CN106886023A - A kind of Radar Echo Extrapolation method based on dynamic convolutional neural networks - Google Patents
A kind of Radar Echo Extrapolation method based on dynamic convolutional neural networks Download PDFInfo
- Publication number
- CN106886023A CN106886023A CN201710110183.6A CN201710110183A CN106886023A CN 106886023 A CN106886023 A CN 106886023A CN 201710110183 A CN201710110183 A CN 201710110183A CN 106886023 A CN106886023 A CN 106886023A
- Authority
- CN
- China
- Prior art keywords
- layer
- image
- error term
- output characteristic
- convolution
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/95—Radar or analogous systems specially adapted for specific applications for meteorological use
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Abstract
The invention discloses a kind of Radar Echo Extrapolation method based on dynamic convolutional neural networks, including:Offline convolution neural metwork training:To the training image collection for giving, training sample set is obtained by data prediction, the dynamic convolution neural network model of initialization, and dynamic convolutional neural networks are trained using training sample set, calculating output valve, the process of back-propagating renewal network parameter by network propagated forward restrains dynamic convolutional neural networks.Online Radar Echo Extrapolation:Test chart image set is converted into by test sample collection by data prediction, tested using the trained dynamic convolutional neural networks of test sample set pair, the probability vector phase convolution that will be obtained in last width radar return image in input image sequence and network propagated forward, the Radar Echo Extrapolation image predicted.
Description
Technical field
The invention belongs to surface weather observation technical field in Atmospheric Survey, more particularly to one kind is based on dynamic convolutional Neural
The Radar Echo Extrapolation method of network.
Background technology
Nowcasting refers mainly to the weather forecast of the high-spatial and temporal resolution of 0~3 hour, and Main Prediction object includes strong drop
The diastrous weathers such as water, strong wind, hail.At present, many forecast systems all use Numerical Prediction Models, but due to numerical forecast
In the presence of return slow (spin-up) has been forecast, its Nowcasting is limited in one's ability.New Generation Doppler Weather Radar has very high
Sensitivity and resolution ratio, the spatial resolution of its data information can reach 200~1000m, and temporal resolution can reach 2
~15min.Additionally, Doppler radar also has rational mode of operation, comprehensive condition monitoring and fault warning, advanced
Real-time Calibration System and abundant Radar meteorology product algorithm, the reliability of Nowcasting can be greatly improved.Nowadays,
New Generation Doppler Weather Radar has become one of most effective instrument of nowcasting, is faced using Doppler radar
Nearly forecast is based primarily upon Radar Echo Extrapolation technology, i.e., according to current time radar observation result, thus it is speculated that radar return future
Position and intensity, to realize the track prediction to strong convection system.
Traditional Radar Echo Extrapolation method is centroid tracking method and the cross-correlation technique based on maximum correlation coefficient
(Tracking Radar Echoes by Correlation, TREC), but all there is certain deficiency, barycenter in conventional method
Tracing is only applicable to echo compared with strong, the less storm monomer of scope, and the forecast for precipitation on a large scale is unreliable;TREC is general
Echo is considered as linear change, and echo change is increasingly complex in reality, while such method is easily by vector field
Unordered vector disturbance.Additionally, existing method is low to the utilization rate of Radar Data, and history Radar Data includes local weather system
The key character of system change, with researching value very high.
To improve the ageing of Radar Echo Extrapolation, and the change of radar return is studied from substantial amounts of history Radar Data
Rule, machine learning method is introduced into Radar Echo Extrapolation.Convolutional neural networks (Convolutional Neural
Network, CNN) as the important branch of deep learning, it is widely used in the fields such as image procossing, pattern-recognition.The network is most
It is using local connection, weights are shared, the method for down-sampling, deformation, translation and upset tool to input picture the characteristics of big
There is stronger adaptability.For the strong temporal correlation existed between radar return image, dynamic convolution of the design based on input
Neutral net, the network can dynamically change weighting parameter according to the radar echo map of input, and then predict extrapolated image.Profit
Dynamic convolutional neural networks are trained with history Radar Data, network is more fully extracted echo character, study echo change
Rule, for improving Radar Echo Extrapolation accuracy, optimization nowcasting effect is significant.
The content of the invention
Goal of the invention:When the technical problems to be solved by the invention are directed to the extrapolation of existing Radar Echo Extrapolation method
Effect is short, not enough to Radar Data utilization rate, it is proposed that a kind of Radar Echo Extrapolation method based on dynamic convolutional neural networks, real
Now to the plane high such as radar echo intensity display CAPPI (Constant AltitudePlan Position Indicator,
CAPPI) outside forecast of image, comprises the following steps:
Step 1, trains offline convolutional neural networks:Input training image collection, data prediction is carried out to training image collection,
Training sample set, the dynamic convolution neural network structure of design are obtained, and initializes network training parameter;Assembled for training using training sample
Practice dynamic convolutional neural networks, the orderly image sequence of input obtains width prediction by dynamic convolutional neural networks propagated forward
Image, calculates the error between prognostic chart picture and control label, updates the weighting parameter of network by backpropagation and biasing is joined
Number, repeats this process until reaching training termination condition, obtains convergent dynamic convolutional neural networks;
Step 2, online Radar Echo Extrapolation:Input test image set, data prediction is carried out to test chart image set, is obtained
Test sample collection, then will in test sample collection input step 1 obtain dynamic convolutional neural networks in, by before network to biography
Calculating probability vector is broadcast, and last width radar return image in input image sequence is mutually rolled up with the probability vector for obtaining
Product, the Radar Echo Extrapolation image predicted.
Step 1 of the present invention is comprised the following steps:
Step 1-1, data prediction:Input training image collection, specification is carried out to every piece image that training image is concentrated
Change is processed, and 280 × 280 floating number image will be converted into per piece image, floating number image collection is obtained, to floating number image
Set is divided, training sample set of the construction comprising TrainsetSize group samples;
Step 1-2, the dynamic convolutional neural networks of initialization:The dynamic convolution neural network structure of design, is configured to generation
The sub-network of probability vector, reconstructs the probabilistic forecasting layer for image extrapolation, for the offline neural metwork training stage provides dynamic
State convolutional neural networks initialization model;
Step 1-3, the dynamic convolution Neural Network Training Parameter of initialization:E-learning rate λ=0.0001 is made,
The sample size BatchSize=10 that training stage is input into every time, most large quantities of frequency of training of training sample setCurrently criticize frequency of training BatchNum=1, the maximum iteration of network training
IterationMax=40, current iteration number of times IterationNum=1;
Step 1-4, reads training sample:By the way of batch training, the training sample obtained from step 1-1 is trained every time
Concentrate and read BatchSize group training samples, every group of training sample is { x1,x2,x3,x4, y }, altogether comprising 5 width images, wherein
{x1,x2,x3,x4Used as input image sequence, y is corresponding control label;
Step 1-5, propagated forward:The input image sequence feature that extraction step 1-4 is obtained in a sub-network, obtains level
Probability vector HPV and vertical probability vector VPV;Probabilistic forecasting layer in, by the last piece image in input image sequence according to
It is secondary with VPV, HPV phase convolution, obtain the output prognostic chart picture of propagated forward;
Step 1-6, backpropagation:The error term of probability vector is reversely tried to achieve in probabilistic forecasting layer, according to probability vector
Error term from rear to the preceding error term for successively calculating each Internet in subnet network layers, and then calculate error in each Internet
To weighting parameter and the gradient of offset parameter, using the parameter of the gradient updating dynamic convolutional neural networks for obtaining;
Step 1-7, off-line training step control:Overall control is carried out to the offline neural metwork training stage, is divided into following
Three kinds of situations:
If training sample is concentrated and still suffers from original training sample, i.e. BatchNum < BatchMax, then step is returned
Rapid 1-4 continues to read BatchSize group training samples, carries out network training;
If training sample is concentrated does not exist original training sample, i.e. BatchNum=BatchMax, and current net
Network iterations is less than maximum iteration, i.e. IterationNum < IterationMax, then make BatchNum=1, returns
Step 1-4 continues to read BatchSize group training samples, carries out network training;
If training sample is concentrated does not exist original training sample, i.e. BatchNum=BatchMax, and network changes
Generation number reaches maximum iteration, i.e. IterationNum=IterationMax, then terminate offline neural metwork training rank
Section, the dynamic convolution neural network model for being trained.
Step 1-1 data predictions of the present invention are comprised the following steps:
Step 1-1-1, sampling:The image that training image is concentrated is sequentially arranged, and constant duration is distributed, when
Between at intervals of 6 minutes, altogether comprising NTrainWidth image, TrainsetSize is determined by equation below:
Wherein, Mod (NTrain, 4) and represent NTrainTo 4 modulus, [N] represents the maximum integer for being not more than N, tries to achieve
After TrainsetSize, retain training image by sampling and concentrate preceding 4 × TrainsetSize+1 width image, by deleting during sampling
Concentrate last image to meet amount of images except training image to require;
Step 1-1-2, normalized images:Image conversion, normalization operation, by original point are carried out to the image that sampling is obtained
Resolution is converted into the floating number image that resolution ratio is 280 × 280 for 2000 × 2000 coloured image;
Step 1-1-3, constructs training sample set:
The floating number image set obtained using step 1-1-2 constructs training sample set, by every four in floating number image set
Adjacent image, i.e. { 4N+1,4N+2,4N+3,4N+4 } width image as one group of list entries, [4 × (N+1)+1] width figure
As by cutting, retain central resolution ratio for 240 × 240 part as correspondence sample control label, for N group samplesIts make is as follows:
In above formula, G4N+1Represent floating number image set in 4N+1 width images, N is positive integer, and have N ∈ [0,
TrainsetSize-1], Crop () represents trimming operation, and the portion that original image center size is for 240 × 240 is retained after cutting
Point, finally give the training sample set comprising TrainsetSize group training samples;
Wherein, step 1-1-2 is comprised the following steps:
Step 1-1-2-1, image conversion:To reduce the computation complexity of convolutional neural networks training, step 1-1-1 is adopted
The image that sample is obtained is converted into gray level image, by cutting the part for retaining that original image center resolution ratio is for 560 × 560, will
Image resolution ratio boil down to 280 × 280 after cutting, obtains the gray-scale map that resolution ratio is 280 × 280;
Step 1-1-2-2, data normalization:By each pixel in step 1-1-2-1 in the gray-scale map of acquisition
Value is mapped to [0~1] from [0~255], by obtaining the floating number image that resolution ratio is 280 × 280 after normalization.
Step 1-2 of the present invention is comprised the following steps:
Step 1-2-1, constructs sub-network:
Sub-network is made up of 10 Internets, be followed successively by from front to back convolutional layer C1, down-sampling layer S1, convolutional layer C2, under
Sample level S2, convolutional layer C3, down-sampling layer S3, convolutional layer C4, down-sampling layer S5, convolutional layer C5 and grader layer F1;
Step 1-2-2, construction probabilistic forecasting layer:
Dynamic convolutional layer DC1 and dynamic convolutional layer DC2 is constructed in probabilistic forecasting layer, by the vertical probability of sub-network output to
Amount VPV as dynamic convolutional layer DC1 convolution kernel, level probability vector HPV as dynamic convolutional layer DC2 convolution kernel;
Wherein, step 1-2-1 is comprised the following steps:
Step 1-2-1-1, constructs convolutional layer:Determine herein below:The output characteristic figure quantity OutputMaps of convolutional layer,
Convolution kernel k and offset parameter bias.For convolution kernel, it is thus necessary to determine that the width KernelSize of convolution kernel (convolution kernel point
Resolution is KernelSize × KernelSize), the quantity KernelNumber of convolution kernel (value be the convolutional layer be input into it is defeated
Go out the product of characteristic pattern quantity), and convolution kernel is constructed according to Xavier initial methods;For offset parameter, its quantity with should
The output characteristic figure quantity of layer is identical.For convolutional layer lC,lC∈ { C1, C2, C3, C4, C5 }, the output characteristic figure width of this layer
ForValue together decided on by the size of its input feature vector figure resolution ratio and convolution kernel, i.e.,Represent convolutional layer lCLast layer convolutional layer
Output characteristic figure width;
For convolutional layer C1, C1 layers of output characteristic figure quantity OutputMaps is madeC1=12, the width of output characteristic figure
Degree OutputSizeC1=272, convolution kernel width KernelSizeC1=9, offset parameter biasC1It is initialized as
Zero, C1 layer of convolution kernel kC1Quantity KernelNumberC1=48, the initial value of each parameter is in convolution kernelRand () is used to generate random number;
For convolutional layer C2, C2 layers of output characteristic figure quantity OutputMaps is madeC2=32, the width of output characteristic figure
OutputSizeC2=128, convolution kernel width KernelSizeC2=9, offset parameter is initialized as zero, C2 layers of convolution kernel
kC2Quantity KernelNumberC2=384, the initial value of each parameter is in convolution kernel
For convolutional layer C3, C3 layers of output characteristic figure quantity OutputMaps is madeC3=32, the width of output characteristic figure
OutputSizeC3=56, convolution kernel width KernelSizeC3=9, offset parameter is initialized as zero, C3 layers of convolution kernel kC3
Quantity KernelNumberC3=1024, the initial value of each parameter is in convolution kernel
For convolutional layer C4, C4 layers of output characteristic figure quantity OutputMaps is madeC4=32, the width of output characteristic figure
OutputSizeC4=20, convolution kernel width KernelSizeC4=9, offset parameter is initialized as zero, C4 layers of convolution kernel kC4
Quantity KernelNumberC4=1024, the initial value of each parameter is in convolution kernel
For convolutional layer C5, this layer of OutputMaps is madeC5=32, OutputSizeC5=4, KernelSizeC5=7, partially
Put the convolution kernel k that parameter is initialized as zero, C5 layersC5Quantity KernelNumberC5=1024, each ginseng in convolution kernel
Several initial values are
Step 1-2-1-2, construction down-sampling layer:Down-sampling layer in not comprising need training parameter, by down-sampling layer S1,
The sampling core of S2, S3 and S4 is initialized asFor down-sampling layer lS,lS∈ { S1, S2, S3, S4 }, its output
Characteristic pattern quantityOutput characteristic figure quantity with the convolutional layer of its last layer is consistent, and output characteristic figure is wide
DegreeIt is the 1/2 of the output characteristic figure width of the convolutional layer of its last layer, formula is expressed as follows:
Step 1-2-1-3, structural classification device layer:Grader layer is made up of a full articulamentum F1, F1 layers of weighting parameter
It is horizontal weighting parameter matrix W H and vertical weighting parameter matrix W V, size is 41 × 512, makes every in weighting parameter matrix
The initial value of one parameter isOffset parameter is Horizontal offset parameter BH and vertical off setting parameter BV,
It is initialized as 41 × 1 one-dimensional null vector.
