256 results about "Orthogonal matching pursuit algorithm" patented technology

The invention discloses a compressed sensing-oriented block-sparse signal reconfiguring method, and particularly relates to a block-sparse signal reconfiguring algorithm, which aims to solve the problems that the optimization complexity of a mixed l2 / l1 optimization algorithm in the conventional block-sparse signal reconfiguring algorithm is relatively higher and that overmatching phenomenon is easily caused by a block-sparse matching pursuit algorithm or a block-sparse orthogonal matching pursuit algorithm. The method of the invention comprises the following steps of: correcting labels, in ameasurement matrix, of column vectors of a recovery matrix calculated in the iteration operation of the (l-1)th time by performing the iteration of the lth time, and for a block-sparse signal x with the block sparsity of K, reconfiguring the block-sparse signal x by performing the iteration for not more than K times. The method is applied to the reconfiguration of the block-sparse signal, particularly to the reconfiguration of a binary block-sparse signal.

The invention discloses a sparse OFDM channel estimation method based on a generalized orthogonal matching tracking algorithm. The method comprises the following steps: step one, translating a channelestimation problem into a problem for reconstructing original signal based on the compressed sensing theory; step two, designing a measurement matrix; and step three, reconstructing the original signal by using the generalized orthogonal matching tracking method so as to finish the channel estimation. The sparse OFDM channel estimation method based on the generalized orthogonal matching trackingalgorithm in the compressed sensing disclosed by the invention comprises the steps of translating the channel estimation problem into the problem for reconstructing original signal based on the compressed sensing theory, designing the measurement matrix and reconstructing the original signal by using the generalized orthogonal matching tracking algorithm. The operation complexity, namely the running time, are greatly reduced, the impulse response of the channel is precisely estimated, the system performance of the OFDM sparse channel estimation is improved so as to improve the signal demodulation quality, and the method has high application value.

The invention discloses a voice secret communication system design method based on compressive sensing and information hiding, comprising the following steps: embedding secret voice into carrier voice by an embedded system to obtain mixed voice; designing a compressive sensing overcomplete dictionary aiming at the voice signal; sampling the secret voice by a compressive sensing self-adaption observation matrix to obtain a observation vector for reducing dimensions; quantizing the observation vector by an LBG (Linde-Buzo-Gray algorithm) vector, taking the quantized observation vector to serve as secret information to embed into the carrier voice, and carrying out two-stage transform on the carrier voice to obtain mixed voice; extracting the secret voice from the mixed voice by an extraction system; carrying out discrete cosine transform on mixed voice, and improving wavelet transform two-stage transform to obtain a wavelet transform coefficient; obtaining a secret bit stream by a scalar Costa decoding algorithm; obtaining a reconstructing observation vector by an LBG vector quantization decoder; reconstructing the secret voice by a compressive sensing orthogonal matching pursuit algorithm; and improving the quality of the reconstructed secret voice with a wavelet denoising method.

The invention discloses a sparse representation and dictionary learning-based bearing fault classification method. The method is characterized by comprising the following steps of: preprocessing acquired history bearing vibration signals to obtain vibration signals under different working states, respectively constructing corresponding sub-dictionaries and combining the sub-dictionaries into a redundant dictionary; and online acquiring a bearing vibration signal by using a sensor, solving a sparse coefficient of the signal under the redundant dictionary by using a generalized orthogonal matching pursuit algorithm, and classifying the vibration signal through a reconstruction error so as to identify the bearing working state. According to the method, relatively good classification effect can be obtained, the dictionary training process can be accelerated and the ability, of being adapted to target signals, of the dictionary is improved, so that the method is capable of realizing the sparse decomposition of complex vibration signals more efficiently and is used for fault identification.

