratio of the reconstructed data is improved, thereby meeting the demand of exploration production.

The invention provides a noise channel identification and reconstruction method and device based on multi-attribute fusion. The method comprises the following steps: configuring a seed gun according to a geological task, dividing a plurality of region sets by taking an excitation point as a center, determining a time window range of a seismic channel in a region through spatial interpolation of a seed gun first arrival wave time window, and generating seismic channel data. And analyzing the signal energy of the first arrival wave time window and the noise time window, carrying out anomaly identification in combination with statistical characteristics, and if abnormal, dividing the abnormal channels into abnormal channels and carrying out zero filling processing. Constructing a three-dimensional data volume, and performing multi-scale curvelet transformation on the three-dimensional data volume to obtain a curvelet coefficient; and carrying out iteration on the curvelet coefficient to realize reconstruction of the seismic channel, and correcting the amplitude of the reconstructed seismic channel through spatial interpolation to obtain reconstructed shot domain data so as to complete identification and reconstruction of the noise channel. According to the method, the abnormal channels are effectively eliminated, the quality of the seismic data with the low signal-to-noise ratio is improved, and a reliable data basis is provided for high-precision imaging under complex geological conditions.

This invention pertains to data processing technology for seismic exploration of sandstone-type uranium deposits, specifically an adaptive denoising method for sandstone-type uranium deposit seismic data based on sparse transform. The method includes: inputting noisy raw two-dimensional seismic data y; initializing the maximum number of iterations I, setting the outer iterative accumulator i=1, and designing a search range J for the denoising coefficient truncation percentage; for each i, calculating the curvelet transform coefficients s of y; for each j within the search range, applying hard thresholding for denoising, and performing an inverse curvelet transform on the denoised coefficients to obtain the denoised seismic data y. j ; Calculate the noise removal ε corresponding to each j j The entropy; compare the calculated noise ε removed for each j. j The entropy is calculated, and the percentage corresponding to the maximum entropy within the search range is taken as the optimal denoising effect, and the corresponding seismic data is output. If i reaches the maximum number of iterations, the loop is exited, and the denoised seismic data corresponding to the current maximum entropy value is used as the output result. This invention can protect the effective signal while suppressing random noise.

A method for accurately segmenting submarine cables from side-scan sonar images comprises the following steps: inputting an image I1, processing the image I1 using a cable feature automatic extraction filter based on a curvelet transform, extracting cable-related information, and obtaining a feature image I2; pixel-by-pixel multiplying the image I1 by the feature image I2 to obtain a feature image I3; using a Hough transform to find no less than three straight line segments in the feature image I3, recording the angle α of the longest line segment in the image, constructing linear structural units in the same direction as the cables in the image, performing noise elimination and binarization processing on the feature image I3 to obtain a binary image I4; performing cable morphology restoration on the binary image I4 to obtain a cable morphology restoration image I5, and retaining only the largest connected area in the image as the cable segmentation result for the submarine side-scan sonar image. The method solves the problems in side-scan sonar image segmentation that a large amount of data is required for training, high image clarity requirements are required, target color information needs to be obtained, and high image size requirements are required.

The invention relates to a method for inverting ground visibility based on curvelet satellite data. The method comprises the following steps: a, carrying out gridding processing on satellite waveband data with the maximum geometric resolution; b, performing curvelet transformation on the satellite wave band data subjected to gridding processing; c, counting the mean square error of the direction coefficient matrix of each direction of the secondary maximum scale; d, calculating a mean value of mean square errors of all direction coefficient matrixes of the second maximum scale after curvelet transformation; e, carrying out regression analysis on a mean value of mean square errors of all direction coefficient matrixes and a visibility measured value; and f, according to a regression analysis result, establishing an estimation function of a mean value of mean square errors of the direction coefficient matrix and visibility. The method improves the precision and effect of ground visibility inversion based on satellite data, and is mainly applied to the fields of regional atmospheric environmental pollution monitoring, aviation, navigation, road traffic, military activities and the like.

