CN115544309B - Improved nearest neighbor data interconnection method based on GCN - Google Patents
Improved nearest neighbor data interconnection method based on GCN Download PDFInfo
- Publication number
- CN115544309B CN115544309B CN202211157579.3A CN202211157579A CN115544309B CN 115544309 B CN115544309 B CN 115544309B CN 202211157579 A CN202211157579 A CN 202211157579A CN 115544309 B CN115544309 B CN 115544309B
- Authority
- CN
- China
- Prior art keywords
- gcn
- node
- nodes
- prediction model
- matrix
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000005259 measurement Methods 0.000 claims abstract description 60
- 238000011156 evaluation Methods 0.000 claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims description 42
- 239000013598 vector Substances 0.000 claims description 20
- 238000012549 training Methods 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 3
- 238000012512 characterization method Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 6
- 238000000605 extraction Methods 0.000 abstract description 5
- 230000009897 systematic effect Effects 0.000 abstract description 5
- 238000013528 artificial neural network Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 238000011161 development Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 101100134058 Caenorhabditis elegans nth-1 gene Proteins 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 235000006629 Prosopis spicigera Nutrition 0.000 description 1
- 240000000037 Prosopis spicigera Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9024—Graphs; Linked lists
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention discloses an improved nearest neighbor data interconnection method based on GCN, which comprises the following steps: step 1, obtaining sensor measurement values, wherein each sensor measurement value corresponds to one node to establish a graph network structure, and a plurality of sensors correspondingly obtain a plurality of graph networks; step 2, constructing a GCN prediction model, and extracting characteristic information of the network structure nodes of the graph by using the GCN prediction model; step 3, calculating node distance based on node characteristics output by the GCN prediction model; and 4, solving an interconnection result by taking the characteristic distance between the nodes output by the GCN prediction model as an interconnection evaluation standard. The invention mainly solves the deep feature extraction problem of multi-sensor measurement, utilizes GCN to automatically learn deep feature information and unchanged topology information between measurement from original measurement data, and improves the data interconnection effect of the traditional nearest neighbor method under the scenes of dense clutter, random error and systematic deviation.
Description
Technical Field
The invention belongs to the data processing technology, and particularly relates to an improved nearest neighbor data interconnection method based on GCN.
Background
The characteristics of diversity, density, maneuver, stealth, fine guidance and the like of the offshore targets are increasingly highlighted, the diversity development of the offshore sensing means, the continuous expansion of the detection range and the phenomena of redundancy, blurring, conflict and the like of multisource sensing data information are more serious, and the phenomena of missing, misplacement, multiple heels and the like of the targets are frequently generated. The offshore target perception data is affected by sea clutter, atmospheric waveguide, platform errors and other factors, and the traditional target tracking technology utilizes ideas of probability theory, fuzzy set or random set and other attempts to accurately model, so that unavoidable and non-negligible errors exist, and actual engineering requirements are difficult to meet.
In recent years, with the rapid development of deep neural networks, deep learning technology is becoming a research hotspot in the field of offshore target tracking in complex environments. However, the existing target tracking depth neural network is simple in design, and cannot overcome the deviation of the traditional target tracking model mechanism, so that the bottleneck problem of point-point/point-navigation data interconnection exists when the existing target tracking depth neural network is applied to actual engineering.
The nearest neighbor method is used as a basic method in the multi-sensor multi-target data interconnection method, is suitable for static target measurement data interconnection in a sparse clutter environment, but for dense clutter scenes, scenes with a large amount of measurement errors and systematic deviation, the correlation accuracy is greatly reduced, and the application scenes of the algorithm are severely limited.
Based on the above, an improved nearest neighbor data interconnection method based on GCN (Graph Convolutional Network, graph convolutional neural network) is provided.
Disclosure of Invention
Aiming at the technical problems, the invention provides an improved nearest neighbor data interconnection method based on GCN.
The technical scheme for solving the technical problems is as follows:
an improved nearest neighbor data interconnection method based on GCN comprises the following steps:
step 1, obtaining sensor measurement values, wherein each sensor measurement value corresponds to one node to establish a graph network structure, and a plurality of sensors correspondingly obtain a plurality of graph networks;
step 2, constructing a GCN prediction model, and extracting characteristic information of the network structure nodes of the graph by using the GCN prediction model;
step 3, calculating node distance based on node characteristics output by the GCN prediction model;
and 4, solving an interconnection result by taking the characteristic distance between the nodes output by the GCN prediction model as an interconnection evaluation standard.