The invention discloses a multi-source image fusion method based on the synchronous orthogonal matching pursuit algorithm. The multi-source image fusion method comprises the following steps: sampling a source image pixel by pixel in an overlapping manner into image blocks of the same size by a sliding window of the fixed size and expanding each image block by columns into column vectors; obtaining the sparse representation coefficient corresponding to each vector on the over-complete dictionary by the synchronous orthogonal matching pursuit algorithm; fusing the corresponding coefficient by the maximum absolute value method; inverse-transforming the fused sparse representation coefficient into the fusion result vector of corresponding to the vectors according to the over-complete dictionary; and restoring all the fusion result vectors to image blocks and re-constructing to obtain the fused image. The invention fully considers the intrinsic characteristics of the image sparsity and the method using sparse representation can more effectively present the useful information of each source image and achieve better fusion effect, therefore, the invention is of great significance and practical value to the post-processing and image display of various application systems.

Disclosed is a shaft sleeve part surface defect on-line detection method based on compressed sensing. Compressed sensing description of a part surface defect image is built through a machine vision and compressed sensing method, and an optical imaging and defect detection model highlighting surface defects is built; a part sample image of typical defects is collected, after denoising and necessary image preprocessing are carried out, sampling frequency adjustment and size normalization are carried out, a sample is trained and a redundant dictionary is built; a proper orthogonal basis decomposition matrix and a random observation matrix are designed, a combined orthogonal matching pursuit algorithm is selected, solution of the minimum norm l0 is converted into the problem of solving the optimal solution to reconstruct a defect image, spare representation of the image to be detected is calculated, and defect recognition is carried out on a part to be detected according to built judgment and recognition standards. An on-line detection system with the functions of feeding, positioning and adjustment, image collection, image processing, defect detection and recognition, part separation and the like is built, and rapid detection on the surface defects of the shaft sleeve part is achieved.

The invention discloses an orthogonal-matching-pursuit-based sparse channel estimation method, which comprises the following steps of: taking an input vector signal as a channel impulse response sequence, constructing a measurement matrix by utilizing a know input training sequence, and constructing a sparse channel impulse response model for recovering and reconstructing the input vector signal by receiving a vector signal; and estimating the position and amplitude of each channel by adopting an orthogonal matching pursuit algorithm to obtain a channel estimation value. In the method, the precision and anti-interference performance of channel parameter estimation can be effectively improved because the orthogonal matching pursuit algorithm is easy to realize and high in proximity; and the sparseness of the channels is taken into account and iteration is performed for relatively fewer times by adopting the orthogonal matching pursuit algorithm to estimate the position and amplitude of a plurality of sparse channels, so that the complexity of a channel estimation method can be effectively reduced.

A low level wind shear velocity estimation method based on compressed sensing comprises the following steps: 1), utilizing Doppler vectors to establish a redundant dictionary so as to achieve sparse representation of an echo signal; 2), establishing a measurement matrix; 3), computing a measured value after the signal is compressed; 4), utilizing a regularization orthogonal matching pursuit algorithm to reconfigure a sparse signal; 5), utilizing a non-zero value in a complex amplitude estimation value to obtain Doppler frequency estimation of the radar echo signal; 6), computing a position of a maximum value in the complex amplitude estimation value after clutter rejection is performed, namely a velocity estimation result of a wind shear signal; and 7), repeating the step 3) to the step 6), judging whether the velocity estimation is accomplished according to the number of range units, and sequentially performing velocity estimation on echo data of all the range units within the scope. Aiming at the problem that velocity estimation precision is poor when the number of pulses is small and a signal to noise ratio (SNR) is low, the low level wind shear velocity estimation method based on the compressed sensing is provided. The low level wind shear velocity estimation method based on the compressed sensing can enable frequency spectrum resolution to be greatly improved while the accurate velocity estimation can be obtained, namely, the wind shear signal and the ground clutter signal can be well distinguished, wherein an interval between a frequency domain of the wind shear signal and a frequency domain of the ground clutter signal is very close.

