Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

69 results about "Variational model" patented technology

In the probability model framework, a variational autoencoder contains a specific probability model of data and latent variables . We can write the joint probability of the model as . The generative process can be written as follows.

Rolling bearing fault diagnosis method based on improved variational model decomposition and extreme learning machine

The invention discloses a rolling bearing fault diagnosis method based on improved variational model decomposition and an extreme learning machine. The method comprises: vibration signals of a rollingbearing under different types of faults are collected, the vibration signals are filtered by means of maximum correlation kurtosis deconvolution, parameter optimization is carried out on the maximumcorrelation kurtosis deconvolution method by using a particle swarm algorithm, and an enveloped energy entropy after signal deconvolution is used as a fitness function; the mode number of variationalmodel decomposition is improved by an energy threshold and improved variational model decomposition of the filtered vibration signals is realized to obtain mode matrixes of the corresponding vibrationsignals; singular value decomposition is carried out on the mode matrixes to obtain a singular value vector and a rolling bearing fault feature set is constructed; and the fault feature set is trained by using an extreme learning machine and a rolling bearing fault diagnosis model is established. Therefore, stable feature extraction of the complex vibration signal of the rolling bearing is realized, so that the diagnostic accuracy is improved.
Owner:HEFEI UNIV OF TECH

Physics-based mosfet model for variational modeling

A method of optimizing MOSFET device production which includes defining key independent parameters, formulating those key independent parameters into a canonical variational form, calculating theoretical extracted parameters using at least one of key independent parameters in canonical variational form, physics-based analytical models, or corner models. The method also includes calculating simulated characteristics of a device using the key independent parameters and extracting target data parameters based on at least one of measured data and predicted data, comparing the simulated characteristics to the target data parameters, and modifying the theoretical extracted parameters or key independent parameters in canonical form as a result of the comparison. Then, calculating and outputting the simulated characteristics based on the modified theoretical extracted parameters and the modified key independent parameters in canonical form.
Owner:ORACLE INT CORP

Multi-process constraint and multi-terminal constraint terminal guidance law based on generalized target control miss distance concept

The invention discloses a multi-process constraint and multi-terminal constraint terminal guidance law based on a generalized target control miss distance concept. The multi-process constraint and multi-terminal constraint terminal guidance law comprises the following steps that 1, stored standard control and an aircraft state initial value are uploaded; 2, a ballistic integral is predicated; 3, process constraints are corrected; 4, a target control trajectory one-order variational model is solved; 5, replacement criteria control is solved. Compared with a traditional terminal guidance law, the guidance law can meet the traditional terminal constraint requirements for the miss distance, the falling angle, the falling point attack angle and the like, a strong-nonlinearity state variable V can be restrained to a needed value, and meanwhile the process constraints can be processed in a loop-locked mode.
Owner:BEIHANG UNIV

Image super-resolution reconstruction method and system

The present invention discloses an image super-resolution reconstruction method and a system. According to the method, firstly, the pixel data of a plurality of low-resolution images having the complementary information are acquired. Secondly, according to the pixel data, data consistency items selectively integrating brightness constancy constraints with gradient constraints are constructed and matched feature points are subjected to the initial flow vector updating treatment. Initial flow vectors are substituted into a motion estimation variation model constructed based on data consistency, and then the optimal motion flow vector of each pixel is obtained. Simultaneously, an optimal fuzzy core is constructed based on the brightness of the L0 norm and the fuzzy estimation energy function of a gradient-combined constraint image prior model. Finally, according to the optimal motion flow vector and the optimal fuzzy core, a reconstructed high-resolution image of a variational model is established. Based on the method or the system, the accuracy of motion displacement vectors and the accuracy of fuzzy parameters during the super-resolution reconstruction process can be effectively improved. As a result, an optimal super-resolution reconstruction result is obtained.
Owner:CAPITAL NORMAL UNIVERSITY

A cervical cell pathological slice pathological cell segmentation method and system

