172results about How to "Good refactoring" patented technology

The invention provides a multi-vision-based bridge three-dimensional deformation monitoring method which comprises the following steps of: (1) obtaining the images of calibration plates and extracting the feature points on the multiple calibration plates by multiple video cameras respectively; (2) establishing a mapping model based on a BP neural network; (3) obtaining each video camera image at the bridge edge feature points; (4) eliminating the error point pairs by use of RANSAC to obtain the correct matching point pairs; (5) extracting the two-dimensional coordinates of each video camera image at the bridge edge feature points, obtaining the three-dimensional world coordinates of the feature points according to the mapping model based on BP neutral network, and drawing a three-dimensional curve of the bridge surface; and (6) extracting the bridge deformation rule according to the bridge curves of the bridge at different moments, and judging the bridge deformation trend. The method provided by the invention can perform non-contact three-dimensional measurement on the bridge deformation, and has the advantages of continuous measurement, instant measurement, synchronous measurement of multiple points, high precision, repeatability, low cost and the like.

The present invention provides a motion capturing system and method based on a CAN bus and an inertial sensor, for tracking and capturing motion information of a human body in a real-time manner. The system comprises an inertial sensing node combination, a data aggregation node, and a central computer. The inertial sensing node combination contains 18 inertial sensing nodes, collects an accelerated speed, an angular speed, and geomagnetic information data of each bone point part in a real-time manner, and performs attitude estimation by using a complementary integration filtering algorithm, so as to obtain data, such as a quaternion and an euler angle, of an three-dimensional attitude of the current bone point. The data aggregation node is connected to the 18 inertial sensing nodes, is configured to collect, in a time-sharing manner, three-dimensional attitude data of each bone point obtained by the 18 inertial sensing nodes, and send the collected data to the central computer by using a wireless wifi module. A main task of the central computer is to complete functions of collecting motion data transmitted from the data aggregation node and driving a virtual 3D person model by using the motion data.

The invention discloses a self-adaption online dictionary learning super-resolution method. The method firstly selects one group of high-resolution training image set and one group of low-resolution training image set,then establishes a relationship on the high-resolution training image set and the low-resolution training image set, and provides more prior information for image reconstructing through high-resolution/low-resolution dictionary to obtain a better reconstructing effect under the condition of high amplification factor, at the same time, the method uses SP (Self Propelled) sparse coding algorithm, overcomes the defect of low reconstructing accuracy due to traditional greedy algorithm and keeps a lower computation complexity. The method has the advantages of overcoming the defects of complex computation and low training speed in the current mainstream dictionary learning algorithms such as MOD (Multimedia On Demand) and K-SVD(k-singular value decomposition) and shortening image reconstructing time in general. Compared with the prior art based on learning image super-resolution, the method of the invention has a higher reconstructing accuracy and a shorter algorithm time.

The embodiments of the invention relate to a method for projecting virtual reality panoramic video stream to a user, comprising the steps of dividing the panoramic video stream into multiple spherical subareas, providing different video qualities for the spherical subareas related to a user perspective and the spherical subareas unrelated to the user perspective according to user perspective information tracked in real time, and moving user observation points, thus realizing an optimal video reproduction effect and reducing the transmission bandwidth. The invention further relates to equipment for projecting virtual reality panoramic video stream to a user.

The invention discloses a digital holography reconstruction method based on an iterated denoising shrinkage-thresholding algorithm. The method comprises the following steps of 1, reading an original image, and acquiring diffraction light complex amplitude O(x, y) by using object light information of the original image, wherein the amplitude of the original image is O<0>(Xi, Eta); 2, intervening the diffraction light complex amplitude O(x, y) obtained through fresnel diffraction and reference light R(x, y) added with o and Pi second phase shift quantity to separately achieve light wave complex amplitude U<1>(x, y)/U<2>(x, y) on a holographic plate, laminating the two holographic images I<(1)> and I<(2)> to form a phase shift holographic image I-bar (x, y), constructing a sensing matrix A, and allowing y=Pi I-bar=Pi RO=AO, wherein Pi is a known measurement matrix, R is a spare basis matrix, and y expresses obtained observation data; and 4, figuring out and acquiring a reconstructed image of the original object by using the iterated denoising shrinkage-thresholding algorithm (IDNST). On the basis of TwIST algorithm, dual shrinkage of a threshold value and a regularization parameter are introduced to the IDNST algorithm, the signal-to-noise ratio can be improved, moreover, the convergence speed is increased, the reconstruction precision is improved, and the reproduction quality reaches a more excellent level.