The invention provides a hard threshold OMP (orthogonal matching pursuit)-based linear array SAR (synthetic aperture radar) sparse imaging method. A linear measurement matrix of original echo signals of a linear array SAR and scattering coefficients in a target space of an observed scene is established for the characteristic that main scattering targets are spatially sparse in the target space of the observed scene of the linear array SAR, and contrast between maximum and minimum target scattering coefficients and a target scattering coefficient change rate are used as iteration ending conditions for the iteration processing of a hard threshold OMP algorithm, so that the dependence of a conventional OMP algorithm on the number of the main scattering targets in linear array SAR sparse imaging is overcome. Compared with a conventional OMP algorithm-based linear array SAR sparse imaging method, so that the method has the advantages that the number of the main scattering targets in the target space of the observed scene is not required to be known, and the method is more applicable to the linear array SAR sparse imaging when the number of the main scattering targets is unknown under actual conditions; the imaging accuracy of the linear array SAR is improved. The method can be applied to the fields of SAR imaging, earth remote sensing and the like.

The invention discloses an electric power system data reconfiguration decompressing method based on orthogonal matching pursuit. A compressed sensing theory is adopted for conducting parallel data compression of sampling and compression on an electric power quality signal. The method includes that first, a line with the largest correlation with margin is selected in a sensing matrix, and simultaneously a selected space is updated. By solving a least square problem, residual is guaranteed to be minimum, sparse vector elements are obtained, then, the residual is updated, selected lines in the sensing matrix are removed, and finally sparse elements are obtained through loop iteration. By means of the method, the compressed sensing theory is adopted for conducting sparse decomposition on electric power quality data, then gauss measuring coding is conducted on sparse signals, and finally a signal is reconfigured through an orthogonal matching pursuit algorithm. The method breaks through a traditional data compression method framework of first sampling and then compressing, sampling and compression are conducted parallelly, a small amount of sampling can recover an original electric power quality signal well, a requirement for hardware is reduced, and compression efficiency is improved.

The invention discloses a channel estimation method for the sparse characteristic of underwater acoustic block structure, which includes three modules: an OFDM system overall working module; a block orthogonal matching tracking algorithm flow module; and a block sparse compressed sensing channel estimation module. This channel estimation method can effectively reduce signal sparsity, thereby reducing the number of pilots required for estimation, greatly reducing the time required for algorithm estimation, and improving the accuracy of reconstruction.

The invention relates to a mechanical vibration fault characteristic time domain blind extraction method, and belongs to the technical field of mechanical equipment status monitor and fault diagnosis. The mechanical vibration fault characteristic time domain blind extraction method includes: firstly, expanding a vibration observation signal into a high dimension signal subspace; then, obtaining a low dimension signal; afterwards, performing FastICA independent component analysis, calculating normalization kurtosis of all independent components, figuring out a component signal corresponding to the minimum normalization kurtosis, and using an orthogonal matching pursuit algorithm to reconstitute periodic signals; subsequently, removing the reconstituted periodic signal from each independent component, and then using an improved KL distance algorithm to calculate a distance matrix among the independent components after the periodic signals are removed from the independent components, and performing dynamic particle swarm clustering so as to obtain an estimation signal; finally, analyzing an envelope demodulation spectrum of the estimation signal, and performing fault diagnosis. The mechanical vibration fault characteristic time domain blind extraction method is suitable for processing a long convolution data problem, can effectively reduce influences from periodic ingredients on a blind separation result, and simultaneously can solve blind separation result order uncertainty problems, and finally achieves bearing fault characteristic extraction.

The invention relates to a FiberBragg grating(FBG) signal self-adapting restoration method based on compressed sensing, and belongs to a signal restoration technology field of an optical fiber sensing system. The FBG signal self-adapting restoration method comprises steps that step 1: EMD combination mutual information is used for self-adapting denoising processing of spectral signals; step 2, segmented testing of a denoising signal is carried out, and the signal is divided into k segments, and sample databases corresponding to the signals are acquired by calculating Euclidean distances among various segments of signals and samples, and self-adapting dictionaries D corresponding to the signals are acquired by adopting a K-SVD dictionary learning method; step 3, measured signals are used to acquire observation matrixes R and observation signals xi; step 4, the observation signals are reconstructed by adopting an improved regularized orthogonal matching pursuit algorithm to acquire complete reconstructed signals. The FBG signal self-adapting restoration method is advantageous in that problems such as interferences of noises on the signals, targeted dictionary learning, and the signal self-adapting reconstruction are considered, and each part represents the self-adaptability of the algorithm, and can be flexibly used in practical engineering, and then influences caused by manual misoperation are reduced.