The invention discloses a cervical cell pathological slice pathological cell segmentation method and system, in particular, an off-line training sample set of cervical cell pathological slice pathological cell is established, a multi-scale cavity convolution structure semantic segmentation network is introduced on the basis of a depth residual network, and the semantic segmentation model is trained. The unit to be identified is extracted from the cervical pathological slice image, and different types of pathological cells are segmented in the unit to be identified by using the trained semanticsegmentation model. Based on the morphological features of pathological cells, a contour deformation model is established to further optimize the semantic segmentation results. According to the cellnumber and confidence level of different pathological types, the pathological types of the whole slice are predicted. The invention utilizes semantic segmentation network model and deformation variational model to accurately segment different types of pathological cells on the cell pathological slice image, and simultaneously improves the recognition accuracy and the recognition efficiency.
Owner:怀光智能科技(武汉)有限公司

Physics-based MOSFET model for variational modeling

A method of optimizing MOSFET device production which includes defining key independent parameters, formulating those key independent parameters into a canonical variational form, calculating theoretical extracted parameters using at least one of key independent parameters in canonical variational form, physics-based analytical models, or corner models. The method also includes calculating simulated characteristics of a device using the key independent parameters and extracting target data parameters based on at least one of measured data and predicted data, comparing the simulated characteristics to the target data parameters, and modifying the theoretical extracted parameters or key independent parameters in canonical form as a result of the comparison. Then, calculating and outputting the simulated characteristics based on the modified theoretical extracted parameters and the modified key independent parameters in canonical form.
Owner:ORACLE INT CORP

Joint denoising method for seismic data

The invention discloses a joint denoising method for seismic data, comprising the following steps: step 1, by utilizing the variation pattern decomposition of seismic data is decomposed to get new data; step 2, the invention uses the revised total variation method to denoise the new data; step 3, combining and refactoring the new data after the denoising, and getting the seismic section data. The invention utilizes a variation mode decomposition has good characteristic of multi-scale and the edge preserving property of total variation regularization method, thus the boundary feature of desired signal can be protected in suppressing random noise at the same time. The joint denoising method of the present invention based on the variational model decomposition and the total variation-based regularization, according to variational model, combining with modified total variation-based regularization, which further increases the quality of the seismic profile, and gets the high signal-noise ratio edge-preserving seismographic record. The joint denoising method for seismic data is used for processing and explanation of subsequent seismic data, and makes the results more reliable.
Owner:CHINA UNIV OF PETROLEUM (EAST CHINA)

Green tide information extraction method

The invention belongs to the technical field of remote sensing image processing, and relates to a green tide information extraction method. The method comprises: obtaining a sea area satellite remote sensing image with a conventional method, screening images of green tide regions, and performing preprocessing operations of geometric correction and image mosaicking; calculating and determining a green tide range by adopting a normalized difference vegetation index; cutting a region of interest in any shape to generate an irregular image file containing the green tide regions; extracting green tide information by adopting a split Bregman fast projection algorithm of a variational level set two-phase image segmentation based Chen-Wees model; and finally, quantitatively calculating the green tide regions by adopting a quantitative formula. The image segmentation based variational model accurately extracts and quantifies the green tide information on the satellite remote sensing image, so that a conventional manual threshold method can be completely replaced and operational application can be realized; and method is scientific and reasonable in principle, low in human factor quantity, high in calculation speed, accurate and stable in result, high in operability, friendly in application environment, high in practicality and easy to popularize.
Owner:国家海洋局北海预报中心

Foreground target extraction method and apparatus, device and storage medium

ActiveCN107895377AAchieve full extractionImage analysisPattern recognitionShape change
Embodiments of the invention disclose a foreground target extraction method and apparatus, a device and a storage medium. The method includes the steps of obtaining a non-foreground target image and aforeground target image obtained by shooting the same region under the same environment; initializing an objective function, constructing a total variational model based on the non-foreground targetimage, the foreground target image and the objective function; updating the objective function iteratively, solving the total variational model and obtaining an optimal objective function; and performing threshold filtering on the optimal objective function according to preset filtering rules, and determining a foreground target corresponding to the foreground target image according to the filtering result and the foreground target image. The technical scheme of the embodiments of the invention overcomes the defects that the existing method can perform detection only at a specified location and cannot adapt to the shape change in the foreground target, and the extraction process is vulnerable to the influence of uneven lighting and environmental changes, and achieves complete extraction ofthe foreground target.
Owner:GUOGUANG ELECTRIC +1

