39results about How to "Improve sparsity" patented technology

The invention relates to the technical field of mechanical fault diagnosis, in particular to a mechanical fault diagnosis method and system based on a pulse neural network. The method comprises the following steps: constructing a fault diagnosis network, wherein the fault diagnosis network comprises an improved gating attention coding module and a pulse residual network; the method comprises the following steps: replacing a common convolution layer of a gating attention coding module with a one-dimensional sparse convolution layer, and replacing a two-dimensional batch normalization layer with a one-dimensional batch normalization layer to obtain an improved gating attention coding module; setting a pulse neuron in the fault diagnosis network as a PSN neuron, and embedding a time step shrinkage layer in the pulse residual network; and inputting the mechanical vibration signal of the current sampling period into a fault diagnosis network, and outputting a predicted fault category label of the current sampling period. According to the invention, high-efficiency and high-precision mechanical fault intelligent diagnosis is realized.

The invention relates to the technical field of information retrieval, in particular to an interdisciplinary knowledge graph construction system and method, and the system comprises a relation matrix construction module which analyzes a node relation based on nodes in a knowledge graph, carries out the statistics according to an association frequency, selects interdisciplinary nodes to construct an association matrix, calculates a node fitting degree, and determines an association node set; and generating an interdisciplinary incidence relation matrix. According to the method, through correlation frequency statistics and fitting degree calculation of interdisciplinary nodes, effective construction of comprehensive correlation between the nodes, decomposition of the feature matrix and elimination of non-key feature items, the sparsity of the matrix is improved, data are more concise and clear, and through screening of low-weight elements, reservation of important node relation and further optimization of matrix feature expression, the method has the advantages of being high in robustness and high in reliability. In the coupling analysis and partitioning process of the nodes, redundant connection of the low-coupling-degree nodes is reduced, the core relation is highlighted, the node structure is recombined, and the accuracy and compactness of the multidisciplinary knowledge graph are remarkably improved.

The invention discloses a signal transmission method, system and device based on generalized chirp division multiplexing and a medium, and the method comprises the steps: employing a specific pulse forming window for each chirp subcarrier at a transmitting end, and employing a matching filtering window for each chirp matching subcarrier at a receiving end; fractional time delay and fractional Doppler effects introduced by a double-dispersion channel in a communication system, namely diffusion of received symbols, are suppressed, sparsity of an equivalent channel of the system is improved, and interference between symbols received by a communication receiving end is greatly reduced. The diffusion degree of a receiving pilot symbol for channel estimation is slowed down, so that the channel estimation is more accurate, the number of protection symbols required by a sending end is reduced, and the reachable capacity of a communication system is improved; compared with the prior art, an equivalent channel is more sparse, interference between a pilot symbol and a data symbol of a receiving end is reduced, so that pilot-data separability of the system is higher, signal detection complexity of the communication system is lower, and the detection effect is better.

The invention provides a bearing compound fault sparse classification method and system based on adaptive feature selection, and the method comprises the steps: firstly, enhancing sample features through intra-class and inter-class distance measurement, and removing a redundant feature set to achieve the adaptive fault feature screening, and effectively improving the fault recognition performance; constructing a single-class fault initialization dictionary by using the feature training set, and enhancing sparsity and keeping model convexity through a norm constraint term; and iteratively training the over-complete dictionary by using the corresponding sample set of each fault category, and finally realizing the discriminant classification of the bearing compound fault through a discriminant vector. Compared with a traditional method in simulation and test, the method can effectively solve the feature adaptive selection problem in the LC-KSVD algorithm, improves the accuracy and robustness of fault classification under the small dictionary condition and the strong noise condition, reduces the time cost, and provides valuable technical reference for engineering application.

The present invention provides a method for visual relocation of unmanned aerial vehicles based on object semantics, including: constructing a visual relocation model of unmanned aerial vehicles: an image processing and representation module, a scene matching module and a relocation module. The current scene is photographed by a camera of the unmanned aerial vehicle, and the semantic information of the object contained in the picture is obtained by using the YOLOv3 target detection network in the image processing module, and a topological map is constructed using the semantic information as a sparse description of the environment; in the case of prior knowledge of the map, an accompanying map of the picture to be matched is constructed, and a random walk algorithm is used to match the semantic features and the scene; the EPnP algorithm is used to solve the position and posture of the unmanned aerial vehicle to complete the relocation. The present invention can realize real-time relocation of unmanned aerial vehicles with strong robustness when the scene lighting conditions change dynamically, and is particularly suitable for automation tasks such as industrial inspections.