Step 1-5 of the present invention is comprised the following steps:
Step 1-5-1, sub-network calculates probability vector:In a sub-network by convolutional layer and the alternate treatment of down-sampling layer
Extract input image sequence characteristic, grader layer in processed by Softmax functions, obtain level probability vector HPV with
Vertical probability vector VPV;
Step 1-5-2, calculates probabilistic forecasting layer output image:HPV that step 1-5-1 is obtained and VPV are used as probabilistic forecasting
Layer convolution kernel, by the last piece image in input image sequence successively with VPV, HPV phase convolution, obtain the defeated of propagated forward
Go out prognostic chart picture.
Step 1-5-1 of the present invention is comprised the following steps:
Step 1-5-1-1, judges network layer type:The Internet in the sub-network being presently in, l initial values are represented with l
It is C1, judges the type of Internet l, if l ∈ { C1, C2, C3, C4, C5 }, then l is convolutional layer, step 1-5-1-2 is performed, if l
∈ { S1, S2, S3, S4 }, then l is down-sampling layer, performs step 1-5-1-4;
Step 1-5-1-2, processes convolutional layer:Now there is l=lC,lC∈ { C1, C2, C3, C4, C5 }, calculates l firstCLayer
J-th output characteristic figureBy lCThe input feature vector figure convolution nuclear phase convolution corresponding with this layer respectively of layer, convolution results are asked
With summed result adds lCJ-th offset parameter of layerProcessed by ReLU activation primitives again, obtainedCalculate public
Formula is as follows:
Wherein,It is lCI-th input feature vector figure of layer convolution kernel corresponding with j-th output characteristic figure, n is current
The output characteristic figure number of the previous down-sampling layer of convolutional layer, Represent lCI-th input of layer
Characteristic pattern, while being also lC- 1 layer of i-th output characteristic figure, * representing matrix convolution, if lC=C1, then lC- 1 layer is input
Layer;
All of output characteristic figure is calculated successively, obtains lCThe output characteristic figure of layerL is updated to l+1, and returns to step
Rapid 1-5-1-1 judges network type, carries out the operation of next Internet;
Step 1-5-1-3, treatment down-sampling layer:Now there is l=lS,lS∈ { S1, S2, S3, S4 }, step 1-5-1-2 is obtained
The output characteristic figure of the convolutional layer for arriving respectively withPhase convolution, then sampled with step-length as 2, sampling obtains lSLayer
Output characteristic figureComputing formula is as follows:
In above formula, Sample () represents the sampling processing that step-length is 2, lS- 1 previous convolution for representing current down-sampling layer
Layer,Represent lSThe output characteristic figure of layerIn j-th output characteristic figure, obtain lSThe output characteristic figure of layerAfterwards, by l more
It is newly l+1, and return to step 1-5-1-1 judges network type, carries out the operation of next Internet;
Step 1-5-1-4, calculates F1 layers of probability vector:If Internet l is grader layer, i.e. l=F1 is become by matrix
Change, be 4 × 4 output characteristic figure with row sequential deployment by the 32 width resolution ratio of C5, obtain that resolution ratio is 512 × 1 F1 layers
Output characteristic vector aF1, difference calculated level weighting parameter matrix W H and aF1Apposition, vertical weighting parameter matrix W V and aF1's
Apposition, result of calculation is sued for peace with Horizontal offset parameter BH, vertical off setting parameter BV respectively, after being processed through Softmax functions
To level probability vector HPV and vertical probability vector VPV, specific formula for calculation is as follows:
By its vertical probability vector VPV transposition, final vertical probability vector is obtained;
Step 1-5-2 is comprised the following steps:
Step 1-5-2-1, predicts DC1 layers of vertical direction:By last width input picture of input layer and vertical probability to
Amount VPV phase convolution, obtains the DC1 layers of output characteristic figure a that resolution ratio is 240 × 280DC1;
Step 1-5-2-2, predicts DC2 layers of vertical direction:By DC1 layers of output characteristic figure aDC1With vertical probability vector HPV phases
Convolution, obtains the output prognostic chart picture of propagated forward, and its resolution ratio is 240 × 240.
Step 1-6 of the present invention is comprised the following steps:
Step 1-6-1, calculates probabilistic forecasting layer error term:The prognostic chart picture that step 1-5-2-2 is obtained and the instruction being input into
The control label practiced in sample asks poor, calculates DC2 layers, DC1 layers of error term, finally tries to achieve the error term of level probability vector
δHPVWith the error term δ of vertical probability vectorVPV;
Step 1-6-2, calculates sub-network error term:According to the error term δ of level probability vectorHPVWith vertical probability vector
Error term δVPV, from it is rear to preceding successively calculate classification layer F1, convolutional layer C5, C4, C3, C2, C1 and down-sampling layer S4, S3, S2,
The error term of S1, the resolution ratio of any layer error term matrix tried to achieve is consistent with the resolution ratio of the output characteristic figure of this layer;
Step 1-6-3, calculates gradient:The error term obtained according to step 1-6-2 calculates the mistake of each Internet of sub-network
Difference item is to this layer of weighting parameter and the Grad of offset parameter;
Step 1-6-4, undated parameter:The weighting parameter and the ladder of offset parameter of each Internet that step 1-6-3 is obtained
Angle value is multiplied by the learning rate of dynamic convolutional neural networks, the renewal of each Internet weighting parameter and offset parameter is obtained, by original
Weighting parameter and offset parameter ask poor with the renewal respectively, weighting parameter and offset parameter after being updated.
Step 1-6-1 of the present invention is comprised the following steps:
Step 1-6-1-1, calculates dynamic convolutional layer DC2 error terms:The prognostic chart picture that step 1-5-2-2 is obtained and the group
The control label of sample asks poor, obtains the error term matrix delta that size is 240 × 240DC2;
Step 1-6-1-2, calculates dynamic convolutional layer DC1 error terms:By zero padding by DC2 layers of error term matrix deltaDC2
It is 240 × 320 to expand, by level probability Vector rotation 180 degree, by the level probability after the error term matrix after expansion and upset
Vectorial phase convolution, obtains DC1 layers of error term matrix deltaDC1, its size is 240 × 280, and formula is as follows:
δDC1=Expand_Zero (δDC2) * rot180 (HPV),
Wherein, Expand_Zero () represents zero extended function, and rot180 () represents the rotation letter that angle is 180 °
Number, 4 × 4 matrix, the matrix after zero expansion, region and former square of the central resolution ratio for 2 × 2 are extended for by 2 × 2 matrix zero
Consistent, zero pixel filling of remaining position of battle array;
Step 1-6-1-3, calculates probability vector error term:The error term of calculated level probability vector HPV, by DC1 layers
Output characteristic figure and error term matrix deltaDC2Phase convolution, obtains 1 × 41 row vector after convolution, the vector is the error term of HPV
δHPV, formula is as follows:
δHPV=aDC1*δDC2,
The error term of vertical probability vector VPV is calculated, by the input feature vector figure of input layer and error term matrix deltaDC1Mutually roll up
Product, obtains 41 × 1 column vector after convolution, the vector is the error term δ of VPVVPV, formula is as follows:
Wherein,Last piece image in for the input image sequence of training sample;
Step 1-6-2 is comprised the following steps:
Step 1-6-2-1, calculates grader layer F1 error terms:The error term of the probability vector that step 1-6-1-3 is obtained
δVPVAnd δHPVWeighting parameter matrix W V vertical with F1 layers and horizontal weighting parameter matrix W H carry out matrix multiple respectively, then by square
The apposition of battle array is sued for peace and is averaged, and obtains F1 layers of error term δF1, formula is as follows:
Wherein, × representing matrix apposition, ()TThe transposition of matrix is represented, the δ for obtainingF1Size be 512 × 1;
Step 1-6-2-2, calculates convolutional layer C5 error terms:By matrixing, F1 layers for being obtained in step 1-6-2-1
Error term δF1It is transformed to the matrix that 32 resolution ratio are 4 × 4Obtain C5 layers of error term δC5,Table
Show the matrix that the 32nd resolution ratio after conversion is 4 × 4;Matrixing in this time matrixing and step 1-5-1-6 is each other
Inverse transformation,
Step 1-6-2-3, judges network layer type:Internet (the l initial values in the sub-network being presently in are represented with l
It is S4), judge the type of Internet l, if l ∈ { S4, S3, S2, S1 }, then l is down-sampling layer, step 1-6-2-4 is performed, if l
∈ { C4, C3, C2, C1 }, then l is convolutional layer, performs step 1-6-2-5;
Step 1-6-2-4, calculates down-sampling layer error term:If l layers is down-sampling layer, now there is l=lS,lS∈{S1,S2,
S3, S4 }, for lSI-th error term matrix of layerBy zero padding respectively by lS+ 1 layer each error term matrix
Expand to width and beWherein lS+ 1 layer of latter convolutional layer for representing current down-sampling layer, j represents lS+ 1 layer
J-th error term, and haveAgain by corresponding convolution kernelRotation 180 degree, after then expanding
Matrix and upset after convolution nuclear phase convolution, and convolution results are sued for peace, obtain lSI-th error term matrix of layerIt is public
Formula is as follows:
Wherein,
All of error term matrix is calculated successively, obtains lSThe output characteristic figure of layerL is updated to l-1, and returns to step
Rapid 1-6-2-3 judges network type, calculates the error term of a upper Internet;
Step 1-6-2-5, calculates convolutional layer error term:If l layers is convolutional layer, now there is l=lC,lC∈{C1,C2,C3,
C4, C5 }, because the initial value of l in step 1-6-2-3 is S4, therefore be not in lCThe situation of=C5, for lCThe i-th of layer
Individual error term matrixFirst to lCCorresponding i-th error term matrix in+1 layerUp-sampled, will during up-samplingIn each element error entry value average mark to sample area in, obtaining resolution ratio is
Up-sampling matrix, then calculate activation primitive in lCDerivative and the inner product for up-sampling matrix tried to achieve at layer character pair figure,
Obtain lCI-th error term matrix of layerFormula is as follows:
Wherein, representing matrix inner product, ReLU'() derivative of ReLU activation primitives is represented, its form is as follows:
UpSample () represents up-sampling function, one up-sampling of each pixel correspondence after up-sampling in original image
Region, in each pixel in original pixel value mean allocation to sample area, calculates all of error term matrix successively, obtains
To lCThe output characteristic figure of layer
Step 1-6-2-6, now l layers is convolutional layer, i.e. l=lC, two kinds of situations are divided into afterwards:
If l ≠ C1, l is updated to l-1, and return to step 1-6-2-3 judges network type, calculates a upper Internet
Error term;
If l=C1, step 1-6-2 sub-networks error term is calculated and terminated;
Step 1-6-3 is comprised the following steps:
Step 1-6-3-1, calculates gradient of the convolutional layer error term to convolution kernel:Use lCRepresent currently processed convolutional layer, lC
∈ { C1, C2, C3, C4, C5 }, successively calculates gradient of each convolutional layer error term to convolution kernel, by l since C1 layersC- 1 layer
I-th output characteristic figureWith lCJ-th error term matrix of layerPhase convolution, convolution results are the ladder of correspondence convolution kernel
Angle valueFormula is as follows:
In above formula,WithL is represented respectivelyCThe output characteristic figure number of layer and lC-
1 layer of output characteristic figure number, ▽ k are Grad of the error term to convolution kernel;
Step 1-6-3-2, calculates gradient of each convolutional layer error term to biasing:Use lCRepresent currently processed convolutional layer, lC
∈ { C1, C2, C3, C4, C5 }, successively calculates gradient of each convolutional layer error term to biasing, by l since C1 layersCJ-th of layer
Error term matrixIn all elements sued for peace, obtain the Grad of j-th of this layer biasingThe following institute of formula
Show:
Wherein, Sum () is represented and all elements of matrix is sued for peace;
Step 1-6-3-3, calculates gradient of the F1 layers of error term to weighting parameter:Respectively calculated level probability vector with it is vertical
The error term δ of probability vectorHPV、δVPVWith F1 layers of error term δF1Inner product, result of calculation be F1 layers of error term to weighting parameter WH,
The Grad of WV, formula is as follows:
▽ WH=(δHPV)T×(δF1)T,
▽ WV=δVPV×(δF1)T,
In above formula, ▽ WH are Grad of the error term to horizontal weighting parameter, and ▽ WV are error term to vertical weighting parameter
Grad;
Step 1-6-3-4, calculates gradient of the F1 layers of error term to offset parameter:By level probability it is vectorial with vertical probability to
The error term δ of amountHPV、δVPVRespectively as F1 layers of error term to Horizontal offset parameter BH and the Grad of vertical off setting parameter BV,
Formula is as follows:
▽ BH=(δHPV)T,
▽ BV=δVPV,
Wherein, ▽ BH are Grad of the error term to Horizontal offset parameter, and ▽ BV are error term to vertical off setting parameter
Grad;
Step 1-6-4 is comprised the following steps:
Step 1-6-4-1, updates each convolutional layer weighting parameter:Each convolutional layer error term pair that step 1-6-3-1 is obtained
The gradient of convolution kernel is multiplied by the learning rate of dynamic convolutional neural networks, obtains the correction term of convolution kernel, then by former convolution kernel and be somebody's turn to do
Correction term asks poor, the convolution kernel for being updatedFormula is as follows:
Wherein, λ is the e-learning rate of determination in step 1-3, λ=0.0001;
Step 1-6-4-2, updates each convolutional layer offset parameter:Each convolutional layer error term pair that step 1-6-3-2 is obtained
The gradient of biasing is multiplied by the learning rate of dynamic convolutional neural networks, obtains the correction term of offset parameter, then by former bias term and be somebody's turn to do
Correction term asks poor, the bias term for being updatedFormula is as follows:
Step 1-6-4-3, updates F1 layers of weighting parameter:The F1 layers of error term that step 1-6-3-3 is obtained is to weighting parameter
The Grad of WH and WV is multiplied by the learning rate of dynamic convolutional neural networks, obtains the correction term of weighting parameter, then former weights are joined
Number WH and WV asks poor with the correction term tried to achieve respectively, and the WH and WV for being updated, formula are as follows:
WH=WH- λ ▽ WH,
WV=WV- λ ▽ WV;
Step 1-6-4-4, updates F1 layers of offset parameter:The F1 layers of error term that step 1-6-3-4 is obtained is to offset parameter
The Grad of BH and BV is multiplied by the learning rate of dynamic convolutional neural networks, obtains the correction term of offset parameter, then original biasing is joined
Number BH and BV asks poor with the correction term tried to achieve respectively, and the BH and BV for being updated, formula are as follows:
BH=BH- λ ▽ BH,
BV=BV- λ ▽ BV.
Step 2 of the present invention is comprised the following steps:
Step 2-1, data prediction:Input test image set, specification is carried out to every piece image that test image is concentrated
Change is processed, and 280 × 280 floating number image will be converted into per piece image, then floating number image collection is divided, and is constructed
Test sample collection comprising TestsetSize group samples;
Step 2-2, read test sample:The TestsetSize groups test sample input that step 2-1 is obtained is by training
Dynamic convolutional neural networks in;
Step 2-3, propagated forward:The image sequence characteristic of input is extracted in a sub-network, obtains level probability vector
HPVtestWith vertical probability vector VPVtest;Probabilistic forecasting layer in, by the last piece image in input image sequence successively with
VPVtest、HPVtestPhase convolution, obtains the final extrapolated image of dynamic convolutional neural networks.