The invention provides a multiple suppression method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring seismic data; curvelet transformation parameters are obtained; transforming the seismic data into a curvelet domain by using a curvelet transformation technology based on the curvelet transformation parameters; according to the distribution positions of the multiples in the curvelet domain, curvelet plates of the multiples distributed in the curvelet domain are removed, and primary curvelet plates are obtained; and performing anti-curvelet transformation on the primary curvelet plate to obtain seismic data after multiple suppression. The method comprises the following steps: transforming seismic data into a curvelet domain by utilizing a curvelet transformation technology, and separating and suppressing multiples in the curvelet domain by utilizing curvelet transformation parameters according to the principle that the characteristics of the multiples are different from the characteristics of primary waves. Compared with other multiple suppression methods, the method has the advantages of being high in operation efficiency, good in suppression effect, accurate in multiple recognition and high in denoising fidelity.

The application discloses an InSAR interference phase adaptive filtering method based on Curvelet transform, and steps are as follows: 1, preprocessing the original InSAR interference phase diagram to obtain a standardized interference phase diagram; 2, converting the original InSAR interference phase diagram to a complex domain, and then performing Curvelet transform on the real part and the imaginary part respectively to obtain the coefficient distribution of the real part and the imaginary part in different scales and directions; 3, the noise standard deviation is estimated by using the median absolute deviation (MAD) method based on the fixed threshold method of Donoho for the coefficients of Curvelet of each scale layer; 4, the threshold size of the real part and the imaginary part is calculated based on the improved threshold formula; 5, inverse Curvelet transform is performed on the coefficients of the real part and the imaginary part after soft threshold processing, and the filtered interference phase diagram is obtained by taking the phase angle. The application can effectively suppress noise, significantly retain interference fringe edges and detail structures, and has lower calculation complexity and higher precision.

The present invention relates to a seismic data denoising method and system based on adaptive equalization. The method comprises acquiring seismic data, performing two-dimensional smoothing on the acquired seismic data to obtain a smoothed profile; performing equalization on the acquired seismic data using the smoothed profile; and sequentially performing curvelet transform, curvelet domain denoising, inverse curvelet transform, and inverse equalization on the equalized seismic data to obtain denoised target seismic data. This method solves the existing problem of poor denoising due to poor energy equalization in shallow, medium, and deep depth seismic data.

The invention is suitable for the technical field of disease detection, and provides an automatic detection method of a corn leaf disease target, comprising the following steps: acquiring an original corn leaf image; extracting parallel features of the image, and respectively capturing global structure features and local texture features; fusing and coding the global structure features and the local texture features; coding is carried out by using a Transform encoder; target query is introduced, and decoding prediction is carried out; the prediction result is matched with a real label, distribution of each query is optimized, a detection result with the highest matching confidence coefficient is reserved, and bounding boxes and corresponding classifications of all scab areas in the image are output. According to the method, Curvelet transformation and a Transform architecture are combined, a multi-scale and multi-direction structural information modeling mechanism is introduced, and the multi-scale and multi-direction structural information modeling mechanism and the local texture features extracted by the CNN are fused and modeled in the Transform, so that the perception capability of a disease area is improved.

This invention relates to a noise reduction method for removing stripes along the survey line direction in airborne geophysical data, comprising the following steps: a. gridding the airborne geophysical data; b. performing curvelet transform on the gridded airborne geophysical data; c. calculating the direction coefficient indices to be processed at each scale; d. thresholding, setting all direction coefficients to be processed calculated in step c to zero; e. performing inverse curvelet transform on the curvelet coefficients after thresholding; f. discretizing the inverse curvelet transform results into airborne geophysical data. This invention can quickly and effectively remove stripes along the survey line direction while retaining anomaly information, thus improving the processing efficiency of airborne geophysical data. This invention is mainly applied to airborne geophysical exploration for mineral exploration, searching for metallic minerals, oil and gas, and radioactive minerals. It can also be applied to tasks such as emergency airborne monitoring without upper crystal nuclei, airborne geophysical exploration, and environmental monitoring.