The invention, which belongs to the communication technology, especially relates to a compression-sensing-based direction-of-arrival (DOA) estimation method for an intelligent antenna. According to a sparse characteristic of a target signal in a natural space in an airspace, a CS-theory-based DOA estimation model under single sampling is constructed; on the basis of the constructed DOA estimation model, CS-theory-based DOA estimation is realized, thereby obtaining a DOA of the target signal; and with an orthogonal matching pursuit algorithm, a reconstruction problem of a s spatial sparse signal in the CS-theory-based DOA estimation is solved, and iteration is carried out by multiple times and thus a residual error between an estimation signal and an original signal is reduced continuously, so that the estimation signal is close to the original signal and DOA estimation is completed. The method has the following beneficial effects: relevant information can be estimated accurately based on a little sampling information; and thus novel DOA estimation method is provided.

The invention discloses a sparse representation-based deblocking method, which mainly solves the problem of the presence of a blocking effect in a block discrete cosine transform (BDCT) compressed image. The method comprises the following implementation steps of: (1) selecting a clean training image set and training a general dictionary with a kernel singular value decomposition (KSVD) algorithm and a batch processing orthogonal matching pursuit algorithm; (2) compressing a test image by controlling a quality factor during joint photographic experts group (JPEG) compression so as to obtain a JPEG compressed image; (3) calculating the noise standard deviation of the JPEG compressed image; (4) automatically estimating an error threshold according to the quality factor and the noise standarddeviation; (5) constructing an image block matrix of the JPEG compressed image so as to obtain a de-noised sparse representation matrix; and (6) obtaining a deblocking result image by using the general dictionary and the sparse representation matrix. Compared with the prior art, the invention has the advantages that: a higher or similar peak signal to noise ratio can be obtained, the visual effect of a deblocked image is good, computation complexity is low, and a blocking effect in a BDCT compressed image can be eliminated.

The invention discloses a multiband signal reconstruction method based on a clustering sparse regularization orthogonal matching tracking algorithm, relates to the technical field of information and communication, and aims at solving a problem of restoring an original multiband signal from an unknown-sparseness multi-observation-value vector in a Xampling frame after the conversion through a continuous-finite module and sampling through a modulation bandwidth converter. Because many simulation signals met a multiband signal model in a process of signal processing, the method plays an important role in enabling the compressed sensing theory to be used for simulation signals. The basic idea of the method is to convert an infinite observation value vector problem into a single-observation-value vector problem. The method comprises the following steps: carrying out the column vectorization of observation values; carrying out the extension of an observation matrix through a Kronecker product; estimating a support set of the original signal through employing the above two results and signal sparseness estimation; finally reconstructing the signal, wherein the regularization idea is used in an estimation process of the support set.

The invention discloses a target high-definition Doppler information reconstruction method of a frequency-agile radar based on a sparse reconstruction algorithm, and mainly solves the problem that a traditional signal phase-coherent accumulation method cannot be used to extract the target Doppler information from a radar in the frequency agile system. The method comprises that distance compensation is carried out on echoes by means of a distance compensation matrix and a dictionary matrix; a compensated echo signal is correlated to the dictionary matrix, optimal compensation is searched for, high-definition distance of the target is obtained, and the dimension of the signal is reduced; and high-definition Doppler information reconstruction is carried out on the compensated signal by meansof an orthogonal matching and tracking algorithm. Compared with a present method, the target characteristic is reconstructed in higher precision and stability, the calculating speed and signal processing instantaneity in the reconstruction process are improved, and the method can be used to fields including design of a real-time radar signal processing system, radar anti-interference and electronic reconnoitering and confrontation.