Method for restoring motion blurred image based on total variational model and neutral network

The invention discloses a method for restoring a motion blurred image based on a total variational model and a neutral network, which mainly solves the problem existing in the prior method that the image cannot be accurately restored. The implementation process comprises the following steps of: (1) constructing a Toeplitz matrix; (2) working out gradients in horizontal and vertical directions; (3) initializing the neutral network; (4) working out a neuron output; (5) working out the output of the neutral network; (6) working out a first variation delta E1 of a network energy function; (7) if the neurons are completely updated, jumping to the step (4); otherwise, jumping to a step (8); (8) if the set iterations are reached, outputting a restoration result, otherwise jumping to a step (9); (9) working out a restoration error; (10) if the restoration error is smaller than the set error, outputting the restoration result, otherwise jumping to a step (11); (11) working out the current input bias matrix; (12) working out a second variation delta E2 of the network energy function; and (13) if the summation of the delta E1 and the delta E2 is less than 0, jumping to a step (2); if the summation of the delta E1 and the delta E2 is more than or equal to 0, jumping to the step (3); and if the delta E1 is equal to 0, outputting the restoration result. The method can obtain relatively accurate restored images, and be applied to the restoration of motion blurred images.
Owner:XIDIAN UNIV

Image restoration method and device based on Shannon-Nuttall wavelet multi-scale expression

ActiveCN104574294AHigh-resolutionReduce resolutionImage enhancementSparse gridInterpolating wavelets
The invention discloses an image restoration method and device based on a Shannon-Nuttall wavelet multi-scale expression. The method comprises the following steps: obtaining a to-be-restored image; shielding a to-be-restored area in the to-be-restored image; constructing a multi-scale interpolation operator based on a Shannon-Nuttall interpolating wavelet and realizing the geometric description of the to-be-restored image under an interpolating wavelet framework; constructing an image restoration variational model under the interpolating wavelet framework according to the geometric description of the to-be-restored image under the interpolating wavelet framework; solving the image restoration variational model under the interpolating wavelet framework and obtaining a sharp image. According to the image restoration method and device disclosed by the invention, the image is reconstructed through the multi-scale interpolation operator, the image restoration variational model is constructed under the interpolating wavelet framework, and the image is solved through a sparse grid algorithm to obtain the restored image, so that the efficiency and the precision of image restoration are improved.
Owner:CHINA AGRI UNIV

Light and face identifying method based on light estimation

ActiveCN107239729AMeet the principle of imagingSatisfy the slowly changing characteristicCharacter and pattern recognitionEffect lightIntrinsics
The invention discloses a light and face identifying method based on light estimation. The method includes the following steps: firstly, in accordance with an image imaging model, estimating a preliminary light from an original face image, then in accordance with the preliminary light, computing a light shielding correction light and light fusion factor, constructing a light fusion model, realizing light shielding light correction, then using a total variational model to conduct smoothing filtering on the light after the light shielding correction to acquire the final light of the original face image, finally using a light invariant extraction model to acquire the light invariant of the original image. According to the invention, the method can better process the problem of shadow edge, extracts the light invariant which contains rich and detailed face characteristics and is closer to original face intrinsic, can meet imaging theories of images, and is applied to the complex lighting environment by means of slow changing characteristics of light.
Owner:NANJING INST OF TECH

Different resolution infrared and visible light image fusion method and system based on total variation

The invention discloses a different resolution infrared and visible light image fusion method and system based on total variation, which comprises establishing a total variation model of image fusion,respectively constraining the fusion image and the infrared image with a data fidelity term and a regularization term, and fusing the relationship between the image and the visible light image, according to the source image, solving the optimal solution of the total variational model and obtaining the fusion result. The invention builds a model according to the situation that the original infrared image and the visible image have different resolutions, and the obtained fusion image has high resolution and clear texture details, while retaining the characteristic that the pixel intensity distribution of the infrared image protrudes out of the thermal target, and the fusion result is convenient for human eyes to discriminate the thermal target and has good visual effect.
Owner:WUHAN UNIV