The embodiment of the invention relates to the technical field of data processing, and discloses a data denoising method and device, electronic equipment, a storage medium and a program product, and the method comprises the steps: obtaining to-be-processed DAS data in a mining area environment; curvelet transformation is carried out on the to-be-processed DAS data to obtain a curvelet coefficient set, the curvelet coefficient set corresponds to multiple scales, the multiple scales comprise at least one key scale, and the at least one key scale is determined based on a frequency band to which an effective signal in the mining area environment belongs; the density of multiple directions corresponding to each key scale in a key direction range is higher than the density in a non-key direction range, and the key directions are determined based on geological features of a mining area environment; performing denoising processing on the curvelet coefficient set to obtain a denoised curvelet coefficient set; and generating first DAS data based on the de-noised curvelet coefficient set. In this way, the quality of the DAS data is improved.

The invention provides an infrared image enhancement method and device based on histogram matching and curvelet transformation. The method comprises the following steps: acquiring a to-be-enhanced infrared image and a reference image; respectively calculating a gray average value and a gray standard deviation of the infrared image and the reference image; calculating a correction coefficient and an average correction value based on the gray average value and the gray standard deviation; generating a histogram matching enhanced image of the infrared image based on the correction coefficient and the average correction value; curvelet transform decomposition is carried out on the histogram matching enhanced image and the infrared image to obtain corresponding multi-scale and multi-direction sub-band coefficients, and fusion is carried out on the corresponding sub-band coefficients under the same scale and the same direction to obtain a fusion sub-band coefficient; and performing curvelet inverse transformation on the fusion sub-band coefficient to obtain a fusion enhanced image. According to the invention, through the cooperation of the global gray correction of histogram matching and the local detail fusion of curvelet transformation, the contrast, definition and detail information of the infrared image are synchronously improved.

The invention discloses a data processing method, electronic equipment, a storage medium and a computer program product, and the method comprises the steps: converting first seismic data into corresponding first wave field data; curvelet transformation is carried out on the first wave field data to obtain a first curvelet coefficient set; extracting one or more first curvelet coefficients corresponding to a first diffracted wave from the first curvelet coefficient set through a first model; and processing the one or more first curvelet coefficients to obtain an image of the first diffracted wave. Through the method and the device, the diffracted waves in the seismic data can be separated more accurately, so that the imaging precision of the diffracted waves is improved.

The present invention discloses a method for predicting alcohol consumption by combining curvelet transform and convolutional neural network, which belongs to the technical field of specific computing models. The method comprises: obtaining a time-feature data set based on user basic profiles, daily behavior trajectories, and purchase information results of alcohol consumption users; obtaining a plurality of curvelet coefficients based on the time-feature data set by performing curvelet transform in combination with the curvelet characteristic frequency, the plurality of curvelet coefficients including low-frequency curvelet coefficients, medium-frequency curvelet coefficients, high-frequency curvelet coefficients, and ultra-high-frequency curvelet coefficients; obtaining a plurality of feature information based on the low-frequency curvelet coefficients, medium-frequency curvelet coefficients, and high-frequency curvelet coefficients by performing feature decomposition; obtaining a plurality of convolution operation results based on the plurality of feature information by performing convolutional neural network, obtaining a pooling operation result by performing pooling operation, and obtaining an alcohol consumption prediction by performing linear correction activation.

