206 results about "Multiplier method" patented technology

The invention discloses a method for compressed sensing synthetic aperture radar imaging based on dual sparse constraints. The method mainly solves the problems that a complete SAR image cannot be obtained in imaging through a traditional synthetic aperture radar imaging method and time loss is large. The method comprises a first step of utilizing radar to irradiate a ground sparse scene D and obtain echo signals r of the scene; a second step of constructing an orientation direction base matrix A of the scene and a distance direction base matrix B of the scene; a third step of obtaining an orientation direction measurement matrix theta, a distance direction measurement matrix omega and an echo measurement matrix S according to the orientation direction base matrix and the distance direction base matrix; a fourth step of constructing a Lagrangian function f(Y,U) according to the orientation direction measurement matrix, the distance direction measurement matrix and the echo measurement matrix; and a fifth step of using an alternate iteration multiplier method to solve the Lagrangian function f(Y,U) in the fourth step and obtaining an image of the scene D. The method can achieve high-resolution imaging of synthetic aperture radar in downsampling, is quick in imaging speed and can be used for large-area topographic mapping and cartography.

The invention discloses a self-adaptive regularized super-resolution image reconstruction method which can keep marginal definition, which mainly solves the problem that the prior method has edge fog in reconstruction of a degraded image. The method comprises the following steps: an imaging model is constructed; on the basis of an unconstrained objective function constructed by a Lagrangian multiplier method, gradient is increased to approach a bound term; the objective function is expanded; L1 norm is adopted to measure a data approximation term; a self-adaptive bilateral total variation model which can carry out local adaptive control on the smoothing effect is utilized to construct a self-adaptive regular term; a gradient approximation term is added to be as constraint of gradient consistency; edge information is kept; the self-adaptive regular term and a gradient consistency bound term are introduced as constraint conditions; an expanded Lagrangian objective function is constructed and optimized; and an optimized unconstrained objective function is utilized to reconstruct an image, thereby obtaining a high-resolution image of which the edge is kept. The method can keep image edge clear, can inhibit noise and is suitable for restoration treatment on the degraded image.

The invention discloses a hyperspectral image super-resolution reconstruction algorithm based on matrix structure sparse non-negative decomposition. According to the reconstruction algorithm, a low-spatial-resolution hyperspectral image and a high-resolution color image are united to reconstruct a high-resolution hyperspectral image, and the problem that an existing algorithm can not accurately restore the high-resolution hyperspectral image. The method comprises the realizing steps that (1) the low-resolution hyperspectral image and the corresponding high-resolution color image are input; (2) local and non-local self-similarity of the hyperspectral images is utilized for constructing a spectrum reconstruction target function based on the matrix structure sparse non-negative decomposition; (3) an alternating direction multiplier method is adopted for alternative solving to obtain an optimized spectrum material coefficient and a spectrum material base; (4) a matrix of the optimized spectrum material coefficient and a matrix of the optimized spectrum material base are utilized to reconstruct the high-resolution hyperspectral image. According to the method, the restored hyperspectral image is clearer, the image edge is sharper, and the spatial resolution of the hyperspectral image can be effectively increased.

The invention discloses a large power grid parameter recognition and estimation method. The large power grid parameter recognition and estimation method specially comprises the following steps that (1) real-time data are obtained; (2) topology decomposition is conducted on the real-time data to decompose the real-time data into a plurality of partitions according to voltage classes; (3) state estimation calculation is conducted on all the partitions sequentially and bad data are recognized; (4) parameter recognition and estimation calculation is conducted on all the partitions sequentially, wherein firstly, parameter recognition is conducted, and if suspicious parameters exist, parameter estimation is conducted and the parameters are corrected; then parameter recognition is conducted again and if no suspicious parameters exist, parameter recognition results and parameter estimation results of this time are stored; (5) average values of the stored parameter recognition results and the stored parameter estimation results are counted and serve as parameter recognition and estimation output results, and after a period interval passes, the steps (1) is conducted. The lagrangian multiplier method is adopted in the large power grid parameter recognition and estimation method, the calculation method is simple and the calculation efficiency is high. The large power grid parameter recognition and estimation method can be successfully applied to large-scale power grids.

