30results about How to "Reduce space complexity" patented technology

The invention belongs to the technical field of power grid main equipment diagnosis, and provides a power grid main equipment fault diagnosis method and system based on voiceprint cloud edge collaboration, which adopts a new cloud edge collaboration method and divides the analysis of voiceprint monitoring data into two stages. According to the method, the exception detection model is deployed on the edge side, exception analysis of mass voiceprint data is achieved, concurrent requests of a large number of concurrent signals at the cloud side are avoided, and the timeliness of voiceprint monitoring and analysis is improved. In the cloud, a multi-modal defect identification method fusing voiceprint, visible light, infrared and other multi-source data is deployed, secondary classification of sound abnormal signals is realized, and the reliability of equipment operation, maintenance and management is improved.

The invention discloses an optical fiber terahertz communication system optimization method based on artificial intelligence, and belongs to the technical field of communication. According to the method, four-dimensional constellation shaping is constructed by utilizing additional dimensions provided by two independent sidebands, and an end-to-end optimization framework is divided into a training stage and a deployment stage; in the training stage, firstly, an equalizer based on a pruning neural network and an equalized channel model based on a mixed density network are trained in sequence through receiving and transmitting of a random training sequence; then connecting a constellation shaper at a transmitting end with a demapper at a receiving end through an equalized channel model with fixed parameters, thereby considering full-link residual distortion and noise after equalization, and carrying out back propagation gradient to learn a globally optimal four-dimensional joint geometric probability constellation diagram to further approach the Shannon limit; and in the deployment stage, transmission of the bit stream is realized through coordination of the four-dimensional geometric shaping constellation table and the probability distribution obtained in the training stage and an equalizer at the receiving end and a demapper.

The invention discloses an end-side cloud system-oriented multi-task joint scheduling method, which comprises the following steps of: firstly, constructing an end-side cloud system multi-task joint scheduling model on the basis of uniformly describing nodes and tasks of an end-side cloud system; secondly, by taking maximization of the expected benefit of the multi-task joint scheduling as an optimization target, establishing an optimization problem solving constraint, and realizing construction of the optimization problem of the multi-task joint scheduling of the end-side cloud system; and finally, a mode of combining a heuristic strategy and nonlinear optimization solution is adopted to realize step-by-step decision of task execution nodes and task allocation resource quantity, and final generation of a multi-task joint scheduling scheme of the end-side cloud system is completed. According to the method, concurrent computation-intensive task joint scheduling highly suitable for the end-side cloud system can be realized from three aspects of multi-task joint scheduling model construction, multi-task joint scheduling optimization problem establishment and multi-task joint scheduling scheme generation.

The invention relates to the technical field of wiring, and discloses a wiring method, electronic equipment and a computer readable storage medium, and the method comprises the steps: converting an initial layout layout of a circuit board into a grid chart; according to the grid chart, a coarse wiring result of the initial layout layout is determined, and the coarse wiring result is an initial wiring path meeting a wiring task and achieving a wiring cross minimization target; generating a wiring geometric figure according to the safety spacing and the initial wiring path, and generating a device geometric figure according to the safety spacing and component information in the initial layout layout; and under the limitation of the geometric figure of the device, optimizing the position of the geometric figure of the wiring by taking the wiring constraint as a target to obtain a fine wiring result. According to the method, the wiring process is divided into a coarse wiring stage meeting the goals of connectivity and minimum wiring intersection number and a fine wiring stage meeting the wiring constraint, so that the solving complexity can be reduced, and a wiring scheme with the minimum wiring intersection number can be obtained in a relatively short time.

The invention relates to a multi-sub-mirror structured short video batch mixing and cutting method with controllable repetition degree of materials, which comprises the following steps of: separately managing sub-mirrors and the materials, establishing a mapping relation between the materials and the videos, introducing a control value of an allowable maximum repetition sub-mirror number during calculation and combination, and performing gradient repetition degree operation from low to high. Through tracking and statistics of historical materials, a material minimum usage priority principle is applied, and video library repetition comparison is converted into material repetition comparison; the method has the advantages that the historical use records of the materials are persistent, and the use condition of each material can be tracked; the output video combinations are output according to the gradient of the repetition degree from low to high, the material repetition index of the material combination of each video is visualized, and the risk of preferential release is reduced to the minimum; the situation that all combinations are compared with all videos in a video library one by one is effectively avoided, and the time and space complexity of a combination calculation algorithm is reduced.

