32 results about "Singular value thresholding" patented technology

The invention discloses a low-rank structure-based sparse compressive sensing MRI (Magnetic Resonance Imaging) image reconstruction method, which mainly solves the problem of difficulty in accurate recovery of an MRI image existing in the conventional technology. The method comprises the following implementation steps: initially recovering the MRI image by using the conventional compressive sensing and looking for a similar block matrix from the image to form an index set; performing singular value decomposition on the similar block matrix and calculating a threshold, and performing threshold calculation on a singular value by using the threshold to obtain an after-threshold singular value; and optimizing the MRI image by using the after-threshold singular value, i.e., circularly performing processes of updating of the similar block matrix and the index thereof, similar block matrix singular value decomposition and threshold and singular value threshold calculation on the MRI image to obtain a final recovery image. The recovered MRI image is clearer and has sharper edges; and the method can be used for processing a medical image.

The invention relates to a periodic long-short code direct sequence spread spectrum code division multiple access signal multi-pseudo-code estimation method. The existing direct sequence spread spectrum code division multiple access signal pseudo code blind estimation technology cannot be applied to periodic long-short code direct sequence spread spectrum code division multiple access signals adopting short code spread spectrum and long code scrambling. The method comprises the steps of: constructing periodic long-short code direct sequence spread spectrum code division multiple access signals with complicated structures into a deficit matrix model of short code direct sequence spread spectrum code division multiple access signals, and modeling compound code matrix estimation as a blind source signal separation problem; applying a matrix filling theory to the compound code matrix estimation, and estimating user compound code sequences based on a singular value threshold algorithm and a fast independent component analysis algorithm; and finally proposing a delay triple correlation algorithm by utilizing a displacement stacking feature of an m sequence, and estimating long-short pseudo code sequences contained in the user compound code sequences. The periodic long-short code direct sequence spread spectrum code division multiple access signal multi-pseudo-code estimation method fully utilizes the matrix filling mathematic model and the triple correlation peak feature of the m sequence, and achieves blind estimation of the user compound code sequences, a long scrambling code sequence and a short spread spectrum code sequence of the signals.

The invention relates to a finite angle projection reconstruction method based on double regular item optimization of L0 norm and singular value threshold decomposition, and belongs to the field of image processing. The method specifically comprises the following steps: S1, establishing an optimization problem target equation according to a CT imaging principle, a regularization framework and a projection data set P; S2, initializing parameters; S3, performing iteration by adopting a SART algorithm to obtain an image X, and correcting the X through error feedback; S4, performing gradient L0 norm optimization on the corrected image X to obtain an image XL0, and updating an error d1; S5, performing singular value decomposition on the image optimized in the step S4, adding a soft threshold constraint to optimize the image to obtain an XSVT, and updating an error d2; and S6, carrying out next round of iteration on the image obtained in the step S5 according to the step S3 until an iteration termination condition is met. According to the method, the CT image contour can be effectively recovered, finite angle artifacts are reduced, and therefore the finite angle CT imaging quality and applicability are improved.

The invention discloses a super-resolution image reconstruction method and system based on multi-feature learning, and the method makes full use of the rich information contained in a single input image for reconstruction, and does not depend on an external database. According to the method, a mapping relation between image features is established based on cross-scale similarity of images, and a high-resolution image containing high-frequency information is reconstructed for an input image directly by using the mapping relation, so that the defect of high-frequency information loss caused by image reconstruction by using an interpolation amplification method is well overcome. According to the method, effective high-frequency information is acquired by using singular value thresholding, andthe high-frequency information is amplified by using a gradient feature mapping relation and then is overlapped on a high-resolution image in a blocking manner, so that a final image reconstruction result is obtained. According to the method for reconstructing the image by utilizing the image feature combination, noise points of the reconstructed image are effectively inhibited, image edge and texture information is well kept, and detail enhancement of the image is realized.