The invention relates to a method for estimating state of a distribution network for supporting large-scale current measurement, and belongs to the automatic dispatching field of power systems. The method comprises the following steps of: squaring a branch head end power and a branch current amplitude to obtain an evaluated variable, and establishing a new state estimation model of an exponentialtarget function according to a power grid model of an actual power grid and real-time measurement data; and estimating the branch head end power and the branch current amplitude by adopting a Lagrangian multiplier method so as to calculate active power and passive power of equipment such as a circuit, a transformer, a generator, a load and the like. The method can directly establish measurement equations for branch current amplitude measurement, branch power measurement, node voltage measurement and injection power measurement, does not need measurement conversion, does not introduce measurement conversion errors, is convenient to implement, and has high calculation efficiency.

The invention discloses a human face confirming method based on similarity measurement. The method comprises the following steps that features of N pairs of human face images are extracted, each pair of human face features are subjected to PCA transform, and a feature vector of human face sample pairs is obtained; within-class weighed subspace is obtained through a weighted covariance matrix and subjected to weighted subspace projection, and feature vectors of the human face pairs obtained after projection are obtained; in combination of the knowledge of priori measurement, a priori similarity matrix and a priori distance matrix are obtained; a lagrangian multiplier method and a paired method are used for solving a target function of similarity measurement learning, and a measurement matrix is obtained; the similarity of human face pairs is calculated in combination with a similarity model, and therefore an optimal threshold value is obtained and used for human face confirmation. The problems that under the constraint-free condition, due to interference of expressions, postures, illumination, backgrounds and other factors, the difference of different human face images of the same person is large, and extracted human face feature vectors are greatly different are solved, and therefore the human face confirmation accuracy is improved.

The invention discloses a joint path selection and power distribution method for energy collection nodes in a wireless sensor network, and belongs to the field of cooperative communication technologies. The method comprises the steps of analyzing a system scene, describing problems; establishing a system mathematic model; and then finding the optimal solution by using an optimization method. The method aims at the special application scene and is derived from the actual application, and being different from the traditional independent sensor node or gateway resource distribution, the method comprehensively considers the joint power distribution and path selection of sensor nodes and gateways, and maximizes the handling capacity performance of the communication nodes by using the gateways as relay stations. According to the method of the invention, the solution of an optimization problem is processed by using convex optimization to convert a target function of the optimization problem without approximate calculation, so that computation complexity is greatly reduced while accuracy of the problem is not influenced, and delay generated by system overheads is reduced; a Lagrangian multiplier method is used in an optimizing process, and thus an optimizing speed is rapid; a subgradient method and an incremental step length are used in an iterative process, so that optimizing is more accurate.

The invention provides a power resource management method for opportunistic array radar multi-target tracking in a clutter environment. The power resource management method comprises the steps of measuring measurement origin uncertainty caused by clutters by means of an information reduction factor, expressing RCS of targets by using random variables, establishing a random chance-constraint processing model, relaxing the random chance-constraint processing model by adopting a conditional value at risk, converting the random chance-constraint processing model into a certainty model which is easy to solve, further solving the certainty model by adopting a Lagrangian multiplier method, regarding a Bayesian Cramer Rao Lower Bound as a measurement standard for power distribution so as to solvethe optimal power distribution at a next moment, and adopting a probability data association filter for tracking the targets in the clutter environment. The power resource management method considersthe target tracking clutter environment, adopts the information reduction factor for measuring magnitude of clutter influence, further considers a target information uncertainty caused by the target RCS, introduces the random chance-constraint processing model, comprehensively considers the relationship between resources and tracking precision, and makes the model closer to reality.

The invention discloses a compressed sensing signal reconstruction method, which belongs to the technical field of signal processing. The compressed sensing signal reconstruction method aims to solve a regularization problem that is more difficult to solve; the regularization problem in a sparse domain of compressed sensing signals is converted to a constraint-regularization problem through a variable division technology, and the constraint-regularization problem is equivalent to the regularization problem and can better present the sparse characteristics of the signals compared with the regularization problem; and higher accuracy for reconstructing the signals can be achieved. The compressed sensing signal reconstruction method utilizes a rapid alternating direction multiplier method further so as to solve the constraint-regularization problem, updates variables in an algorithm of the alternating direction multiplier method for a second time, updates multipliers, and increases the convergence rate of optimization solution. Compared with the prior art, the compressed sensing signal reconstruction method has the advantages of higher reconstruction accuracy and higher convergence rate.