A plant community multi-objective network optimization method in an intelligent logistics edge device completes intelligent logistics network multi-objective optimization by simulating behaviors of sowing, growing, flowering and fruiting of a plant community on the intelligent logistics edge device, and comprises the following steps: step 1, solving intelligent logistics network planning optimization by the plant community; step 2, optimizing a plant community solution intelligent logistics plan; step 3, performing optimization on plant community solution intelligent logistics execution; and step 4, optimization of the intelligent logistics reverse network for solving the plant community. According to the method, multi-objective optimization of the intelligent logistics network can be efficiently completed on the edge device.

The present invention relates to the technical field of sequence recommendation, and provides a sequence recommendation method and system based on a semantic generalization diffusion model, including: obtaining a user's behavior sequence; extracting the user's prior semantic behavior from an item-semetic dependency matrix based on the behavior sequence; statistically analyzing the occurrence frequency of each semantic behavior based on the prior semantic behavior to obtain a statistical sequence, and combining the time interval reweighting strategy to obtain the user's original interest distribution for the semantics; performing a reverse conversion on the user's original interest distribution for the semantics to obtain a generalized interest distribution; and predicting and obtaining a recommended item based on the generalized interest distribution. The space and time complexity of DM is reduced.

The application discloses a kind of relative radiation calibration methods and systems of remote sensing image based on quantile, it is related to remote sensing image processing technical field, the method includes: collecting satellite historical image data and separating according to spectral band;The value of each column of single-band image detector is sequentially arranged in ascending order, and quantile value is uniformly extracted to generate the quantile feature sequence of each column;The arithmetic mean value of all columns under the same quantile level is calculated, and the global reference quantile distribution is constructed;Based on least square method, the linear mapping relationship between each column quantile sequence and reference distribution is constructed, and the radiation correction gain and bias coefficient of each detector column are obtained;The original image is corrected column by column using the coefficient;The present application effectively eliminates the vertical strip noise caused by the non-uniformity of sensor response, overcomes the limitations of traditional calibration under the meteorological conditions and platform maneuvering, significantly improves the timeliness and image radiation quality of high-frequency and high-precision calibration.

The invention discloses a high-precision radar target recognition system based on orthogonal waveform design and a convolutional neural network. The high-precision radar target recognition system comprises a radar orthogonal waveform design module, a waveform generation module, a transmitting antenna module, a receiving antenna module, a waveform receiving module and a convolutional neural network module, by designing an orthogonal radar transmitting pulse string, the receiving entropy is increased, so that a radar echo pulse string received by a receiving antenna contains enough target information, and strong difference is presented between a target and clutter as well as multipath interference characteristics; therefore, the convolutional neural network can extract target key features from a radar echo pulse string received by an antenna and suppress clutter and multipath interference at the same time. According to the method, radar echoes are directly processed in the electromagnetic original domain, features are efficiently extracted, high-precision recognition is achieved, traditional multi-step signal processing is integrated into one step, the process is simplified, the hardware complexity of a receiver can be reduced possibly, and the method has great significance in improving the performance of an electronic information system.

The present invention discloses a method, system and device for underwater image target detection, including: acquiring an underwater image, preprocessing the underwater image; inputting the preprocessed underwater image into a feature extraction network for feature extraction to obtain underwater image features; sending the preprocessed underwater image into a detection model to obtain a mask of the target in the underwater image; inputting the underwater image features and the mask of the target in the underwater image into a pre-trained generative adversarial model to obtain a pseudo underwater image; sending the pseudo underwater image into the detection model to obtain the position information, mask and category of the target in the underwater image to complete target detection. The present invention can realize underwater image target detection.