Step 2-1 of the present invention is comprised the following steps:
Step 2-1-1, sampling:The image that test image is concentrated is sequentially arranged, and constant duration is distributed, when
Between at intervals of 6 minutes, altogether comprising NTestWidth image, TestsetSize is determined by equation below:
If Mod (NTest, 4)=0
If Mod (NTest, 4) ≠ 0
After trying to achieve TestsetSize, retain test image by sampling and concentrate preceding 4 × TestsetSize+1 width image, adopt
Amount of images is set to meet requirement by deleting the last image of test image concentration during sample;
Step 2-1-2, image normalization:Image conversion, normalization operation, by original point are carried out to the image that sampling is obtained
Resolution is converted into the floating number image that resolution ratio is 280 × 280 for 2000 × 2000 coloured image;
Step 2-1-3, constructs test sample collection:The floating number image set obtained using step 2-1-2 constructs test sample
Collection, by every four adjacent images in floating number image set, i.e. { 4M+1,4M+2,4M+3,4M+4 } width image is defeated as one group
Enter sequence, by cutting, the central resolution ratio of reservation is 240 × 240 part as correspondence sample to [4 × (M+1)+1] width image
This control label, wherein being positive integer, and has M ∈ [0, TestsetSize-1] to obtain comprising TestsetSize group test specimens
This test sample collection;
Step 2-1-2 is comprised the following steps:
Step 2-1-2-1, image conversion:To reduce the computation complexity of convolutional neural networks training, by colored echo
Intensity CAPPI images are converted into gray level image, then by cutting the part for retaining that original image center resolution ratio is for 560 × 560,
By the image resolution ratio boil down to 280 × 280 after cutting, the gray-scale map that resolution ratio is 280 × 280 is obtained;
Step 2-1-2-2, data normalization:By each pixel in step 1-1-2-1 in the gray-scale map of acquisition
Value is mapped to [0~1] from [0~255], by obtaining the floating number image that resolution ratio is 280 × 280 after normalization;
Step 2-3 is comprised the following steps:
Step 2-3-1, calculates sub-network probability vector:In a sub-network by convolutional layer and the alternate treatment of down-sampling layer
The image sequence characteristic of input is extracted, is then processed by Softmax functions in grader layer, obtain level probability vector
HPVtestWith vertical probability vector VPVtest;
Step 2-3-2, calculates probabilistic forecasting layer output image:The VPV that step 2-3-1 is obtainedtestAnd HPVtestAs probability
The convolution kernel of prediction interval, by the last piece image in input image sequence successively with VPVtestAnd HPVtestPhase convolution, is moved
The final extrapolated image of state convolutional neural networks;
Step 2-3-1 is comprised the following steps:
Step 2-3-1-1, judges network layer type:The Internet in the sub-network being presently in is represented with p, network is judged
The type of layer p, p initial values are C1, if p ∈ { C1, C2, C3, C4, C5 }, then p is convolutional layer, step 2-3-1-2 is performed, if p ∈
{ S1, S2, S3, S4 }, then p is down-sampling layer, performs step 2-3-1-4;
Step 2-3-1-2, processes convolutional layer:Now there is p=pC,pC∈ { C1, C2, C3, C4, C5 }, calculates p firstCLayer
V-th output characteristic figureBy pCThe input feature vector figure convolution nuclear phase convolution corresponding with this layer respectively of layer, then by convolution knot
Fruit is sued for peace, and summed result adds pCV-th biasing of layerProcessed by ReLU activation primitives again, obtain pCThe v of layer
Individual output characteristic figureComputing formula is as follows:
Wherein,It is pCU-th input feature vector figure of layer convolution kernel corresponding with v-th output characteristic figure, m is current
The output characteristic figure number of the previous down-sampling layer of convolutional layer,Represent pCU-th of layer is defeated
Enter characteristic pattern, while being also pC- 1 layer of u-th output characteristic figure, * representing matrix convolution works as pCDuring=C1, then pC- 1 layer is defeated
Enter layer.
All of output characteristic figure is calculated successively, obtains pCThe output characteristic figure of layerP is updated to p+1, and returns to step
Rapid 2-3-1-1 judges network type, carries out the operation of next Internet;
Step 2-3-1-3, treatment down-sampling layer:Now there is p=pS,pS∈ { S1, S2, S3, S4 }, step 2-3-1-2 is obtained
The output characteristic figure of the convolutional layer for arriving respectively withPhase convolution, then sampled with step-length as 2, sampling obtains pSLayer
Output characteristic figureComputing formula is as follows:
Wherein, Sample () represents the sampling processing that step-length is 2, pS- 1 previous convolution for representing current down-sampling layer
Layer,Represent pSThe output characteristic figure of layerIn j-th output characteristic figure, obtain pSThe output characteristic figure of layerAfterwards, by p
P+1 is updated to, and return to step 2-3-1-1 judges network type, carries out the operation of next Internet;
Step 2-3-1-4, calculates F1 layers of probability vector:If Internet p is grader layer, i.e. p=F1 is become by matrix
Change, be 4 × 4 output characteristic figure with row sequential deployment by the 32 width resolution ratio of C5, obtain that resolution ratio is 512 × 1 F1 layers
Output characteristic vectorThen respectively calculated level parameter matrix WH, Vertical Parameters matrix W V withApposition, by calculate tie
Fruit is sued for peace with Horizontal offset parameter BH, vertical off setting parameter BV respectively, and summed result obtains level after being processed through Softmax functions
Probability vector HPVtest, vertical probability vector VPVtest, computing formula is as follows:
By its vertical probability vector VPVtestTransposition, obtains final vertical probability vector;
Step 2-3-2 is comprised the following steps:
Step 2-3-2-1, predicts DC1 layers of vertical direction:By last width input picture of input layer and vertical probability to
Amount VPVtestPhase convolution, obtains the DC1 layers of output characteristic figure that resolution ratio is 240 × 280
Step 2-3-2-2, predicts DC2 layers of vertical direction:Step 2-3-2-1 is obtainedWith level probability vector
HPVtestPhase convolution, obtains the final extrapolated image of dynamic convolutional neural networks, and its resolution ratio is 240 × 240, by extrapolated image
Compared with the control label of corresponding test sample, judge that dynamic convolutional neural networks carry out the accuracy of Radar Echo Extrapolation.
Beneficial effect:Based on dynamic convolutional neural networks, trained on substantial amounts of radar return image set, and utilize
The network for training carries out Radar Echo Extrapolation, effectively increases the degree of accuracy of Radar Echo Extrapolation.
It is specifically of the invention to be had the advantage that compared with existing method:1. the utilization rate of Radar Data is high, with traditional thunder
Only compared using a small amount of Radar Data up to echo extrapolation technique, the present invention is trained on substantial amounts of radar return image set, number
It is higher according to utilization rate;2. extrapolation timeliness is long, can improve outer by the time interval between sample in adjusting training sample set
Push away timeliness.
Brief description of the drawings
The present invention is done with reference to the accompanying drawings and detailed description further is illustrated, it is of the invention above-mentioned or
Otherwise advantage will become apparent.
Fig. 1 is flow chart of the present invention.
Fig. 2 is dynamic convolutional neural networks initialization model structure chart.
Fig. 3 is sub-network structural map.
Fig. 4 is probabilistic forecasting layer structural map.
Fig. 5 is that matrix zero expands schematic diagram.
Fig. 6 is the process schematic up-sampled to 2 × 2 matrix.
Specific embodiment
Below in conjunction with the accompanying drawings and embodiment the present invention will be further described.
As shown in figure 1, the present invention is comprised the following steps:
Step 1, as shown in Fig. 2 training offline convolutional neural networks:Input training image collection, is carried out to training image collection
Data prediction, obtains training sample set, the dynamic convolution neural network structure of design, and initialize network training parameter;Utilize
The dynamic convolutional neural networks of training sample set training, the orderly image sequence of input is by dynamic convolutional neural networks propagated forward
A width prognostic chart picture is obtained, the error between prognostic chart picture and control label is calculated, the weights of network are updated by backpropagation
Parameter and offset parameter, repeat this process until reaching training termination condition, obtain convergent dynamic convolutional neural networks;
Step 2, online Radar Echo Extrapolation:Input test image set, data prediction is carried out to test chart image set, is obtained
Test sample collection, then will in test sample collection input step 1 obtain dynamic convolutional neural networks in, by before network to biography
Calculating probability vector is broadcast, and last width radar return image in input image sequence is mutually rolled up with the probability vector for obtaining
Product, the Radar Echo Extrapolation image predicted.
Step 1 of the present invention is comprised the following steps:
Step 1-1, data prediction:Input training image collection, specification is carried out to every piece image that training image is concentrated
Change is processed, and 280 × 280 floating number image will be converted into per piece image, floating number image collection is obtained, to floating number image
Set is divided, training sample set of the construction comprising TrainsetSize group samples;
Step 1-2, the dynamic convolutional neural networks of initialization:The dynamic convolution neural network structure of design, is configured to generation
The sub-network of probability vector, reconstructs the probabilistic forecasting layer for image extrapolation, for the offline neural metwork training stage provides dynamic
State convolutional neural networks initialization model;
Step 1-3, the dynamic convolution Neural Network Training Parameter of initialization:E-learning rate λ=0.0001 is made, is instructed
Practice the sample size BatchSize=10 of stage each input, most large quantities of frequency of training of training sample setCurrently criticize frequency of training BatchNum=1, the maximum iteration of network training
IterationMax=40, current iteration number of times IterationNum=1;
Step 1-4, reads training sample:By the way of batch training, the training sample obtained from step 1-1 is trained every time
Concentrate and read BatchSize group training samples, every group of training sample is { x1,x2,x3,x4, y }, altogether comprising 5 width images, wherein
{x1,x2,x3,x4As input image sequence, x1Piece image is represented, y is corresponding control label;
Step 1-5, propagated forward:The input image sequence feature that extraction step 1-4 is obtained in a sub-network, obtains level
Probability vector HPV and vertical probability vector VPV;Probabilistic forecasting layer in, by the last piece image in input image sequence according to
It is secondary with VPV, HPV phase convolution, obtain the output prognostic chart picture of propagated forward;
Step 1-6, backpropagation:The error term of probability vector is reversely tried to achieve in probabilistic forecasting layer, according to probability vector
Error term from rear to the preceding error term for successively calculating each Internet in subnet network layers, and then calculate error in each Internet
To weighting parameter and the gradient of offset parameter, using the parameter of the gradient updating dynamic convolutional neural networks for obtaining;
Step 1-7, off-line training step control:Overall control is carried out to the offline neural metwork training stage, is divided into following
Three kinds of situations:
If training sample is concentrated and still suffers from original training sample, i.e. BatchNum < BatchMax, then step is returned
Rapid 1-4 continues to read BatchSize group training samples, carries out network training;
If training sample is concentrated does not exist original training sample, i.e. BatchNum=BatchMax, and current net
Network iterations is less than maximum iteration, i.e. IterationNum < IterationMax, then make BatchNum=1, returns
Step 1-4 continues to read BatchSize group training samples, carries out network training;
If training sample is concentrated does not exist original training sample, i.e. BatchNum=BatchMax, and network changes
Generation number reaches maximum iteration, i.e. IterationNum=IterationMax, then terminate offline neural metwork training rank
Section, the dynamic convolution neural network model for being trained.
Step 1-1 data predictions of the present invention are comprised the following steps:
Step 1-1-1, sampling:The image that training image is concentrated is sequentially arranged, and constant duration is distributed, when
Between at intervals of 6 minutes, altogether comprising NTrainWidth image, TrainsetSize is determined by equation below:
Wherein, Mod (NTrain, 4) and represent NTrainTo 4 modulus, [N] represents the maximum integer for being not more than N, tries to achieve
After TrainsetSize, retain training image by sampling and concentrate preceding 4 × TrainsetSize+1 width image, by deleting during sampling
Concentrate last image to meet amount of images except training image to require;
Step 1-1-2, normalized images:Image conversion, normalization operation, by original point are carried out to the image that sampling is obtained
Resolution is converted into the floating number image that resolution ratio is 280 × 280 for 2000 × 2000 coloured image;
Step 1-1-3, constructs training sample set:
The floating number image set obtained using step 1-1-2 constructs training sample set, by every four in floating number image set
Adjacent image, i.e. { 4N+1,4N+2,4N+3,4N+4 } width image as one group of list entries, [4 × (N+1)+1] width figure
As by cutting, retain central resolution ratio for 240 × 240 part as correspondence sample control label, for N group samplesIts make is as follows:
In above formula, G4N+1Represent floating number image set in 4N+1 width images, N is positive integer, and have N ∈ [0,
TrainsetSize-1], Crop () represents trimming operation, and the portion that original image center size is for 240 × 240 is retained after cutting
Point, finally give the training sample set comprising TrainsetSize group training samples;
Wherein, step 1-1-2 is comprised the following steps:
Step 1-1-2-1, image conversion:To reduce the computation complexity of convolutional neural networks training, step 1-1-1 is adopted
The image that sample is obtained is converted into gray level image, by cutting the part for retaining that original image center resolution ratio is for 560 × 560, will
Image resolution ratio boil down to 280 × 280 after cutting, obtains the gray-scale map that resolution ratio is 280 × 280;
Step 1-1-2-2, data normalization:By each pixel in step 1-1-2-1 in the gray-scale map of acquisition
Value is mapped to [0~1] from [0~255], by obtaining the floating number image that resolution ratio is 280 × 280 after normalization.