Further, the step 1 of obtaining the sensor measurement value, and constructing a graph network structure according to the sensor measurement value includes the following steps:
acquiring sensor measurement values, wherein each sensor measurement value corresponds to one node, calculating the relation between any two sensor measurement values, namely the relation between the nodes, and constructing a relation edge according to the relation of the nodes;
and constructing a graph network structure according to the nodes and the associated nodes and the relation edges.
Further, calculating a relation between any two sensor measurement values, namely a relation between nodes, and constructing a relation edge according to the relation of the nodes, wherein the relation edge comprises the following steps:
calculating the distance d between any two measurement values n,ij If the distance meets the relation of the formula (1), building an edge between two nodes corresponding to the two measurement values, otherwise, not building the edge;
wherein d n,ij Representing a distance between an ith measurement value and a jth measurement value of an nth sensor; z n,i 、z n,j Respectively representing the ith measurement value and the jth measurement value of the nth sensorMeasuring a value; t represents a transpose; gamma is the threshold for edge building.
Further, before constructing the GCN prediction model, adopting a matrix form for the nodes and the relation edges;
given a total of N nodes in the graph network structure, the input characteristics of the nodes are as followsWherein z is n,i ∈R D D is the input feature number of each node, and l layers of all node feature vectors are +.>Expressed in matrix form, get-> For an NxD dimension feature matrix, each row in the matrix corresponds to the feature characterization of each node;
using an N-dimensional adjacency matrix, A represents the edge relationship between nodes, where A ij =1 means node i is connected to node j, there is an edge, otherwise a ij =0。
Further, in the step 2, a two-layer GCN network structure is used for constructing a GCN prediction model, the GCN prediction model is input into a node feature matrix and an adjacent matrix, and the propagation modes between layers are as follows:
wherein,i is an identity matrix; />Is->Degree matrix of->Normalizing the Laplace matrix; />For the first layer feature matrix, the feature vector of all nodes of the nth sensor on the first layer is represented, l=0, 1 for the input layer is +.>W (l) Representing the weight of the first layer, which weight is shared for different sensors n; sigma is a nonlinear activation function.
Further, in the step 3, the node distance is calculated based on the node characteristics output by the GCN prediction model, and the method includes the following steps:
node characteristic matrix based on GCN prediction model outputObtaining a feature vector h of each node n,i Calculating the characteristic distance between any two nodes in different graphs:
wherein h is n-1,i A feature vector h representing an ith node in an nth-1 sensor based on GCN output n,j A feature vector representing a jth node in an nth sensor based on the GCN output;representing the distance of the output features of the GCN predictive model between node i and node j.
Further, in the step 4, the data interconnection result that minimizes the global feature distance is solved as follows:
wherein ρ is ij Is a binary data interconnection variable, expressed by 0 or 1, and if the ith node is interconnected with the jth node, ρ is ij =1; otherwise ρ ij =0;N n-1 、N n Respectively represents the common N of the N-1 th sensor n-1 N is shared between the node and the nth sensor n And each node.
Further, the training of the GCN prediction model comprises the following steps:
interconnecting the obtained output result rho with the real data of the target to obtain a result rho * And taking the matrix norms to establish a loss function, and training a GCN prediction model through back propagation.
Compared with the prior art, the invention has the following technical effects:
1. constructing a sensor measurement value into a graph network structure, so that the GCN can extract depth characteristic information and topology characteristic information between target measurements;
2. the similarity of homologous target measurement is measured by using the distance between the features, so that the data interconnection effect under the scenes of dense clutter, random errors and systematic deviation is improved;
3. and constructing a loss function by using the output result of the optimal distribution model and the real data interconnection result of the target, so as to realize network training.
Drawings
Fig. 1 is a schematic flow chart of an improved nearest neighbor data interconnection method based on GCN according to the present invention.
Detailed Description
The principles and features of the present invention are described below with reference to the drawings, the examples are illustrated for the purpose of illustrating the invention and are not to be construed as limiting the scope of the invention.