Disclosed is a single image super-resolution method based on identical scale structure self-similarity and compressed sensing. Firstly, the interpolation is performed for a low-resolution image and a quasi-high-resolution image is obtained; then, the quasi-high-resolution image is divided into quasi-high-resolution image blocks, vectors corresponding to the quasi-high-resolution image blocks serve as a training sample, a sample matrix is assembled, a K-SVD dictionary studying method is used for a solution and a dictionary is obtained; the low-resolution image is divided into low-resolution image blocks; by the aid of a down-sampling matrix, the dictionary and vectors corresponding to all low-resolution image blocks, an orthogonal matching pursuit (OMP) method is used for a solution, and vectors corresponding to high-resolution reconstruction image blocks; and finally, vectors corresponding to high-resolution reconstruction image blocks are assembled and a high-resolution reconstruction image is formed. According to the super-resolution method based on the identical scale structure self-similarity and the compressed sensing, additional information is added in the high-resolution reconstruction image through a compressed sensing frame, and the space resolution is improved.

The invention discloses a compressed sensing image reconstructing method based on a prior model and 10 norms, mainly used for solving the defects of poor visual effect and long operation time existing in image reconstruction in the prior art. In the technical scheme of the invention, a compressed sensing image reconstruction frame with 10 norms is optimized by utilizing a prior model; and the positioning of sparsity coefficient and solution of the sparsity coefficient value are achieved through two effective steps: step 1, establishing the prior model, and carrying out low frequency coefficient inverse wavelet transform so as to obtain an image with a fuzzy edge, determining the position of the edge by edge detection, and searching the position of wavelet high frequency subband sparsity coefficient through an immunization genetic algorithm by using the prior model of which the wavelet coefficient has inter-scale aggregation; and step 2, solving a corresponding high frequency subband by using an improved clone selective algorithm, and then carrying out the inverse wavelet transform so as to obtain a reconstructed image. Compared with the prior art, the method has the advantages of good visual effect and low calculation complexity, and can be used in the fields of image processing and computer visual.

The invention discloses a human face expression recognition method based on Curvelet transform and sparse learning. The method comprises the following steps: 1, inputting a human face expression image, carrying out the preprocessing of the human face expression image, and cutting and obtaining an eye region and a mouth region from the human face expression image after processing; 2, extracting the human face expression features through Curvelet transform, carrying out the Curvelet transform and feature extraction of the human face expression image after preprocessing, the eye region and the mouth region, carrying out the serial fusion of the three features, and obtaining fusion features; 3, carrying out the classification recognition based on the sparse learning, and respectively employing SRC for classification and recognition of the human face Curvelet features and fusion features; or respectively employing FDDL for classification and recognition of the human face Curvelet features and fusion features. The Curvelet transform employed in the method is a multi-scale geometric analysis tool, and can extract the multi-scale and multi-direction features. Meanwhile, the method employs a local region fusion method, and enables the fusion features to be better in imaging representing capability and feature discrimination capability.

The invention discloses an end member learning based hyperspectral image sparse unmixing method, which mainly solves the problems of low hyperspectral image unmixing precision, poor reconstruction effect, long consumed time and low efficiency during a low-signal-noise-ratio hyperspectral image unmixing process in the prior art. The method comprises the steps of: inputting hyperspectral data; synthesizing hyperspectral base data; performing end member learning; solving a hyperspectral data abundance matrix; calculating a reconstruction error of the hyperspectral data abundance matrix; and outputting an unmixing result. The method adopts a new solving mode, introduces an end member learning thought, has the advantages of high unmixing precision, good reconstruction effect and high efficiency, is simple in solving step and explicit in principle, and can be used for understanding interpretation of hyperspectral images.