Step 1-2 of the present invention is comprised the following steps:
Step 1-2-1, constructs sub-network:Sub-network is configured to structure as shown in Figure 3:
Sub-network is made up of 10 Internets, be followed successively by from front to back convolutional layer C1, down-sampling layer S1, convolutional layer C2, under
Sample level S2, convolutional layer C3, down-sampling layer S3, convolutional layer C4, down-sampling layer S5, convolutional layer C5 and grader layer F1;
Step 1-2-2, construction probabilistic forecasting layer:Probabilistic forecasting layer is configured to structure as shown in Figure 4:
Dynamic convolutional layer DC1 and dynamic convolutional layer DC2 is constructed in probabilistic forecasting layer, by the vertical probability of sub-network output to
Amount VPV as dynamic convolutional layer DC1 convolution kernel, level probability vector HPV as dynamic convolutional layer DC2 convolution kernel;
Wherein, step 1-2-1 is comprised the following steps:
Step 1-2-1-1, constructs convolutional layer:Determine herein below:The output characteristic figure quantity OutputMaps of convolutional layer,
Convolution kernel k and offset parameter bias.For convolution kernel, it is thus necessary to determine that the width KernelSize of convolution kernel (convolution kernel point
Resolution is KernelSize × KernelSize), the quantity KernelNumber of convolution kernel (value be the convolutional layer be input into it is defeated
Go out the product of characteristic pattern quantity), and convolution kernel is constructed according to Xavier initial methods;For offset parameter, its quantity with should
The output characteristic figure quantity of layer is identical.For convolutional layer lC,lC∈ { C1, C2, C3, C4, C5 }, the output characteristic figure width of this layer
ForValue together decided on by the size of its input feature vector figure resolution ratio and convolution kernel, i.e.,
For convolutional layer C1, C1 layers of output characteristic figure quantity OutputMaps is madeC1=12, the width of output characteristic figure
Degree OutputSizeC1=272, convolution kernel width KernelSizeC1=9, offset parameter biasC1Zero is initialized as,
C1 layers of convolution kernel kC1Quantity KernelNumberC1=48, the initial value of each parameter is in convolution kernelRand () is used to generate random number;
For convolutional layer C2, this layer of OutputMaps is madeC2=32, OutputSizeC2=128, KernelSizeC2=9,
Offset parameter is initialized as zero, C2 layers of convolution kernel kC2Quantity KernelNumberC2=384, each ginseng in convolution kernel
Several initial values are
For convolutional layer C3, this layer of OutputMaps is madeC3=32, OutputSizeC3=56, KernelSizeC3=9, partially
Put the convolution kernel k that parameter is initialized as zero, C3 layersC3Quantity KernelNumberC3=1024, each ginseng in convolution kernel
Several initial values are
For convolutional layer C4, this layer of OutputMaps is madeC4=32, OutputSizeC4=20, KernelSizeC4=9, partially
Put the convolution kernel k that parameter is initialized as zero, C4 layersC4Quantity KernelNumberC4=1024, each ginseng in convolution kernel
Several initial values are
For convolutional layer C5, this layer of OutputMaps is madeC5=32, OutputSizeC5=4, KernelSizeC5=7, partially
Put the convolution kernel k that parameter is initialized as zero, C5 layersC5Quantity KernelNumberC5=1024, each ginseng in convolution kernel
Several initial values are
Step 1-2-1-2, construction down-sampling layer:Down-sampling layer in not comprising need training parameter, by down-sampling layer S1,
The sampling core of S2, S3 and S4 is initialized asFor down-sampling layer lS,lS∈ { S1, S2, S3, S4 }, its output
Characteristic pattern quantityOutput characteristic figure quantity with the convolutional layer of its last layer is consistent, and output characteristic figure is wide
DegreeIt is the 1/2 of the output characteristic figure width of the convolutional layer of its last layer, formula is expressed as follows:
Step 1-2-1-3, structural classification device layer:Grader layer is made up of a full articulamentum F1, F1 layers of weighting parameter
It is horizontal weighting parameter matrix W H and vertical weighting parameter matrix W V, size is 41 × 512, makes every in weighting parameter matrix
The initial value of one parameter isOffset parameter is Horizontal offset parameter BH and vertical off setting parameter BV,
It is initialized as 41 × 1 one-dimensional null vector.
Step 1-5 of the present invention is comprised the following steps:
Step 1-5-1, sub-network calculates probability vector:In a sub-network by convolutional layer and the alternate treatment of down-sampling layer
Extract input image sequence characteristic, grader layer in processed by Softmax functions, obtain level probability vector HPV with
Vertical probability vector VPV;
Step 1-5-2, calculates probabilistic forecasting layer output image:HPV that step 1-5-1 is obtained and VPV are used as probabilistic forecasting
Layer convolution kernel, by the last piece image in input image sequence successively with VPV, HPV phase convolution, obtain the defeated of propagated forward
Go out prognostic chart picture.
Step 1-5-1 of the present invention is comprised the following steps:
Step 1-5-1-1, judges network layer type:The Internet in the sub-network being presently in, l initial values are represented with l
It is C1, judges the type of Internet l, if l ∈ { C1, C2, C3, C4, C5 }, then l is convolutional layer, step 1-5-1-2 is performed, if l
∈ { S1, S2, S3, S4 }, then l is down-sampling layer, performs step 1-5-1-4;
Step 1-5-1-2, processes convolutional layer:Now there is l=lC,lC∈ { C1, C2, C3, C4, C5 }, calculates l firstCLayer
J-th output characteristic figureBy lCThe input feature vector figure convolution nuclear phase convolution corresponding with this layer respectively of layer, convolution results are asked
With summed result adds lCJ-th offset parameter of layerProcessed by ReLU activation primitives again, obtainedCalculate public
Formula is as follows:
Wherein,It is lCI-th input feature vector figure of layer convolution kernel corresponding with j-th output characteristic figure, n is current
The output characteristic figure number of the previous down-sampling layer of convolutional layer,Represent lCI-th input of layer
Characteristic pattern, while being also lC- 1 layer of i-th output characteristic figure, * representing matrix convolution, if lC=C1, then lC- 1 layer is input
Layer;
All of output characteristic figure is calculated successively, obtains lCThe output characteristic figure of layerL is updated to l+1, and returns to step
Rapid 1-5-1-1 judges network type, carries out the operation of next Internet;
Step 1-5-1-3, treatment down-sampling layer:Now there is l=lS,lS∈ { S1, S2, S3, S4 }, step 1-5-1-2 is obtained
The output characteristic figure of the convolutional layer for arriving respectively withPhase convolution, then sampled with step-length as 2, sampling obtains lS
The output characteristic figure of layerComputing formula is as follows:
In above formula, Sample () represents the sampling processing that step-length is 2, lS- 1 previous convolution for representing current down-sampling layer
Layer,Represent lSThe output characteristic figure of layerIn j-th output characteristic figure, obtain lSThe output characteristic figure of layerAfterwards, by l more
It is newly l+1, and return to step 1-5-1-1 judges network type, carries out the operation of next Internet;
Step 1-5-1-4, calculates F1 layers of probability vector:If Internet l is grader layer, i.e. l=F1 is become by matrix
Change, be 4 × 4 output characteristic figure with row sequential deployment by the 32 width resolution ratio of C5, obtain that resolution ratio is 512 × 1 F1 layers
Output characteristic vector aF1, difference calculated level weighting parameter matrix W H and aF1Apposition, vertical weighting parameter matrix W V and aF1's
Apposition, result of calculation is sued for peace with Horizontal offset parameter BH, vertical off setting parameter BV respectively, after being processed through Softmax functions
To level probability vector HPV and vertical probability vector VPV, specific formula for calculation is as follows:
By its vertical probability vector VPV transposition, final vertical probability vector is obtained;
Step 1-5-2 is comprised the following steps:
Step 1-5-2-1, predicts DC1 layers of vertical direction:By last width input picture of input layer and vertical probability to
Amount VPV phase convolution, obtains the DC1 layers of output characteristic figure a that resolution ratio is 240 × 280DC1;
Step 1-5-2-2, predicts DC2 layers of vertical direction:By DC1 layers of output characteristic figure aDC1With vertical probability vector HPV phases
Convolution, obtains the output prognostic chart picture of propagated forward, and its resolution ratio is 240 × 240.
Step 1-6 of the present invention is comprised the following steps:
Step 1-6-1, calculates probabilistic forecasting layer error term:The prognostic chart picture that step 1-5-2-2 is obtained and the instruction being input into
The control label practiced in sample asks poor, calculates DC2 layers, DC1 layers of error term, finally tries to achieve the error term of level probability vector
δHPVWith the error term δ of vertical probability vectorVPV;
Step 1-6-2, calculates sub-network error term:According to the error term δ of level probability vectorHPVWith vertical probability vector
Error term δVPV, from it is rear to preceding successively calculate classification layer F1, convolutional layer C5, C4, C3, C2, C1 and down-sampling layer S4, S3, S2,
The error term of S1, the resolution ratio of any layer error term matrix tried to achieve is consistent with the resolution ratio of the output characteristic figure of this layer;
Step 1-6-3, calculates gradient:The error term obtained according to step 1-6-2 calculates the mistake of each Internet of sub-network
Difference item is to this layer of weighting parameter and the Grad of offset parameter;
Step 1-6-4, undated parameter:The weighting parameter and the ladder of offset parameter of each Internet that step 1-6-3 is obtained
Angle value is multiplied by the learning rate of dynamic convolutional neural networks, the renewal of each Internet weighting parameter and offset parameter is obtained, by original
Weighting parameter and offset parameter ask poor with the renewal respectively, weighting parameter and offset parameter after being updated.
Step 1-6-1 of the present invention is comprised the following steps:
Step 1-6-1-1, calculates dynamic convolutional layer DC2 error terms:The prognostic chart picture that step 1-5-2-2 is obtained and the group
The control label of sample asks poor, obtains the error term matrix delta that size is 240 × 240DC2;
Step 1-6-1-2, calculates dynamic convolutional layer DC1 error terms:By zero padding by DC2 layers of error term matrix deltaDC2
It is 240 × 320 to expand, by level probability Vector rotation 180 degree, by the level probability after the error term matrix after expansion and upset
Vectorial phase convolution, obtains DC1 layers of error term matrix deltaDC1, its size is 240 × 280, and formula is as follows:
δDC1=Expand_Zero (δDC2) * rot180 (HPV),
Wherein, Expand_Zero () represents zero extended function, and rot180 () represents the rotation letter that angle is 180 °
Number, 2 × 2 matrix zero is extended for 4 × 4 process as shown in figure 5, zero expand after matrix, central resolution ratio is 2 × 2
Region is consistent with original matrix, zero pixel filling of remaining position;
Step 1-6-1-3, calculates probability vector error term:The error term of calculated level probability vector HPV, by DC1 layers
Output characteristic figure and error term matrix deltaDC2Phase convolution, obtains 1 × 41 row vector after convolution, the vector is the error term of HPV
δHPV, formula is as follows:
δHPV=aDC1*δDC2,
The error term of vertical probability vector VPV is calculated, by the input feature vector figure of input layer and error term matrix deltaDC1Mutually roll up
Product, obtains 41 × 1 column vector after convolution, the vector is the error term δ of VPVVPV, formula is as follows:
Wherein,Last piece image in for the input image sequence of training sample;
Step 1-6-2 is comprised the following steps:
Step 1-6-2-1, calculates grader layer F1 error terms:The error term of the probability vector that step 1-6-1-3 is obtained
δVPVAnd δHPVWeighting parameter matrix W V vertical with F1 layers and horizontal weighting parameter matrix W H carry out matrix multiple respectively, then by square
The apposition of battle array is sued for peace and is averaged, and obtains F1 layers of error term δF1, formula is as follows:
Wherein, × representing matrix apposition, ()TThe transposition of matrix is represented, the δ for obtainingF1Size be 512 × 1;
Step 1-6-2-2, calculates convolutional layer C5 error terms:By matrixing, F1 layers for being obtained in step 1-6-2-1
Error term δF1It is transformed to the matrix that 32 resolution ratio are 4 × 4Obtain C5 layers of error term
Represent the matrix that the 32nd resolution ratio after conversion is 4 × 4;This time matrixing is mutual with matrixing in step 1-5-1-6
It is inverse transformation,
Step 1-6-2-3, judges network layer type:Internet (the l initial values in the sub-network being presently in are represented with l
It is S4), judge the type of Internet l, if l ∈ { S4, S3, S2, S1 }, then l is down-sampling layer, step 1-6-2-4 is performed, if l
∈ { C4, C3, C2, C1 }, then l is convolutional layer, performs step 1-6-2-5;
Step 1-6-2-4, calculates down-sampling layer error term:If l layers is down-sampling layer, now there is l=lS,lS∈{S1,S2,
S3, S4 }, for lSI-th error term matrix of layerBy zero padding respectively by lS+ 1 layer each error term matrix
Expand to width and beWherein lS+ 1 layer of latter convolutional layer for representing current down-sampling layer, j represents lS+ 1 layer
J-th error term, and haveAgain by corresponding convolution kernelRotation 180 degree, after then expanding
Matrix and upset after convolution nuclear phase convolution, and convolution results are sued for peace, obtain lSI-th error term matrix of layerIt is public
Formula is as follows:
Wherein,
All of error term matrix is calculated successively, obtains lSThe output characteristic figure of layerL is updated to l-1, and returns to step
Rapid 1-6-2-3 judges network type, calculates the error term of a upper Internet;
Step 1-6-2-5, calculates convolutional layer error term:If l layers is convolutional layer, now there is l=lC,lC∈{C1,C2,C3,
C4, C5 }, because the initial value of l in step 1-6-2-3 is S4, therefore be not in lCThe situation of=C5, for lCThe i-th of layer
Individual error term matrixFirst to lCCorresponding i-th error term matrix in+1 layerUp-sampled, will during up-samplingIn each element error entry value average mark to sample area in, obtaining resolution ratio is
Up-sampling matrix, then calculate activation primitive in lCDerivative and the inner product for up-sampling matrix tried to achieve at layer character pair figure,
Obtain lCI-th error term matrix of layerFormula is as follows:
Wherein, representing matrix inner product, ReLU'() derivative of ReLU activation primitives is represented, its form is as follows:
UpSample () represents up-sampling function, and Fig. 6 is the process up-sampled to 2 × 2 matrix, after up-sampling
One up-sampling region of each pixel correspondence in original image, each picture in original pixel value mean allocation to sample area
In vegetarian refreshments, all of error term matrix is calculated successively, obtain lCThe output characteristic figure of layer
Step 1-6-2-6, now l layers is convolutional layer, i.e. l=lC, two kinds of situations are divided into afterwards:
If l ≠ C1, l is updated to l-1, and return to step 1-6-2-3 judges network type, calculates a upper Internet
Error term;
If l=C1, step 1-6-2 sub-networks error term is calculated and terminated;
Step 1-6-3 is comprised the following steps:
Step 1-6-3-1, calculates gradient of the convolutional layer error term to convolution kernel:Use lCRepresent currently processed convolutional layer, lC
∈ { C1, C2, C3, C4, C5 }, successively calculates gradient of each convolutional layer error term to convolution kernel, by l since C1 layersC- 1 layer
I-th output characteristic figureWith lCJ-th error term matrix of layerPhase convolution, convolution results are the ladder of correspondence convolution kernel
Angle valueFormula is as follows:
In above formula,WithL is represented respectivelyCLayer and lC- 1 layer of output characteristic figure
Number, ▽ k are Grad of the error term to convolution kernel;
Step 1-6-3-2, calculates gradient of each convolutional layer error term to biasing:Use lCRepresent currently processed convolutional layer, lC
∈ { C1, C2, C3, C4, C5 }, successively calculates gradient of each convolutional layer error term to biasing, by l since C1 layersCJ-th of layer
Error term matrixIn all elements sued for peace, obtain the Grad of j-th of this layer biasingThe following institute of formula
Show:
Wherein, Sum () is represented and all elements of matrix is sued for peace;
Step 1-6-3-3, calculates gradient of the F1 layers of error term to weighting parameter:Respectively calculated level probability vector with it is vertical
The error term δ of probability vectorHPV、δVPVWith F1 layers of error term δF1Inner product, result of calculation be F1 layers of error term to weighting parameter WH,
The Grad of WV, formula is as follows:
▽ WH=(δHPV)T×(δF1)T,
▽ WV=δVPV×(δF1)T,
In above formula, ▽ WH are Grad of the error term to horizontal weighting parameter, and ▽ WV are error term to vertical weighting parameter
Grad;
Step 1-6-3-4, calculates gradient of the F1 layers of error term to offset parameter:By level probability it is vectorial with vertical probability to
The error term δ of amountHPV、δVPVRespectively as F1 layers of error term to Horizontal offset parameter BH and the Grad of vertical off setting parameter BV,
Formula is as follows:
▽ BH=(δHPV)T,
▽ BV=δVPV,
Wherein, ▽ BH are Grad of the error term to Horizontal offset parameter, and ▽ BV are error term to vertical off setting parameter
Grad;
Step 1-6-4 is comprised the following steps:
Step 1-6-4-1, updates each convolutional layer weighting parameter:Each convolutional layer error term pair that step 1-6-3-1 is obtained
The gradient of convolution kernel is multiplied by the learning rate of dynamic convolutional neural networks, obtains the correction term of convolution kernel, then by former convolution kernel and be somebody's turn to do
Correction term asks poor, the convolution kernel for being updatedFormula is as follows:
Wherein, λ is the e-learning rate of determination in step 1-3, λ=0.0001;
Step 1-6-4-2, updates each convolutional layer offset parameter:Each convolutional layer error term pair that step 1-6-3-2 is obtained
The gradient of biasing is multiplied by the learning rate of dynamic convolutional neural networks, obtains the correction term of offset parameter, then by former bias term and be somebody's turn to do
Correction term asks poor, the bias term for being updatedFormula is as follows:
Step 1-6-4-3, updates F1 layers of weighting parameter:The F1 layers of error term that step 1-6-3-3 is obtained is to weighting parameter
The Grad of WH and WV is multiplied by the learning rate of dynamic convolutional neural networks, obtains the correction term of weighting parameter, then former weights are joined
Number WH and WV asks poor with the correction term tried to achieve respectively, and the WH and WV for being updated, formula are as follows:
WH=WH- λ ▽ WH,
WV=WV- λ ▽ WV;
Step 1-6-4-4, updates F1 layers of offset parameter:The F1 layers of error term that step 1-6-3-4 is obtained is to offset parameter
The Grad of BH and BV is multiplied by the learning rate of dynamic convolutional neural networks, obtains the correction term of offset parameter, then original biasing is joined
Number BH and BV asks poor with the correction term tried to achieve respectively, and the BH and BV for being updated, formula are as follows:
BH=BH- λ ▽ BH,
BV=BV- λ ▽ BV.