In order to improve the application range and the association effect of the nearest neighbor method in multi-sensor multi-target data interconnection, the scheme of the invention introduces the GCN method to improve the traditional nearest neighbor method, and utilizes the GCN to automatically learn deep characteristic information and unchanged topology information between measurement from original measurement data, thereby remarkably improving the data interconnection effect under the scenes of dense clutter, random errors and systematic deviation.
Fig. 1 is a schematic flow chart of an improved nearest neighbor data interconnection method based on GCN according to the present invention. Referring to fig. 1, an improved nearest neighbor data interconnection method based on GCN includes the steps of: step 1, obtaining sensor measurement values, wherein each sensor measurement value corresponds to one node to establish a graph network structure, and a plurality of sensors correspondingly obtain a plurality of graph networks; step 2, constructing a GCN prediction model, and extracting characteristic information of the network structure nodes of the graph by using the GCN prediction model; step 3, calculating node distance based on node characteristics output by the GCN prediction model; and 4, solving an interconnection result by taking the characteristic distance between the nodes output by the GCN prediction model as an interconnection evaluation standard.
The following detailed development of each step is performed:
step 1, obtaining sensor measurement values, wherein each sensor measurement value corresponds to one node to establish a graph network structure, and a plurality of sensors correspondingly obtain a plurality of graph networks.
Acquiring sensor measurement values, wherein each sensor measurement value corresponds to one node, calculating the relation between any two sensor measurement values, namely the relation between the nodes, and constructing a relation edge according to the relation of the nodes;
calculating the relation between any two sensor measurement values, namely the relation between nodes, and constructing a relation edge according to the relation of the nodes, wherein the relation edge comprises the following steps:
calculating the distance d between any two measurement values n,ij If the distance meets the relation of the formula (1), building an edge between two nodes corresponding to the two measurement values, otherwise, not building the edge;
wherein d n,ij Representing a distance between an ith measurement value and a jth measurement value of an nth sensor; z n,i 、z n,j Respectively representing the ith measurement of the nth sensorA value and a j-th measurement value; t represents a transpose; gamma is the threshold for edge building.
And constructing a graph network structure according to the nodes and the associated nodes and the relation edges. The Graph structure is denoted as g= (V, E), where V represents Node (Node) and E represents Edge (Edge), so as to obtain local topology information between the measurements.
And 2, constructing a GCN prediction model, and extracting characteristic information of the network structure nodes of the graph by using the GCN prediction model.
The GCN is a basic convolutional graph neural network, which belongs to a model in deep learning, and is a feature extractor, the object of which is graph data, which acts virtually as a convolutional neural CNN. The GCN directly acts on the neural network on the graph structure, and feature vectors representing the relation between the nodes and the adjacent nodes are obtained through feature extraction, so that preparation is made for node interconnection analysis.
Let N nodes in a given Graph, the input characteristics of the nodes are as followsWherein z is n,i ∈R D D is the input feature number (including multidimensional features such as position, speed, course, attribute and type) of each node, and all node feature vectors of layer I are +.>Expressed in matrix form, get-> For an NxD dimension feature matrix, each row in the matrix corresponds to the feature characterization of each node; using an N-dimensional Adjacency Matrix (Adjacent Matrix) A to represent the side relationships between nodes, where A ij =1 means node i is connected to node j, there is an edge, otherwise a ij =0。
In this embodiment, a two-layer GCN network structure is used to extract feature information of the graph structure, the input of the network is a feature matrix and an adjacent matrix, and the propagation modes between the layers are as follows:
wherein,i is an identity matrix; />Is->Degree matrix of->Normalizing the Laplace matrix; />For the first layer feature matrix, the feature vector of all nodes of the nth sensor on the first layer is represented, l=0, 1 for the input layer is +.>W (l) Representing the weight of the first layer, which weight is shared for different sensors n; sigma is a nonlinear activation function.