The invention relates to a brittleness parameter direct inversion method and system based on reweighting, and belongs to the technical field of oil-gas exploration seismic inversion. The method comprises the following steps: performing angle superposition on pre-stack seismic data; establishing a forward modeling model based on the initial model, the difference matrix and the wavelet kernel matrix; on the basis of the forward modeling model, constructing a brittleness parameter inversion objective function based on reweighting; and inverting the objective function based on the seismic data obtained by angle superposition and the brittleness parameter based on reweighting to obtain an inversion result. According to the method, the accumulative error is reduced by adopting direct inversion, and the accumulative error is continuously increased by a conventional method for converting the brittleness index through pre-stack inversion, so that the prediction of the brittleness index is not accurate. According to the brittleness parameter direct inversion method, the problem can be well solved, the brittleness index is represented through direct inversion of E < rho >, accumulated errors are reduced to a great extent, and the brittleness index prediction precision is improved.

The present invention provides a training method, a recommendation method, a device, an electronic device and a storage medium for a click rate prediction model; the training method for the click rate prediction model includes: obtaining sample features and the click results of the sample features corresponding to the user; initializing the click rate prediction model according to the set weight parameters; processing the sample features through the click rate prediction model to obtain the predicted click rate; constructing an objective function according to the error between the click result and the predicted click rate, and a zero-norm regularization term; back-propagating the error in the click rate prediction model through the objective function, and updating the weight parameters of the click rate prediction model during the propagation process. Through the present invention, the sparsity of the weight parameters can be improved, the file size of the generated click rate prediction model can be reduced, and the accuracy of click rate prediction can be improved.

The invention belongs to the technical field of time-frequency analysis, and particularly relates to a rolling bearing fault method and system based on Riesz atlas sparse decomposition, which utilizes a new time-frequency analysis method-Riesz four-parameter spectrum time-frequency analysis method to obtain four-parameter information of amplitude, frequency, phase and direction of a vibration signal in a time-frequency domain. The time-frequency aggregation and the anti-noise performance of an analysis result are improved; in order to screen out an optimal parameter spectrum and expand the symbol dynamics entropy to a two-dimensional scale, a new entropy calculation method, namely cosine symbol dynamics entropy, is provided; and reconstructing components of the parameter spectrum obtained by entropy screening, and constructing a sparse filter parameter optimization component amplitude, so that a decomposition result has physical significance. The bearing vibration signal under the variable speed working condition and the bearing vibration signal under the constant speed are analyzed, the RSSD method is provided on the basis, and feature extraction and diagnosis of rolling bearing faults under the constant rotating speed working condition and the variable rotating speed working condition are achieved.

The application discloses a machine learning algorithm for screening of metal elements in a diatomic catalyst full cycle based on product yield prediction and application thereof, and the algorithm comprises the following steps: (1) collecting attribute features of metal elements in the diatomic catalyst as a feature space 1; (2) preliminarily screening the attribute features according to comprehensive scores of feature importance to obtain a feature space 2; (3) performing feature operation processing on the feature space 2 to generate a multivariate descriptor space matrix; (4) performing dimension reduction processing on the multivariate descriptor space matrix by combining a gradient boosting regression model with a recursive feature elimination algorithm; (5) further screening to obtain an interpretable descriptor by using a least absolute shrinkage and selection operator; and (6) applying the interpretable descriptor to a random forest regression model to screen a catalyst for a propane dehydrogenation reaction to prepare propylene. The application can solve the problem that a traditional descriptor is difficult to directly predict a catalytic product yield.