Step 2 of the present invention is comprised the following steps:
Step 2-1, data prediction:Input test image set, specification is carried out to every piece image that test image is concentrated
Change is processed, and 280 × 280 floating number image will be converted into per piece image, then floating number image collection is divided, and is constructed
Test sample collection comprising TestsetSize group samples;
Step 2-2, read test sample:The TestsetSize groups test sample input that step 2-1 is obtained is by training
Dynamic convolutional neural networks in;
Step 2-3, propagated forward:The image sequence characteristic of input is extracted in a sub-network, obtains level probability vector
HPVtestWith vertical probability vector VPVtest;Probabilistic forecasting layer in, by the last piece image in input image sequence successively with
VPVtest、HPVtestPhase convolution, obtains the final extrapolated image of dynamic convolutional neural networks.
Step 2-1 of the present invention is comprised the following steps:
Step 2-1-1, sampling:The image that test image is concentrated is sequentially arranged, and constant duration is distributed, when
Between at intervals of 6 minutes, altogether comprising NTestWidth image, TestsetSize is determined by equation below:
If Mod (NTest, 4)=0
If Mod (NTest, 4) ≠ 0
After trying to achieve TestsetSize, retain test image by sampling and concentrate preceding 4 × TestsetSize+1 width image, adopt
Amount of images is set to meet requirement by deleting the last image of test image concentration during sample;
Step 2-1-2, image normalization:Image conversion, normalization operation, by original point are carried out to the image that sampling is obtained
Resolution is converted into the floating number image that resolution ratio is 280 × 280 for 2000 × 2000 coloured image;
Step 2-1-3, constructs test sample collection:The floating number image set obtained using step 2-1-2 constructs test sample
Collection, by every four adjacent images in floating number image set, i.e. { 4M+1,4M+2,4M+3,4M+4 } width image is defeated as one group
Enter sequence, by cutting, the central resolution ratio of reservation is 240 × 240 part as correspondence sample to [4 × (M+1)+1] width image
This control label, wherein being positive integer, and has M ∈ [0, TestsetSize-1] to obtain comprising TestsetSize group test specimens
This test sample collection;
Step 2-1-2 is comprised the following steps:
Step 2-1-2-1, image conversion:To reduce the computation complexity of convolutional neural networks training, by colored echo
Intensity CAPPI images are converted into gray level image, then by cutting the part for retaining that original image center resolution ratio is for 560 × 560,
By the image resolution ratio boil down to 280 × 280 after cutting, the gray-scale map that resolution ratio is 280 × 280 is obtained;
Step 2-1-2-2, data normalization:By each pixel in step 1-1-2-1 in the gray-scale map of acquisition
Value is mapped to [0~1] from [0~255], by obtaining the floating number image that resolution ratio is 280 × 280 after normalization;
Step 2-3 is comprised the following steps:
Step 2-3-1, calculates sub-network probability vector:In a sub-network by convolutional layer and the alternate treatment of down-sampling layer
The image sequence characteristic of input is extracted, is then processed by Softmax functions in grader layer, obtain level probability vector
HPVtestWith vertical probability vector VPVtest;
Step 2-3-2, calculates probabilistic forecasting layer output image:The VPV that step 2-3-1 is obtainedtestAnd HPVtestAs probability
The convolution kernel of prediction interval, by the last piece image in input image sequence successively with VPVtestAnd HPVtestPhase convolution, is moved
The final extrapolated image of state convolutional neural networks;
Step 2-3-1 is comprised the following steps:
Step 2-3-1-1, judges network layer type:The Internet in the sub-network being presently in is represented with p, network is judged
The type of layer p, p initial values are C1, if p ∈ { C1, C2, C3, C4, C5 }, then p is convolutional layer, step 2-3-1-2 is performed, if p ∈
{ S1, S2, S3, S4 }, then p is down-sampling layer, performs step 2-3-1-4;
Step 2-3-1-2, processes convolutional layer:Now there is p=pC,pC∈ { C1, C2, C3, C4, C5 }, calculates p firstCLayer
V-th output characteristic figureBy pCThe input feature vector figure convolution nuclear phase convolution corresponding with this layer respectively of layer, then by convolution results
Summation, summed result adds pCV-th biasing of layerProcessed by ReLU activation primitives again, obtain pCV-th of layer
Output characteristic figureComputing formula is as follows:
Wherein,It is pCU-th input feature vector figure of layer convolution kernel corresponding with v-th output characteristic figure, m is current
The output characteristic figure number of the previous down-sampling layer of convolutional layer,Represent pCU-th of layer is defeated
Enter characteristic pattern, while being also pC- 1 layer of u-th output characteristic figure, * representing matrix convolution works as pCDuring=C1, then pC- 1 layer is defeated
Enter layer.
All of output characteristic figure is calculated successively, obtains pCThe output characteristic figure of layerP is updated to p+1, and returns to step
Rapid 2-3-1-1 judges network type, carries out the operation of next Internet;
Step 2-3-1-3, treatment down-sampling layer:Now there is p=pS,pS∈ { S1, S2, S3, S4 }, step 2-3-1-2 is obtained
The output characteristic figure of the convolutional layer for arriving respectively withPhase convolution, then sampled with step-length as 2, sampling obtains pS
The output characteristic figure of layerComputing formula is as follows:
Wherein, Sample () represents the sampling processing that step-length is 2, pS- 1 previous convolution for representing current down-sampling layer
Layer,Represent pSThe output characteristic figure of layerIn j-th output characteristic figure, obtain pSThe output characteristic figure of layerAfterwards, by p
P+1 is updated to, and return to step 2-3-1-1 judges network type, carries out the operation of next Internet;
Step 2-3-1-4, calculates F1 layers of probability vector:If Internet p is grader layer, i.e. p=F1 is become by matrix
Change, be 4 × 4 output characteristic figure with row sequential deployment by the 32 width resolution ratio of C5, obtain that resolution ratio is 512 × 1 F1 layers
Output characteristic vectorThen respectively calculated level parameter matrix WH, Vertical Parameters matrix W V withApposition, by calculate tie
Fruit is sued for peace with Horizontal offset parameter BH, vertical off setting parameter BV respectively, and summed result obtains level after being processed through Softmax functions
Probability vector HPVtest, vertical probability vector VPVtest, computing formula is as follows:
By its vertical probability vector VPVtestTransposition, obtains final vertical probability vector;
Step 2-3-2 is comprised the following steps:
Step 2-3-2-1, predicts DC1 layers of vertical direction:By last width input picture of input layer and vertical probability to
Amount VPVtestPhase convolution, obtains the DC1 layers of output characteristic figure that resolution ratio is 240 × 280
Step 2-3-2-2, predicts DC2 layers of vertical direction:Step 2-3-2-1 is obtainedWith level probability vector
HPVtestPhase convolution, obtains the final extrapolated image of dynamic convolutional neural networks, and its resolution ratio is 240 × 240, by extrapolated image
Compared with the control label of corresponding test sample, judge that dynamic convolutional neural networks carry out the accuracy of Radar Echo Extrapolation.
The invention provides a kind of Radar Echo Extrapolation method based on dynamic convolutional neural networks, the technology is implemented
The method and approach of scheme are a lot, and the above is only the preferred embodiment of the present invention, it is noted that for the art
Those of ordinary skill for, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these change
Enter and retouch and also should be regarded as protection scope of the present invention.Each part being not known in the present embodiment can use prior art to add
To realize.
Claims (10)
1. a kind of Radar Echo Extrapolation method based on dynamic convolutional neural networks, it is characterised in that comprise the following steps:
Step 1, trains offline convolutional neural networks:Input training image collection, data prediction is carried out to training image collection, is obtained
Training sample set, the dynamic convolution neural network structure of design, and initialize network training parameter;It is dynamic using training sample set training
State convolutional neural networks, the orderly image sequence of input obtains a width prognostic chart by dynamic convolutional neural networks propagated forward
Picture, calculates the error between prognostic chart picture and control label, and the weighting parameter and offset parameter of network are updated by backpropagation,
This process is repeated until reaching training termination condition, convergent dynamic convolutional neural networks are obtained;
Step 2, online Radar Echo Extrapolation:Input test image set, data prediction is carried out to test chart image set, is tested
Sample set, in the dynamic convolutional neural networks that then will be obtained in test sample collection input step 1, by network propagated forward meter
Probability vector is calculated, and by last width radar return image in input image sequence and the probability vector phase convolution for obtaining, is obtained
To the Radar Echo Extrapolation image of prediction.
2. method according to claim 1, it is characterised in that step 1 is comprised the following steps:
Step 1-1, data prediction:Input training image collection, is carried out at standardization to every piece image that training image is concentrated
Reason, will be converted into 280 × 280 floating number image per piece image, floating number image collection be obtained, to floating number image collection
Divided, training sample set of the construction comprising TrainsetSize group samples;
Step 1-2, the dynamic convolutional neural networks of initialization:The dynamic convolution neural network structure of design, is configured to generating probability
The sub-network of vector, reconstructs the probabilistic forecasting layer for image extrapolation, for the offline neural metwork training stage provides dynamic volume
Product neutral net initialization model;
Step 1-3, the dynamic convolution Neural Network Training Parameter of initialization:E-learning rate λ=0.0001 is made, the training stage is each
The sample size BatchSize=10 of input, most large quantities of frequency of training of training sample set
Currently criticize frequency of training BatchNum=1, the maximum iteration IterationMax=40 of network training, current iteration number of times
IterationNum=1;
Step 1-4, reads training sample:By the way of batch training, the training sample that training is obtained from step 1-1 every time is concentrated
BatchSize group training samples are read, every group of training sample is { x1,x2,x3,x4, y }, altogether comprising 5 width images, wherein { x1,x2,
x3,x4Used as input image sequence, y is corresponding control label;
Step 1-5, propagated forward:The input image sequence feature that extraction step 1-4 is obtained in a sub-network, obtains level probability
Vectorial HPV and vertical probability vector VPV;Probabilistic forecasting layer in, by the last piece image in input image sequence successively with
VPV, HPV phase convolution, obtain the output prognostic chart picture of propagated forward;
Step 1-6, backpropagation:The error term of probability vector is reversely tried to achieve in probabilistic forecasting layer, according to the mistake of probability vector
Difference Xiang Conghou successively calculates the error term of each Internet in subnet network layers to preceding, and then calculates error term pair in each Internet
The gradient of weighting parameter and offset parameter, using the parameter of the gradient updating dynamic convolutional neural networks for obtaining;
Step 1-7, off-line training step control:Overall control is carried out to the offline neural metwork training stage, is divided into following three kinds
Situation:
If training sample is concentrated and still suffers from original training sample, i.e. BatchNum < BatchMax, then return to step 1-4
Continue to read BatchSize group training samples, carry out network training;
If training sample is concentrated does not exist original training sample, i.e. BatchNum=BatchMax, and current network changes
Generation number is less than maximum iteration, i.e. IterationNum < IterationMax, then make BatchNum=1, return to step
1-4 continues to read BatchSize group training samples, carries out network training;
If training sample is concentrated does not exist original training sample, i.e. BatchNum=BatchMax, and network iteration time
Number reaches maximum iteration, i.e. IterationNum=IterationMax, then terminate the offline neural metwork training stage, obtains
To the dynamic convolution neural network model for training.
3. method according to claim 2, it is characterised in that step 1-1 data predictions are comprised the following steps:
Step 1-1-1, sampling:The image that training image is concentrated is sequentially arranged, and constant duration is distributed, between the time
It is divided into 6 minutes, altogether comprising NTrainWidth image, TrainsetSize is determined by equation below:
Wherein, Mod (NTrain, 4) and represent NTrainTo 4 modulus,Expression is not more thanMaximum integer, try to achieve
After TrainsetSize, retain training image by sampling and concentrate preceding 4 × TrainsetSize+1 width image, by deleting during sampling
Concentrate last image to meet amount of images except training image to require;
Step 1-1-2, normalized images:Image conversion, normalization operation, by original resolution are carried out to the image that sampling is obtained
Coloured image for 2000 × 2000 is converted into the floating number image that resolution ratio is 280 × 280;
Step 1-1-3, constructs training sample set:The floating number image set obtained using step 1-1-2 constructs training sample set, will
Every four adjacent images in floating number image set, i.e., { 4N+1,4N+2,4N+3,4N+4 } width image is used as one group of input sequence
Row, by cutting, the central resolution ratio of reservation is 240 × 240 part as correspondence sample to [4 × (N+1)+1] width image
Control label, for N group samplesIts make is as follows:
In above formula, G4N+1Represent floating number image set in 4N+1 width images, N is positive integer, and have N ∈ [0,
TrainsetSize-1], Crop () represents trimming operation, and the portion that original image center size is for 240 × 240 is retained after cutting
Point, finally give the training sample set comprising TrainsetSize group training samples;
Wherein, step 1-1-2 is comprised the following steps:
Step 1-1-2-1, image conversion:The image that step 1-1-1 samplings are obtained is converted into gray level image, is retained by cutting
Original image center resolution ratio is 560 × 560 part, and the image resolution ratio boil down to 280 × 280 after cutting is divided
Resolution is 280 × 280 gray-scale map;
Step 1-1-2-2, data normalization:By in step 1-1-2-1 obtain gray-scale map in each pixel value from
[0~255] is mapped to [0~1], by obtaining the floating number image that resolution ratio is 280 × 280 after normalization.