Decomposition method of image texture and structure based on a total variational model of a guide map

InactiveCN109272539AGood texture/structure decompositionExact decompositionImage enhancementImage analysisImaging processingDecomposition
The invention discloses a decomposition method of image texture and structure based on a total variational model of a guide map, belonging to the technical field of image processing. According to theinvention, firstly, the local main structure of the image is reconstructed based on the guide image filter, then the texture descriptor is calculated according to the reconstructed local structure image, finally, the accuracy and computational efficiency of texture and structure decomposition are improved by combining the multi-scale total variational model and block translation method. The technical scheme of the invention can obtain better texture / structure decomposition effect for noisy images, the textures and structures of different scales can also be decomposed accurately, and the decomposed structure layers can keep the original shading of the image, avoiding the blurring of the structure caused by local smoothing or the color block effect caused by the global optimization method. In addition, the technical proposal of the invention needs to extract simple features, and does not depend on learning a large number of image samples.
Owner:YUNNAN UNIV

Single image super-resolution reconstruction method based on enhanced non-local total variational model a priori

The invention discloses a single image super-resolution reconstruction method based on an enhanced non-local total variational model a priori. The method mainly comprises the following steps of: carrying out bicubic interpolation on an input low-resolution image to obtain an initial high-resolution estimation; the non-local similar pixel group corresponding to each pixel is obtained by using the multi-offset search strategy. Based on the attenuation kernel strategy, the weights of the offset target image blocks attenuated with the offset distance are assigned in the multi-offset search process. Based on the stable group similarity reliability strategy, the similarity reliability between each group of similar pixels and the target pixel is obtained. The super-resolution cost function basedon enhanced non-local total variational is constructed and the high-resolution image is solved. Repeat these steps until the number of iterations reaches a preset value. The image reconstructed by theinvention has obvious advantages in both subjective and objective effects, so the invention is an effective single image super-resolution reconstruction method, and can be widely used in the fields of military, medical, agriculture and the like.
Owner:SICHUAN UNIV

Fault diagnosis method under action of central frequency convergence trend

The invention discloses a fault diagnosis method under the action of a central frequency convergence trend. The method comprises the following steps: (1) collecting a dynamic signal x (t) of rotary mechanical equipment; (2) setting initial decomposition parameters of the variational model; (3) decomposing the dynamic signal x (t) by using a variational model with set initial decomposition parameters, and traversing the signal analysis frequency band under the guidance of a central frequency convergence trend to iteratively decompose the dynamic signal x (t) to obtain an optimized mode {m1... Mn... MN} and a corresponding central frequency {omega1... Omegan... OmegaN}; (4) searching a fault correlation mode mI, guiding parameter optimization by the central frequency omega I of the fault correlation mode mI, and extracting an optimal target component containing fault information; and (5) performing envelope analysis on the optimal target component, and diagnosing the rotating mechanicalequipment according to an envelope spectrum. According to the fault diagnosis method provided by the invention, intelligent decomposition of the original dynamic signal of the diagnosis target equipment is realized by adopting a decomposition mode guided by a central frequency convergence trend, the acquired equipment dynamic signal can be adaptively analyzed, and the difficulty of performing mechanical fault diagnosis by a technician by using a variational mode decomposition method is reduced.
Owner:SUZHOU UNIV

Scallop shell growing pattern segmentation and recognition method

The invention provides a scallop shell growing pattern segmentation and recognition method. After a scallop shell CT image is acquired, Gaussian kernel matching filter is used to enhance the tubular pattern matching. The enhanced image is iterated. Each time of iteration retains the pattern characteristic information, and acquired fine patterns are segmented. According to the invention, the comb shell CT image is used as an input image; Gaussian kernel matching filter is used to enhance the tubular pattern matching in the image; the image segmentation of full variational model image binarization is carried out; intersections in an image pattern network are recognized; the information of a large number of scallop growing points is acquired; and the premise of the scallop growth rate is further accurately calculated. Compared with a traditional time measurement method, the method has the advantages that simple and rapid measurement and calculation are realized through computer analysis; and the method has the advantages of high accuracy and large flux; and the method provides a basis for acquiring accurate breeding phenotype information for shellfish breeding in the future.
Owner:OCEAN UNIV OF CHINA