The invention discloses a digital predistorter design method and device based on power amplifier model cutting, and the method comprises the steps: transmitting an original input signal to a hardwarecommunication system, and obtaining an output signal of a radio frequency power amplifier through a hardware feedback channel; carrying out down-conversion operation and digital sampling on the outputsignal, and then carrying out frequency alignment on the output signal and the input signal; carrying out autocorrelation synchronization processing and normalization operation on the output signal and the input signal; establishing a power amplifier model between the input signal and the output signal by using the generalized memory polynomial model; cutting the power amplifier model by adoptinga blind sparse segmented weak orthogonal matching pursuit algorithm to obtain a simplified power amplifier model; and performing inversion operation on the simplified power amplifier module to obtainthe power amplifier digital predistorter. According to the method, the GMP model is cut through the blind sparse SWOMP algorithm, and the power amplifier sparse model with high sparse capability andaccuracy is constructed, so that the digital predistorter with low complexity and high accuracy is obtained.

The invention discloses a high-resolution rapid deconvolution sound source imaging algorithm which is characterized by comprising the following steps: calculating a point spread function of a sound source at the central position of a sound source calculation plane by utilizing approximation space translation invariance of the point spread function in the process of constructing a point spread function matrix of the deconvolution sound source imaging algorithm; constructing the point spread function matrix through a method for circularly shifting the point spread function at the central position upwards and downwards, so that calculation of the total point spread functions is avoided, the calculated amount, and the calculation speed and efficiency are improved; realizing rapid sparse deconvolution reconstruction of sound source intensity energy distribution through an orthogonal matching pursuit algorithm by utilizing space sparse prior of the sound source and combining compressed sensing in the deconvolution reconstruction process of the sound source intensity energy distribution, so that the iterations are reduced, and the calculation efficiency and resolution ratio are improved. The algorithm disclosed by the invention has high calculation efficiency and spatial resolution and is capable of well rapidly identifying and locating the position of the sound source in the space.

The invention discloses an image sparse representation method based on a Curvelet redundant dictionary, mainly aiming to solve the problems that in the existing method, the redundant dictionary has large scale, the calculation complexity is high, and sparse representation can not be effectively carried out on the rich border outline details in the image. The invention is realized through the following steps: (1) selecting the tight frame of Curvelet as an atomic model; (2) determining the numeric areas of the scale parameter j, direction parameter theta and displacement parameter k in the frame, carrying out discretization on each parameter to form the Curvelet redundant dictionary; and (3) blocking each input image, carrying out sparse decomposition on each sub-image by utilizing an orthogonal matching pursuit (OMP) algorithm sparse decomposition to solve sparse coefficient vectors, combining all the sparse coefficient vectors to obtain the sparse matrix, and multiplying the sparse matrix by the Curvelet redundant dictionary to obtain the sparse representation results of the input image. Compared with the prior art, the invention has the advantages of low calculation complexity, high quality of sparse representation image, especially can better capture the singularity of curves in the image, and can be applied to the fields of image processing and computer vision.

The invention discloses a nuclear magnetic resonance T2 spectrum inversion method based on an orthogonal matching pursuit algorithm. A T2 spectrum non-zero value range is determined by utilizing the orthogonal matching pursuit algorithm and then a T2 spectrum is solved in the non-zero value range by adopting a following singular value decomposition algorithm improved by a regularization method; firstly, a problem initial solution is calculated by adopting the regularization method and then the singular value decomposition algorithm is used for carrying out non-negative iteration. The nuclear magnetic resonance T2 spectrum inversion method comprises the following steps: (1): reading original well logging data Y; (2): carrying out median filtering processing on the data and calculating a signal matrix A; (3): calculating a T2 spectrum non-zero zone by utilizing the orthogonal matching pursuit algorithm; (4): calculating the T2 spectrum of a problem in the T2 spectrum non-zero zone by utilizing the singular value decomposition algorithm improved by the regularization method; (5): calculating whether an error is in an allowable range or not; if so, outputting the final T2 spectrum; otherwise, taking the calculated T2 spectrum as the initial solution and returning back to STEP 4 for recalculating; and (6): outputting the final T2 spectrum. The regularization method is introduced and the initial solution is provided, so that the calculation precision of the algorithm can be improved to the great extent.