4. method according to claim 3, it is characterised in that step 1-2 is comprised the following steps:
Step 1-2-1, constructs sub-network:
Sub-network is made up of 10 Internets, and convolutional layer C1, down-sampling layer S1, convolutional layer C2, down-sampling are followed successively by from front to back
Layer S2, convolutional layer C3, down-sampling layer S3, convolutional layer C4, down-sampling layer S5, convolutional layer C5 and grader layer F1;
Step 1-2-2, construction probabilistic forecasting layer:
Dynamic convolutional layer DC1 and dynamic convolutional layer DC2, the vertical probability vector that sub-network is exported are constructed in probabilistic forecasting layer
VPV as dynamic convolutional layer DC1 convolution kernel, level probability vector HPV as dynamic convolutional layer DC2 convolution kernel;
Wherein, step 1-2-1 is comprised the following steps:
Step 1-2-1-1, constructs convolutional layer:Determine herein below:The output characteristic figure quantity OutputMaps of convolutional layer, convolution
Core k and offset parameter bias, for convolution kernel, it is thus necessary to determine that the width KernelSize of convolution kernel, the quantity of convolution kernel
KernelNumber, the value is the product of convolutional layer input and output characteristic figure quantity, and the resolution ratio of convolution kernel is
KernelSize × KernelSize, and convolution kernel is constructed according to Xavier initial methods;For offset parameter, its quantity with
The output characteristic figure quantity of this layer is identical;For convolutional layer lC,lC∈ { C1, C2, C3, C4, C5 }, the output characteristic figure of this layer is wide
Spend and beValue by convolutional layer lCInput feature vector figure resolution ratio and convolution kernel widthTogether decide on, i.e., Represent volume
Lamination lCLast layer convolutional layer output characteristic figure width;
For convolutional layer C1, C1 layers of output characteristic figure quantity OutputMaps is madeC1=12, the C1 layer of width of output characteristic figure
OutputSizeC1=272, C1 layer of convolution kernel width KernelSizeC1=9, C1 layer of offset parameter biasC1Zero is initialized as,
C1 layers of convolution kernel kC1Quantity KernelNumberC1=48, the initial value of each parameter is in convolution kernelRand () is used to generate random number;
For convolutional layer C2, C2 layers of output characteristic figure quantity OutputMaps is madeC2=32, the C2 layer of width of output characteristic figure
OutputSizeC2=128, C2 layer of convolution kernel width KernelSizeC2=9, C2 layer of offset parameter is initialized as zero, C2 layers
Convolution kernel kC2Quantity KernelNumberC2=384, the initial value of each parameter is in convolution kernel
For convolutional layer C3, C3 layers of output characteristic figure quantity OutputMaps is madeC3=32, the C3 layer of width of output characteristic figure
OutputSizeC3=56, C3 layer of convolution kernel width KernelSizeC3=9, C3 layer of offset parameter is initialized as zero, C3 layers
Convolution kernel kC3Quantity KernelNumberC3=1024, the initial value of each parameter is in convolution kernel
For convolutional layer C4, C4 layers of output characteristic figure quantity OutputMaps is madeC4=32, the C4 layer of width of output characteristic figure
OutputSizeC4=20, C4 layer of convolution kernel width KernelSizeC4=9, C4 layer of offset parameter is initialized as zero, C4 layers
Convolution kernel kC4Quantity KernelNumberC4=1024, the initial value of each parameter is in convolution kernel
For convolutional layer C5, C5 layers of output characteristic figure quantity OutputMaps is madeC5=32, the C5 layer of width of output characteristic figure
OutputSizeC5=4, C5 layer of convolution kernel width KernelSizeC5=7, C5 layer of offset parameter is initialized as zero, C5 layers
Convolution kernel kC5Quantity KernelNumberC5=1024, the initial value of each parameter is in convolution kernel
Step 1-2-1-2, construction down-sampling layer:Down-sampling layer in not comprising need training parameter, by down-sampling layer S1, S2,
The sampling core of S3 and S4 is initialized asFor down-sampling layer lS,lS∈ { S1, S2, S3, S4 }, its output is special
Levy figure quantityOutput characteristic figure quantity with the convolutional layer of its last layer is consistent, output characteristic figure widthIt is the 1/2 of the output characteristic figure width of the convolutional layer of its last layer, formula is expressed as follows:
Step 1-2-1-3, structural classification device layer:Grader layer is made up of a full articulamentum F1, and F1 layers of weighting parameter is water
Equal rights value parameter matrix W H and vertical weighting parameter matrix W V, size is 41 × 512, makes each in weighting parameter matrix
The initial value of parameter isOffset parameter is Horizontal offset parameter BH and vertical off setting parameter BV, just
Beginning turns to 41 × 1 one-dimensional null vector.
5. method according to claim 4, it is characterised in that step 1-5 is comprised the following steps:
Step 1-5-1, sub-network calculates probability vector:Extracted by the alternate treatment of convolutional layer and down-sampling layer in a sub-network
The image sequence characteristic of input, grader layer in processed by Softmax functions, obtain level probability vector HPV with vertically
Probability vector VPV;
Step 1-5-2, calculates probabilistic forecasting layer output image:HPV that step 1-5-1 is obtained and VPV is used as probabilistic forecasting layer
Convolution kernel, by the last piece image in input image sequence successively with VPV, HPV phase convolution, the output for obtaining propagated forward is pre-
Altimetric image.
6. method according to claim 5, it is characterised in that step 1-5-1 is comprised the following steps:
Step 1-5-1-1, judges network layer type:The Internet in the sub-network being presently in is represented with l, l initial values are C1,
Judge the type of Internet l, if l ∈ { C1, C2, C3, C4, C5 }, then l is convolutional layer, performs step 1-5-1-2, if l ∈ S1,
S2, S3, S4 }, then l is down-sampling layer, performs step 1-5-1-4;
Step 1-5-1-2, processes convolutional layer:Now there is l=lC,lC∈ { C1, C2, C3, C4, C5 }, calculates l firstCThe jth of layer
Individual output characteristic figureBy lCThe input feature vector figure convolution nuclear phase convolution corresponding with this layer respectively of layer, convolution results are sued for peace,
Summed result adds lCJ-th offset parameter of layerProcessed by ReLU activation primitives again, obtainedComputing formula is such as
Shown in lower:
Wherein,It is lCI-th input feature vector figure of layer convolution kernel corresponding with j-th output characteristic figure, n is current convolution
The output characteristic figure number of the previous down-sampling layer of layer, Represent lCI-th input feature vector of layer
Figure, while being also lC- 1 layer of i-th output characteristic figure, * representing matrix convolution, if lC=C1, then lC- 1 layer is input layer;
All of output characteristic figure is calculated successively, obtains lCThe output characteristic figure of layerL is updated to l+1, and return to step 1-
5-1-1 judges network type, carries out the operation of next Internet;
Step 1-5-1-3, treatment down-sampling layer:Now there is l=lS,lS∈ { S1, S2, S3, S4 }, step 1-5-1-2 is obtained
The output characteristic figure of convolutional layer respectively withPhase convolution, then sampled with step-length as 2, sampling obtains lSLayer it is defeated
Go out characteristic patternComputing formula is as follows:
In above formula, Sample () represents the sampling processing that step-length is 2, lS- 1 previous convolutional layer for representing current down-sampling layer,
Represent lSThe output characteristic figure of layerIn j-th output characteristic figure, obtain lSThe output characteristic figure of layerAfterwards, l is updated to l
+ 1, and return to step 1-5-1-1 judges network type, carries out the operation of next Internet;
Step 1-5-1-4, calculates F1 layers of probability vector:If Internet l is grader layer, i.e. l=F1, by matrixing, will
The 32 width resolution ratio of C5 are 4 × 4 output characteristic figure with row sequential deployment, obtain F1 layer that resolution ratio is 512 × 1 of output spy
Levy vectorial aF1, difference calculated level weighting parameter matrix W H and aF1Apposition, vertical weighting parameter matrix W V and aF1Apposition,
Result of calculation is sued for peace with Horizontal offset parameter BH, vertical off setting parameter BV respectively, level is obtained after being processed through Softmax functions
Probability vector HPV and vertical probability vector VPV, specific formula for calculation is as follows:
By its vertical probability vector VPV transposition, final vertical probability vector is obtained;
Step 1-5-2 is comprised the following steps:
Step 1-5-2-1, predicts DC1 layers of vertical direction:By last width input picture of input layer and vertical probability vector VPV
Phase convolution, obtains the DC1 layers of output characteristic figure a that resolution ratio is 240 × 280DC1;
Step 1-5-2-2, predicts DC2 layers of vertical direction:By DC1 layers of output characteristic figure aDC1Rolled up with level probability vector HPV phases
Product, obtains the output prognostic chart picture of propagated forward, and its resolution ratio is 240 × 240.
7. method according to claim 6, it is characterised in that step 1-6 is comprised the following steps:
Step 1-6-1, calculates probabilistic forecasting layer error term:The prognostic chart picture that step 1-5-2-2 is obtained and the training sample being input into
Control label in this asks poor, calculates DC2 layers, DC1 layers of error term, finally tries to achieve the error term δ of level probability vectorHPVWith
The error term δ of vertical probability vectorVPV;
Step 1-6-2, calculates sub-network error term:According to the error term δ of level probability vectorHPVWith the mistake of vertical probability vector
Difference item δVPV, classification layer F1, convolutional layer C5, C4, C3, C2, C1 and down-sampling layer S4, S3, S2, S1 are calculated successively from rear to preceding
Error term, the resolution ratio of any layer error term matrix tried to achieve is consistent with the resolution ratio of the output characteristic figure of this layer;
Step 1-6-3, calculates gradient:The error term obtained according to step 1-6-2 calculates the error term of each Internet of sub-network
To this layer of weighting parameter and the Grad of offset parameter;
Step 1-6-4, undated parameter:The weighting parameter and the Grad of offset parameter of each Internet that step 1-6-3 is obtained
The learning rate of dynamic convolutional neural networks is multiplied by, the renewal of each Internet weighting parameter and offset parameter is obtained, by former weights
Parameter and offset parameter ask poor with the renewal respectively, weighting parameter and offset parameter after being updated.