This disclosure relates to the field of computer technology, and particularly to a gather repair method, apparatus, device, storage medium, and computer program. The gather repair method includes: acquiring seismic data, the seismic data including multiple seismic coordinate points, the seismic coordinate points being the coordinate points where earthquakes occur, and the seismic coordinate points being located in a Cartesian coordinate system; extracting curvelet coefficients corresponding to at least one spatial parameter of the seismic data based on a curvelet transform algorithm to obtain at least one curvelet coefficient; acquiring source-receiver offset vector surface element (OVT) gathers corresponding to the at least one curvelet coefficient, the OVT gathers being seismic data with the same spatial information; extracting anisotropy parameters of each curvelet coefficient, and using the anisotropy parameters of each curvelet coefficient to repair the corresponding OVT gathers, obtaining target OVT gathers corresponding to each OVT gather. The embodiments of this disclosure improve the accuracy of gather repair.

The invention discloses a first arrival refraction area linear interference suppression method, a product, a medium and equipment, and belongs to the technical field of seismic data processing. The method comprises the following steps: designing different parameters aiming at inconsistent linear characteristics of different frequency band ranges; the multi-scale and multi-directional representation capability of curvelet transform is utilized to decompose seismic data into subsets with different scales and directions, and then accurate wave field separation is realized through Radon transform in each curvelet domain, so that refraction interference with relatively strong energy in a first arrival refraction region is eliminated, and effective reflection signals in a large offset range are recovered. The method can accurately suppress first-arrival refraction interference, recover effective signals, and improve seismic imaging quality.

The application discloses a kind of multi-sampling rate seismic data volume reconstruction method and device.The method includes, according to multi-sampling rate seismic data volume, establish observation system regular empty data volume, assign the data of matching way in seismic data volume or set initial value for the way therein, constitute initial data volume;Initial data volume is divided into multiple small data volumes, one-dimensional Fourier transform of time variable is carried out to each small data volume, and the data volume of f-x-y domain is obtained;For each f-x-y domain data volume, the transformed data volume is obtained by three-dimensional curve wave iterative transformation;The initial reconstruction data volume is obtained by combining the transformed data volume;According to seismic data volume, the initial reconstruction data volume is corrected, and the reconstruction data volume is obtained.The curve wave transformation of three-dimensional seismic data volume is carried out in f-x-y domain using split beam, the problem of large memory occupation in large data volume three-dimensional curve wave transformation is solved, and the reconstruction of multi-sampling rate seismic data can be completed with high fidelity and high efficiency.

The present invention relates to a curvelet domain high-pass filtering diffraction separation method. The method comprises obtaining seismic common-offset data for identifying hot dry rock faults; applying a curvelet transform to the seismic common-offset data to obtain multi-scale curvelet domain data, thereby obtaining a curvelet coefficient matrix containing reflection and diffraction top information; performing a two-dimensional fast Fourier transform on the coefficient matrix to obtain its spectrum; extracting the high-frequency portion of the spectrum using a frequency domain bandpass filter to separate the reflection and diffraction curvelet coefficients; and performing an inverse curvelet transform on the separated coefficients to obtain separated common-offset diffraction wave data. The method can solve the energy loss problem of diffraction wave separation imaging, thereby improving the fault resolution of hot dry rock seismic data, and has important application value in geothermal resource exploration and construction site selection.

The invention discloses a weak signal enhancement method, device and equipment based on a three-dimensional curvelet domain in a low signal-to-noise area and a medium, and belongs to the field of seismic exploration, and the method comprises the steps: decomposing seismic data into harmonic signal combinations with different frequencies and amplitudes, and carrying out the three-dimensional curvelet transformation; extracting curvelet coefficient extreme values of the decomposed signals in each scale direction in a three-dimensional curvelet transform domain; establishing a time window in the three-dimensional seismic data, and extracting a curvelet coefficient extreme value by establishing a weak signal threshold value range; screening out weak reflection signals containing effective waves in all seismic data; carrying out inverse transformation on the screened curvelet coefficient, and transforming the curvelet coefficient back to a time domain to obtain a time domain signal of which the noise is filtered; and performing energy enhancement on the time domain signal to obtain a final mid-deep layer weak reflection signal. According to the method, the characteristic of high sparse representation capability of three-dimensional curvelet transformation is utilized, separation of effective signals and noise is completed, and finally the purposes of identifying and enhancing weak reflection signals in a low signal-to-noise area are achieved.