The invention discloses a partial K space sequence image reconstruction method based on self-adapted double-dictionary learning, and the method is mainly used for solving the problems of an existing method that the quality of a reconstructed image is more seriously reduced under the condition of sampling under 10 times. The partial K space sequence image reconstruction method comprises the following main steps of: collecting partial K space data and utilizing the correlation between the partial K space data to be integrated into complete K space data; obtaining a training image by the complete K space data; utilizing a KSVD (Kernel Singular Value Decomposition) algorithm to train the training image to obtain dictionaries with high and low resolution ratios; and utilizing a relation between the dictionaries with the high and low resolution ratios to reconstruct the input partial K space data, and carrying out residual error compensation on the reconstructed image to obtain a more accurate reconstruction result. According to the partial K space sequence image reconstruction method disclosed by the invention, the quality of the reconstructed image can be effectively improved under the condition of sampling under 10 times; and the partial K space sequence image reconstruction method can be used for reconstructing MRI (Magnetic Resonance Imaging) sequence images of a plurality of parts.

The invention provides an object software oriented automatic refactoring method and relates to the technical field of software quality improvement. According to the method, a to-be-refactored software system is established as a class level multilayer dependency directed network model; refactoring preprocessing is carried out; class level network connecting components are combined; each class level network connecting component is converted into an entity set of the same class; semantic and structure coupling relationships among the entity set elements are analyzed; a method level coupling undirected network model is established; weight coefficients of different classes of coupling relationships among the nodes of the undirected network are determined; community division is carried out on each method level network; refactoring suggestions are generated; and the to-be-refactored software system is refactored. According to the method, starting from the angles of global cohesion and coupling of the whole software system, through combination of a semantic similarity, a structure similarity and a hierarchical clustering algorithm, a move function, a move attribute and extraction class refactoring operation suggestions are generated at the same time, and the intelligibility, reusability and maintainability of the code are effectively improved.

The invention discloses a finite impulse response (FIR) filter, which consists of a filtering coefficient buffer, a to-be-filtered data buffer, a multiplier, an accumulator, an accumulating register and a comparator, wherein the filtering coefficient buffer is used for buffering coefficients which are used in the filter; the to-be-filtered data buffer is used for buffering data to be filtered; the multiplier is used for computing a product of a single filter coefficient multiplied by BS data to be filtered; the accumulator is used for computing a sum of the output of the multiplier and the output of the accumulating register; the accumulating register is used for storing a result of the accumulator; and the output of the comparator is used for indicating whether the current output of the accumulator is filtering output. By adoption of the FIR filter, high efficiency of parallel computation can be realized.

The invention provides a fault sensor information reconstruction method based on measured value association degree. The method comprises the following steps: S1, aimed at a fault sensor, calculating association degree of response of a measuring point where a sensor is on in normal operation and other measuring points; S2, through comparing association degree values, determining an association model, establishing required response variables; S3, and using a partial least squares to establish a reconstruction model of reconstruction variables and the response variables, and using measured data of structural health monitoring, performing fault sensor response information reconstruction on the fault sensor. The method has very good effect on fault sensor information reconstruction, obviously reduces reconstruction errors, and ensures variation trend of the reconstruction values to keep consistent with variation trends of practical values. The method has obvious integrated structure reliability.

The invention discloses an aerospace engine abnormity intelligent detection method based on hierarchical adversarial training, and the method comprises the steps: employing a plurality of sensors to collect original signals of an aerospace engine in an operation state as multi-source data, intercepting a time sequence at a fixed length to obtain a multi-channel data sample set, and converting a one-dimensional sequence into a two-dimensional image; dividing the two-dimensional image sample into a training set and a test set; constructing a relative generative adversarial network as an anomalydetection model, and performing hierarchical adversarial training by using the training set; using the training model to evaluate the state of the training set sample, modeling the obtained evaluationscore distribution, and calculating the score threshold of the normal sample; using the model for evaluating the state of a test set, aggregating neighborhood information during testing, and conducting anomaly detection according to a score threshold value. According to the method, the model detection capability is improved through hierarchical adversarial training, multi-source information is fused, neighborhood information is aggregated to improve the result reliability, and finally, intelligent detection of abnormal operation of the aerospace engine can be realized.

The invention relates to an image super-resolution reconstruction algorithm based on a jump connection residual network, and belongs to the technical field of image processing. The method comprises the steps of selecting a training data set, and performing bicubic interpolation on a low-resolution image; constructing a specific structure of the network, and formulating a network training strategy;performing detail extraction on the interpolated image; the dimension of the total feature is reduced, and the single-pixel receptive field is widened; performing multiple iterations on the network training until the maximum number of iterations is reached; and final high-resolution image reconstruction is completed in a global residual error learning mode. According to the method, the residual network based on jumping is combined with the sub-network based on the parallel channels, detail features of the shallow network are fully utilized, rapid learning of the detail features of the depth image is achieved, and rapid convergence of the network is achieved through adjustable gradient clipping.