8. method according to claim 7, it is characterised in that step 1-6-1 is comprised the following steps:
Step 1-6-1-1, calculates dynamic convolutional layer DC2 error terms:The prognostic chart picture that step 1-5-2-2 is obtained and this group of sample
Control label ask poor, obtain size be 240 × 240 error term matrix deltaDC2;
Step 1-6-1-2, calculates dynamic convolutional layer DC1 error terms:By zero padding by DC2 layers of error term matrix deltaDC2Expand
It is 240 × 320, by level probability Vector rotation 180 degree, by the level probability vector after the error term matrix after expansion and upset
Phase convolution, obtains DC1 layers of error term matrix deltaDC1, its size is 240 × 280, and formula is as follows:
δDC1=Expand_Zero (δDC2) * rot180 (HPV),
Wherein, Expand_Zero () represents zero extended function, and rot180 () represents the rotation function that angle is 180 °, by 2
× 2 matrix zero is extended for 4 × 4 matrix, the matrix after zero expansion, region and original matrix phase one of the central resolution ratio for 2 × 2
Cause, zero pixel filling of remaining position;
Step 1-6-1-3, calculates probability vector error term:The error term of calculated level probability vector HPV, by DC1 layers of output
Characteristic pattern and error term matrix deltaDC2Phase convolution, obtains 1 × 41 row vector after convolution, the vector is the error term δ of HPVHPV, it is public
Formula is as follows:
δHPV=aDC1*δDC2,
The error term of vertical probability vector VPV is calculated, by the input feature vector figure of input layer and error term matrix deltaDC1Phase convolution, volume
41 × 1 column vector is obtained after product, the vector is the error term δ of VPVVPV, formula is as follows:
Wherein,Last piece image in for the input image sequence of training sample;
Step 1-6-2 is comprised the following steps:
Step 1-6-2-1, calculates grader layer F1 error terms:The error term δ of the probability vector that step 1-6-1-3 is obtainedVPVWith
δHPVWeighting parameter matrix W V vertical with F1 layers and horizontal weighting parameter matrix W H carry out matrix multiple respectively, then by matrix
Apposition is sued for peace and is averaged, and obtains F1 layers of error term δF1, formula is as follows:
Wherein, × representing matrix apposition, ()TThe transposition of matrix is represented, the δ for obtainingF1Size be 512 × 1;
Step 1-6-2-2, calculates convolutional layer C5 error terms:By matrixing, the mistake of F1 layers for being obtained in step 1-6-2-1
Difference item δF1It is transformed to the matrix that 32 resolution ratio are 4 × 4Obtain C5 layers of error term δC5,Represent and become
The 32nd resolution ratio after changing is 4 × 4 matrix;
Step 1-6-2-3, judges network layer type:The Internet in the sub-network being presently in is represented with l, l initial values are S4,
Judge the type of Internet l, if l ∈ { S4, S3, S2, S1 }, then l is down-sampling layer, performs step 1-6-2-4, if l ∈ C4,
C3, C2, C1 }, then l is convolutional layer, performs step 1-6-2-5;
Step 1-6-2-4, calculates down-sampling layer error term:If l layers is down-sampling layer, now there is l=lS,lS∈{S1,S2,S3,
S4 }, for lSI-th error term matrix of layerBy zero padding respectively by lS+ 1 layer each error term matrixExpand
It is to widthWherein lS+ 1 layer of latter convolutional layer for representing current down-sampling layer, j represents lSJ-th of+1 layer
Error term, and haveAgain by corresponding convolution kernelRotation 180 degree, then by the matrix after expansion
With the convolution nuclear phase convolution after upset, and convolution results are sued for peace, obtain lSI-th error term matrix of layerFormula is as follows
It is shown:
Wherein,
All of error term matrix is calculated successively, obtains lSThe output characteristic figure of layerL is updated to l-1, and return to step 1-
6-2-3 judges network type, calculates the error term of a upper Internet;
Step 1-6-2-5, calculates convolutional layer error term:If l layers is convolutional layer, now there is l=lC,lC∈{C1,C2,C3,C4,
C5 }, because the initial value of l in step 1-6-2-3 is S4, therefore be not in lCThe situation of=C5, for lCI-th of layer
Error term matrixFirst to lCCorresponding i-th error term matrix in+1 layerUp-sampled, will during up-sampling
In each element error entry value average mark to sample area in, obtaining resolution ratio isIt is upper
Sampling matrix, then activation primitive is calculated in lCDerivative and the inner product of the up-sampling matrix tried to achieve at layer character pair figure, obtain lC
I-th error term matrix of layerFormula is as follows:
Wherein, representing matrix inner product, ReLU'() derivative of ReLU activation primitives is represented, its form is as follows:
UpSample () represents up-sampling function, the one up-sampling area of each pixel correspondence after up-sampling in original image
Domain, in each pixel in original pixel value mean allocation to sample area, calculates all of error term matrix successively, obtains
lCThe output characteristic figure of layer
Step 1-6-2-6, now l layers is convolutional layer, i.e. l=lC, two kinds of situations are divided into afterwards:
If l ≠ C1, l is updated to l-1, and return to step 1-6-2-3 judges network type, calculates the mistake of a upper Internet
Difference item;
If l=C1, step 1-6-2 sub-networks error term is calculated and terminated;
Step 1-6-3 is comprised the following steps:
Step 1-6-3-1, calculates gradient of the convolutional layer error term to convolution kernel:Use lCRepresent currently processed convolutional layer, lC∈
{ C1, C2, C3, C4, C5 }, successively calculates gradient of each convolutional layer error term to convolution kernel, by l since C1 layersCThe i-th of -1 layer
Individual output characteristic figureWith lCJ-th error term matrix of layerPhase convolution, convolution results are the gradient of correspondence convolution kernel
ValueFormula is as follows:
In above formula,WithL is represented respectivelyCThe output characteristic figure number of layer and lC- 1 layer
Output characteristic figure number, ▽ k are Grad of the error term to convolution kernel;
Step 1-6-3-2, calculates gradient of each convolutional layer error term to biasing:Use lCRepresent currently processed convolutional layer, lC∈
{ C1, C2, C3, C4, C5 }, successively calculates gradient of each convolutional layer error term to biasing, by l since C1 layersCJ-th mistake of layer
Difference item matrixIn all elements sued for peace, obtain the Grad of j-th of this layer biasingFormula is as follows:
Wherein, Sum () is represented and all elements of matrix is sued for peace;
Step 1-6-3-3, calculates gradient of the F1 layers of error term to weighting parameter:Difference calculated level probability vector and vertical probability
The error term δ of vectorHPV、δVPVWith F1 layers of error term δF1Inner product, result of calculation be F1 layers of error term to weighting parameter WH, WV
Grad, formula is as follows:
▽ WH=(δHPV)T×(δF1)T,
▽ WV=δVPV×(δF1)T,
In above formula, ▽ WH are Grad of the error term to horizontal weighting parameter, and ▽ WV are ladder of the error term to vertical weighting parameter
Angle value;
Step 1-6-3-4, calculates gradient of the F1 layers of error term to offset parameter:Level probability is vectorial with vertical probability vector
Error term δHPV、δVPVRespectively as F1 layers of error term to Horizontal offset parameter BH and the Grad of vertical off setting parameter BV, formula
It is as follows:
▽ BH=(δHPV)T,
▽ BV=δVPV,
Wherein, ▽ BH are Grad of the error term to Horizontal offset parameter, and ▽ BV are gradient of the error term to vertical off setting parameter
Value;
Step 1-6-4 is comprised the following steps:
Step 1-6-4-1, updates each convolutional layer weighting parameter:Each convolutional layer error term that step 1-6-3-1 is obtained is to convolution
The gradient of core is multiplied by the learning rate of dynamic convolutional neural networks, obtains the correction term of convolution kernel, then by former convolution kernel and the amendment
Item asks poor, the convolution kernel for being updatedFormula is as follows:
Wherein, λ is the e-learning rate of determination in step 1-3, λ=0.0001;
Step 1-6-4-2, updates each convolutional layer offset parameter:Each convolutional layer error term that step 1-6-3-2 is obtained is to biasing
Gradient be multiplied by the learning rate of dynamic convolutional neural networks, obtain the correction term of offset parameter, then by former bias term and the amendment
Item asks poor, the bias term for being updatedFormula is as follows:
Step 1-6-4-3, updates F1 layers of weighting parameter:By step 1-6-3-3 obtain F1 layer error term to weighting parameter WH with
The Grad of WV is multiplied by the learning rate of dynamic convolutional neural networks, obtains the correction term of weighting parameter, then by former weighting parameter WH
Ask poor with the correction term tried to achieve respectively with WV, the WH and WV for being updated, formula are as follows:
WH=WH- λ ▽ WH,
WV=WV- λ ▽ WV;
Step 1-6-4-4, updates F1 layers of offset parameter:By step 1-6-3-4 obtain F1 layer error term to offset parameter BH with
The Grad of BV is multiplied by the learning rate of dynamic convolutional neural networks, obtains the correction term of offset parameter, then by former offset parameter BH
Ask poor with the correction term tried to achieve respectively with BV, the BH and BV for being updated, formula are as follows:
BH=BH- λ ▽ BH,
BV=BV- λ ▽ BV.
9. method according to claim 8, it is characterised in that characterized in that, step 2 is comprised the following steps:
Step 2-1, data prediction:Input test image set, is carried out at standardization to every piece image that test image is concentrated
Reason, will be converted into 280 × 280 floating number image per piece image, then floating number image collection is divided, and construction is included
The test sample collection of TestsetSize group samples;
Step 2-2, read test sample:The TestsetSize groups test sample input that step 2-1 is obtained is trained dynamic
In state convolutional neural networks;
Step 2-3, propagated forward:The image sequence characteristic of input is extracted in a sub-network, obtains level probability vector HPVtestWith
Vertical probability vector VPVtest;Probabilistic forecasting layer in, by the last piece image in input image sequence successively with VPVtest、
HPVtestPhase convolution, obtains the final extrapolated image of dynamic convolutional neural networks.
10. method according to claim 9, it is characterised in that step 2-1 is comprised the following steps:
Step 2-1-1, sampling:The image that test image is concentrated is sequentially arranged, and constant duration is distributed, between the time
It is divided into 6 minutes, altogether comprising NTestWidth image, TestsetSize is determined by equation below:
If Mod (NTest, 4)=0
If Mod (NTest, 4) ≠ 0
After trying to achieve TestsetSize, retain test image by sampling and concentrate preceding 4 × TestsetSize+1 width image, during sampling
Amount of images is set to meet requirement by deleting the last image of test image concentration;
Step 2-1-2, image normalization:Image conversion, normalization operation, by original resolution are carried out to the image that sampling is obtained
Coloured image for 2000 × 2000 is converted into the floating number image that resolution ratio is 280 × 280;
Step 2-1-3, constructs test sample collection:The floating number image set obtained using step 2-1-2 constructs test sample collection, will
Every four adjacent images in floating number image set, i.e., { 4M+1,4M+2,4M+3,4M+4 } width image is used as one group of input sequence
Row, by cutting, the central resolution ratio of reservation is 240 × 240 part as correspondence sample to [4 × (M+1)+1] width image
Control label, wherein being positive integer, and has M ∈ [0, TestsetSize-1] to obtain comprising TestsetSize group test samples
Test sample collection;
Step 2-1-2 is comprised the following steps:
Step 2-1-2-1, image conversion:Colored echo strength CAPPI images are converted into gray level image, then are protected by cutting
It is 560 × 560 part to stay original image center resolution ratio, and the image resolution ratio boil down to 280 × 280 after cutting is obtained
Resolution ratio is 280 × 280 gray-scale map;
Step 2-1-2-2, data normalization:By in step 1-1-2-1 obtain gray-scale map in each pixel value from
[0~255] is mapped to [0~1], by obtaining the floating number image that resolution ratio is 280 × 280 after normalization;
Step 2-3 is comprised the following steps:
Step 2-3-1, calculates sub-network probability vector:Extracted by the alternate treatment of convolutional layer and down-sampling layer in a sub-network
The image sequence characteristic of input, is then processed in grader layer by Softmax functions, obtains level probability vector HPVtest
With vertical probability vector VPVtest;
Step 2-3-2, calculates probabilistic forecasting layer output image:The VPV that step 2-3-1 is obtainedtestAnd HPVtestAs probabilistic forecasting
Layer convolution kernel, by the last piece image in input image sequence successively with VPVtestAnd HPVtestPhase convolution, obtains dynamic volume
The final extrapolated image of product neutral net;
Step 2-3-1 is comprised the following steps:
Step 2-3-1-1, judges network layer type:The Internet in the sub-network being presently in is represented with p, Internet p is judged
Type, p initial values be C1, if p ∈ { C1, C2, C3, C4, C5 }, then p be convolutional layer, perform step 2-3-1-2, if p ∈
{ S1, S2, S3, S4 }, then p is down-sampling layer, performs step 2-3-1-4;
Step 2-3-1-2, processes convolutional layer:Now there is p=pC,pC∈ { C1, C2, C3, C4, C5 }, calculates p firstCThe v of layer
Individual output characteristic figureBy pCThe input feature vector figure convolution nuclear phase convolution corresponding with this layer respectively of layer, then convolution results are asked
With summed result adds pCV-th biasing of layerProcessed by ReLU activation primitives again, obtain pCV-th of layer is defeated
Go out characteristic patternComputing formula is as follows:
Wherein,It is pCU-th input feature vector figure of layer convolution kernel corresponding with v-th output characteristic figure, m is current convolution
The output characteristic figure number of the previous down-sampling layer of layer, Represent pCU-th input feature vector of layer
Figure, while being also pC- 1 layer of u-th output characteristic figure, * representing matrix convolution works as pCDuring=C1, then pC- 1 layer is input layer.
All of output characteristic figure is calculated successively, obtains pCThe output characteristic figure of layerP is updated to p+1, and return to step 2-
3-1-1 judges network type, carries out the operation of next Internet;
Step 2-3-1-3, treatment down-sampling layer:Now there is p=pS,pS∈ { S1, S2, S3, S4 }, step 2-3-1-2 is obtained
The output characteristic figure of convolutional layer respectively withPhase convolution, then sampled with step-length as 2, sampling obtains pSLayer
Output characteristic figureComputing formula is as follows:
Wherein, Sample () represents the sampling processing that step-length is 2, pS- 1 previous convolutional layer for representing current down-sampling layer,
Represent pSThe output characteristic figure of layerIn j-th output characteristic figure, obtain pSThe output characteristic figure of layerAfterwards, p is updated to
P+1, and return to step 2-3-1-1 judges network type, carries out the operation of next Internet;
Step 2-3-1-4, calculates F1 layers of probability vector:If Internet p is grader layer, i.e. p=F1, by matrixing, will
The 32 width resolution ratio of C5 are 4 × 4 output characteristic figure with row sequential deployment, obtain F1 layer that resolution ratio is 512 × 1 of output spy
Levy vectorThen respectively calculated level parameter matrix WH, Vertical Parameters matrix W V withApposition, result of calculation is distinguished
With Horizontal offset parameter BH, vertical off setting parameter BV summation, summed result through Softmax functions process after obtain level probability to
Amount HPVtest, vertical probability vector VPVtest, computing formula is as follows:
By its vertical probability vector VPVtestTransposition, obtains final vertical probability vector;
Step 2-3-2 is comprised the following steps:
Step 2-3-2-1, predicts DC1 layers of vertical direction:By last width input picture and vertical probability vector of input layer
VPVtestPhase convolution, obtains the DC1 layers of output characteristic figure that resolution ratio is 240 × 280
Step 2-3-2-2, predicts DC2 layers of vertical direction:Step 2-3-2-1 is obtainedWith level probability vector HPVtestPhase