The invention discloses a virtual multi-view fusion millimeter wave point cloud imaging method, which belongs to the field of millimeter wave radars, and realizes three-dimensional point cloud imaging of a target by constructing a virtual observation view angle and fusing perception data of different view angles: establishing a multi-view synthetic aperture radar imaging model; a single-bit compressed sensing framework is introduced to construct a sparse imaging optimization problem, an accurate relay angle is solved by minimizing a model error, and a relay plane model error is corrected; and high-precision 3D point cloud imaging of the region of interest is realized based on the accurate angle. According to the method, a three-dimensional point cloud of a target is constructed by fusing virtual visual angle detection shielding characteristics formed by reflection of a plurality of relay surfaces, joint estimation of relay surface angle prior and 3D point cloud is realized by a nested iteration method, and artifacts and offset caused by prior errors are eliminated; the sparsity and the target continuity of the three-dimensional point cloud can be improved through the additional composite norm constraint, and the accuracy of virtual multi-view 3D point cloud imaging is effectively improved.

The present invention discloses a method for estimating blood flow velocity by transmitting sparse pulses, and specifically designs a blood flow velocity estimation method based on sparse pulses. The present invention transmits sparse pulses to obtain echo signals, and then performs autocorrelation operation and vectorization on the echo signals to obtain virtual echo signals with equivalent uniform emission and a larger number of pulses. The result of spectral estimation of the echo signals is compared with the result of spectral estimation of the virtual echo signals, and finally the value of blood flow velocity estimation is obtained. The method used in the present invention can obtain high-resolution, low-artifact blood flow velocity estimation results with a small number of pulses, which is beneficial for medical ultrasound systems to further improve the refresh rate and real-time performance of B-mode imaging.

The present invention relates to the field of image communication technology, and in particular to a construction progress monitoring method for a highway construction project monitoring platform, comprising: using a segmentation threshold to segment a differential image between a target image and a reference image to obtain a target pixel on the target image and a reference pixel on the reference image; grayscale values ​​of the target pixel are grayscale transformed, the sparse fusion degree of the transformation curve used and the historical curve is the largest, the sparse fusion degree is obtained by the sparsity of the difference between the grayscale value after the grayscale transformation and the grayscale value of the reference pixel, the similarity between the transformation curve and the historical curve, and the similarity of the grayscale distribution of the target pixel at the first two positions, grayscale transformation is performed on the target image using the transformation curve, and the target image is compressed and transmitted using the difference between the target image after the grayscale transformation and the reference image; the segmentation threshold is corrected according to the error of the transformation curve. The present invention increases the compression transmission efficiency of the target image and saves network resources.

The present invention discloses a blast furnace sinter ore particle size detection method and system based on RGB and laser feature fusion. By acquiring an RGB data set and a laser data set, a deep learning network of an encoder-decoder structure is constructed to obtain an RGB initial feature tensor and a laser initial feature tensor. A multi-source feature weighted fusion subnetwork is constructed at the last layer of a fusion branch. Based on the multi-source feature weighted fusion subnetwork, the RGB initial feature tensor and the laser initial feature tensor are adaptively weighted fused to obtain a fused feature tensor, and the fused feature tensor is input into a decoding process to obtain a segmented image. The particle size of the blast furnace sinter ore is detected according to the segmented image, and the technical problem of low detection accuracy of the existing blast furnace sinter ore particle size is solved. By adding a multi-source feature weighted fusion subnetwork to the deep learning network of the encoder-decoder structure, the adaptive weighted fusion of multi-source features can be achieved, and the supplementary role of the laser is fully utilized, which is beneficial to improving the detection accuracy of the blast furnace sinter ore particle size.

This invention relates to a power data sparsity compression observation method for non-intrusive load monitoring, comprising the following steps: Step 1, acquiring power monitoring data and determining its electricity consumption behavior pattern; Step 2, initializing wavelet sparse basis for sample data based on simple power behavior patterns; Step 3, based on the determination results of the electricity consumption behavior pattern of the completed training samples and the results after wavelet basis initialization in Steps 1 and 2, performing improved K-SVD training and generating sparse basis; Step 4, training the sample dataset Y based on Step 3 to obtain a sparse dictionary D, and performing power monitoring data compression observation based on the improved sparse basis.