The invention discloses a MIMO (Multiple-input Multiple-output) radar angle estimation algorithm based on tensor subspace and rotation invariant. Through building a three-order tensor model for receiving data, a higher-order covariance tensor model for tensor data is further built, and the internal correlation structure of array signals is thoroughly exploited; HOSVD (Higher Order Singular Value Decomposition) is then carried out on the tensor data, new signal subspace is built, and high-precision noise subspace is thus acquired; and finally, by using the rotation invariant of array data, a DOD (Direction of Departure) and a DOA (Direction of Arrival) which are paired are acquired through a constrained optimization method and a lagrangian multiplier method, and further pairing calculation is not needed. According to the MIMO radar angle estimation algorithm of the invention, by using the internal correlation structure of the received signals, the angle estimation precision is higher, the target DOD and the target DOA which are more accurate can be acquired, a more reasonable reference is provided for further processing on a detected target, spectral peak search is not needed, and the computation complexity is low.

The invention relates to a bilinearization-based all-distributed robust state estimation method for a multi-regional power network, and belongs to the fields of power system operation and control. The method is divided into three states: preprocessing a measurement to obtain pre-estimation of an intermediate state variable in the first stage; carrying out nonlinear conversion on the intermediate state variable in the second stage to obtain new measurements, namely a phase-angle difference between voltages at two ends of each branch and twice of the sum of logarithmic magnitudes of the voltages; and estimating the final state variable on the basis of these new measurements in the third stage. Through nonlinear conversion in the second stage, parallel solving can be carried out in various regions and communication with the other regions is not needed. State estimation in the first stage and the third stage is aimed at a linearized system; and each control region determines the local state quantity by communication of the local and a neighbor by an alternating direction multiplier method. The bilinearization-based all-distributed robust state estimation method provided by the invention is suitable for large-scale multi-regional state estimation, can well protect the data privacy of various regions and has very high agility, flexibility and robustness.

The invention discloses a nonlinear amplification-retransmission optimal power allocation method. The method comprises the following steps: an ideal amplitude-limiting amplifier model is constructed into a linear model with a Gaussian distortion noise option by using a linearized theory; according to the linear model, the chain equivalent single-to-noise rate of each nonlinear relay, devoted to a target receiver, is calculated; in order to obtain the optimal signal-to-noise rate performance, a maximum-combination-ratio receiver which is capable of realizing the nonlinear distortion is also designed; then, an optimal relay power allocation mathematical model is built, although the optimal power allocation problem is not a convex problem is found by analyzing a curve structure of signal to noise rate and power, the optimal relay power allocation problem can be converted into a convex optimization problem by limiting the maximum emission power; and finally, an optimal power allocation scheme is provided by utilizing a lagrangian multiplier method. A theoretical analysis and simulation result indicates that the method disclosed by the invention is superior to an optimal power allocation algorithm has no sense to nonlinear distortion, and has the capability of effectively improving the system capacity.

The present invention relates to a separation method of the foreground and the background of a video based on low rank and structure sparseness. The method comprises: the step 1, reading a video sequence; the step 2, performing low-rank and structure sparse matrix decomposition through adoption of a low-rank and structure sparse non-precision lagrangian multiplier method; the step 3, estimating the rank r of a background matrix A in advance; the step 4, reconstructing the background matrix A, namely, performing decomposition through a singular value, and calculating the background matrix A by using the rank r and a singular value contraction operator; and the step 5, reconstructing a foreground matrix E, obtaining the background matrix A through an observation matrix D and the step 4, and calculating a foreground matrix E; the step 6, determining the iteration times and the errors, and if the iteration times are equal to or smaller than iteration errors, the iteration is finished; and the step 7, recovering the background and the foreground images. The separation method of the foreground and the background of a video based on low rank and structure sparseness is able to directly realize the separation of the background and the foreground of the video sequence.