Convolution, obtains the final extrapolated image of dynamic convolutional neural networks, and its resolution ratio is 240 × 240.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710110183.6A CN106886023B (en) | 2017-02-27 | 2017-02-27 | A kind of Radar Echo Extrapolation method based on dynamic convolutional neural networks |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710110183.6A CN106886023B (en) | 2017-02-27 | 2017-02-27 | A kind of Radar Echo Extrapolation method based on dynamic convolutional neural networks |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106886023A true CN106886023A (en) | 2017-06-23 |
CN106886023B CN106886023B (en) | 2019-04-02 |
Family
ID=59180072
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710110183.6A Active CN106886023B (en) | 2017-02-27 | 2017-02-27 | A kind of Radar Echo Extrapolation method based on dynamic convolutional neural networks |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106886023B (en) |
Cited By (32)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107748942A (en) * | 2017-11-24 | 2018-03-02 | 清华大学 | Radar Echo Extrapolation Forecasting Methodology and system based on velocity field sensing network |
CN108229404A (en) * | 2018-01-09 | 2018-06-29 | 东南大学 | A kind of radar echo signal target identification method based on deep learning |
CN108427989A (en) * | 2018-06-12 | 2018-08-21 | 中国人民解放军国防科技大学 | Deep space-time prediction neural network training method for radar echo extrapolation |
CN108508505A (en) * | 2018-02-05 | 2018-09-07 | 南京云思创智信息科技有限公司 | Heavy showers and thunderstorm forecasting procedure based on multiple dimensioned convolutional neural networks and system |
CN108648457A (en) * | 2018-06-28 | 2018-10-12 | 苏州大学 | A kind of method, apparatus and computer readable storage medium of prediction of speed |
CN108732550A (en) * | 2018-08-01 | 2018-11-02 | 北京百度网讯科技有限公司 | Method and apparatus for predicting radar return |
CN108734357A (en) * | 2018-05-29 | 2018-11-02 | 北京佳格天地科技有限公司 | Weather prognosis system and method |
CN108846409A (en) * | 2018-04-28 | 2018-11-20 | 中国人民解放军国防科技大学 | Radar echo extrapolation model training method based on cyclic dynamic convolution neural network |
CN108872993A (en) * | 2018-04-28 | 2018-11-23 | 中国人民解放军国防科技大学 | Radar echo extrapolation method based on cyclic dynamic convolution neural network |
CN109001736A (en) * | 2018-06-12 | 2018-12-14 | 中国人民解放军国防科技大学 | Radar echo extrapolation method based on deep space-time prediction neural network |
CN109283505A (en) * | 2018-09-03 | 2019-01-29 | 南京信息工程大学 | A method of amendment Radar Echo Extrapolation image Divergent Phenomenon |
CN109444863A (en) * | 2018-10-23 | 2019-03-08 | 广西民族大学 | A kind of estimation method of the narrowband ultrasonic echo number based on convolutional neural networks |
CN109754357A (en) * | 2018-01-26 | 2019-05-14 | 京东方科技集团股份有限公司 | Image processing method, processing unit and processing equipment |
CN109871829A (en) * | 2019-03-15 | 2019-06-11 | 北京行易道科技有限公司 | A kind of detection model training method and device based on deep learning |
CN109884625A (en) * | 2019-02-22 | 2019-06-14 | 中国人民解放军军事科学院国防科技创新研究院 | Radar relevance imaging method based on convolutional neural networks |
CN109919295A (en) * | 2017-12-12 | 2019-06-21 | 北京大学深圳研究生院 | A kind of embedded audio event detecting method based on lightweight convolutional neural networks |
CN110009150A (en) * | 2019-04-02 | 2019-07-12 | 中国石油大学(华东) | The compact reservoir physical parameter intelligent Forecasting of data-driven |
CN110568442A (en) * | 2019-10-15 | 2019-12-13 | 中国人民解放军国防科技大学 | Radar echo extrapolation method based on confrontation extrapolation neural network |
CN111062410A (en) * | 2019-11-05 | 2020-04-24 | 复旦大学 | Star information bridge weather prediction method based on deep learning |
CN111142109A (en) * | 2019-12-30 | 2020-05-12 | 上海眼控科技股份有限公司 | Marking method, marking device, computer equipment and storage medium |
WO2020124948A1 (en) * | 2018-12-21 | 2020-06-25 | 中科寒武纪科技股份有限公司 | Network offline model processing method, artificial intelligence processing device, and related product |
CN111488887A (en) * | 2020-04-09 | 2020-08-04 | 腾讯科技(深圳)有限公司 | Image processing method and device based on artificial intelligence |
CN111507474A (en) * | 2020-06-18 | 2020-08-07 | 四川大学 | Neural network distributed training method for dynamically adjusting Batch-size |
CN111736157A (en) * | 2020-08-26 | 2020-10-02 | 蔻斯科技(上海)有限公司 | PPI data-based prediction method and device for nowcasting |
CN112116060A (en) * | 2019-06-21 | 2020-12-22 | 杭州海康威视数字技术股份有限公司 | Network configuration implementation method and device |
CN112232361A (en) * | 2020-10-13 | 2021-01-15 | 国网电子商务有限公司 | Image processing method and device, electronic equipment and computer readable storage medium |
CN113239722A (en) * | 2021-03-31 | 2021-08-10 | 成都信息工程大学 | Deep learning based strong convection extrapolation method and system under multi-scale |
CN113640769A (en) * | 2021-08-27 | 2021-11-12 | 南京信息工程大学 | Weather radar basic reflectivity prediction method based on deep neural network |
CN113657477A (en) * | 2021-08-10 | 2021-11-16 | 南宁五加五科技有限公司 | Method, device and system for forecasting short-term rainfall |
CN115390164A (en) * | 2022-10-27 | 2022-11-25 | 南京信息工程大学 | Radar echo extrapolation forecasting method and system |
CN115755227A (en) * | 2023-01-10 | 2023-03-07 | 南京信大气象科学技术研究院有限公司 | Three-dimensional radar extrapolation method based on deep neural network model |
US11699073B2 (en) | 2018-12-29 | 2023-07-11 | Cambricon Technologies Corporation Limited | Network off-line model processing method, artificial intelligence processing device and related products |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100321229A1 (en) * | 2005-10-28 | 2010-12-23 | Raytheon Company | Biometric radar system and method for identifying persons and positional states of persons |
CN104537387A (en) * | 2014-12-16 | 2015-04-22 | 广州中国科学院先进技术研究所 | Method and system for classifying automobile types based on neural network |
CN104811276A (en) * | 2015-05-04 | 2015-07-29 | 东南大学 | DL-CNN (deep leaning-convolutional neutral network) demodulator for super-Nyquist rate communication |
CN106127725A (en) * | 2016-05-16 | 2016-11-16 | 北京工业大学 | A kind of millimetre-wave radar cloud atlas dividing method based on multiresolution CNN |
CN106228201A (en) * | 2016-06-20 | 2016-12-14 | 电子科技大学 | A kind of anti-Deceiving interference method of synthetic aperture radar based on shade characteristic |
-
2017
- 2017-02-27 CN CN201710110183.6A patent/CN106886023B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100321229A1 (en) * | 2005-10-28 | 2010-12-23 | Raytheon Company | Biometric radar system and method for identifying persons and positional states of persons |
CN104537387A (en) * | 2014-12-16 | 2015-04-22 | 广州中国科学院先进技术研究所 | Method and system for classifying automobile types based on neural network |
CN104811276A (en) * | 2015-05-04 | 2015-07-29 | 东南大学 | DL-CNN (deep leaning-convolutional neutral network) demodulator for super-Nyquist rate communication |
CN106127725A (en) * | 2016-05-16 | 2016-11-16 | 北京工业大学 | A kind of millimetre-wave radar cloud atlas dividing method based on multiresolution CNN |
CN106228201A (en) * | 2016-06-20 | 2016-12-14 | 电子科技大学 | A kind of anti-Deceiving interference method of synthetic aperture radar based on shade characteristic |
Cited By (46)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107748942B (en) * | 2017-11-24 | 2018-12-11 | 清华大学 | Radar Echo Extrapolation prediction technique and system based on velocity field sensing network |
CN107748942A (en) * | 2017-11-24 | 2018-03-02 | 清华大学 | Radar Echo Extrapolation Forecasting Methodology and system based on velocity field sensing network |
CN109919295A (en) * | 2017-12-12 | 2019-06-21 | 北京大学深圳研究生院 | A kind of embedded audio event detecting method based on lightweight convolutional neural networks |
CN109919295B (en) * | 2017-12-12 | 2022-10-28 | 北京大学深圳研究生院 | Embedded audio event detection method based on lightweight convolutional neural network |
CN108229404A (en) * | 2018-01-09 | 2018-06-29 | 东南大学 | A kind of radar echo signal target identification method based on deep learning |
CN108229404B (en) * | 2018-01-09 | 2022-03-08 | 东南大学 | Radar echo signal target identification method based on deep learning |
CN109754357A (en) * | 2018-01-26 | 2019-05-14 | 京东方科技集团股份有限公司 | Image processing method, processing unit and processing equipment |
US11281938B2 (en) | 2018-01-26 | 2022-03-22 | Boe Technology Group Co., Ltd. | Image processing method, processing apparatus and processing device |
CN108508505A (en) * | 2018-02-05 | 2018-09-07 | 南京云思创智信息科技有限公司 | Heavy showers and thunderstorm forecasting procedure based on multiple dimensioned convolutional neural networks and system |
CN108846409A (en) * | 2018-04-28 | 2018-11-20 | 中国人民解放军国防科技大学 | Radar echo extrapolation model training method based on cyclic dynamic convolution neural network |
CN108872993A (en) * | 2018-04-28 | 2018-11-23 | 中国人民解放军国防科技大学 | Radar echo extrapolation method based on cyclic dynamic convolution neural network |
CN108734357A (en) * | 2018-05-29 | 2018-11-02 | 北京佳格天地科技有限公司 | Weather prognosis system and method |
CN109001736B (en) * | 2018-06-12 | 2022-04-05 | 中国人民解放军国防科技大学 | Radar echo extrapolation method based on deep space-time prediction neural network |
CN109001736A (en) * | 2018-06-12 | 2018-12-14 | 中国人民解放军国防科技大学 | Radar echo extrapolation method based on deep space-time prediction neural network |
CN108427989B (en) * | 2018-06-12 | 2019-10-11 | 中国人民解放军国防科技大学 | Deep space-time prediction neural network training method for radar echo extrapolation |
CN108427989A (en) * | 2018-06-12 | 2018-08-21 | 中国人民解放军国防科技大学 | Deep space-time prediction neural network training method for radar echo extrapolation |
CN108648457A (en) * | 2018-06-28 | 2018-10-12 | 苏州大学 | A kind of method, apparatus and computer readable storage medium of prediction of speed |
CN108648457B (en) * | 2018-06-28 | 2021-07-13 | 苏州大学 | Method, device and computer readable storage medium for speed prediction |
CN108732550A (en) * | 2018-08-01 | 2018-11-02 | 北京百度网讯科技有限公司 | Method and apparatus for predicting radar return |
CN109283505A (en) * | 2018-09-03 | 2019-01-29 | 南京信息工程大学 | A method of amendment Radar Echo Extrapolation image Divergent Phenomenon |
CN109444863A (en) * | 2018-10-23 | 2019-03-08 | 广西民族大学 | A kind of estimation method of the narrowband ultrasonic echo number based on convolutional neural networks |
WO2020124948A1 (en) * | 2018-12-21 | 2020-06-25 | 中科寒武纪科技股份有限公司 | Network offline model processing method, artificial intelligence processing device, and related product |
US11699073B2 (en) | 2018-12-29 | 2023-07-11 | Cambricon Technologies Corporation Limited | Network off-line model processing method, artificial intelligence processing device and related products |
CN109884625A (en) * | 2019-02-22 | 2019-06-14 | 中国人民解放军军事科学院国防科技创新研究院 | Radar relevance imaging method based on convolutional neural networks |
CN109871829A (en) * | 2019-03-15 | 2019-06-11 | 北京行易道科技有限公司 | A kind of detection model training method and device based on deep learning |
CN109871829B (en) * | 2019-03-15 | 2021-06-04 | 北京行易道科技有限公司 | Detection model training method and device based on deep learning |
CN110009150A (en) * | 2019-04-02 | 2019-07-12 | 中国石油大学(华东) | The compact reservoir physical parameter intelligent Forecasting of data-driven |
CN112116060B (en) * | 2019-06-21 | 2023-07-25 | 杭州海康威视数字技术股份有限公司 | Network configuration implementation method and device |
CN112116060A (en) * | 2019-06-21 | 2020-12-22 | 杭州海康威视数字技术股份有限公司 | Network configuration implementation method and device |
CN110568442A (en) * | 2019-10-15 | 2019-12-13 | 中国人民解放军国防科技大学 | Radar echo extrapolation method based on confrontation extrapolation neural network |
CN110568442B (en) * | 2019-10-15 | 2021-08-20 | 中国人民解放军国防科技大学 | Radar echo extrapolation method based on confrontation extrapolation neural network |
CN111062410A (en) * | 2019-11-05 | 2020-04-24 | 复旦大学 | Star information bridge weather prediction method based on deep learning |
CN111142109A (en) * | 2019-12-30 | 2020-05-12 | 上海眼控科技股份有限公司 | Marking method, marking device, computer equipment and storage medium |
CN111488887B (en) * | 2020-04-09 | 2023-04-18 | 腾讯科技(深圳)有限公司 | Image processing method and device based on artificial intelligence |
CN111488887A (en) * | 2020-04-09 | 2020-08-04 | 腾讯科技(深圳)有限公司 | Image processing method and device based on artificial intelligence |
CN111507474B (en) * | 2020-06-18 | 2022-07-01 | 四川大学 | Neural network distributed training method for dynamically adjusting Batch-size |
CN111507474A (en) * | 2020-06-18 | 2020-08-07 | 四川大学 | Neural network distributed training method for dynamically adjusting Batch-size |
CN111736157B (en) * | 2020-08-26 | 2021-01-05 | 蔻斯科技(上海)有限公司 | PPI data-based prediction method and device for nowcasting |
CN111736157A (en) * | 2020-08-26 | 2020-10-02 | 蔻斯科技(上海)有限公司 | PPI data-based prediction method and device for nowcasting |
CN112232361A (en) * | 2020-10-13 | 2021-01-15 | 国网电子商务有限公司 | Image processing method and device, electronic equipment and computer readable storage medium |
CN113239722A (en) * | 2021-03-31 | 2021-08-10 | 成都信息工程大学 | Deep learning based strong convection extrapolation method and system under multi-scale |
CN113657477B (en) * | 2021-08-10 | 2022-04-08 | 南宁五加五科技有限公司 | Method, device and system for forecasting short-term rainfall |
CN113657477A (en) * | 2021-08-10 | 2021-11-16 | 南宁五加五科技有限公司 | Method, device and system for forecasting short-term rainfall |
CN113640769A (en) * | 2021-08-27 | 2021-11-12 | 南京信息工程大学 | Weather radar basic reflectivity prediction method based on deep neural network |
CN115390164A (en) * | 2022-10-27 | 2022-11-25 | 南京信息工程大学 | Radar echo extrapolation forecasting method and system |
CN115755227A (en) * | 2023-01-10 | 2023-03-07 | 南京信大气象科学技术研究院有限公司 | Three-dimensional radar extrapolation method based on deep neural network model |
Also Published As
Publication number | Publication date |
---|---|
CN106886023B (en) | 2019-04-02 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106886023B (en) | A kind of Radar Echo Extrapolation method based on dynamic convolutional neural networks | |
Zhou et al. | Forecasting different types of convective weather: A deep learning approach | |
Yu et al. | Comparison of random forests and support vector machine for real-time radar-derived rainfall forecasting | |
Ustaoglu et al. | Forecast of daily mean, maximum and minimum temperature time series by three artificial neural network methods | |
CN108846409A (en) | Radar echo extrapolation model training method based on cyclic dynamic convolution neural network | |
CN106355151A (en) | Recognition method, based on deep belief network, of three-dimensional SAR images | |
CN109800628A (en) | A kind of network structure and detection method for reinforcing SSD Small object pedestrian detection performance | |
CN105787439A (en) | Depth image human body joint positioning method based on convolution nerve network | |
CN113496104B (en) | Precipitation prediction correction method and system based on deep learning | |
CN107016357A (en) | A kind of video pedestrian detection method based on time-domain convolutional neural networks | |
CN103258324B (en) | Based on the method for detecting change of remote sensing image that controlled kernel regression and super-pixel are split | |
Kovordányi et al. | Cyclone track forecasting based on satellite images using artificial neural networks | |
CN109766936A (en) | Image change detection method based on information transmitting and attention mechanism | |
CN107563411A (en) | Online SAR target detection method based on deep learning | |
CN107967474A (en) | A kind of sea-surface target conspicuousness detection method based on convolutional neural networks | |
CN103366371B (en) | Based on K distribution and the SAR image segmentation method of textural characteristics | |
CN105741309B (en) | A kind of remote sensing image variation detection method based on the conversion of card side and samples selection | |
CN111881971A (en) | Power transmission line fault type identification method based on deep learning LSTM model | |
CN104680151B (en) | A kind of panchromatic remote sensing image variation detection method of high-resolution for taking snow covering influence into account | |
CN103366365A (en) | SAR image varying detecting method based on artificial immunity multi-target clustering | |
CN104361351A (en) | Synthetic aperture radar (SAR) image classification method on basis of range statistics similarity | |
CN107607942A (en) | Based on the large scale electromagnetic scattering of deep learning model and the Forecasting Methodology of back scattering | |
CN104751185A (en) | SAR image change detection method based on mean shift genetic clustering | |
CN109143408A (en) | Combine short-term precipitation forecasting procedure in dynamic area based on MLP | |
CN105117736A (en) | Polarized SAR image classification method based on sparse depth stack network |
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 |