The present disclosure relates to the technical field of image analysis, and provides an image analysis method and device. The method comprises: acquiring an original image containing a target to be recognized; introducing the original image into a preset recognition model to recognize a target rectangular region; cutting a target sub-image from the original image based on the target rectangular region; performing multi-resolution curvelet transform on the target sub-image to generate a target feature vector; and comparing the target feature vector with a preset template feature vector to determine whether the target to be recognized in the original image is abnormal. The embodiments of the present disclosure can determine the state of the insulator in the whole and in the details synchronously, thereby greatly improving the accuracy and effectiveness of the analysis result.

Methods and systems for denoising media frames captured in low-light environment are provided. A method includes obtaining a media frame in pixel space using adaptive trimming of Fourier coefficients of an input media frame based on a threshold criterion, and obtaining a media frame in curvelet space by performing curvelet transform on the media frame in pixel space, extracting a texture frame based on texture details and a homogeneous frame based on homogeneous details, present in the media frame in pixel space, performing longitudinal and latitudinal filtering of the texture and homogeneous frames downscaled by a same factor, integrating the longitudinal and latitudinal filtered texture and homogeneous frames, upscaling the integration to obtain a curvelet media frame, and obtaining an output media frame by sharpening the curvelet media frame in multiple iterations based on partially sharpened curvelet media frames, the curvelet media frame, and the media frame in curvelet space.

The invention discloses a seismic signal denoising method based on direction disturbance and sparse transformation response, and relates to the technical field of seismic signal processing. The method comprises the following steps: extracting a two-dimensional seismic section from three-dimensional post-stack seismic data to obtain two-dimensional post-stack seismic data, constructing a direction disturbance mask, and performing direction disturbance on the two-dimensional post-stack seismic data by using the direction disturbance mask to obtain disturbed two-dimensional post-stack seismic data. Synchronously carrying out synchronous compression curvelet transformation on the two-dimensional post-stack seismic data before and after disturbance in a sparse representation domain, obtaining sparse disturbance response indexes of synchronous compression curvelet coefficients of the two-dimensional post-stack seismic data, cooperatively calculating and self-adapting threshold values, screening the synchronous compression curvelet coefficients, obtaining the synchronous compression curvelet coefficients of the seismic signals, and finally obtaining the sparse disturbance response indexes of the synchronous compression curvelet coefficients of the seismic signals. Based on a multi-round disturbance and fusion strategy, multi-round disturbance results are fused to obtain denoised two-dimensional post-stack seismic data, random noise is weakened, weak reflection signals are enhanced, and the signal-to-noise ratio and structural definition of the seismic data are improved.

The application provides an infrared image enhancement method and device based on histogram matching and curvelet transform. The method comprises: acquiring an infrared image to be enhanced and a reference image; calculating the gray mean value and the gray standard deviation of the infrared image and the reference image respectively; calculating a correction coefficient and an average correction amount based on the gray mean value and the gray standard deviation; generating a histogram matching enhanced image of the infrared image based on the correction coefficient and the average correction amount; respectively performing curvelet transform decomposition on the histogram matching enhanced image and the infrared image to obtain corresponding multi-scale and multi-direction sub-band coefficients, and fusing corresponding sub-band coefficients at the same scale and the same direction to obtain fused sub-band coefficients; and performing curvelet inverse transform on the fused sub-band coefficients to obtain a fused enhanced image. Through the cooperation of the global gray correction of histogram matching and the local detail fusion of curvelet transform, the application realizes the synchronous improvement of the contrast, the definition and the detail information of the infrared image.