The invention relates to the field of the identification and positioning of a noise source and particularly relates to a sound source positioning method based on a novel orthogonal matching pursuit algorithm. The method comprises a step of forming a measuring surface, a step of collecting sound pressure data at each sensor, a step of forming a focusing surface and obtaining a focusing point, a step of establishing a relationship between a sound source source-intensity vector and a microphone array sound pressure measurement value, a step of solving a sound source recognition model and obtaining the sound source source-intensity of each grid point of the focusing surface, and a step of identifying and positioning a sound source according to a module of the obtained sound source source-intensity of each grid point. According to the invention, the atomic selection process of the orthogonal matching pursuit algorithm is improved, therefore, the algorithm reconstruction performance under the condition of strong correlation between atoms is improved, the resolution of the sound source recognition is further improved, the high-resolution positioning of the sound source in a strong correlation environment is achieved, and at the same time, the ability to recognize low frequency and medium frequency signals is also effectively improved.

The invention relates to a K-SVD learning dictionary based woven fabric texture flaw detection method. The whole woven fabric texture image is decomposed into multiple sub-images, a flaw contained sub-image is obtained by discrimination, and the position of a flaw of a fabric is determined according to the position of the flaw contained sub-image; and flaw discrimination is realized by comparing and reconstructing the sub-images, all the sub-images are unfolded into column vectors and then combined to obtain a test sample image matrix, discrete cosine transform is selected for an initial dictionary, an initial sparse coefficient matrix is solved from the initial dictionary and a training sample image matrix via an orthogonal matching and tracking algorithm, K-SVD dictionary learning is carried out on the training sample image matrix to obtain a dictionary, a sparse coefficient matrix is solved from the dictionary and the test sample image matrix via the orthogonal matching and trackingalgorithm, the test sample image matrix is reconstructed, and column vectors of a reconstructed sample image matrix are converted into reconstructed sub-images. According to the method of the invention, detection is rapid and accurate, and a detection result is stable and highly adaptive.

The invention relates to a traffic signal lamp recognition method based on Gabor and sparse representation, comprising the following steps: (1) carrying out 2D Gabor wavelet filtering on a region of interest, and acquiring a filtered Gabor image as a test sample, wherein the region of interest is a traffic signal lamp region; (2) based on a trained over-complete dictionary, calculating the sparse coefficient of the test sample using an orthogonal matching pursuit algorithm, wherein the over-complete dictionary is trained based on a K-SVD algorithm; and (3) carrying out image reconstruction using the sparse coefficient and the over-complete dictionary to get reconstructed images corresponding to the traffic signal lamp categories, and judging to which traffic signal lamp category the test sample belongs according to the residual error between the test sample and each reconstructed image. Compared with the prior art, traffic signal lamp recognition is carried out based on Gabor and sparse representation, big error caused by arrow signal lamp direction recognition is overcome, and small round and arrow signal lamps in an image can be recognized correctly.

The invention discloses a rolling bearing fault sparseness diagnosis method based on average random weak orthogonal matching pursuit. The method comprises the steps of firstly, constructing an overcomplete dictionary according to a collected rolling bearing vibration signal, completing initialized setting of algorithm parameters, and estimating sparseness of an original signal; secondly, adoptingan average random weak orthogonal matching pursuit algorithm to update a sparseness dictionary and residual errors; finally, and using the obtained sparseness dictionary to calculate sparseness representation coefficients, so that a fault signal is obtained through reconstruction. The steps are repeated N times, and the final processing result is obtained through set average. By means of the rolling bearing fault sparseness diagnosis method, through a residual error updating mode of estimating and improving atomicity, the influence of artificial setting of the sparseness on the decomposition result is avoided; through an improved simulated annealing algorithm, the probability that small-amplitude fault components are extracted is increased, the problem that weak periodic impact features are difficult to extract effectively is solved, and the method is significant in achieving weak fault diagnosis of a rolling bearing in the early period.