The invention discloses an auto-fluorescence tomography re-establishing method based on a multiplier method. The auto-fluorescence tomography re-establishing method based on the multiplier method comprises the following steps of: carrying out discretization by using a finite element method diffusion equation, and establishing an optimization problem model without constraint conditions based on a penalty term of an L1 norm; obtaining a dual model of the optimization problem model without constraint conditions; establishing an augmentation lagrange function of the dual model; simplifying the maximum function of the augmentation lagrange function; solving the maximum value of the augmentation lagrange function by using a truncated-Newton algorithm; upgrading the target vector by using the gradient of the augmentation lagrange function as the steepest descent direction of a target vector; upgrading a penalty vector; and calculating an objective function value J(w), calculating k=k+1 if the ratio of the norm of J(w)k-J(w)(k-1) to the norm of Pai m being not smaller than t0l is real, and jumping to the step S4, otherwise, ending the calculation, wherein t0l is the convergence efficiency threshold value of the target function. The auto-fluorescence tomography re-establishing method provided by the invention can quickly obtain accurate and reliable light source distribution information within a large image-forming region, so that other parameters except from the regularization parameter can realize self-adaptive adjustment for improving the image-forming robustness.

The invention discloses a low-side-lobe beam pattern integrated design method based on convex optimization. Auxiliary variables are introduced to achieve variable separation by adopting an alternative direction multiplier method (ADMM), a large amount of original inequality constraint limits are converted into solvable problems, iteration solution is performed by applying an ADMM thought, and accordingly parameters are determined to obtain an ideal wave beam, namely side-lobe values are reduced. The robustness of a wave beam forming device is good in the process by optimizing wave beam side lobes, the side effect of the side lobes on array gains is low, the obtained side lobe values are low, and the low-side-lobe beam pattern integrated design method is simple and simple in operation and has a very good practical value.

The invention discloses a Retinex low-illumination color image enhancement method based on variation constraint, and relates to the technical field of color image processing, the method comprises thefollowing steps: converting an original color image from an RGB space to an HSV space, and extracting an image of an intensity channel V as a grayscale image; constructing a conductivity weight matrixby using the extracted grayscale image; constructing an objective function by using an energy minimization principle, inputting the grayscale image and the conductance weight matrix into the objective function, and solving the objective function by using an alternating direction multiplier method to obtain an illumination image; Separating a reflection component from the original color image according to the illumination image to obtain a reflection image; The gray level image is reconstructed by using the illumination image and the reflection image; According to the method, the boundary of the image is well reserved, so that the estimated illumination image is more accurate, the reflection image contains more internal details and texture features of the image, the contrast ratio of the enhanced image is improved, and the details of the high-brightness position of the image are reserved.

The invention relates to a mobile edge calculation scheme in an Internet of Vehicles scene, and the method optimizes the calculation task allocation and transmission power control problems of user equipment in a vehicle on the premise of meeting the time delay requirement. The energy loss of the equipment under the weighted calculation task allocation rate is taken as an objective function, a datatransmission model of the user equipment and the edge calculation node are obtained by using a queuing theory method, and the optimization problem is solved through non-linear fractional optimizationand iteration of an alternating direction multiplier method. And in each round of loop, the outer layer loop solves a nonlinear fractal programming problem, the inner layer loop updates an initial value and a variable until an iteration result meets a set threshold value, the task amount distribution ratio of each user equipment is determined, and the minimum energy consumption is obtained. According to the technical scheme provided by the invention, the energy consumption of the user equipment can be effectively reduced, the delay requirement is met, and the calculation capability of the whole network is improved.