The present invention provides a total variation regularized direction-finding method and system based on acoustic features, comprising obtaining an output acoustic signal of an incident sound source based on a planar array; calculating an estimated value of the direction of arrival of the incident sound source based on a pre-constructed two-dimensional total variation sparse direction of arrival estimation model; wherein the construction process of the two-dimensional total variation sparse direction of arrival estimation model is specifically as follows: establishing a spatially extended sound source model according to the spatial distribution characteristics of the sound source; constructing a two-dimensional array manifold matrix to obtain an overcomplete representation of the two-dimensional array manifold matrix, and determining a two-dimensional sparse representation of the incident sound source based on the overcomplete representation of the two-dimensional array manifold matrix; constructing a two-dimensional total variation structure matrix, and constructing a two-dimensional total variation sparse direction of arrival estimation model based on the two-dimensional total variation structure matrix; the present invention promotes piecewise constant solutions and embodies spatial smoothness, and thus effectively improves DOA estimation performance in situations involving block sparsity or a mixture of peaks and flat areas.

This invention discloses a text summarization method and system based on sparse attention acceleration, comprising: reading long text data, performing word segmentation and embedding encoding processing, extracting sequence feature vectors, and mapping them to a query matrix Q, a key matrix K, and a value matrix V; constructing a summarization generation network, including a sparse attention calculation module, a feedforward calculation module, a prediction head module, and a key-value caching module; inputting the query matrix Q, key matrix K, and value matrix V into the summarization generation network, first passing through a backbone network composed of multiple stacks of the sparse attention calculation module and the feedforward calculation module, and then being processed by the prediction head module, while the key-value caching module caches historical decoding states in real time to obtain an initial text feature vector; based on the initial text feature vector, performing an autoregressive decoding process through the summarization generation network to output the final summary result. This invention can accelerate the decoding process and ensure the semantic integrity and coherence of the generated summary.

The present invention discloses a hybrid dynamic pruning method for large medical model compression, which adopts a sparse predictor based on hybrid sparse prediction, and has the advantage of reducing performance loss compared with other dynamic pruning methods based solely on dynamic sparse prediction. It effectively solves the problem of serious performance degradation after compression of large medical models, improves the model reasoning speed, and provides a feasible solution for the promotion and application of high-sparse, high-performance large models in practical applications.

The invention provides a global sparse texture filtering method based on edge structure preservation, which comprises the following steps of: introducing a texture suppression function into a penalty term, constraining the gradient of an output image, and suppressing texture, noise and unnecessary detail information in the image by setting two threshold values through the texture suppression function, then, the suppressed gradient is used as input of a denominator of a penalty term, so that the penalty term can fully distinguish the texture from the structure; a penalty term is constrained by adopting a sparse regular L1 norm, non-convex optimization is converted into convex optimization by introducing subgradient, and iterative solution is carried out by adopting an alternating direction multiplier method, so that better edge preservation is achieved; and an Lp norm with relatively high sparsity is adopted to constrain a penalty term, and a preprocessing conjugate gradient method is adopted to accelerate and improve the calculation efficiency, so that a more robust and sparsity image smoothing effect is achieved. According to the method, the robustness of an algorithm for distinguishing textures and structures can be improved, relatively good semantic information is reserved, and relatively good smooth performance of edge structure maintenance is achieved.

This invention belongs to the field of radar signal processing technology and discloses a method for detecting small sea targets based on optimized feature mode decomposition, including: S1: acquiring the signal data to be detected; S2: decomposing the original signal into several modal components using FMD, and selecting envelope spectral entropy as the fitness function; S3: using the SOS algorithm to globally optimize the fitness function in FMD; S4: introducing the PSO algorithm to locally optimize the key parameters of FMD; S5: retaining components with low envelope spectral entropy values ​​and correlation coefficients greater than a threshold; S6: extracting envelope spectral entropy and frequency band energy proportion features from the selected modal components, introducing the Gini coefficient as a weighting factor, and constructing GSEBE joint features; S7: inputting the envelope spectral entropy value into a DELM classifier with controllable false alarm rate, and achieving target detection based on the comparison between the predicted value and the decision threshold. This invention enhances the ability to distinguish between sea clutter and target echoes, achieving more accurate classification and detection.