The invention discloses a statistical compressed sensing image reconstruction method based on a layered Gauss mixing model, specifically relates to a statistical model learning and signal reconstruction algorithm based on image block data, and mainly solves the problem of an existing statistical compressed sensing image reconstruction method that a single Gauss prior model fails to accurately depict non-Gauss statistical property of sub image blocks and the quality of a reconstructed image is not high. According to the invention, on a first layer of the model, a global Gauss mixing model is learned, hard clustering is carried out on the sub image blocks, and the similar structures of the image blocks are fully utilized globally; and on a second layer of the model, a local Gauss mixing model of the sub image blocks are learned, and the differences among the sub image blocks are distinguished and models locally. Compared with an existing conventional compressed sensing and statistical compressed sensing reconstruction technology, higher reconstruction precision is obtained by less measurement, and the statistical compressed sensing image reconstruction method is suitable for the reconstruction of a natural image.

The invention discloses a partitioning compressive sensing reconstruction method based on image block clustering and sparse dictionary learning, and belongs to the technical field of image processing. The method comprises the following steps that an image is read in, and is divided into sub-image blocks; compressive sampling is carried out on the sub-image blocks to achieve measurement; edge images in (K-1) directions are generated, PCA transformation is carried out on the edge images to generate (K-1) PCA bases, and then, a DCT base is taken for forming a union dictionary of K initial direction bases; the typical correlation coefficients between the measurement and the direction bases are calculated, and the sub-images are clustered to form K classes; the sub-image blocks in the K classes are reconstructed through a multivariable tracking algorithm; the reconstructed sub-image blocks are used for updating the K direction bases; whether the maximum number of times of iteration reconstruction is reached or not is judged; the reconstructed sub-image blocks are spliced together to obtain a reconstructed image of the original image; the image is output. According to the partitioning compressive sensing reconstruction method based on image block clustering and sparse dictionary learning, the blocking effect in the reconstructed image can be obviously weakened or removed in two reconstruction modes, and the method has a reconstruction effect on a natural image.

The invention discloses a measurement matrix design method based on an LDPC matrix. The method comprises the following steps: the step one in which the number L of continuous '1' at the beginning of the first row of the matrix is determined at first, and positions of '1' in the following row of the matrix are subjected to translation to the right for L positions successively according to positions of '1' in the previous row so as to ensure that the positions of '1' between every two rows or every two columns in a cycle submatrix are different to construct the cycle submatrix; the step two in which required rows or columns are randomly selected from the cycle submatrix to construct a measurement matrix based on the LDPC matrix; and the step three in which four-side loops are found in the whole matrix through searches of a limited number, and the four-side loops are eliminated to construct the measurement matrix based on the LDPC matrix. The measurement matrix inherits advantages, such as good sparsity, small column correlation values, etc., of the LDPC matrix, and at the same time, the shortage that optimal d values (the average number of '1' in each column) need to be determined for different dimensions of measurement matrixs in advance when the LDPC matrix is used as the compression sensing measurement matrix can be overcome. The matrix has advantages of good sparsity, simple structure, good orthogonality, easy hardware implementation and good reconstruction effect.

Instant noodles are produced, which have an excellent reconstitution property, can be reconstituted by pouring of boiling water even if the noodles are thicker than before, and have excellent taste and texture. The instant noodles are produced by: making raw noodle strings each having a multilayer structure including three or more layers; spraying superheated steam to the raw noodle strings; gelatinizing the noodle strings to which the superheated steam has been sprayed; and drying the noodle strings which have been gelatinized. It is preferable to gelatinize the noodle strings by spraying the superheated steam to the noodle strings, supplying moisture in liquid form to the noodle strings, and further heating the noodle strings by using the superheated steam and/or saturated steam.

The invention discloses an erasure code storage and reconfiguration optimization method based on redirect-on-write. The method includes: redirecting data, written in failure data nodes, to an RS array composed of survival node free spaces. The data redirecting process and the data migration process, including failure data node reconfiguration, redirected data migration and check data updating caused by migration, are processed independently, the two processes have fewer competitions for resources such as disk, memory and network bandwidth, the background reconfiguration process is accelerated, and foreground user access performance is optimized.