The application discloses a strip ridge regression input optimization method and device, equipment, medium and product, and relates to the technical field of machine learning and data processing. The method comprises the following steps: performing feature extraction on obtained target stimulation data to obtain a plurality of original feature space data; constructing a kernel representation matrix for each original feature space data, and performing normalization processing on each kernel representation matrix; calculating similarity according to the normalized kernel matrices of each pair of original feature space data, and converting the similarity into a distance matrix between each pair of original feature space data; performing hierarchical clustering according to the distance matrix between all pairs of original feature space data to obtain a plurality of cluster-level feature spaces corresponding to feature clusters; jointly modeling based on the cluster-level feature spaces corresponding to all feature clusters and target response data; and solving a strip ridge regression input optimization model obtained by modeling to obtain a strip ridge regression input optimization result. The application can improve the calculation efficiency, prediction performance and interpretability.

This application provides a method, device, and storage medium for obtaining multi-party aggregated credit enhancement, relating to the field of data security, and is used by the i-th participating party; including: obtaining the credit enhancement query request and shared base sent by the querying party; sending the encryption index r to the querying party. i The enhanced credit query request is used to request enhanced credit information from the querying user; the encryption index of other participants sent by the querying party is obtained, and the encryption noise R of the i-th participant is obtained based on the common base and the encryption index. i , where; according to the encryption noise R i To obtain the ciphertext E of the user's plaintext security information from the i-th participant, encrypt the security information in plaintext. i The encrypted information E will be used to enhance the security information. i Send to the querying party; wherein, the encrypted information E is the enhanced information. i This invention is used by the querying party to obtain aggregated credit enhancement information, which is the sum of the plaintext credit enhancement information of the user from N participating parties. This invention enables the rapid acquisition of aggregated credit enhancement information while ensuring data security.

This invention provides a multi-channel continuous fast packet reception, sorting, and lock-free asynchronous read / write method, relating to the fields of computer communication and storage technology. The method includes: employing a dynamic expansion mechanism to continuously receive and store multi-channel data packets; classifying and sorting the stored data packets; and asynchronously reading and writing the parsed data packets in a lock-free manner. This invention generates an independent message stack for each channel, enabling continuous reception and storage of messages from multiple channels without pre-allocating excessive memory space. The speed of receiving buffer expansion is dynamically adjusted based on data flow rate. Fast classification and sorting of data packets is achieved using a hash table or sorted binary tree. A circular stack is created to sequentially store the parsed data content of each packet, updating the stack top position after data content is stored. When reading data, the starting position for reading is found using the stack top position and the number of packets read, and then the data is copied, thereby achieving lock-free asynchronous parallel data read / write.

The application discloses an ICS intrusion detection system and method fusing reinforcement learning and feature selection optimization, carries out binary coding and population initialization on data feature selection of an industrial control system (ICS) historical data set, carries out offline training through SVM-reinforcement learning, takes the accuracy obtained on a verification set as a fitness function, designs cross operation and mutation operation based on cumulative probability to update the population, and obtains an optimal feature set after iterative optimization; feature selection is carried out on an ICS real-time data set based on the optimal feature set, online intrusion detection testing is carried out on the real-time data set through support vector machine (SVM)-reinforcement learning, and thus an intrusion detection performance index is obtained. The application adopts a new mode based on SVM-reinforcement learning, and fuses intelligent optimization of optimal feature selection on this basis, and improves the intelligent design level and the precision of intrusion detection of the ICS intrusion detection system.

This invention discloses a signal detection method for an OTFS system based on an adaptive wavelet scattering convolutional neural network, comprising: converting the data symbols to be transmitted to the DD domain, and then obtaining an OTFS signal through an OTFS modulator; the OTFS signal reaching the receiver through a wireless channel; at the receiver, transforming the received signal back to the DD domain using an OTFS demodulator; extracting the real and imaginary signal matrices from the received OTFS signal, performing a three-layer wavelet scattering network transformation on these two matrices to obtain scattering feature coefficients for each layer, concatenating the scattering feature coefficients along the feature channel direction to reconstruct a high-dimensional feature matrix of the signal, and using it as the input to an adaptive wavelet convolutional neural network 2D-AWCNN to obtain the possible values ​​corresponding to each QAM symbol; and performing 16-QAM demodulation on the output of the adaptive wavelet convolutional neural network 2D-AWCNN using a demodulator to obtain the output bits used for signal detection.