The invention relates to a health state recognition method based on image statistical characteristics. Comprising the following steps of: firstly, acquiring an original vibration signal of a bearing by using a sensor, converting the original vibration signal into a time-frequency domain color image by adopting short-time Fourier transform, and then converting the time-frequency domain color image into a grayscale image by using grayscale transform; secondly, extracting six types of image statistical characteristics from the grayscale image, wherein the six types of image statistical characteristics comprise a mean value, a variance, a contrast ratio, correlation, energy and homogeneity; then, six types of image statistical features are used as input, and a fusion feature vector is output through a designed stability-enhanced sparse PCA (Principal Component Analysis) module; and finally, performing health state recognition on the fused feature vector by using a state change detection function. The method is good in recognition effect and high in industrial scene adaptability, and provides high-precision and high-reliability technical support for predictive maintenance of equipment; the method is high in calculation efficiency, good in real-time performance and high in robustness, unification of precision, efficiency and robustness is achieved, and the method is very suitable for online real-time monitoring application in engineering practice.

This application provides a training method, a spiking neural network, and an apparatus for a spiking neural network. The method includes: iteratively training the spiking neural network using training data of the object to be processed, wherein the spiking neural network contains at least one neuron model; calculating a task loss function based on the training results of the spiking neural network and updating the weight parameters; updating the current discharge threshold parameter of the neuron model based on the historical discharge frequency of the neuron model, wherein the value of the current discharge threshold parameter increases with the increase of the historical discharge frequency; and obtaining the trained spiking neural network model using the updated weight parameters and the updated discharge threshold parameter after multiple iterations of training. This method facilitates the training of a sparsely discharging spiking neural network, reduces the energy consumption during spiking neural network applications, and improves model energy efficiency by using the trained spiking neural network model to process the data of the object to be processed.

The application discloses an image rain removing method and device for unmanned aerial vehicle power grid inspection, and belongs to the technical field of unmanned aerial vehicle aerial image rain removing.The image rain removing method provided by the application comprises the following steps: acquiring an aerial image collected by an unmanned aerial vehicle, performing step-by-step information mining operation on the aerial image to obtain a denoising sub-image; in at least one information mining operation process, a spatial dimension reduction operation and a spatial dimension expansion operation are performed in sequence; and a target image with rain streaks removed is constructed based on the denoising sub-image, so that the image rain removing operation is completed.In the feature information mining process of the aerial image, the spatial dimension reduction operation is performed first, the content of rain streak noise is greatly compressed, and the influence on other objects in the image is relatively small, then the spatial dimension expansion operation is performed, the sparseness and the discrete degree of residual rain streak information are improved, the rain streaks are efficiently removed, and other image features are well reserved.

The invention provides a structural modal analysis method based on improved blind source separation. The method comprises the following steps: firstly, obtaining a sparse matrix and a time-frequency coefficient of a response signal through multiple synchronous extraction transformation with a sliding window; summing all the time-frequency coefficients along the time direction so as to obtain time-frequency total energy; then locking the position of the maximum time-frequency energy scatter point by adopting a peak value pickup method; calculating energy points having the same distance with each peak value position by using a cosine distance so as to estimate each source frequency band; and finally, performing multiple synchronous extraction inverse transformation with a sliding window on each source frequency band, and reconstructing a single-mode signal.

The invention discloses an intelligent factory intrusion equipment detection method based on a long and short term memory network. According to the method, the time sequence relation between the characteristics of the signals emitted by the equipment in the intelligent factory at different moments is analyzed based on the long-short-term memory network, and deeper information related to the signal characteristics is extracted from the time sequence relation, so that the detection efficiency of the illegal invasion equipment is improved. According to the invention, training is carried out based on a specific loss function, so that the feature space of the neural network has good sparsity. And by using the sparsity, the identified features of the illegal invasion equipment can be clustered, so that the identification of the number of the illegal invasion equipment in the intelligent factory is realized. The method is of great significance in finding illegal invasion equipment and excluding the illegal invasion equipment in practical application.

This application provides a source code generation method and apparatus. The method includes: determining target features based on a first fragment, wherein the first fragment includes one or more codes input by a user before a first moment, and the target features are used to indicate code attributes of the codes in the fragment; generating multiple candidate fragments; and selecting a second fragment from the multiple candidate fragments based on the target features, wherein the features of the second fragment have the highest similarity to the target features among the features of the multiple candidate fragments. The target features are obtained based on the source code input by the user and serve as the user's sequence preference. The source code suggestions provided to the user are generated based on the user's sequence preference. In other words, the generated source code is related to the source code input by the user. Therefore, the accuracy of the generated source code can be improved.