An image denoising method based on an adaptive weighted total variational model

The invention discloses an image denoising method based on an adaptive weight total variational model. An adaptive weighted total variational model is constructed. A non-local mean filter is used to process the noisy image to obtain an estimated de-noisy image. After preprocessing, four edge operators are used to detect the noisy image so as to obtain and edge detection image, and the adaptive normal parameters are defined according to the four edge operators. The pixels of the edge detection image are reset according to the adaptive normal form parameters to obtain the weight matrix of the noisy image. The denoised image is obtained by substituting the estimated denoised image, the adaptive normal form parameters and the weight matrix into the adaptive weighted total variational model. The invention can self-adaptively separate the edge region and the smooth region of the image according to the image characteristics, and the image texture and the edge are kept while noise is removed,thereby avoiding the generation of the step effect.
Owner:XIAN UNIV OF TECH

Object material classification method for pulse type ToF depth camera

The invention provides an object material classification method for a pulse type ToF camera. The object material classification method belongs to the field of pattern recognition and image processing.The object material classification method comprises the following steps: de-noising original data of a pulse type ToF camera; converting the image data into a feature vector; and optimizing a radialbasis function neural network classifier. The object material classification method is characterized in that an object material classification method is provided for a pulse type ToF imaging model; asecond-order full generalized variational model and a semi-blind deconvolution method are used to carry out denoising processing on the original image; and then sampling, curve fitting and resamplingare performed on the denoised image so as to obtain the feature vector of the object material, and finally optimization of singular value decomposition on a radial basis function neural network classifier is facilitated so that high accuracy and robustness of object material classification can be met.
Owner:BEIJING UNIV OF TECH

Rolling bearing fault diagnosis method based on variational Hilbert-Huang transform

ActiveCN112326245ARealize the effect of noise suppressionAccurately extract fault characteristic frequencyMachine part testingCharacter and pattern recognitionRolling-element bearingHilbert huang transformation
The invention discloses a rolling bearing fault diagnosis method based on variational Hilbert-Huang transform, and the method comprises the following steps: S1, carrying out processing of measurementdata of a rolling bearing through empirical mode decomposition to obtain a plurality of intrinsic mode functions IMFi; S2, screening the obtained IMFi by using a sensitivity criterion to obtain a sensitive mode containing rolling bearing fault information; S3, constructing each sensitive mode into a variational model taking the minimum sum of bandwidths as a target, and solving the variational model to obtain a plurality of finite bandwidth modes; S4, according to the signal characteristics, reconstructing the finite bandwidth modes according to a specified sequence to obtain rolling bearing fault components; S5, using Hilbert-Huang transform for detecting fault components, comparing the fault components with the theoretical fault characteristic frequency, and determining the fault position of the rolling bearing. Rolling bearing data are processed through variational Hilbert-Huang transform, and the effect of rolling bearing noise suppression is achieved.
Owner:AVIC SHANGHAI AERONAUTICAL MEASUREMENT CONTROLLING RES INST

Water quality prediction and early warning system based on anisotropic Delaunay subdivision and implementation method thereof

The invention belongs to the technical field of water quality prediction and early warning systems, in particular to a water quality prediction and early warning system based on the anisotropic Delaunay subdivision. The method comprises the following steps of extracting the river network information through an elevation map, and establishing a graph structure among a plurality of monitoring stations and a graph convolutional network, training and predicting, adopting the hessian matrix filtering for an elevation map enhanced river structure, obtaining a tubular measure and a tubular vector field, constructing a variational model, combining the original gray values, the tubular measure and anisotropism of the elevation map to provide a new energy model; and according to the graph structureand the tubular measurement distance, determining a mark node and a neighborhood thereof, training the graph convolutional neural network by adopting the water quality historical data and the river network information, and predicting the water quality data.
Owner:江西博微新技术有限公司