The invention discloses a power distribution network load missing data recovery method based on approximate low-rank matrix completion. The power distribution network load missing data recovery methodcomprises the following steps: S1, dividing load historical data into two types including a data loss vector and a data integrity vector; S2, forming an original matrix; S3, performing interpolationon the missing elements to form a preliminary recovery matrix; S4, solving a Pearson correlation coefficient matrix for the preliminary recovery matrix, and screening out a vector with high correlation with a data missing vector; s5, performing Pearson correlation coefficient matrix screening on the original vectors containing data loss to obtain vectors; s6, performing recovery completion on missing elements in the recovery matrix by using a singular value threshold shrinkage algorithm; and S7, taking out the next vector with data loss, and repeating the above steps. Compared with a traditional low-rank matrix completion algorithm, the improved algorithm provided by the invention has the advantages that a better overall recovery effect and stability are achieved, and the improved algorithm is more excellently applied to completion of the load missing data of the power distribution network.

The invention relates to an intersection arrival rate estimation method based on sampling trajectory data. The method comprises the following steps: 1) obtaining sampling vehicle trajectory data, andpreprocessing the arrival rate between queuing trajectories into an initial arrival rate matrix; 2) according to the initial arrival rate matrix, filling null values in the initial arrival rate matrixthrough a singular value threshold algorithm to obtain a filled arrival rate matrix; and 3) correcting and repairing the filled arrival rate matrix to obtain a final arrival rate matrix, and accumulating and multiplying the arrival rate vector of each period by the length of the unit time interval to obtain the flow of each period. Compared with the prior art, the method has the advantages of being high in universality, good in expansibility, wide in applicability and the like.

The invention relates to a video foreground separation method and system based on adaptive robust principal component analysis. The method comprises the following steps: firstly, each image in an image sequence is pulled into a column vector according to a row, and then the column vector is combined into a new matrix M. The matrix M is decomposed by singular value decomposition, and the rth singular value is used as the initial threshold of singular value threshold operation. Then, the first r singular values are reconstructed to form a new matrix Mr, and the ratio of the information containedin the matrix Mr reconstructed by each singular value to the information contained in the matrix Mr is calculated. The singular values are adaptively sparsed according to their proportional magnitude. Finally, according to the inexact augmented Lagrange multiplier method, the matrix M is decomposed into a low rank matrix and a sparse matrix by the singular value threshold operation model. Experiments show that the method of the invention takes into account the influence of information fusion between the foreground target and the background, and accurately separates the low-rank background part and the sparse foreground part.

The invention discloses a wireless sensor node task time slot scheduling method based on matrix completion, and the method comprises the steps: dividing the continuous working time of a node into different task time slots through taking a sampling time slot as a unit based on the time correlation and low-rank characteristics of sampling data of a wireless sensor node in an environment monitoring application; the energy consumption of the node is reduced and the battery life of the node is prolonged by reducing the frequency of executing a sampling task time slot by the node; constructing a sampling matrix of wireless sensor nodes at a coordinator end, reconstructing a node sampling matrix subjected to data reduction by utilizing a singular value threshold algorithm, ensuring that completenode sampling data is provided for a user, and meanwhile, meeting the requirement of low-power-consumption operation of the wireless sensor network nodes; designing a singular value threshold algorithm of dynamic correction amplitude, adjusting the correction amplitude according to the reconstruction recovery precision of the sampling matrix, reducing the iteration frequency of matrix reconstruction, shortening the reconstruction time of the sampling matrix, and improving the algorithm performance.

The invention discloses a thick cloud remote sensing removal method and device and belongs to the technical field of image processing. The thick cloud remote sensing removal method comprises the following steps of s101, acquiring a reliable element set from a matrix of images to be recovered through deterministic sampling; s102, setting step length and soft threshold parameters according to a singular value and threshold value shrinkage iterative algorithm; s103, performing shrinkage iteration through the singular value and threshold value shrinkage iterative algorithm to reconstruct an original image matrix; s104, judging whether a relative reconstruction error is smaller than the preset error; s105, adding one time on iterations; s106, judging whether the number of iterations is the largest; s107, outputting the reconstructed image matrix. By means of the thick cloud remote sensing removal method, defect information caused by pollution, damage, blocking or the like can be repaired well, information can be recovered, detail texture information is remained, and consistency of image structures is kept.