The invention discloses a data processing delay optimization method and system based on edge computing. The method comprises the following steps: building a network architecture model; determining thecomputation delay of an edge computing layer by using a Lagrangian multiplier method; determining the communication delay of the edge computing layer by adopting a Kruskal method; determining the computation delay of a cloud computing layer; determining the communication delay of the cloud computing layer by adopting a balanced transmission method; determining a data processing delay optimizationmodel according to the computation delay and the communication delay of the edge computing layer, and the computation delay and the communication delay of the cloud computing layer; and determining an optimal data processing delay value according to the data processing delay optimization model. The data processing delay optimization method based on the edge computing provided by the invention firstly determines the data processing delay of the edge computing layer, then determines the data processing delay of the cloud computing layer, and finally, determines the optimal data processing delayvalue according to the data processing delay of the edge computing layer and the data processing delay of the cloud computing layer, therefore the system data processing delay is reduced, and the data processing efficiency is improved.

The invention discloses a sea-surface infrared small object detection method. The method comprises the steps that (1) a non-local block is used to construct a new image matrix by decomposing an original image matrix into superposed subblocks, expanding the subblocks into column vectors and combining the column vectors to obtain the new image matrix; (2) sparse and low-rank matrix decomposition is carried out by decomposing the constructed new image matrix into sparse small object components, low-rank background image components and noise components; (3) a convex optimization model is established by constructing a convex optimization constrained energy function related to small object and background images; (4) high-efficiency optimized solution is carried out by solving the constructed convex optimization constrained energy function in an alternative direction multiplier method in which parameters are updated adaptively in high efficiency; and (5) the small object image of interest is obtained by iteration. The detection method provided by the invention is low in calculation cost, high in parameter adaption, high in detection efficiency, and capable of resisting background noises and clutter interference, and can be applied to infrared small-object detection in complex sea-surface background.

The invention discloses a generalized tree sparse-based weight nuclear norm magnetic resonance imaging reconstruction method. The method comprises the following steps of: firstly obtaining a test magnetic resonance imaging sampling data sample to carry out Fourier transform; constructing a sparse of a tree structure according to a sampled signal, and approaching to sparse expression of a constrained target function by utilizing a nuclear norm with a weight; optimizing the constrained target function through an augmented lagrangian multiplier method, and carrying out iterative updating on the test data through an alternating direction search algorithm until estimated recovery data is obtained; and obtaining a final recovery image through constructing tree sparse inverse transformation. According to the method, an internal structure relationship between image signals is sufficiently mined, the generalized tree sparse structure characteristics of image blocks are combined with weight nuclear norms, and the calculation process is simplified by utilizing an ADMM algorithm, so that the complexity of the algorithm is reduced, the performance of a part of spatial data reconstruction images is improved, images can be reconstructed more accurately under less scanning and measurement, fake shadows of the reconstructed images can be decreased and rapid magnetic resonance imaging is realized.

The invention discloses a signal angle-of-arrival high-precision estimation method under high sampling 1 bit quantification conditions. The method includes the steps: firstly, generating a snapshot signal received by a uniform array comprising M array elements and performing 1 bit quantification sampling on the signal; secondly, building a sparse signal representation model taking a rho norm as a sparse constraint item and combining an insensitive epsilon-SVR (support vector regression) model, and performing sparse representation in original signal information by the aid of a non-convex optimization model; finally, resolving the non-convex optimization model by an ADMM (alternative direction multiplier method) to obtain a sparse representation coefficient for determining the arriving direction of the signal. According to the method, the arriving direction of the signal can be estimated in a high-precision manner without calculating an autocorrelation matrix and without reference to coherence of information sources when sampling quantity is small (such as single-snapshot), and signal arriving direction estimation based on the 1 bit quantification sampling conditions is more easily and rapidly implemented in engineering.

The invention discloses a rapid compression perception reconstruction method facing to a wearable device, and belongs to the signal processing field. The method comprises the steps of: performing compression measurement to an original electrocardiogram signal x through a constructed m*n dimension random sparse binary observation matrix phi, so as to get an observation vector y, the length of which is m; and based on the observation matrix phi and the observation vector y, reconstructing the original signal by the compression perception reconstruction method learned from the alternating direction multiplier method and the block sparse Bayes, so as to get an estimation vector of the original signal, which is as shown in the description. Compared with the conventional compression perception reconstruction method learned from the block sparse Bayes, the method provided by the invention is advantaged in reaching quicker reconstruction speed on the premise of ensuring the reconstruction precision.