The present application provides a vehicle railway flatbed detection method based on laser radar, the method comprising: obtaining two-dimensional point cloud data including a plurality of two-dimensional point data of the railway flatbed sensed by the laser radar. Determine the two-dimensional point data with a scanning angle of a preset angle as the orthographic projection point, and traverse the two-dimensional point cloud data starting from the orthographic projection point according to the angle difference of the scanning angle between adjacent two-dimensional point data, if the difference of the scanning distance of adjacent two-dimensional point data is greater than the distance threshold, determine the two-dimensional point data corresponding to the smaller scanning distance in the adjacent two-dimensional point data as the first edge point and the second edge point of the railway flatbed. Based on the trigonometric function, determine the first distance between the first edge point and the orthographic projection point, and the second distance between the second edge point and the orthographic projection point according to the scanning distance and scanning angle of the first edge point and the second edge point. Detect whether the vehicle deviates from the center of the railway flatbed according to the first distance and the second distance.

The invention relates to a semiconductor switching device (1), comprising a switchable power semiconductor (T 2 ) and at least one regulating circuit (RS 1 , R.S. 2 ), the regulation circuit includes a current sink (S 1 , S 2 ) and the current amplifier (V 1 , V 2 ). Current sink (S 1 , S 2 ) through the adjustment capacitor (C 1 ) and power semiconductors (T 2 ) is capacitively coupled to the collector / drain of the MOSFET and is designed to derive the current flowing through the regulating capacitor (C 1 ) of the current until the maximum current that can be set. Current amplifier (V 1 , V 2 ) on the input side and the current sink (S 1 , S 2 ) and the regulating capacitor (C 1 ) is connected to the end of the capacitive coupling and is connected to a positive or negative supply voltage and to a power semiconductor (T 2 ) is connected to the control electrode and is designed to amplify the current flowing through the regulating capacitor (C 1 ) of the current not flowing through the current sink (S 1 , S 2 ) and conducts the partial current into the control electrode of the power semiconductor (T2), so that the partial current in the power semiconductor (T2) is offset during the switching process. 2 ) on the collector-emitter-path or on the drain-source-path. In order to improve the semiconductor switching device (1), it is proposed that the semiconductor switching device (1) also has an additional circuit (1.1) for adaptive switching transition, which is connected to the power semiconductor (T 2 ) and the additional circuit is designed to, when the collector-emitter-path or the drain-source-path of the power semiconductor is switched from a blocked state to a conducting state, the power semiconductor (T 2 )’s collector voltage or drain voltage.

The present invention belongs to the field of tunnel advance detection, and specifically, is a tunnel advance detection imaging method and device, which can effectively process tunnel advance detection seismic data in the harsh environment of tunnel construction. The source wave field is extended to obtain the source extension wave field; seismic data on both sides of the tunnel and the face are collected; the seismic data is used as the boundary condition to extend the detector wave field to obtain the detector extension wave field; imaging condition calculation is performed on the source extension wave field and the detector extension wave field. The calculation efficiency is improved, the space occupation is reduced, and the availability and real-time performance of tunnel advance detection are guaranteed.

This invention discloses a tree-based encoding method and system, belonging to the field of data structure representation technology. The method defines the encoding structure using letters and numbers; it binds encoding relationships by inheriting the parent node's encoding as a prefix for the child node's encoding, and concatenates this prefix with the child node's unique node to form the complete child node encoding; when generating encodings for added nodes, according to the encoding generation logic, it first determines the parent node and then the sibling nodes, and generates encodings in the order of the largest sibling node by sorting the sibling nodes; based on reflection, it calls the encoding function, establishes an Entity base class with necessary attributes and a rule class with generics, and implements the Class class through reflection. <iservice>The initialization of the tree structure. The application reduces the algorithm complexity of the tree structure inquiring all child nodes, and does not need to modify other nodes when adding and deleting, and guarantees the persistence of the database.< / iservice>

This invention provides a method and apparatus for selecting combined electronic coupons, applicable to the field of artificial intelligence technology. The method includes: splitting an order corresponding to the total order amount into final sub-orders corresponding to different electronic coupon denominations based on the obtained electronic coupon denominations, the number of electronic coupons corresponding to each denomination, and the total order amount; generating combined electronic coupon selection information based on the minimum number of coupons used in the stored final sub-orders corresponding to different electronic coupon denominations and the total order amount. This method can greatly reduce time and space complexity, thereby improving the efficiency of electronic coupon selection.