Pansharking method for gray scale amplitude modulation based on spectrum mapping

ActiveCN109785281AKeep Spectral InformationSpectral information increasesImage enhancementHigh spatial resolutionImage resolution
The invention discloses a pansharking method for gray scale amplitude modulation based on spectrum mapping. The method comprises: firstly, a spectral mapping conversion matrix T is solved by establishing a variational model, and a high spatial resolution image (P _ HCM) with a certain spectral resolution is obtained. Then, carrying out spatial filtering on the image obtained after spectrum mappingto generate spatial details with spectrum information, and injecting the spatial details into the up-sampled MS image to obtain an HR _ MS image with higher resolution; and finally, based on the brightness amplitude modulation (SFIM) principle of smooth filtering, obtaining a final fusion image (HRMS) with high spatial and spectral resolution. On the basis of a traditional pansharking method, thespectrum and spatial information of the HRPAN image and the LRMS image are fully mined, on the premise that the spatial structure information is well kept, good spectrum fidelity is achieved, and a fused image with a good visual effect can be obtained.
Owner:NANJING UNIV OF SCI & TECH

Infrared image enhancement method based on transmission image fusion

The invention discloses an infrared image enhancement method based on transmission image fusion. The method comprises: carrying out under-segmentation and super-segmentation on an image under different scales through a clustering algorithm, and generating two clustering segmentation results; for the clustering segmentation result, generating two transmission images by using image pure pixel priori; constructing a fusion weight map for each transmission map, constructing an image pyramid structure model for each transmission map and the fusion weight map thereof, and performing layer-by-layer fusion to obtain a fused transmission map; based on the variational model, carrying out edge-preserving smoothing processing on the fused transmission map to obtain an optimized transmission map; and obtaining an enhanced infrared image by using the optimized transmission image. According to the method, a large number of originally hidden scene details in the infrared image can be recovered, and negative effects cannot be introduced.
Owner:NANJING COLLEGE OF INFORMATION TECH

Global image segmentation method of multi-resolution analysis

ActiveCN106296668AHandle Boundaries EfficientlySmooth and complete object boundariesImage enhancementImage analysisNMR - Nuclear magnetic resonanceMulti resolution analysis
The invention discloses a global image segmentation method of multi-resolution analysis. The method comprises the following steps that firstly, compactly-supported wavelet transformation is used to decompose an image; then, taking a decomposed coefficient matrix which carries out reconstruction as a filter; secondly, the filter acquired through reconstruction is applied to a gradient item of a horizontal set function so as to form a new regular term; and then, based on an ideal condition, boundaries of an object and a background are fully fit, and a condition that a boundary regular item is used for maintaining a global optimal segmentation result is provided; and finally, a rapid dual algorithm is used to minimize variation models of a regular item containing multi-resolution information and a boundary regular item so as to realize evolution of a horizontal set curve. A medical nuclear magnetic resonance image segmentation experiment shows that an obvious capturing capability and high-efficient calculating time are possessed by using the method of the invention for a narrow and long topology structure, compared to a general active contour model.
Owner:苏州伟易佳电子有限公司

Non-stationary signal state monitoring method and system for nuclear power equipment

ActiveCN111881594AImprove feature expressivenessSolve the problem of noise removalMeasurement devicesDesign optimisation/simulationFeature extractionNuclear power
The invention relates to a non-stationary signal state monitoring method and system for nuclear power equipment. The monitoring method comprises the following steps: acquiring a characteristic signalof each measuring point of the nuclear power equipment in normal operation; carrying out variational mode decomposition on each characteristic signal, and construcitng a constrained variational model;introducing a penalty parameter and a penalty factor, and determining an optimal solution of the constrained variational model; determining modal components according to the optimal solution, screening the modal components, and determining an intrinsic modal function; performing feature extraction on the time sequence by adopting multi-scale weighted permutation entropy, and determining real-timeweighted permutation entropies under different scales; determining a weighted permutation entropy statistical threshold of each weighted permutation entropy under different scales according to the real-time weighted permutation entropies under different scales; and comparing the real-time weighted permutation entropies under different scales with the weighted permutation entropy statistical threshold, and determining the non-stationary signal state of the nuclear power equipment. The detection precision can be improved.
Owner:HARBIN ENG UNIV

Renewable energy short-term prediction method based on data mining and variational mode decomposition

The invention discloses a renewable energy short-term prediction method based on data mining and variational mode decomposition. The renewable energy short-term prediction method comprises the following steps: 1), dividing a clustering process into data clusters: dividing the data clusters into some different groups in a given data set through a clustering algorithm, and extracting historical power generation data similar to environmental features into the same class; 2) processing historical data by variational modal decomposition: determining the bandwidth and the center frequency of each modal component by continuously iterating and searching the optimal solution of a variational model, and performing adaptive subdivision on the signal in a frequency domain; and 3) utilizing an extremelearning machine to predict future data. The technical scheme of the embodiment of the invention is combined with the K-means clustering technology, utilizes the variational mode decomposition and theextreme learning machine to predict the power generation amount of renewable energy, and has better accuracy compared with a traditional prediction method.
Owner:ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER COMPANY +1

Variation-based multi-radar fusion track state estimation method

The invention relates to a variation-based multi-radar fusion track state estimation method, belongs to the field of multi-sensor target tracking and data fusion, and aims to overcome the defect thata radar information processing system often swings back and forth among a plurality of original tracks during track fusion to form a saw-tooth shape which influences of track quality, and the solve the problem that turning bulging and even tracking failure easily occur in a conventional method. Observation information of multiple radars is managed in a unified mode, a variation model containing atarget motion speed transformation curve is constructed, a rapid solving mode of the variation model is given, the solved motion speed curve is used for track extrapolation of all the observation radars, and finally a target state estimation value is estimated and updated. According to the method, correlation of observation information is considered to the greatest extent, the adverse effects of track relative deviation and sudden maneuvering of the target on target estimation are reduced, the target motion speed and the fusion position estimation precision are improved, the track quality is improved, and reliable tracking of the target is obtained.
Owner:STRATEGIC EARLY WARNING RES INST OF THE PEOPLES LIBERATION ARMY AIR FORCE RES INST

Weight-adaptive mixed-order total variation image denoising algorithm

The invention discloses a weight-adaptive mixed-order total variation image denoising algorithm, which comprises the following steps of: preprocessing an image, converting a color image into a grayscale image, and adding Gaussian white noise to the image; constructing an image denoising model by taking the total variation model as a basic framework, and solving minimization under a constraint model; putting forward a high-order total variation model, fusing the high-order total variation model into an image denoising model, dividing an edge texture region and a flat region by the whole algorithm model through the structural information of a noisy image, constructing a weight function, combining the total variation model and the high-order total variation model, establishing a mixed-order total variation image denoising model, and obtaining a minimizedsolution. A gradient constraint term is proposed, a denoising model is introduced, the structural information of the image is ensured, a final weight adaptive mixed-order total variation image denoising model is established, and the solution is minimized. Compared with a traditional algorithm model, the algorithm provided by the invention has the advantages that the peak signal-to-noise ratio is improved by 8-13dB, and the numerical value of the structural similarity is superior to that of the previous algorithm.
Owner:NANJING UNIV OF INFORMATION SCI & TECH

Hyper-spectral image super-resolution method based on hyper-parameter fidelity and depth prior joint learning

The invention discloses a hyper-spectral image super-resolution method based on hyper-parameter fidelity and depth prior joint learning. The method comprises the following steps: establishing a hyper-spectral and multi-spectral image fusion variational model based on depth prior regularization of a hyper-parameter fidelity model; optimizing a hyperspectral multispectral image fusion variation model; carrying out tensor representation on the model optimization iteration process; performing network expansion on the iterative process of variational model optimization, and executing the iterative process of optimization; and training the network by using the L1 norm as a loss function. The method has the capability of representing the hyperspectral image degradation model and the data prior in the network at the same time, and has excellent performance when being applied to hyperspectral multispectral image fusion.
Owner:NANJING UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products