The present invention relates to the technical field of electric power systems, and discloses a GIS fault diagnosis method and system based on acoustic monitoring, including: obtaining initial sound data of each monitoring point in the GIS; performing a first fault location according to the initial sound data to obtain the area where the fault source in the GIS is located; and extracting the positioning sound data; performing a second fault location according to the positioning sound data to obtain the location information of multiple fault sources in the GIS; and performing voiceprint feature analysis according to the positioning sound data to obtain the fault types of multiple fault sources in the GIS; the fault types include partial discharge faults and mechanical faults; and performing GIS fault diagnosis according to the location information of the multiple fault sources and the fault types of the multiple fault sources. The GIS fault diagnosis method and system based on acoustic monitoring of the present invention significantly improves the accuracy of determining the location of the fault source, and simultaneously realizes the timeliness diagnosis of multiple fault sources.

This invention relates to the field of remote sensing image compression technology, specifically to a remote sensing image compression method based on dual-tree complex wavelet convolution and a frequency dictionary entropy model. The method includes: compressing remote sensing images using a remote sensing image compression model based on dual-tree complex wavelet convolution and a frequency dictionary entropy model; the remote sensing image compression model includes a dual-tree complex wavelet convolution module, a frequency dictionary entropy model module, and a residual feature extraction module; the dual-tree complex wavelet convolution module is used to perform downsampling and upsampling feature processing on the remote sensing image, effectively removing frequency domain redundancy in the latent representation; the frequency dictionary entropy model module is used to perform frequency division on the obtained latent representation of the remote sensing image to establish a probability model; the residual feature extraction module is used to extract features from the remote sensing image, effectively capturing long-range contextual information. This invention enhances the model's image compression effect by introducing dual-tree complex wavelet convolution and a frequency dictionary entropy model, significantly improving the fidelity of remote sensing images at high compression ratios.

The present invention discloses a radar target recognition method based on a multi-scale convolutional capsule network, which uses an improved deep learning network, namely a multi-scale convolutional capsule network, to identify and classify target HRRP, integrates a network with strong generalization ability to fully capture effective structural information in HRRP, and optimizes the parameters in the network while building a complex model, so as to achieve faster convergence and higher recognition performance. Based on the ability of CNN to obtain local structural information and the Inception structure, a multi-scale CNN structure is proposed, which captures the separable information of different levels of HRRP targets in detail from multiple angles, and increases the sparsity of the network. At the same time, a more robust capsule network is used as a supplement to CNN, which makes up for the deficiency of CNN in reflecting the target HRRP sequence structural information, and exerts greater advantages in the field of radar HRRP recognition with stronger fitting ability.

A low-complexity OTFS channel estimation method based on windowing processing, a storage medium and a computer are provided, which solve the problems of high complexity and large pilot overhead of the cross-correlation channel estimation method. The method comprises the following steps: constructing a Doppler axis function in a time-delay-Doppler domain equivalent channel processed by a window function, and performing interception to obtain a matching function; setting a threshold coefficient, a Doppler axis resolution, an initial path cycle number and a pilot area; obtaining an estimated time-delay-Doppler domain impulse response matrix according to the pilot area position relationship; performing a cross-correlation operation on the matching function and the impulse response matrix according to the Doppler axis resolution to obtain a maximum cross-correlation value, a Doppler grid point value, a time-delay grid point value and a cross-correlation value thereof; estimating corresponding channel parameters; updating all values on the lth time-delay grid point of the impulse response matrix; setting an iteration stopping condition, and outputting the estimated value when the stopping condition is met. The method is applied to the field of high-speed mobile communication.

The invention discloses a sparse array beam reconfigurable design method based on minimum spacing constraint. According to the method, firstly, an off-grid group sparse reconstruction model of multiple directional diagrams is established at a half-wavelength interval for an array with a given aperture length, then an inequality constraint condition for minimum adjacent array element interval control is further introduced into the group sparse reconstruction model, and a 10 problem of a joint performance function is converted into a 10 / 11 combined optimization problem. The invention provides an SAI (Sequential Alternating Iteration) algorithm for solving an excitation vector and a position of each array element of a sparse array. Simulation experiments show that the method can effectively realize sparse array antenna pattern reconstruction, and can control the minimum spacing between adjacent array elements, and the designed sparse array is easy for engineering realization.