The invention discloses a harmonic detection method based on compressive sampling orthogonal matching pursuit. The harmonic detection method comprises the following steps of: firstly, carrying out compressive sampling on an original harmonic signal, and subsequently carrying out harmonic detection and separation on a sampling sequence value directly by using an orthogonal matching pursuit algorithm. The sparseness does not need to be estimated, the frequency characteristic corresponding to each harmonic component comprises two spectral lines, and the sparseness in a compressive sampling orthogonal matching pursuit harmonic detection algorithm is a definite quantity, so that errors caused by the sparseness estimation are avoided; the characteristic quantity is modified in each iteration, and a redundancy error value is updated so as to establish a new redundancy signal proxy, and furthermore identify a maximum element in a present component; and the reconstruction of the original signal does not need to be carried out, however the original harmonic signal can be precisely detected with very few signal sampling number, so that the burden of sampling equipment is reduced, the storage space of intermediate variables is saved, and interested fundamental waves and sub-harmonic components are directly detected from the compressive signal.

The invention discloses a bearing fault intelligent diagnosis method based on compressed sensing and a correlation vector machine. According to the method, fault diagnosis is achieved through vibration signal analysis. The method comprises the steps of firstly selecting a Gaussian random matrix as a measurement matrix based on a compressed sensing theory to realize compressed sampling of signals,secondly constructing an over-complete redundant dictionary to perform sparse representation on the signals, then utilizing an orthogonal matching pursuit algorithm to realize signal reconstruction, and selecting a time domain index sensitive to fault features as a feature vector for the reconstructed signals; and finally, selecting a Gaussian function as a kernel function, dividing a training sample and a test sample by utilizing a feature vector, importing the training sample into an intelligent recognizer of a relevance vector machine model constructed by a relevance vector machine, and comparing a test result with an actual fault type and degree to obtain the effectiveness of the diagnosis model. According to the invention, the problems of difficult transmission and processing of massdata and low signal-to-noise ratio of bearing vibration signals can be solved, and qualitative and quantitative identification of bearing faults can be realized more accurately.

The invention discloses a handwriting digital recognition method and a handwriting digital recognition device. The method comprises the following steps of acquiring a target figure and an image data set in a target image which are to be recognized, wherein the image data set comprises at least one piece of image data; performing vector conversion on each piece of image data in the image data set to obtain vector data corresponding to each piece of image data, wherein the vector data forms a vector data set; performing dimension reduction on the vector data set by using a local sparse linear embedding dimension reduction method based on an orthogonal matching tracing algorithm to obtain a dimension reduction vector data set; and according to the dimension reduction vector data set, identifying a digital image corresponding to the target figure in the target image. Not all of computational solutions obtained by the orthogonal matching tracing algorithm in the method are zero. Compared with the prior art, the method and the device have the advantages that local sparsity can be obviously embodied, stability is realized, and the accuracy of the digital image which is finally recognized is relatively high.

The invention discloses a material surface defect detection method based on compressed sensing. The method comprises the following steps: first, a to-be-detected material surface low-resolution image acquisition model is established; then, super-resolution reconstruction is carried out on a material surface image, a Haar wavelet is adopted as the basis function of a representation function, and transform basis matrix design is carried out on the basis function by the use of discrete cosine transform and discrete wavelet transform; next, during super-resolution reconstruction of the material surface image, an orthogonal matching pursuit algorithm is adopted to orthogonalize a selected bionic cell in the iterative process to enable the reconstruction process to be converged in finite steps; and finally, macro block processing is introduced to complete reconstruction of the to-be-detected material surface image. Based on the principle of compressed sensing, super-resolution reconstruction is carried out by the use of a general sparse transform domain, and target recognition and detection are achieved on the basis.

The invention provides a signal reconstruction method of the OMP based on rapid inner product calculation under a distributed type CS framework, relates to the technical field of multichannel CS, and aims to solve the problem that the reconstruction speed of a conventional signal reconstruction method is low. According to the signal reconstruction method, after multichannel signals are subjected to compression observation, signal reconstruction is performed with the OMP, rapid calculation of the inner product in the OMP is realized by the aid of the internal sparsity of a joint observation matrix, and signal reconstruction is realized further. The signal reconstruction method is applicable to signal reconstruction of the OMP based on the rapid inner product calculation under the distributed type CS framework.