In contrast to the structured feature vector input by the conventional prediction method, the prediction model is input into an unstructured graph, the traditional deep convolution network can perform feature extraction on structured data (images, voices, sequences and the like), but the graph data has the characteristics of unstructured, disordered and random, the construction nodes and the relation numbers of the embodiment are not fixed, the expression form is more flexible, the characteristic matrix which cannot be aligned into a fixed scale is subjected to feature extraction by using the traditional convolution network, and therefore, the convolution network for the graph is required to be adopted, the dependency relationship on a two-dimensional structure is not limited when the graph convolution network captures the features, and the richer associated node information can be aggregated.
And 3, calculating the node distance based on the node characteristics output by the GCN prediction model.
Node characteristic matrix based on GCN prediction model outputObtaining a feature vector h of each node n,i Calculating the characteristic distance between any two nodes in different graphs:
wherein h is n-1,i A feature vector h representing an ith node in an nth-1 sensor based on GCN output n,j A feature vector representing a jth node in an nth sensor based on the GCN output;representing the distance of the output features of the GCN predictive model between node i and node j.
Specifically, taking two sensors as an example, according to the constructed graph network structure, a node characteristic matrix based on GCN output is taken as an exampleFeature vector h of each node can be obtained n,i Calculating the characteristic distance between any two nodes in different graphs:
wherein h is 1,i 、h 2,j Respectively representing the characteristic vector of the ith node in the 1 st sensor and the characteristic vector of the jth node in the 2 nd sensor which are output based on the GCN;representing GCN pre-emption between node i and node jAnd outputting the characteristic distance by the measuring model.
And 4, solving an interconnection result by taking the characteristic distance between the nodes output by the GCN prediction model as an interconnection evaluation standard.
In order to obtain the data interconnection relation among different sensor measurement sets, the feature distance among nodes output by the GCN prediction model is used as an interconnection evaluation standard, the data interconnection result which enables the global feature distance to be minimum is solved, and an optimal distribution model is established as follows:
wherein ρ is ij Is a binary data interconnection variable, represented by 0 or 1, ρ is the case if the i-th node is interconnected with the j-th node ij =1; otherwise ρ ij =0;N n-1 、N n Respectively represents the common N of the N-1 th sensor n-1 N is shared between the node and the nth sensor n And each node.
Specifically, taking two sensors as an example, in order to obtain a data interconnection relationship between two sensor measurement sets, a feature distance between nodes output by a GCN prediction model is taken as an interconnection evaluation standard, a data interconnection result which minimizes a global feature distance is solved, and an optimal distribution model is established as follows:
wherein ρ is ij Is a binary data interconnection variable, represented by 0 or 1, ρ is the case if the i-th node is interconnected with the j-th node ij =1; otherwise ρ ij =0。N 1 、N 2 Respectively represent the common N in the 1 st sensor 1 N is shared between the node and the 2 nd sensor 2 And each node.
Training of a GCN predictive model, comprising the steps of: interconnecting the obtained output result rho with the real data of the target to obtain a result rho * Performing difference, taking matrix norm and establishing loss functionThe GCN predictive model is trained by back propagation.
Specifically, the output result ρ of the obtained optimal distribution model is interconnected with the data of the target reality to obtain the result ρ * Taking the matrix norm to establish the loss function is as follows:
Loss=||ρ-ρ * || (5)
after the Loss is calculated, the GCN network model is trained by back propagation.
The invention mainly solves the deep feature extraction problem of multi-sensor measurement, utilizes GCN to automatically learn deep feature information and unchanged topology information between measurement from original measurement data, and improves the data interconnection effect of the traditional nearest neighbor method under the scenes of dense clutter, random error and systematic deviation.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (7)
1. An improved nearest neighbor data interconnection method based on GCN is characterized by comprising the following steps:
step 1, obtaining sensor measurement values, wherein each sensor measurement value corresponds to one node to establish a graph network structure, and a plurality of sensors correspondingly obtain a plurality of graph network structures;
step 2, constructing a GCN prediction model, and extracting characteristic information of the network structure nodes of the graph by using the GCN prediction model;
step 3, calculating node distance based on node characteristics output by the GCN prediction model;
step 4, the characteristic distance between the nodes output by the GCN prediction model is used as an interconnection evaluation standard, and an interconnection result is solved;
in the step 4, the data interconnection result which minimizes the global feature distance is solved as follows:
wherein ρ is ij Is a binary data interconnection variable, expressed by 0 or 1, and if the ith node is interconnected with the jth node, ρ is ij =1; otherwise ρ ij =0;N n-1 、N n Respectively represents the common N of the N-1 th sensor n-1 N is shared between the node and the nth sensor n And each node.