This disclosure provides a data storage method and apparatus. The method includes: acquiring target data to be stored; reading a bitmap structure from a database based on a target identifier in the target data; and determining the mode of the bitmap structure. When the bitmap structure is determined to be in a variable-length mode, writing a target time from the target data into a newly added time unit in the data portion of the bitmap structure, wherein the data portion of the variable-length bitmap structure contains at least one time unit. When the bitmap structure is determined to be in a fixed-length mode, setting corresponding data bits in the data portion of the bitmap structure based on the target time in the target data, wherein the storage location of each data bit in the data portion of the fixed-length bitmap structure corresponds to a time point. This method can reduce the space occupied by data storage, thereby reducing storage costs, and improve the efficiency of data retrieval, thereby reducing query latency.

The present invention discloses a network structure and method for multi-scale super-resolution reconstruction of a single image. The network structure of the present invention comprises: a feature extraction layer, the input of which is the original image, and the output of which is connected to the input layer of a dilated convolutional neural network; a cascade expansion layer, the input of which is connected to the output layer of a dilated convolutional neural network; an aggregation layer, which is used to fuse the low-level coding information of different scales of the feature extraction layer and the decoding information of the cascade expansion layer; a reconstruction module, the input of which is the output of the cascade expansion layer, and the output of which is the reconstructed image. The present invention adopts a back-end upsampling network to perform image mapping transformations in a low-resolution space, thereby reducing the computational complexity and spatial complexity.

The invention provides an autonomous navigation laser radar-camera fast fusion target detection method and device and a storage medium, and belongs to the field of multi-modal target detection, and the method comprises the steps: obtaining a laser radar point cloud and a camera image; obtaining point cloud BEV features according to the laser radar point cloud; performing feature extraction on the camera image to obtain a first image feature, and converting the first image feature into an image BEV feature by using a dynamic view angle converter; fusing the point cloud BEV features and the image BEV features to obtain fused BEV features; and a final target detection result is obtained. According to the invention, the operand of BEVPooling is reduced through the dynamic view angle converter, and the time and space complexity of the network structure is reduced, so that the reasoning time of the laser radar-camera fusion perception neural network is reduced, and the real-time performance of target detection is improved.

The invention provides an efficient space optimization dynamic programming algorithm for large-scale sequence matching. The efficient space optimization dynamic programming algorithm comprises the following steps: inputting and preprocessing; carrying out sequence blocking; initializing a global score matrix; initializing a one-dimensional dynamic programming array, performing dynamic programming calculation in a block, calculating a matching score, performing pruning judgment, updating the one-dimensional dynamic programming array, and updating a global score matrix; according to the sequence of the blocks, combining scores, processing matching dependence among the blocks, and calculating a global optimal score; and outputting a result. According to the method, the space complexity can be reduced from O (mn) to O (n), the calculation efficiency is remarkably improved, the requirement for calculation resources is greatly reduced on the premise that the matching precision is guaranteed, and the method is suitable for DNA sequence alignment, text similarity calculation and other scenes needing to process super-long sequences.

The invention provides a heterogeneous graph node classification method based on a double-layer variational auto-encoder, which effectively improves the node classification effect of a heterogeneous graph under the condition of feature deficiency and inaccuracy. The method comprises the following steps: firstly, determining the meaning of a heterogeneous graph and the definition of inaccurate and missing features, and initializing the missing features through one-hot coding; secondly, mapping the high-dimensional features into a low-dimensional space; secondly, representing the features in a node-level mode and an attribute-level mode; in the encoder, a double-layer variational auto-encoder is used for generating embedding for the features of the two modes; in the decoder, through an inner product form, the embedding of the two modes is restored respectively; then, aligning the restored attribute-level representation with the initial feature through a multi-layer perceptron; and finally, applying the generated features to a downstream classification task. According to the method, the cost of time, space and manual annotation is remarkably reduced, good generalization ability is achieved, and the classification effect under the condition of feature missing and inaccuracy in the heterogeneous graph is improved.

The invention relates to a target alarm method, in particular to a target alarm method for a refrigeration type infrared detector image. Comprising the following steps: S1, constructing a target alarm background library; s2, target restoration parameters are calculated, and a target tracking restoration model is established; s3, filtering out an explosion alarm point through the threshold value of the central point and the threshold value near the central point; s4, filtering the explosion alarm points in combination with the intensity proportional relation near the central point to obtain alarm points; s5, updating the alarm points into a historical alarm point tracking library, and listing the alarm points into a pre-alarm queue; s6, judging whether the alarm point is a new outbreak alarm point or not; s7, duplicate checking is carried out on the newly-exploded alarm point and the pre-alarm queue, and whether the newly-exploded alarm point is re-exploded of the historically-exploded alarm point or not is determined; and S8, if the number of reproduction times recorded by the counter exceeds 20, the refrigeration type infrared detector gives an alarm. The method can be applied to alarm of any target; the false alarm rate can be reduced, and the alarm accuracy can be improved.

The application discloses a REC signal detection method based on OTFS modulation and deep learning, and the implementation steps are as follows: OTFS demodulation is performed on a REC signal; a target state information in the REC signal is detected by using an approximate message passing algorithm of sparse Bayesian learning; a channel is sequentially subjected to equalization operation and preprocessing by using the state information; a conditional generative adversarial network constructed is trained by using the preprocessed received signal; and the preprocessed received signal is sent into the trained conditional generative adversarial network to perform communication decision, so that a recovered REC communication signal transmitted by a UAV label of our side is obtained. The approximate message passing algorithm of sparse Bayesian learning is used to reduce the time complexity and space complexity of an algorithm in a target detection process, and the special properties of a channel matrix are used to reduce the inverse operation in an equalization process, so that the calculation complexity is further reduced, and the system operation efficiency is improved. The deep learning model is introduced into the communication receiver design, and the communication reliability can still be maintained when a high-speed moving target is used.

The present invention provides a virus propagation control method based on bounded percolation-greedy algorithm. First, extract the target network information to understand the target network nodes and edge attributes; then, based on the percolation theory, continuously occupy specific nodes on the target network to achieve the reverse process of removing key nodes; each time occupy the node that minimizes the objective function, limit the size of the maximum connected component and the degree of the remaining network nodes to control the virus propagation; set the key indicator boundary value, when the key indicators of all candidate nodes exceed the boundary value, update the boundary value to occupy more candidate nodes; finally, use order parameters and network resilience to represent the effect of controlling virus propagation. The present invention can achieve fast and efficient decomposition of large networks, thereby timely controlling virus propagation and reducing the losses caused by it.

This invention discloses a method, device, and medium for security assessment of number-theoretical transformation modules based on bypass analysis. The method includes: collecting first and second bypass curves generated by the target device during number-theoretical transformation and subsequent polynomial multiplication, and constructing a leakage template accordingly; constructing a number-theoretical transformation factor graph based on the inter-layer mapping relationship of butterfly operations from layer i to L of the number-theoretical transformation; obtaining first and second probability distributions through template matching using the first and second bypass curves respectively; merging the two probability distributions as prior conditions to update and iterate the number-theoretical transformation factor graph; constructing a lookup table for the layer i butterfly operation; reversing the variable node values ​​of the preceding butterfly operation based on the updated factor graph, and recursively backtracking to the first layer input to recover the key; and finally assessing the security of the number-theoretical transformation module based on the recovered key. This method combines number-theoretical transformation factor graphs, probability inference, and lookup table reverse reasoning, improving assessment efficiency and accuracy.

The application discloses a lithium ion battery state of charge estimation method based on an Informer, which comprises the following steps: collecting battery sample data, constructing an Informer network model, processing the battery sample data by using the Informer network model, and obtaining a state of charge estimation value of the lithium ion battery. The Informer network model used in the application reduces the point product calculation of the attention mechanism and improves the encoder-decoder architecture, thereby reducing the calculation complexity and spatial complexity, improving the model training and calculation speed, and to some extent, solving the cumulative error problem existing in the recurrent neural network, achieving higher SOC estimation accuracy, showing superiority in MAE and RMSE error indicators, and better solving the estimation difficulty under low temperature conditions and the problem of insufficient temperature expansion capability. The application is widely applied in the technical field of batteries.