The invention discloses an SAR image compression method based on robust tensor decomposition, and the method comprises the following steps: converting an original multi-channel SAR image into a tensor; performing mode-n expansion on the tensor to obtain a mode-n matrix, and verifying sparse distribution characteristics of SAR multi-channel image tensor decomposition singular values; setting a singular value threshold value, retaining singular values exceeding the threshold value, performing zero setting on the residual singular values, and performing singular value truncation processing; obtaining approximate tensor representation based on singular value truncation and serving as an initial iteration value of multichannel SAR image compression; calculating robust tensor decomposition based on an augmented Lagrangian multiplier method, and performing dimensionality reduction on the singular value matrix; and performing tensor reconstruction by using the data obtained by dimension reduction to obtain a final multi-channel SAR image compression result. According to the method, the SAR multi-channel image is converted into a tensor form, and the robust tensor decomposition technology is utilized to carry out high-dimensional principal component sparse representation on the image, so that remote sensing SAR image compression is realized, and the influence of an outlier on the image compression effect is effectively inhibited.

The invention discloses a low-voltage distribution network phase-household relation identification method based on matrix completion. The method comprises the following steps of collecting the current time sequence data of a low-voltage bus and all ammeters in multiple time periods in a target station area to form an original current matrix; decomposing, optimizing and combining the original current matrix with the data missing condition by adopting a singular value threshold decomposition method to form a current completion matrix; calculating a relative error of the complemented matrix relative to the original current matrix by solving the ratio of the norm of the difference between the current complementing matrix and the original current matrix to the norm of the original current matrix; and adopting a mathematical optimization method to identify the corresponding phase-household relationship in the transformer area. According to the method, the engineering condition that the acquired data of the low-voltage transformer area is not complete is effectively complemented by means of a data analysis technology, the phase-household identification accuracy is improved, and the intelligent and efficient operation and maintenance of the low-voltage transformer area are facilitated.

The invention discloses an image compression method combining wavelet packet transformation and singular value decomposition, which comprises the following steps of: performing wavelet packet decomposition on an original image to obtain a complete binary tree; taking logarithmic energy entropy as a cost function, sequentially and upwards comparing logarithmic energy entropy values of child nodes and father nodes from the bottom layer of the complete binary tree, retaining nodes with smaller logarithmic energy entropy values, deleting nodes with larger logarithmic energy entropy values, and further obtaining an optimal wavelet tree; reconstructing the optimal wavelet tree by adopting a wavelet packet reconstruction algorithm to obtain a reconstructed image; performing singular value decomposition on the two-dimensional time-frequency information of the reconstructed image to decompose the two-dimensional time-frequency information into texture vectors, geometric vectors and singular values; processing the data in an energy spectrum mode to obtain a characteristic value set; processing the feature value set by adopting a self-adaptive singular value decomposition method to obtain a singular value threshold number K, and selecting the first K feature values in the feature value set to form an approximate matrix; according to the invention, the image compression quality can be improved, and a high compression ratio can be obtained.

The invention discloses an image decompression method combining a prior model and detail enhancement. The method mainly comprises the following steps: for a JPEG compressed image, establishing a Gaussian quantization noise model to construct a data item; constructing a deblocking model based on the shape adaptive low-rank prior and the quantization constraint prior; constructing a detail enhancement method based on sparse representation; under a maximum posteriori framework, establishing an optimization function of image decompression combining a prior model and detail enhancement; and solvingthe optimization function by using a singular value threshold and convex quadratic minimization, and reconstructing a decompressed image. An image reconstructed by the image decompression method provided by the invention not only removes visually unpleasant block effect, but also retains rich detail information and has excellent performance on objective evaluation parameters. The method providedby the invention is an effective image decompression method.