2. The improved nearest neighbor data interconnection method based on GCN according to claim 1, wherein the step 1 of obtaining sensor measurement values, each sensor measurement value corresponding to a node, establishes a graph network structure, includes the steps of:
acquiring sensor measurement values, wherein each sensor measurement value corresponds to one node, calculating the relation between any two sensor measurement values, namely the relation between the nodes, and constructing a relation edge according to the relation of the nodes;
and constructing a graph network structure according to the nodes and the associated nodes and the relation edges.
3. The improved nearest neighbor data interconnection method based on GCN according to claim 2, wherein calculating a relationship between any two sensor measurement values, namely, a relationship between nodes, and constructing a relationship edge according to the relationship of the nodes, comprises the steps of:
calculating the distance d between any two measurement values n,ij If the distance meets the relation of the formula (1), building an edge between two nodes corresponding to the two measurement values, otherwise, not building the edge;
d n,ij =(z n,i -z n,j ) T (z n,i -z n,j )≤γ,n=1,2 (1)
wherein d n,ij Representing a distance between an ith measurement value and a jth measurement value of an nth sensor; z n,i 、z n,j Respectively representing an ith measurement value and a jth measurement value of an nth sensor; t represents a transpose; gamma is the threshold for edge building.
4. The improved nearest neighbor data interconnection method based on GCN according to claim 1, wherein nodes and relation edges are in a matrix form before a GCN prediction model is built;
given a total of N nodes in the graph network structure, the input characteristics of the nodes are as followsWherein z is n,i ∈R D D is the input feature number of each node, and l layers of all node feature vectors are +.>Expressed in matrix form, get-> For an NxD dimension feature matrix, each row in the matrix corresponds to the feature characterization of each node;
using an N-dimensional adjacency matrix, A represents the edge relationship between nodes, where A ij =1 means node i is connected to node j, there is an edge, otherwise a ij =0。
5. The improved nearest neighbor data interconnection method based on GCN according to claim 4, wherein in said step 2, a two-layer GCN network structure is used for constructing a GCN prediction model, the GCN prediction model is input into a node feature matrix and an adjacent matrix, and the propagation manner between layers is as follows:
wherein,i is an identity matrix; />Is->Degree matrix of->Normalizing the Laplace matrix;for the first layer feature matrix, the feature vector of all nodes of the nth sensor on the first layer is represented, l=0, 1 for the input layer is +.>W (l) Representing the weight of the first layer, which weight is shared for different sensors n; sigma is a nonlinear activation function.
6. The improved nearest neighbor data interconnection method based on GCN according to claim 5, wherein the calculating of the node distance based on the node characteristics output by the GCN prediction model in step 3 includes the steps of:
node characteristic matrix based on GCN prediction model outputObtaining a feature vector h of each node n,i Calculating the characteristic distance between any two nodes in different graphs:
wherein h is n-1,i Representing the ith sensor of the n-1 th sensor based on GCN outputFeature vector of node, h n,j A feature vector representing a jth node in an nth sensor based on the GCN output;representing the distance of the output features of the GCN predictive model between node i and node j.
7. The improved nearest neighbor data interconnection method based on GCN according to claim 1, wherein training of GCN predictive model comprises the steps of:
interconnecting the obtained output result rho with the real data of the target to obtain a result rho * And taking the matrix norms to establish a loss function, and training a GCN prediction model through back propagation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211157579.3A CN115544309B (en) | 2022-09-22 | 2022-09-22 | Improved nearest neighbor data interconnection method based on GCN |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211157579.3A CN115544309B (en) | 2022-09-22 | 2022-09-22 | Improved nearest neighbor data interconnection method based on GCN |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115544309A CN115544309A (en) | 2022-12-30 |
CN115544309B true CN115544309B (en) | 2024-03-19 |
Family
ID=84728790
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211157579.3A Active CN115544309B (en) | 2022-09-22 | 2022-09-22 | Improved nearest neighbor data interconnection method based on GCN |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115544309B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113468288A (en) * | 2021-07-23 | 2021-10-01 | 平安国际智慧城市科技股份有限公司 | Content extraction method of text courseware based on artificial intelligence and related equipment |
CN113591380A (en) * | 2021-07-28 | 2021-11-02 | 浙江大学 | Traffic flow prediction method, medium and equipment based on graph Gaussian process |
CN114418174A (en) * | 2021-12-13 | 2022-04-29 | 国网陕西省电力公司电力科学研究院 | Electric vehicle charging load prediction method |
WO2022134576A1 (en) * | 2020-12-23 | 2022-06-30 | 深圳壹账通智能科技有限公司 | Infrared video timing behavior positioning method, apparatus and device, and storage medium |
-
2022
- 2022-09-22 CN CN202211157579.3A patent/CN115544309B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2022134576A1 (en) * | 2020-12-23 | 2022-06-30 | 深圳壹账通智能科技有限公司 | Infrared video timing behavior positioning method, apparatus and device, and storage medium |
CN113468288A (en) * | 2021-07-23 | 2021-10-01 | 平安国际智慧城市科技股份有限公司 | Content extraction method of text courseware based on artificial intelligence and related equipment |
CN113591380A (en) * | 2021-07-28 | 2021-11-02 | 浙江大学 | Traffic flow prediction method, medium and equipment based on graph Gaussian process |
CN114418174A (en) * | 2021-12-13 | 2022-04-29 | 国网陕西省电力公司电力科学研究院 | Electric vehicle charging load prediction method |
Non-Patent Citations (2)
Title |
---|
" Network Link Connectivity Prediction Based on GCN and Differentiable Pooling Model";Naixue Kuang 等;《2022 IEEE 14th International Conference on Advanced Infocomm Technology (ICAIT)》;20220711;1-6 * |
党吉圣 ; 杨军 ; .多特征融合的三维模型识别与分割.西安电子科技大学学报.2020,(第04期),153-161. * |
Also Published As
Publication number | Publication date |
---|---|
CN115544309A (en) | 2022-12-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110245709B (en) | 3D point cloud data semantic segmentation method based on deep learning and self-attention | |
CN111462282B (en) | Scene graph generation method | |
CN110705406B (en) | Face beauty prediction method and device based on transfer learning resistance | |
CN110135459B (en) | Zero sample classification method based on double-triple depth measurement learning network | |
CN111126134B (en) | Radar radiation source deep learning identification method based on non-fingerprint signal eliminator | |
CN110059616A (en) | Pedestrian's weight identification model optimization method based on fusion loss function | |
CN111340076B (en) | Zero sample identification method for unknown mode of radar target of new system | |
US20200125930A1 (en) | Artificial neural network and method of training an artificial neural network with epigenetic neurogenesis | |
CN115730509A (en) | Spacecraft in-cabin temperature field reconstruction task research benchmark method based on machine learning | |
JP6892606B2 (en) | Positioning device, position identification method and computer program | |
CN111624522A (en) | Ant colony optimization-based RBF neural network control transformer fault diagnosis method | |
CN111539422A (en) | Flight target cooperative identification method based on fast RCNN | |
CN112364747A (en) | Target detection method under limited sample | |
CN116580322A (en) | Unmanned aerial vehicle infrared small target detection method under ground background | |
CN115544309B (en) | Improved nearest neighbor data interconnection method based on GCN | |
CN111862260B (en) | Cross-domain dual generation type countermeasure network-based bias eliminating method and device | |
CN111883265A (en) | Target state estimation method applied to fire control system | |
Dai et al. | Clustering of DOA data in radar pulse based on SOFM and CDbw | |
CN115423847A (en) | Twin multi-modal target tracking method based on Transformer | |
CN114819325A (en) | Score prediction method, system, device and storage medium based on graph neural network | |
CN115510597B (en) | Multi-sensor data interconnection method based on generalized S-dimension distribution under GNN framework | |
CN114580285B (en) | Hyperbolic system model reduction method based on CAE network | |
CN116894180B (en) | Product manufacturing quality prediction method based on different composition attention network | |
CN115512221B (en) | GNN-based synchronous track robustness association method | |
CN117077774A (en) | Incomplete situation information reasoning and supplementing method and model training method thereof |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |