CN106611423B - SAR image segmentation method based on ridge ripple filter and deconvolution structural model - Google Patents
SAR image segmentation method based on ridge ripple filter and deconvolution structural model Download PDFInfo
- Publication number
- CN106611423B CN106611423B CN201611269937.4A CN201611269937A CN106611423B CN 106611423 B CN106611423 B CN 106611423B CN 201611269937 A CN201611269937 A CN 201611269937A CN 106611423 B CN106611423 B CN 106611423B
- Authority
- CN
- China
- Prior art keywords
- sketch
- pixel
- sar image
- line
- indicate
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
- G06T2207/10044—Radar image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Image Analysis (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The SAR image segmentation method based on ridge ripple filter and deconvolution structural model that the invention discloses a kind of mainly solves the problems, such as prior art segmentation SAR image inaccuracy.Implementation step is: (1) SAR image sketch obtains sketch map;(2) according to the administrative division map of SAR image, the pixel subspace of SAR image is divided;(3) ridge ripple filter set is constructed;(4) deconvolution structural model is constructed;(5) SAR image segmentation method based on ridge ripple filter and deconvolution structural model, segmentation mixing aggregated structure atural object pixel subspace are used;(6) gather the pinpoint target segmentation of feature based on sketch line;(7) the line target segmentation of view-based access control model semantic rules;(8) homogenous region pixel subspace is split using based on multinomial logistic regression prior model;(9) combination and segmentation is as a result, obtain SAR image segmentation result.Present invention obtains the good segmentation effects of SAR image, can be used for the semantic segmentation of SAR image.
Description
Technical field
The invention belongs to technical field of image processing, further relate to one of target identification technology field and are based on ridge
Image segmentation side synthetic aperture radar SAR (Synthetic Aperture Radar) of wave filter and deconvolution structural model
Method.The present invention can have the region of different characteristic to be accurately split synthetic aperture radar SAR image, and can be used for
The object detection and recognition of subsequent synthetic aperture radar SAR image.
Background technique
Synthetic aperture radar SAR is the impressive progress in remote sensing technology field, for obtaining the full resolution pricture of earth surface.
Compared with other kinds of imaging technique, SAR has very important advantage, it is not by atmosphere such as cloud layer, rainfall or dense fogs
The influence of condition and intensity of illumination, can round-the-clock, round-the-clock obtain high resolution remote sensing data.SAR technology for it is military,
Many fields such as agricultural, geography have great importance.Image segmentation refers to will scheme according to color, gray scale and Texture eigenvalue
Process as being divided into several mutually disjoint regions.Being interpreted by computer to SAR image is one faced at present
A huge challenge, and SAR image segmentation is its steps necessary, it influences further detection, identification very big.
The common method of image segmentation is broadly divided into the method based on feature and the method based on statistical model at present.Due to
SAR unique imaging mechanism leads to the conventional method of many optical imagerys not containing there are many coherent speckle noises in SAR image
The segmentation of SAR image can be directly used in.The conventional segmentation methods of SAR image are mainly to have supervision and semi-supervised method.They
It generally requires manually experience and carries out feature extraction, however the quality for the feature extracted has the segmentation result of SAR image
Key effect.For having supervision and semi-supervised method, label data is needed, the label data of SAR image is seldom, obtains mark
The cost for signing data is very high.Key technology of the machine learning as unsupervised feature learning can be used for SAR image segmentation and appoint
Business.However, traditional deep learning method can only often reach feature level, without preferably excavating SAR image in semantic layer
Information on secondary causes it that can not efficiently accomplish the segmentation to SAR image.
A kind of patent " specific objective contour images based on depth convolutional neural networks of the North China Electric Power University in its application
It is disclosed in dividing method " (number of patent application CN201610109536.6, publication number CN105787482A) a kind of based on depth
The image partition method of the specific objective profile of convolutional neural networks.It is big that training image is normalized to same pixel by this method
It is small, obtained training image is input in a convolutional neural networks, by several layers of convolutional layer and full articulamentum, is being connected entirely
The last layer of layer obtains image expression, and is compared to obtain prediction error with corresponding mark image.Using backpropagation
Algorithm and stochastic gradient descent method reduce prediction error with the training neural network, obtain the segmentation of specific objective contour images
Training pattern.Although this method has achieved the purpose that autonomous learning characteristics of image, still, the deficiency that this method still has
Place is that input picture has been carried out normalized, just destroyed the original of image in this way by the method for the convenience handled
Structural information.Meanwhile the method also marks image, is divided into training sample and test sample, to reach trained volume
The purpose of product network.There is the processing mode of supervision to substantially increase the complexity of dividing method in this way.
Patent " SAR image based on ridge ripple deconvolution network and sparse classification of the Xian Electronics Science and Technology University in its application
It is disclosed in dividing method " (number of patent application CN201510675676.5, publication number CN105374033A) a kind of based on ridge ripple
The SAR image segmentation method of deconvolution network and sparse classification.This method distinguishes the aggregation zone of SAR image and homogenous region
Training ridge ripple deconvolution network RDN obtains the optimal value of filter group in ridge ripple deconvolution network, and using the side of sparse classification
The segmentation of method completion SAR image.Although the method has reached unsupervised study characteristics of image, still, this method still has
Shortcoming be, when initializing filter, the method for the random initializtion ridge ripple filter of use, and have ignored image
Structural information just greatly reduces the accuracy of image segmentation in this way.
Liu Fang, Duan Yiping, Li Lingling, burnt Lee " are based on the semantic and adaptive neighbour of level vision in its paper delivered at equal
The SAR image of the hidden model of domain multinomial is divided " (IEEE Trancactions on Geoscience and Remote
Sensing, 2016,54 (7): 4287-4301.) in propose it is a kind of based on level vision semanteme and adaptive neighborhood multinomial
The SAR image segmentation method of hidden model, this method propose the level vision of SAR image on the basis of SAR image sketch map
It is semantic.SAR image is divided into aggregation zone, structural region and homogenous region by the level vision semanteme.Based on the division, to not
Region with characteristic uses different dividing methods.For aggregation zone, gray level co-occurrence matrixes feature is extracted, and using part
The method of linear restriction coding obtains the expression of each aggregation zone, and then is split using the method for hierarchical clustering.To knot
Structure region devises vision semantic rules positioning boundary and line target by analysis side mode type and line model.In addition, boundary and
Line target contains strong directional information, therefore devises the hidden model of multinomial based on geometry window and be split.It is right
Homogenous region is gone to indicate center pixel, devises the hidden mould of multinomial based on self-adapting window in order to find appropriate neighborhood
Type is split.The segmentation result in these three regions be integrated into together segmentation result to the end.The deficiency of this method
Place is, inaccurate to the boundary alignment of aggregation zone, not reasonable to the determination of homogenous region classification number, the area of segmentation result
Domain consistency is poor, and does not handle pinpoint target in the segmentation of structural region.
Summary of the invention
It is an object of the invention to be directed to the deficiency of above-mentioned prior art, propose a kind of based on ridge ripple filter and deconvolution
The SAR image segmentation method of structural model, with the significantly more efficient segmentation for completing SAR image.
To achieve the above object, technical scheme is as follows:
(1) sketch SAR image:
(1a) obtains it according to the pixel fluctuating characteristic distributions of SAR image to the synthetic aperture radar SAR image of input
Sketch model;
(1b) obtains the sketch map of synthetic aperture radar SAR image from sketch model extraction sketch map;
(2) pixel subspace is divided:
(2a) uses sketch line fields method, divides pixel subspace to synthetic aperture radar SAR image, is synthesized
The administrative division map of aperture radar SAR image;
Administrative division map is mapped in the synthetic aperture radar SAR image of input by (2b), obtains synthetic aperture radar SAR image
Middle mixing aggregated structure atural object pixel subspace, homogenous region pixel subspace and structure-pixel subspace;
(3) ridge ripple filter set is constructed:
It is corresponding that (3a) extracts mixing aggregated structure atural object pixel subspace from the administrative division map of synthetic aperture radar SAR image
Aggregation zone, be divided into 10 ° in [0 °, 180 °] section with and be divided into 18 sections i.e. 18 directions, statistics is each respectively
In section in the aggregation zone sketch line segment line segment item number;
(3b) arranges sketch line segment all in the aggregation zone according to the number of the line segment item number in each interval
Sequence obtains the collating sequence in direction, using the degree in 6 directions preceding in the collating sequence in direction as ridge ripple filter in
Directioin parameter;
(3c) according to the following formula, the ridge ripple function of 9 × 9 ridge ripple filter is calculated according to parameter a, θ and b:
Y=a × (y1×cosθ+y2×sinθ-b)
Wherein, Y indicates that the ridge ripple function of ridge ripple filter, a indicate the scale parameter of ridge ripple filter, the value range of a
For [0,3], the discrete interval of a is 0.2, y1Indicate the abscissa positions of ridge ripple filter pixel, y1Value range be [0,
8], y1Discrete interval be 1, cos indicate cosine operation, θ indicate ridge ripple filter directioin parameter, y2Indicate ridge ripple filter
The ordinate position of pixel, y2Value range be [0,8], y2Discrete interval be 1, sin indicate sinusoidal operation;B indicates ridge
The displacement parameter of wave filter, when directioin parameter θ be [0 °, 90 °) when, b is in the section [0,9 × (sin θ+cos θ)] with interval
For 0.2 carry out discretization, when directioin parameter θ be [90 °, 180 °) when, b is divided into [9 × cos θ, 9 × sin θ] section with
0.2 carries out discretization;
(3d) according to the following formula, calculates each ridge ripple wave filter:
Wherein, c (Y) indicates that the ridge ripple filter using ridge ripple function Y as parameter, K indicate ridge ripple filter frobenius
The inverse of norm, exp are indicated using natural constant e as the index operation at bottom;
Each the ridge ripple filter bank being calculated is become ridge ripple filter set by (3e);
(4) deconvolution structural model is constructed:
(4a) is not connected to each of the mixing aggregated structure atural object pixel subspace in synthetic aperture radar SAR image mutually
Region, by 31 × 31 window carry out every a sliding window sample, the corresponding multiple images block in each region is obtained, by multiple images
Block is sequentially inputted in deconvolution structural model, obtains the input layer of deconvolution structural model;
(4b) carries out convolution operation using characteristic pattern and ridge ripple filter to reconstruct the image block in input layer, obtains warp
The warp lamination of product structural model;
(4c) according to the following formula, calculates data fidelity term:
Wherein, E (c) indicates data fidelity term, and c indicates the ridge ripple filter in deconvolution structural model warp lamination, N table
Showing each of to be learnt the total number for the image block that mutually disconnected region includes, ∑ indicates sum operation, | | | |FExpression is done
The operation of frobenius norm,Indicate the square operation of frobenius norm, xiIt indicates in deconvolution structural model to be constructed
I-th of input picture block, MiIndicate the corresponding ridge ripple filter of i-th of the image block inputted in deconvolution structural model to be constructed
Sum, * indicate convolution operation,Indicate corresponding j-th of the feature of i-th of image block in deconvolution structural model to be constructed
Figure,Indicate corresponding j-th of ridge ripple filter of i-th of image block in deconvolution structural model to be constructed;
(4d) according to the following formula, calculates structure fidelity term:
Wherein, G (c) indicates structure fidelity term, and R () indicates to ask the operation of all sketch line total lengths in sketch map, SM
() indicates to extract the operation with the one-to-one sketch segment of input picture block;
(4e) according to the following formula, calculating target function:
Wherein, L (c) indicates objective function,It indicates in objective function L (c) value minimum, seeks ridge ripple filtering
The operation of device c;
(4f) exports the ridge ripple filter set obtained by objective function guidance learning, obtains the defeated of deconvolution structural model
Layer out;
(5) training deconvolution structural model:
(5a) sets 0.1 for structural failure threshold value;
Step (4a) sampling is obtained image block and is sequentially inputted in deconvolution structural model by (5b);
(5c) randomly selects six filters, directioin parameter is by statistics in step (3b) from ridge ripple filter set
6 directions obtain, displacement parameter and scale parameter random initializtion, the filter that these initial six filters are formed
Set is as selected ridge ripple filter set;
The null matrix that (5d) is 39 × 39 with 6 sizes initializes the characteristic pattern that 6 sizes are 39 × 39, after initialization
6 sizes be 39 × 39 characteristic pattern as feature set of graphs;
Feature set of graphs and selected ridge ripple filter set are carried out convolution operation to reconstruct input picture block by (5e);
(5f) calculates the structure fidelity term of reconstruct input picture block using the structure fidelity term formula in step (4d);
Whether the structure fidelity term of the current reconstruct input picture block of (5g) judgement is less than structural failure threshold value, if so, holding
Row step (5j) otherwise executes step (5h);
(5h) utilizes scale parameter more new formula and displacement parameter more new formula, updates step (4c) data fidelity term respectively
The scale parameter and displacement parameter of formula median ridge wave filter obtain updated ridge ripple filter set, more using characteristic pattern
New formula updates characteristic pattern, obtains updated feature set of graphs;
(5i) using updated ridge ripple filter set as selected ridge ripple filter set, by updated feature
Set of graphs re-starts study to input picture block as feature set of graphs, return step (5e);
The ridge ripple filter that study obtains is saved the ridge ripple filter collection succeeded in school to the reconstruct input picture block by (5j)
In conjunction, the study to the input picture block feature is completed, and export the ridge ripple filter set that the input picture block succeeds in school;
(5k) judges whether all image blocks pass through the study that deconvolution structural model completes feature, if so, terminating journey
Otherwise sequence inputs next image block and executes step (5c);
(6) segmentation SAR image mixes aggregated structure atural object pixel subspace:
(6a) by it is all mutually it is disconnected mixing aggregated structure atural object pixel subspace regional trainings ridge ripple filter collection
It is merged and is connected into code book;
(6b) will be in the mutually ridge ripple filter set of disconnected mixing aggregated structure atural object pixel subspace regional training
All ridge ripple filters, are projected to code book, obtain projection vector;
(6c) carries out the projection vector in each mutually disconnected mixing aggregated structure atural object pixel subspace region maximum
Chi Hua obtains a structural eigenvector;
(6d) propagates AP clustering algorithm using neighbour, clusters to structural eigenvector, obtains and structural eigenvector
The segmentation result of corresponding mixing aggregated structure atural object pixel subspace;
(7) segmenting structure pixel subspace:
(7a) uses vision semantic rules, divides line target;
The feature of gathering of (7b) based on sketch line divides pinpoint target;
The result that (7c) divides line target and pinpoint target merges, and obtains the segmentation knot of structure-pixel subspace
Fruit.
(8) divide homogenous region pixel subspace:
Using the homogenous region dividing method based on multinomial logistic regression prior model, to homogenous region pixel subspace
It is split, obtains the segmentation result of homogenous region pixel subspace.
(9) combination and segmentation result:
The segmentation result of aggregated structure pixel subspace, homogenous region pixel subspace and structure-pixel subspace will be mixed
Merge, obtains the final segmentation result of synthetic aperture radar SAR image.
The invention has the following advantages over the prior art:
First, the present invention is to each of the mixing aggregated structure atural object pixel subspace in synthetic aperture radar SAR image
Mutual disconnected region sample every a sliding window by 31 × 31 window, obtains multiple images block, and obtained image block is defeated
Enter into deconvolution structural model, do not need that the synthetic aperture radar SAR image of input is normalized, overcomes existing
There is technology to need that synthetic aperture radar SAR image is normalized before input and deficiency so that the present invention has
Can directly using the image block obtained after sampling as input, without destroy input picture block prototype structure information it is excellent
Point.
Second, the present invention is closed using the sketch map of synthetic aperture radar SAR image using sketch line fields method
At the administrative division map of aperture radar SAR image, administrative division map is mapped to the synthetic aperture radar SAR image of input, obtains synthesis hole
Mixing aggregated structure atural object pixel subspace, homogeneous texture pixel subspace and structure-pixel in diameter radar SAR image is empty
Between, sampling and feature learning are carried out in the pixel subspace of mixing aggregated structure atural object, do not need to the type of image block into
Rower shows, overcomes the prior art and input picture is denoted as training image and mark image not when learning characteristics of image
Foot, so that the advantages of present invention is not with needing to indicate input picture, and reducing deconvolution structural model complexity.
Third, in the deconvolution structural model that constructs of the present invention, using the energy error function of input picture block and defeated
Enter the structural failure function of image block to extract the structural information of input picture block, overcomes the prior art using random initializtion
Ridge ripple filter and the deficiency for having ignored the structural information of synthetic aperture radar SAR image, so that have can be more preferable by the present invention
The structural information using SAR image complete the segmentation of synthetic aperture radar SAR image, improve synthetic aperture radar SAR
Image segmentation accuracy.
Detailed description of the invention
Fig. 1 is flow chart of the invention;
Fig. 2 is analogous diagram of the invention;
Fig. 3 is extracted in the present invention with the sketch line chart for gathering feature;
Fig. 4 is simulation result schematic diagram of the present invention.
Specific embodiment
The present invention will be further described with reference to the accompanying drawing.
With reference to attached drawing 1, the specific steps of the present invention are as follows.
Step 1, SAR image sketch.
It inputs synthetic aperture radar SAR image and its sketch is obtained into the sketch map of synthetic aperture radar SAR image.
Step 1 constructs a template on the side being made of pixel with different directions and scale, line, utilizes template
Direction and dimensional information structural anisotropy Gaussian function calculate the weighting coefficient of every bit in the template, mesoscale
Number value is 3~5, and direction number value is 18.
Step 2 calculates pixel in synthetic aperture radar SAR image corresponding with template area position according to the following formula
Mean value and variance yields:
Wherein, μ indicates the mean value of pixel in synthetic aperture radar SAR image corresponding with template area position, ∑ table
Show sum operation, g indicates the position of a pixel in the Ω region of template, and ∈ expression belongs to symbol, wgIndicate template the
Weight coefficient of the pixel at the position g, w in Ω regiongValue range be wg∈ [0,1], AgIt indicates and template Ω
Pixel value of the pixel at the position g in corresponding synthetic aperture radar SAR image in region, ν are indicated and template area position
The variance yields of pixel in corresponding synthetic aperture radar SAR image.
Step 3 calculates the response of each pixel reduced value operator in synthetic aperture radar SAR image according to the following formula:
Wherein, R indicates the response of each pixel reduced value operator in synthetic aperture radar SAR image, and min { } is indicated
It minimizes operation, a and b respectively indicate two different zones in template, μaAnd μbIt respectively indicates and template area a and template
The mean value of pixel in the corresponding synthetic aperture radar SAR image in the region position b.
Step 4, according to the following formula, response of each pixel to correlation operator in calculating synthetic aperture radar SAR image:
Wherein, C indicates that the response of correlation operator, a and b distinguish each pixel in synthetic aperture radar SAR image
Indicate two different zones in template, νaAnd νbRespectively indicate synthetic aperture corresponding with template area a and the template area position b
The variance yields of pixel, μ in radar SAR imageaAnd μbRespectively indicate synthesis hole corresponding with template area a and the template area position b
The mean value of pixel in diameter radar SAR image,Indicate square root functions.
Step 5 merges the response and synthesis of pixel reduced value operator in synthetic aperture radar SAR image according to the following formula
Pixel calculates each pixel pair in synthetic aperture radar SAR image to the response of correlation operator in aperture radar SAR image
The response of each template:
Wherein, F indicates that for each pixel to the response of each template, R and C distinguish table in synthetic aperture radar SAR image
Show in synthetic aperture radar SAR image that pixel is to correlation operator in pixel reduced value operator and synthetic aperture radar SAR image
Response,Indicate square root functions.
Step 6 selects template of the template with maximum response as pixel in synthetic aperture radar SAR image, and
It is obtained using maximum response as the intensity of the pixel using the direction of the template with maximum response as the direction of the pixel
Obtain the sideline response diagram and directional diagram of synthetic aperture radar SAR image.
Step 7 obtains synthetic aperture radar using the selected template of pixel each in synthetic aperture radar SAR image
The gradient map of SAR image.
Step 8 according to the following formula carries out the sideline response diagram for normalizing to [0,1] with the gradient map for normalizing to [0,1]
Fusion, obtains intensity map:
Wherein, I indicates that the intensity value in intensity map, x indicate the value in the response diagram of sideline, and y indicates the value in gradient map.
Step 9 detects intensity map using non-maxima suppression method, obtains suggestion sketch.
Step 10 chooses the pixel suggested in sketch with maximum intensity, will suggest the picture in sketch with the maximum intensity
The pixel of element connection connects to form suggestion line segment, obtains suggestion sketch map.
Step 11 calculates the code length gain CLG for suggesting sketch line in sketch map according to the following formula:
Wherein, CLG indicates to suggest the code length gain of sketch line in sketch map, and ∑ indicates sum operation, and m indicates current
The number of pixel in sketch line neighborhood, t indicate the number of pixel in current sketch line neighborhood, AtIt indicates in current sketch line neighborhood
The observation of t-th of pixel, At,0It indicates in the case where current sketch line cannot indicate the hypothesis of structural information, in the sketch line neighborhood
The estimated value of t-th of pixel, ln () are indicated using e as the log operations at bottom, At,1It indicates to indicate structure in current sketch line
Under the hypothesis of information, the estimated value of t-th of pixel in the sketch line neighborhood.
Step 12, the value range of given threshold T, T are 5~50, select the suggestion sketch line of CLG > T as final sketch
Sketch line in figure obtains the corresponding sketch map of input synthetic aperture radar SAR image.
The synthetic aperture radar SAR image sketch model that the present invention uses is that Jie-Wu et al. in 2014 was published in IEEE
Article " Local maximal on Transactions on Geoscience and Remote Sensing magazine
homogenous region search for SAR speckle reduction with sketch-based
Geometrical kernel function " proposed in model.
Step 2, pixel subspace is divided.
Step 1 divides sketch line according to the concentration class of sketch line segment in the sketch map of synthetic aperture radar SAR image
To indicate the aggregation sketch line of aggregation atural object and indicating the sketch line of boundary, line target and isolated target.
Step 2 chooses the sketch line that concentration class is equal to optimal concentration class according to the statistics with histogram of sketch line segment concentration class
Duan Zuowei seed line-segment sets { Ek, k=1,2 ..., m }, wherein EkIndicate any bar sketch line segment in seed line-segment sets, k table
Show that the label of any bar sketch line segment in seed line-segment sets, m indicate the total number of seed line segment, { } indicates set operation.
Step 3, as basic point by the unselected line segment for being added to some seed line-segment sets sum, with this basic point recursive resolve
New line segment aggregate.
Step 4 constructs the round primitive that a radius is the optimal concentration class section upper bound, with the circle primitive to line-segment sets
Line segment in conjunction is expanded, and is corroded to the line segment aggregate ecto-entad after expansion, is obtained in sketch map with sketch point
For the aggregation zone of unit.
Step 5, to the sketch line for indicating boundary, line target and isolated target, with each sketch point of each sketch line
Centered on construction size be 5 × 5 geometry window, obtain structural region.
Step 6 will remove the part other than aggregation zone and structural region as can not sketch region in sketch map.
Step 7, by sketch map aggregation zone, structural region and can not sketch region, be respectively mapped to synthetic aperture
In radar SAR image, mixing aggregated structure atural object pixel subspace, structure-pixel of synthetic aperture radar SAR image are obtained
Space and homogeneous texture pixel subspace.
Step 3, ridge ripple filter set is constructed.
Step 1 extracts mixing aggregated structure atural object pixel subspace pair from the administrative division map of synthetic aperture radar SAR image
The aggregation zone answered is divided into 10 ° in [0 °, 180 °] section with and is divided into 18 sections i.e. 18 directions, and statistics is every respectively
In a section in the aggregation zone sketch line segment line segment item number.
Step 2, to sketch line segment all in the aggregation zone, according to how much progress of the line segment item number in each interval
Sequence, obtain the collating sequence in direction, using the degree in 6 directions preceding in the collating sequence in direction as ridge ripple filter in
Directioin parameter.
Step 3 calculates the ridge ripple function of 9 × 9 ridge ripple filter according to parameter a, θ and b according to the following formula:
Y=a × (y1×cosθ+y2×sinθ-b)
Wherein, Y indicates that the ridge ripple function of ridge ripple filter, a indicate the scale parameter of ridge ripple filter, the value range of a
For [0,3], the discrete interval of a is 0.2, y1Indicate the abscissa positions of ridge ripple filter pixel, y1Value range be [0,
8], y1Discrete interval be 1, cos indicate cosine operation, θ indicate ridge ripple filter directioin parameter, y2Indicate ridge ripple filter
The ordinate position of pixel, y2Value range be [0,8], y2Discrete interval be 1, sin indicate sinusoidal operation;B indicates ridge
The displacement parameter of wave filter, when directioin parameter θ be [0 °, 90 °) when, b is in the section [0,9 × (sin θ+cos θ)] with interval
For 0.2 carry out discretization, when directioin parameter θ be [90 °, 180 °) when, b is divided into [9 × cos θ, 9 × sin θ] section with
0.2 carries out discretization.
Step 4 calculates each ridge ripple wave filter according to the following formula:
Wherein, c (Y) indicates that the ridge ripple filter using ridge ripple function Y as parameter, K indicate ridge ripple filter frobenius
The inverse of norm, exp are indicated using natural constant e as the index operation at bottom.
Each the ridge ripple filter bank being calculated is become ridge ripple filter set by step 5.
Step 4, deconvolution structural model is constructed.
Step 1 does not connect each of the mixing aggregated structure atural object pixel subspace in synthetic aperture radar SAR image mutually
Logical region sample every a sliding window by 31 × 31 window, the corresponding multiple images block in each region is obtained, by multiple figures
As block is sequentially inputted in deconvolution structural model, the input layer of deconvolution structural model is obtained.
Step 2 carries out convolution operation using characteristic pattern and ridge ripple filter to reconstruct the image block in input layer, obtains anti-
The warp lamination of convolutional coding structure model.
Step 3 calculates data fidelity term according to the following formula:
Wherein, E (c) indicates data fidelity term, and c indicates the ridge ripple filter in deconvolution structural model warp lamination, N table
Showing each of to be learnt the total number for the image block that mutually disconnected region includes, ∑ indicates sum operation, | | | |FExpression is done
The operation of frobenius norm,Indicate the square operation of frobenius norm, xiIt indicates in deconvolution structural model to be constructed
I-th of input picture block, MiIndicate the corresponding ridge ripple filter of i-th of the image block inputted in deconvolution structural model to be constructed
Sum, * indicate convolution operation,Indicate corresponding j-th of the feature of i-th of image block in deconvolution structural model to be constructed
Figure,Indicate corresponding j-th of ridge ripple filter of i-th of image block in deconvolution structural model to be constructed.
Step 4 calculates structure fidelity term according to the following formula:
Wherein, G (c) indicates structure fidelity term, and R () indicates to ask the operation of all sketch line total lengths in sketch map, SM
() indicates to extract the operation with the one-to-one sketch segment of input picture block.
Step 5, according to the following formula, calculating target function:
Wherein, L (c) indicates objective function,It indicates in objective function L (c) value minimum, seeks ridge ripple filtering
The operation of device c.
Step 6 exports the ridge ripple filter obtained by objective function guidance learning, obtains the output of deconvolution structural model
Layer.
Step 5, training deconvolution structural model.
Step 1 sets 0.1 for structural failure threshold value.
Step 4 sampling is obtained image block and is sequentially inputted in deconvolution structural model by step 2.
Step 3 randomly selects six filters from ridge ripple filter set, and directioin parameter in step 3 by counting
6 directions obtain, displacement parameter and scale parameter random initializtion, the filter collection that these initial six filters are formed
Cooperation is selected ridge ripple filter set.
The null matrix that step 4 is 39 × 39 with 6 sizes initializes the characteristic pattern that 6 sizes are 39 × 39, will initialize
6 sizes be 39 × 39 characteristic pattern as feature set of graphs;
Feature set of graphs and ridge ripple filter set are carried out convolution operation to reconstruct input picture block by step 5;
Step 6 calculates the structure fidelity term of reconstruct input picture block using structure fidelity term formula.
Step 7, judges whether the structure fidelity term of current reconstruct input picture block is less than structural failure threshold value, if so,
Step 10 is executed, otherwise, executes step 8.
Step 8, using scale parameter more new formula and displacement parameter more new formula, respectively in more new data fidelity term formula
The scale parameter and displacement parameter of ridge ripple filter obtain updated ridge ripple filter set, using characteristic pattern more new formula,
Characteristic pattern is updated, updated feature set of graphs is obtained.
The more new formula of scale parameter is as follows:
Wherein a indicates the scale parameter of ridge ripple filter, atThe scale acquired, a are walked for tt-1It is acquired for t-1 step
Scale, δ indicate coefficient, and value range is [0,1], and ∑ indicates to be added, xiFor i-th piece 31 × 31 of SAR image sampling block,Table
Show corresponding j-th of ridge ripple filter of i-th of image block,Indicate corresponding j-th of feature segment of i-th of image block, γ=
(a, b, θ), K (γ) indicate that the inverse of ridge ripple filter frobenius norm, e indicate that natural constant, Y indicate ridge ripple filter
Ridge ripple function, y1Indicate the abscissa positions of 9 × 9 ridge ripple filter pixel, y2Indicate 9 × 9 ridge ripple filter pixel
The ordinate position of point, θ indicate the directioin parameter of ridge ripple filter.
The more new formula of displacement parameter is as follows:
Wherein btIndicate the displacement parameter for the ridge ripple filter that the t times iteration acquires, bt-1The displacement ginseng acquired for t-1 times
Number, δ indicate coefficient, and value range is [0,1], and ∑ indicates to be added, xiFor i-th piece 31 × 31 of SAR image sampling block,It indicates
Corresponding j-th of ridge ripple filter of i-th of image block,Indicate corresponding j-th of feature segment of i-th of image block, γ=(a,
B, θ), K (γ) indicates that the inverse of ridge ripple filter frobenius norm, e indicate that natural constant, Y indicate the ridge of ridge ripple filter
Wave function.
Characteristic pattern more new formula is as follows:
WhereinIndicate the characteristic pattern that the t times iteration acquires,For the characteristic pattern that t-1 iteration acquires, δ indicates step
Long, value range is [0,1], and ∑ indicates sum operation, xiIndicate i-th of input picture block,Indicate that i-th of image block is corresponding
J-th of ridge ripple filter,Indicate corresponding j-th of the characteristic pattern of i-th of image block in t-1 iteration.
Step 9 returns to step 4 using updated ridge ripple filter set as selected ridge ripple filter set, right
Input picture block re-starts study.
The ridge ripple filter that study obtains is saved the ridge ripple filter set succeeded in school to the input picture block by step 10
In, the study to the input picture block feature is completed, and export the ridge ripple filter set that the input picture block succeeds in school.
Step 11, judges whether all image blocks pass through the study that deconvolution structural model completes feature, if so, terminating
Otherwise program inputs next image block and executes step 3.
Step 6, segmentation SAR image mixes aggregated structure atural object pixel subspace.
By all mutually not characteristic set splicings of connected region, using spliced characteristic set as code book.
To all features of each mutual not connected region, the inner product with each feature in code book is calculated separately, is obtained every
Projection vector of a all features in region on code book.
Maximum pond is carried out to the projection vector of each mutual connected region, obtains the corresponding structure spy in each region
Levy vector.
AP clustering algorithm is propagated using neighbour, the structural eigenvector of all mutual connected regions is not clustered, is obtained
Mix the segmentation result of aggregated structure atural object pixel subspace.
Step 7, segmenting structure pixel subspace.
With vision semantic rules, divide line target.
If i-th sketch line liWith j-th strip sketch line ljThe distance between be Dij, liDirection be Oi, ljDirection be Oj,
I, j ∈ [1,2 ..., S], S are the total number of sketch line.
Width is greater than the line target of 3 pixels with two sketch line liAnd ljIt indicates, liAnd ljThe distance between DijIt is less than
T1And direction difference Oi-OjLess than 10 degree, wherein T1=5.
If the s articles sketch line lsGeometry window wsThe average gray of interior each column is AiIf the gray scale difference of adjacent column is
ADi=| Ai-Ai+1|, if zs=[zs1,zs2,...,zs9] be adjacent column gray scale difference ADiLabel vector.
By width less than the line target of 3 pixels with single sketch line lsIt indicates, lsGeometry window wsIt is interior, calculate phase
The gray scale difference AD of adjacent columniIf ADi> T2, then zsi=1;Otherwise zsi=0, zsIn there are two element value be 1, remaining is 0,
Middle T2=34.
If L1,L2It is the set for indicating the sketch line of line target, if Dij< T1And | Oi-Oj| < 10, then li,lj∈
L1;If sum (zs)=2, then ls∈L2, wherein sum () indicates the sum of parameter element.
In structure-pixel subspace, according to the set L of the sketch line of line target1, by liAnd ljBetween region as line mesh
Mark.
In structure-pixel subspace, according to the set L of the sketch line of line target2, l will be coveredsRegion as line target.
Gather feature based on sketch line, dividing pinpoint target, specific step is as follows:
Step 1 would not indicate all sketch wire tags of line target in the structural region of administrative division map as candidate sketch line
Sketch line in set;
Step 2 randomly selects a sketch line from candidate sketch line set, with an endpoint of selected sketch line
Centered on, construct the geometry window that size is 5 × 5;
Step 3 judges the endpoint that whether there is other sketch lines in geometry window, and if it exists, execute step 4;Otherwise,
Execute step 6;
Step 4 judges whether to only exist an endpoint, if so, sketch line where the endpoint and current sketch line are carried out
Connection;Otherwise, step 5 is executed;
Step 5 connects the sketch line where selected sketch line and each endpoint, wherein angle is chosen from all connecting lines
The sketch line that maximum two sketch lines are completed as connection;
Step 6 judges the endpoint that whether there is other sketch lines in the geometry window of another endpoint of sketch line, if
In the presence of execution step 4;Otherwise, step 7 is executed;
Step 7 chooses the sketch line comprising two and two or more sketch line segments to the sketch line for completing attended operation,
The item number n comprising sketch line segment in selected sketch line is counted, wherein n >=2;
Step 8, judges whether the item number n of sketch line is equal to 2, if so, executing step 9;Otherwise, step 10 is executed;
Sketch line of the angle value on sketch line vertex in the range of [10 °, 140 °] is used as to have and gathers spy by step 9
The sketch line of sign;
Step 10 selects sketch line of the angle value on the corresponding n-1 vertex of sketch line all in [10 °, 140 °] range;
Step 11 is defined as follows two kinds of situations in selected sketch line:
Whether the first situation judges adjacent (i-1)-th, the two sketch line segments of i-th sketch line segment, i+1 item i-th
The same side of straight line, 2≤i≤n-1, if all sketch line segments and adjacent segments on sketch line are all same where sketch line segment
Side, then marking the sketch line is with the sketch line for gathering feature;
Whether second situation judges adjacent (i-1)-th, the two sketch line segments of i-th sketch line segment, i+1 item i-th
The same side of straight line, 2≤i≤n-1, if having n-1 sketch line segment and adjacent segments same on sketch line where sketch line segment
Side, and have a sketch line segment line segment adjacent thereto in non-the same side, also marking the sketch line is with the element for gathering feature
Retouch line;
Step 12, an optional sketch line in there is the sketch line for gathering feature, by two ends of selected sketch line
Point coordinate, determine the distance between two endpoints, if the end-point distances in [0,20] range, then using selected sketch line as table
Show the sketch line of pinpoint target;
Step 13, judge it is untreated have gather the sketch line of feature and whether all selected, if so, executing step 12;
Otherwise, step 14 is executed;
Step 14, with the method for super-pixel segmentation, to the sketch line for indicating pinpoint target in synthetic aperture radar SAR image
The pixel of surrounding carries out super-pixel segmentation, by super-pixel of the gray value of super-pixel after segmentation in [0,45] or [180,255]
As pinpoint target super-pixel;
Step 15 merges pinpoint target super-pixel, using the boundary of the pinpoint target super-pixel after merging as pinpoint target
Boundary, obtain the segmentation result of pinpoint target.
The result divided to line target and pinpoint target merges, and obtains the segmentation result of structure-pixel subspace.
Step 8, divide homogenous region pixel subspace.
Step 1 arbitrarily chooses a pixel, centered on selected pixel from the pixel subspace of homogenous region
The square window for establishing 3 × 3 calculates the standard deviation sigma of the window1。
The side length of square window is increased by 2, obtains new square window, calculate the standard deviation of new square window by step 2
σ2。
Step 3, if standard deviation threshold method T3=3, if | σ1-σ2| < T3, then it is σ by standard deviation2Square window as most
Whole square window executes step 4;Otherwise, step 2 is executed.
Step 4 calculates the prior probability of center pixel in square window according to the following formula:
Wherein, p '1Indicate that the prior probability of center pixel in square window, exp () indicate exponential function operation, η ' table
Show that probabilistic model parameter, η ' value are 1, xk′' indicate to belong to the number of pixels of kth ' class in square window, k' ∈ [1 ...,
K'], K' indicates the classification number of segmentation, and K' value is 5, xi' indicate the pixel for belonging to the i-th ' class in the obtained square window of step 3
Number.
The probability density of pixel grey scale is multiplied with the probability density of texture, obtains likelihood probability p' by step 52, wherein
The probability density of gray scale is distributed to obtain by fading channel Nakagami, and the probability density of texture is distributed to obtain by t.
Step 6, by prior probability p1' and likelihood probability p2' be multiplied, obtain posterior probability p12'。
Step 7 judges whether there are also untreated pixels in the pixel subspace of homogenous region, if so, executing step 1;
Otherwise, step 9 is executed.
Step 8 obtains the segmentation result of homogenous region pixel subspace according to maximum posteriori criterion.
Step 9, combination and segmentation result.
The segmentation of aggregated structure atural object pixel subspace, homogenous region pixel subspace and structure-pixel subspace will be mixed
As a result it merges, obtains the final segmentation result of synthetic aperture radar SAR image.
Effect of the invention is further described below with reference to analogous diagram.
1. simulated conditions:
The hardware condition that the present invention emulates are as follows: Intellisense and image understanding laboratory graphics workstation;Present invention emulation
Used synthetic aperture radar SAR image are as follows: the Piperiver that Ku wave band resolution ratio is 1 meter schemes.
2. emulation content:
Emulation experiment of the invention is split to the Piperiver figure in SAR image, as shown in Fig. 2 (a)
Piperiver figure.The synthetic aperture radar SAR image that the figure is 1 meter from Ku wave band resolution ratio.
Using SAR image sketch step of the invention, to retouching of Piperiver pixel shown in Fig. 2 (a), obtain as
Sketch map shown in Fig. 2 (b), wherein the deeper sketch line of color is to indicate that line target and pinpoint target obtain sketch line.
Sketch map compartmentalization shown in Fig. 2 (b) is obtained such as Fig. 2 using division pixel subspace step of the invention
(c) administrative division map shown in.White space in Fig. 2 (c) indicates mixing aggregated structure semantic space, and others are homogeneous texture language
Adopted space and structure semantics space.Administrative division map shown in Fig. 2 (c) is mapped to original SAR image shown in Fig. 2 (a), is obtained such as Fig. 2 (d)
Shown in SAR image mix aggregated structure atural object pixel subspace figure.
Image Segmentation Methods Based on Features pinpoint target step is gathered based on sketch line using of the invention, it will be in administrative division map shown in Fig. 2 (c)
Structural region is mapped to sketch map shown in Fig. 2 (b), obtains the corresponding sketch line of structural region shown in Fig. 3 (a).Fig. 3 (b) institute
In the corresponding sketch line of the structural region shown, black is the sketch line for representing line target, and structural region shown in Fig. 3 (c) is corresponding
Sketch line in, black is the sketch line for representing pinpoint target.The figure of Piperiver shown in Fig. 2 (a) carries out point of pinpoint target
It cuts, obtains the segmentation result figure of pinpoint target shown in Fig. 3 (d), wherein black region indicates pinpoint target.
Aggregated structure atural object pixel subspace step is mixed using segmentation SAR image of the invention, to shown in Fig. 2 (d)
The mixing aggregated structure atural object pixel subspace figure of Piperiver figure is split, and obtains mixed land cover picture shown in Fig. 4 (a)
Sub-prime space segmentation result figure, grey area indicate untreated ground object space, and the region of remaining same color indicates same
A kind of atural object, the region of different colours indicate different atural object.
Using combination and segmentation result step of the invention, it is empty to merge mixing aggregated structure atural object pixel shown in Fig. 4 (a)
Between segmentation result and homogenous region pixel subspace segmentation result and structure-pixel subspace segmentation result, obtain Fig. 4 (b), Fig. 4
(b) be Fig. 2 (a) Piperiver image final segmentation result, Fig. 4 (c) be based on level vision semanteme and adaptive neighborhood it is more
Final segmentation result figure of the SAR image segmentation method of Xiang Shiyin model to Piperiver image.
3. simulated effect is analyzed:
Fig. 4 (b) is final segmentation result figure of the method for the present invention to Piperiver image, and Fig. 4 (c) is regarded based on level
Feel semantic and the hidden model of adaptive neighborhood multinomial SAR image segmentation method to the final segmentation result of Piperiver image
Figure, by the comparison of segmentation result figure, it could be assumed that, the method for the present invention is for mixing aggregated structure atural object pixel subspace
Boundary determination is more accurate, and the segmentation for homogenous region pixel subspace, region consistency is obviously preferable, classification number more adduction
Reason, and preferable dividing processing has been carried out to the pinpoint target in structure-pixel subspace.Use the method for the present invention pairing pore-forming
Diameter radar SAR image is split, and is effectively divided SAR image, and improves the accuracy of SAR image segmentation.
Claims (10)
1. a kind of SAR image segmentation method based on ridge ripple filter and deconvolution structural model, includes the following steps:
(1) sketch SAR image:
(1a) obtains its sketch according to the pixel fluctuating characteristic distributions of SAR image to the synthetic aperture radar SAR image of input
Model;
(1b) obtains the sketch map of synthetic aperture radar SAR image from sketch model extraction sketch map;
(2) pixel subspace is divided:
(2a) uses sketch line fields method, divides pixel subspace to synthetic aperture radar SAR image, obtains synthetic aperture
The administrative division map of radar SAR image;
Administrative division map is mapped in the synthetic aperture radar SAR image of input by (2b), obtains mixing in synthetic aperture radar SAR image
Close aggregated structure atural object pixel subspace, homogenous region pixel subspace and structure-pixel subspace;
(3) ridge ripple filter set is constructed:
It is corresponding poly- that (3a) extracts mixing aggregated structure atural object pixel subspace from the administrative division map of synthetic aperture radar SAR image
Collect region, is divided into 10 ° in [0 °, 180 °] section with and is divided into 18 sections i.e. 18 directions, count each section respectively
The line segment item number of sketch line segment in the interior aggregation zone;
(3b) is ranked up sketch line segment all in the aggregation zone according to the number of the line segment item number in each interval,
The collating sequence in direction is obtained, using the degree in 6 directions preceding in the collating sequence in direction as the side in ridge ripple filter
To parameter;
(3c) according to the following formula, the ridge ripple function of 9 × 9 ridge ripple filter is calculated according to parameter a, θ and b:
Y=a × (y1×cosθ+y2×sinθ-b)
Wherein, Y indicates the ridge ripple function of ridge ripple filter, and a indicates the scale parameter of ridge ripple filter, the value range of a be [0,
3], the discrete interval of a is 0.2, y1Indicate the abscissa positions of ridge ripple filter pixel, y1Value range be [0,8], y1
Discrete interval be 1, cos indicate cosine operation, θ indicate ridge ripple filter directioin parameter, y2Indicate ridge ripple filter pixel
The ordinate position of point, y2Value range be [0,8], y2Discrete interval be 1, sin indicate sinusoidal operation;B indicates ridge ripple filter
The displacement parameter of wave device, when directioin parameter θ be [0 °, 90 °) when, b is divided into 0.2 in the section [0,9 × (sin θ+cos θ)] with
Carry out discretization, when directioin parameter θ is [90 °, 180 °) when, b be divided into [9 × cos θ, 9 × sin θ] section 0.2 into
Row discretization;
(3d) according to the following formula, calculates each ridge ripple wave filter:
Wherein, c (Y) indicates that the ridge ripple filter using ridge ripple function Y as parameter, K indicate ridge ripple filter frobenius norm
Inverse, exp indicate using natural constant e as the index operation at bottom;
Each the ridge ripple filter bank being calculated is become ridge ripple filter set by (3e);
(4) deconvolution structural model is constructed:
(4a) is to the mutual disconnected area in each of mixing aggregated structure atural object pixel subspace in synthetic aperture radar SAR image
Domain, by 31 × 31 window carry out every a sliding window sample, obtain the corresponding multiple images block in each region, by multiple images block according to
It is secondary to be input in deconvolution structural model, obtain the input layer of deconvolution structural model;
(4b) carries out convolution operation using characteristic pattern and ridge ripple filter to reconstruct the image block in input layer, obtains deconvolution knot
The warp lamination of structure model;
(4c) according to the following formula, calculates data fidelity term:
Wherein, E (c) indicates data fidelity term, and c indicates the ridge ripple filter in deconvolution structural model warp lamination, and N expression is wanted
The total number of each of study image block that mutually disconnected region includes, ∑ indicate sum operation, | | | |FExpression is done
The operation of frobenius norm,Indicate the square operation of frobenius norm, xiIt indicates in deconvolution structural model to be constructed
I-th of input picture block, MiIndicate the corresponding ridge ripple filter of i-th of the image block inputted in deconvolution structural model to be constructed
Sum, * indicate convolution operation,Indicate corresponding j-th of the feature of i-th of image block in deconvolution structural model to be constructed
Figure,Indicate corresponding j-th of ridge ripple filter of i-th of image block in deconvolution structural model to be constructed;
(4d) according to the following formula, calculates structure fidelity term:
Wherein, G (c) indicates structure fidelity term, and R () indicates to ask the operation of all sketch line total lengths in sketch map, SM ()
It indicates to extract the operation with the one-to-one sketch segment of input picture block;
(4e) according to the following formula, calculating target function:
Wherein, L (c) indicates objective function,It indicates in objective function L (c) value minimum, seeks ridge ripple filter c's
Operation;
(4f) exports the ridge ripple filter set obtained by objective function guidance learning, obtains the output of deconvolution structural model
Layer;
(5) training deconvolution structural model:
(5a) sets 0.1 for structural failure threshold value;
Step (4a) sampling is obtained image block and is sequentially inputted in deconvolution structural model by (5b);
(5c) randomly selects six filters, directioin parameter is by middle 6 counted of step (3b) from ridge ripple filter set
Direction obtains, displacement parameter and scale parameter random initializtion, the filter set that these initial six filters are formed
As selected ridge ripple filter set;
The null matrix that (5d) is 39 × 39 with 6 sizes initializes the characteristic pattern that 6 sizes are 39 × 39, by 6 after initialization
The characteristic pattern that a size is 39 × 39 is as feature set of graphs;
Feature set of graphs and selected ridge ripple filter set are carried out convolution operation to reconstruct input picture block by (5e);
(5f) calculates the structure fidelity term of reconstruct input picture block using the structure fidelity term formula in step (4d);
Whether the structure fidelity term of the current reconstruct input picture block of (5g) judgement is less than structural failure threshold value, if so, executing step
Suddenly (5j) is otherwise executed step (5h);
(5h) utilizes scale parameter more new formula and displacement parameter more new formula, updates step (4c) data fidelity term formula respectively
The scale parameter and displacement parameter of median ridge wave filter obtain updated ridge ripple filter set, are updated using characteristic pattern public
Formula updates characteristic pattern, obtains updated feature set of graphs;
(5i) using updated ridge ripple filter set as selected ridge ripple filter set, by updated feature atlas
Cooperation is characterized set of graphs, and return step (5e) re-starts study to input picture block;
(5j) saves the ridge ripple filter that study obtains in the ridge ripple filter set succeeded in school to the reconstruct input picture block,
The study to the input picture block feature is completed, and exports the ridge ripple filter set that the input picture block succeeds in school;
(5k) judges whether all image blocks pass through the study that deconvolution structural model completes feature, if so, terminate program, it is no
Then, it inputs next image block and executes step (5c);
(6) segmentation SAR image mixes aggregated structure atural object pixel subspace:
(6a) spells the ridge ripple filter set of all mutually disconnected mixing aggregated structure atural object pixel subspace regional trainings
It is connected into code book;
(6b) will own in the mutually ridge ripple filter set of disconnected mixing aggregated structure atural object pixel subspace regional training
Ridge ripple filter, projected to code book, obtain projection vector;
(6c) carries out maximum pond to the projection vector in each mutually disconnected mixing aggregated structure atural object pixel subspace region,
Obtain a structural eigenvector;
(6d) propagates AP clustering algorithm using neighbour, clusters, obtains opposite with structural eigenvector to structural eigenvector
The segmentation result for the mixing aggregated structure atural object pixel subspace answered;
(7) segmenting structure pixel subspace:
(7a) uses vision semantic rules, divides line target;
The feature of gathering of (7b) based on sketch line divides pinpoint target;
The result that (7c) divides line target and pinpoint target merges, and obtains the segmentation result of structure-pixel subspace;
(8) divide homogenous region pixel subspace:
Using the homogenous region dividing method based on multinomial logistic regression prior model, homogenous region pixel subspace is carried out
Segmentation, obtains the segmentation result of homogenous region pixel subspace;
(9) combination and segmentation result:
The segmentation result for mixing aggregated structure pixel subspace, homogenous region pixel subspace and structure-pixel subspace is closed
And obtain the final segmentation result of synthetic aperture radar SAR image.
2. the SAR image segmentation method according to claim 1 based on ridge ripple filter and deconvolution structural model, special
Sign is that specific step is as follows for step (1) described sketch:
Step 1 constructs a template on the side being made of pixel with different directions and scale, line, utilizes the side of template
To with dimensional information structural anisotropy's Gaussian function, count the weighting coefficient of every bit in the template, mesoscale number takes
Value is 3~5, and direction number value is 18;
Step 2 calculates the mean value of pixel in synthetic aperture radar SAR image corresponding with template area position according to the following formula
And variance yields:
Wherein, μ indicates the mean value of pixel in synthetic aperture radar SAR image corresponding with template area position, and ∑ expression is asked
And operation, g indicate the position of a pixel in the Ω region of template, ∈ expression belongs to symbol, wgIndicate template Ω
Weight coefficient of the pixel at the position g, w in regiongValue range be wg∈ [0,1], AgIt indicates and the Ω region of template
Pixel value of the middle pixel at the position g in corresponding synthetic aperture radar SAR image, ν indicate opposite with template area position
The variance yields of pixel in the synthetic aperture radar SAR image answered;
Step 3 calculates the response of each pixel reduced value operator in synthetic aperture radar SAR image according to the following formula:
Wherein, R indicates the response of each pixel reduced value operator in synthetic aperture radar SAR image, and min { } expression is asked most
Small Value Operations, a and b respectively indicate two different regions in template, μaAnd μbIt respectively indicates and template area a and template region
The mean value of pixel in the corresponding synthetic aperture radar SAR image in the domain position b;
Step 4, according to the following formula, response of each pixel to correlation operator in calculating synthetic aperture radar SAR image:
Wherein, C indicates response of each pixel to correlation operator in synthetic aperture radar SAR image, νaAnd νbIt respectively indicates
The variance yields of pixel in synthetic aperture radar SAR image corresponding with template area a and the template area position b,Expression square
Root operation;
Step 5 merges the response and synthetic aperture of pixel reduced value operator in synthetic aperture radar SAR image according to the following formula
Pixel is to the response of correlation operator in radar SAR image, calculates in synthetic aperture radar SAR image each pixel to each
The response of template:
Wherein, F indicates that each pixel is to the response of each template in synthetic aperture radar SAR image;
Step 6, selection has the template of maximum response from the response of each template, schemes as synthetic aperture radar SAR
The template of pixel as in, and using maximum response as the intensity of the pixel, the direction of the template with maximum response is made
For the direction of the pixel, the sideline response diagram and directional diagram of synthetic aperture radar SAR image are obtained;
Step 7 is obtained using the selected template with maximum response of each pixel in synthetic aperture radar SAR image
The gradient map of synthetic aperture radar SAR image;
Step 8 merges the response of sideline response diagram and the value of gradient map, intensity value is calculated, by intensity value according to the following formula
Each pixel composition synthetic aperture radar SAR image intensity map:
Wherein, I indicates that intensity value, x indicate the value in the response diagram of synthetic aperture radar SAR image sideline, and y indicates synthetic aperture thunder
Value up in SAR image gradient map;
Step 9 detects intensity map using non-maxima suppression method, obtains suggestion sketch;
Step 10 will suggest the pixel in sketch with the maximum intensity from pixel of the selection with maximum intensity in sketch is suggested
The pixel of connection connects to form suggestion line segment, obtains suggestion sketch map;
Step 11 calculates the code length gain for suggesting sketch line in sketch map according to the following formula:
Wherein, CLG indicates to suggest the code length gain of sketch line in sketch map, and m indicates pixel in current sketch line neighborhood
Number, t indicate the number of pixel in current sketch line neighborhood, AtIndicate the observation of t-th of pixel in current sketch line neighborhood,
At,0It indicates under the premise of current sketch line cannot indicate structural information, the estimated value of t-th of pixel in the sketch line neighborhood,
Ln () is indicated using e as the log operations at bottom, At,1Indicate element under the premise of current sketch line can indicate structural information
Retouch the estimated value of t-th of pixel in line neighborhood;
Step 12, the value range of given threshold T, T are 5~50, select the suggestion sketch line of CLG > T as in final sketch map
Sketch line, obtain the corresponding sketch map of input synthetic aperture radar SAR image.
3. the SAR image segmentation method according to claim 1 based on ridge ripple filter and deconvolution structural model, special
Sign is that specific step is as follows for sketch line fields method described in step (2):
Sketch line is divided into table according to the concentration class of sketch line segment in the sketch map of synthetic aperture radar SAR image by step 1
Show the aggregation sketch line of aggregation atural object and indicates the sketch line of boundary, line target and isolated target;
Step 2 chooses the sketch line segment work that concentration class is equal to optimal concentration class according to the statistics with histogram of sketch line segment concentration class
For seed line-segment sets { Ek, k=1,2 ..., m }, wherein EkIndicate that any bar sketch line segment in seed line-segment sets, k indicate kind
Sub-line section concentrates the label of any bar sketch line segment, and m indicates the total number of seed line segment, and { } indicates set operation;
Step 3, as basic point with the unselected line segment for being added to seed line-segment sets sum, with the new line segment of this basic point recursive resolve
Set;
Step 4 constructs the round primitive that a radius is the optimal concentration class section upper bound, with the circle primitive in line segment aggregate
Line segment expanded, the line segment aggregate ecto-entad after expansion is corroded, is obtained in sketch map with sketch point being single
The aggregation zone of position;
Step 5, to the sketch line for indicating boundary, line target and isolated target, during each sketch point with each sketch line is
The heart constructs the geometry window that size is 5 × 5, obtains structural region;
Step 6 will remove the part other than aggregation zone and structural region as can not sketch region in sketch map;
Step 7, by sketch map aggregation zone, structural region and can not sketch region, be respectively mapped to synthetic aperture radar
In SAR image, mixing aggregated structure atural object pixel subspace, the structure-pixel subspace of synthetic aperture radar SAR image are obtained
With homogeneous texture pixel subspace.
4. the SAR image segmentation method according to claim 1 based on ridge ripple filter and deconvolution structural model, special
Sign is that the more new formula of the scale parameter a of ridge ripple filter described in step (5h) is as follows:
Wherein a indicates the scale parameter of ridge ripple filter, atThe scale acquired, a are walked for tt-1The scale acquired, δ are walked for t-1
Indicate coefficient, value range is [0,1], and ∑ indicates to be added, xiFor i-th piece 31 × 31 of SAR image sampling block,Indicate i-th
Corresponding j-th of ridge ripple filter of a image block,Indicate corresponding j-th of feature segment of i-th of image block, γ=(a, b,
θ), K (γ) indicates that the inverse of ridge ripple filter frobenius norm, e indicate that natural constant, Y indicate the ridge ripple of ridge ripple filter
Function, y1Indicate the abscissa positions of 9 × 9 ridge ripple filter pixel, y2Indicate the vertical of 9 × 9 ridge ripple filter pixel
Coordinate position, θ indicate the directioin parameter of ridge ripple filter.
5. the SAR image segmentation method according to claim 1 based on ridge ripple filter and deconvolution structural model, special
Sign is that the more new formula of displacement parameter b described in step (5h) is as follows:
Wherein btIndicate the displacement parameter for the ridge ripple filter that the t times iteration acquires, bt-1The displacement parameter acquired for t-1 times, δ table
Show coefficient, value range is [0,1], and ∑ indicates to be added, xiFor i-th piece 31 × 31 of SAR image sampling block,It indicates i-th
Corresponding j-th of ridge ripple filter of image block,Indicate corresponding j-th of feature segment of i-th of image block, γ=(a, b, θ),
K (γ) indicates that the inverse of ridge ripple filter frobenius norm, e indicate that natural constant, Y indicate the ridge ripple letter of ridge ripple filter
Number.
6. the SAR image segmentation method according to claim 1 based on ridge ripple filter and deconvolution structural model, special
Sign is that characteristic pattern more new formula described in step (5h) is as follows:
WhereinIndicate the characteristic pattern that the t times iteration acquires,For the characteristic pattern that t-1 iteration acquires, δ indicates step-length, takes
Being worth range is [0,1], and ∑ indicates sum operation, xiIndicate i-th of input picture block,Indicate the corresponding jth of i-th of image block
A ridge ripple filter,Indicate corresponding j-th of feature segment of i-th of image block in t-1 iteration.
7. the SAR image segmentation method according to claim 1 based on ridge ripple filter and deconvolution structural model, special
Sign is that vision semantic rules described in step (7a) are as follows:
If i-th sketch line liWith j-th strip sketch line ljThe distance between be Dij, liDirection be Oi, ljDirection be Oj, i, j
∈ [1,2 ..., S], S are the total number of sketch line;
Width is greater than the line target of 3 pixels with two sketch line liAnd ljIt indicates, liAnd ljThe distance between DijLess than T1And
Direction difference Oi-OjLess than 10 degree, wherein T1=5;
If the s articles sketch line lsGeometry window wsThe average gray of interior each column is AiIf the gray scale difference of adjacent column is ADi=
|Ai-Ai+1|, if zs=[zs1,zs2,...,zs9] be adjacent column gray scale difference ADiLabel vector;
By width less than the line target of 3 pixels with single sketch line lsIt indicates, lsGeometry window wsIt is interior, calculate adjacent column
Gray scale difference ADiIf ADi> T2, then zsi=1;Otherwise zsi=0, zsIn there are two element value be 1, remaining is 0, wherein T2
=34;
If L1,L2It is the set for indicating the sketch line of line target, if Dij< T1And | Oi-Oj| < 10, then li,lj∈L1;If
sum(zs)=2, then ls∈L2, wherein sum () indicate to vector important summation operation.
8. the SAR image segmentation method according to claim 1 based on ridge ripple filter and deconvolution structural model, special
Sign is that specific step is as follows for segmentation line target described in step (7a):
Step 1, in structure-pixel subspace, according to the set L of the sketch line of line target1, by liAnd ljBetween region as line
Target;
Step 2, in structure-pixel subspace, according to the set L of the sketch line of line target2, l will be coveredsRegion as line mesh
Mark.
9. the SAR image segmentation method according to claim 1 based on ridge ripple filter and deconvolution structural model, special
Sign is that specific step is as follows for segmentation pinpoint target described in step (7b):
Step 1 would not indicate all sketch wire tags of line target in the structural region of administrative division map as candidate sketch line set
In sketch line;
Step 2 randomly selects a sketch line from candidate sketch line set, during an endpoint with selected sketch line is
The heart, the geometry window that construction size is 5 × 5;
Step 3 judges the endpoint that whether there is other sketch lines in geometry window, and if it exists, execute step 4;Otherwise, it executes
Step 6;
Step 4 judges whether to only exist an endpoint, if so, sketch line where the endpoint and current sketch line are attached;
Otherwise, step 5 is executed;
Step 5 connects the sketch line where selected sketch line and each endpoint, and it is maximum that wherein angle is chosen from all connecting lines
Two sketch lines as connection complete sketch line;
Step 6 judges the endpoint that whether there is other sketch lines in the geometry window of another endpoint of sketch line, if depositing
Executing step 4;Otherwise, step 7 is executed;
Step 7 chooses the sketch line comprising two and two or more sketch line segments, statistics to the sketch line for completing attended operation
It include the item number n of sketch line segment, wherein n >=2 in selected sketch line;
Step 8, judges whether the item number n of sketch line is equal to 2, if so, executing step 9;Otherwise, step 10 is executed;
Step 9, by sketch line of the angle value on sketch line vertex in the range of [10 °, 140 °] as have gather feature
Sketch line;
Step 10 selects sketch line of the angle value on the corresponding n-1 vertex of sketch line all in [10 °, 140 °] range;
Step 11 is defined as follows two kinds of situations in selected sketch line:
Whether the first situation judges adjacent (i-1)-th, the two sketch line segments of i-th sketch line segment, i+1 item in i-th element
The same side of straight line, 2≤i≤n-1, if all sketch line segments and adjacent segments on sketch line are all same where retouching line segment
Side, then marking the sketch line is with the sketch line for gathering feature;
Whether second situation judges adjacent (i-1)-th, the two sketch line segments of i-th sketch line segment, i+1 item in i-th element
The same side of straight line where line segment, 2≤i≤n-1 are retouched, if having n-1 sketch line segment and adjacent segments on sketch line in the same side,
And have a sketch line segment line segment adjacent thereto in non-the same side, also marking the sketch line is with the sketch line for gathering feature;
Step 12, an optional sketch line in having the sketch line for gathering feature are sat by two endpoints of selected sketch line
Mark, determines the distance between two endpoints, if the end-point distances are in [0,20] range, then only using selected sketch line as expression
The sketch line of vertical target;
Step 13, judge it is untreated have gather the sketch line of feature and whether all selected, if so, executing step 12;Otherwise,
Execute step 14;
Step 14, with the method for super-pixel segmentation, around the sketch line for indicating pinpoint target in synthetic aperture radar SAR image
Pixel carry out super-pixel segmentation, by super-pixel conduct of the gray value of super-pixel after segmentation in [0,45] or [180,255]
Pinpoint target super-pixel;
Step 15 merges pinpoint target super-pixel, using the boundary of the pinpoint target super-pixel after merging as the side of pinpoint target
Boundary obtains the segmentation result of pinpoint target.
10. the SAR image segmentation method according to claim 1 based on ridge ripple filter and deconvolution structural model,
It is characterized in that, the specific steps of the homogenous region dividing method based on multinomial logistic regression prior model described in step (8)
It is as follows:
Step 1 is arbitrarily chosen a pixel from the pixel subspace of homogenous region, is established centered on selected pixel
3 × 3 square window calculates the standard deviation sigma of the window1;
The side length of square window is increased by 2, obtains new square window, calculate the standard deviation sigma of new square window by step 22;
Step 3, if standard deviation threshold method T3=3, if | σ1-σ2| < T3, then it is σ by standard deviation2Square window as final
Square window executes step 4;Otherwise, step 2 is executed;
Step 4 calculates the prior probability of center pixel in square window according to the following formula:
Wherein, p'1Indicate that the prior probability of center pixel in square window, exp () indicate exponential function operation, η ' indicates general
Rate model parameter, η ' value are 1, xk'' indicate the number of pixels for belonging to kth ' class in square window, k' ∈ [1 ..., K'], K'
Indicate the classification number of segmentation, K' value is 5, xi' indicate the number of pixels for belonging to the i-th ' class in the obtained square window of step 3;
The probability density of pixel grey scale is multiplied with the probability density of texture, obtains likelihood probability p' by step 52, wherein gray scale
Probability density is distributed to obtain by fading channel Nakagami, and the probability density of texture is distributed to obtain by t;
Step 6, by prior probability p1' and likelihood probability p2' be multiplied, obtain posterior probability p12';
Step 7 judges whether there are also untreated pixels in the pixel subspace of homogenous region, if so, executing step 1;Otherwise,
Execute step 9;
Step 8 obtains the segmentation result of homogenous region pixel subspace according to maximum posteriori criterion.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611269937.4A CN106611423B (en) | 2016-12-30 | 2016-12-30 | SAR image segmentation method based on ridge ripple filter and deconvolution structural model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611269937.4A CN106611423B (en) | 2016-12-30 | 2016-12-30 | SAR image segmentation method based on ridge ripple filter and deconvolution structural model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106611423A CN106611423A (en) | 2017-05-03 |
CN106611423B true CN106611423B (en) | 2019-05-21 |
Family
ID=58636801
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611269937.4A Active CN106611423B (en) | 2016-12-30 | 2016-12-30 | SAR image segmentation method based on ridge ripple filter and deconvolution structural model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106611423B (en) |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107341813B (en) * | 2017-06-15 | 2019-10-25 | 西安电子科技大学 | SAR image segmentation method based on Structure learning and sketch characteristic inference network |
CN107292268A (en) * | 2017-06-23 | 2017-10-24 | 西安电子科技大学 | The SAR image semantic segmentation method of quick ridge ripple deconvolution Structure learning model |
CN107403434B (en) * | 2017-07-28 | 2019-08-06 | 西安电子科技大学 | SAR image semantic segmentation method based on two-phase analyzing method |
CN107480677B (en) * | 2017-08-07 | 2020-04-28 | 北京深睿博联科技有限责任公司 | Method and device for identifying interest region in three-dimensional CT image |
CN107483058B (en) * | 2017-08-17 | 2019-10-22 | 西安电子科技大学 | A kind of match tracing reconstructing method instructed based on structuring observation and direction |
CN108109153B (en) * | 2018-01-12 | 2019-10-11 | 西安电子科技大学 | SAR image segmentation method based on SAR-KAZE feature extraction |
CN109272515B (en) * | 2018-08-17 | 2020-10-09 | 西安电子科技大学 | Unsupervised SAR image segmentation method based on high-order multi-scale CRF |
CN109344837B (en) * | 2018-10-22 | 2022-03-04 | 西安电子科技大学 | SAR image semantic segmentation method based on deep convolutional network and weak supervised learning |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104346814A (en) * | 2014-11-20 | 2015-02-11 | 西安电子科技大学 | SAR (specific absorption rate) image segmentation method based on hierarchy visual semantics |
CN105374033A (en) * | 2015-10-19 | 2016-03-02 | 西安电子科技大学 | SAR image segmentation method based on ridgelet deconvolution network and sparse classification |
CN105389798A (en) * | 2015-10-19 | 2016-03-09 | 西安电子科技大学 | Synthetic aperture radar (SAR) image segmentation method based on deconvolution network and mapping inference network |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2673949A1 (en) * | 2011-02-11 | 2013-12-18 | Canadian Space Agency | Method and system of increasing spatial resolution of multi-dimensional optical imagery using sensor's intrinsic keystone |
-
2016
- 2016-12-30 CN CN201611269937.4A patent/CN106611423B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104346814A (en) * | 2014-11-20 | 2015-02-11 | 西安电子科技大学 | SAR (specific absorption rate) image segmentation method based on hierarchy visual semantics |
CN105374033A (en) * | 2015-10-19 | 2016-03-02 | 西安电子科技大学 | SAR image segmentation method based on ridgelet deconvolution network and sparse classification |
CN105389798A (en) * | 2015-10-19 | 2016-03-09 | 西安电子科技大学 | Synthetic aperture radar (SAR) image segmentation method based on deconvolution network and mapping inference network |
Non-Patent Citations (2)
Title |
---|
Local Maximal Homogeneous Region Search for SAR Speckle Reduction With Sketch-Based Geometrical Kernel Function;Jie Wu et.al;《IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING》;20140930;第52卷(第9期);第5751-5764页 |
SAR Image Segmentation Based on Hierarchical Visual Semantic and Adaptive Neighborhood Multinomial Latent Model;Fang Liu et.al;《IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING》;20160731;第54卷(第7期);第4287-4301页 |
Also Published As
Publication number | Publication date |
---|---|
CN106611423A (en) | 2017-05-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106611423B (en) | SAR image segmentation method based on ridge ripple filter and deconvolution structural model | |
CN106683102B (en) | SAR image segmentation method based on ridge ripple filter and convolutional coding structure learning model | |
CN106611420B (en) | The SAR image segmentation method constrained based on deconvolution network and sketch map direction | |
CN106651884B (en) | Mean field variation Bayes's SAR image segmentation method based on sketch structure | |
CN106611421B (en) | The SAR image segmentation method constrained based on feature learning and sketch line segment | |
CN107341813B (en) | SAR image segmentation method based on Structure learning and sketch characteristic inference network | |
CN106611422B (en) | Stochastic gradient Bayes's SAR image segmentation method based on sketch structure | |
CN103049763B (en) | Context-constraint-based target identification method | |
CN105608692B (en) | Polarization SAR image segmentation method based on deconvolution network and sparse classification | |
CN106909902B (en) | Remote sensing target detection method based on improved hierarchical significant model | |
CN102013017B (en) | Method for roughly sorting high-resolution remote sensing image scene | |
CN104915676B (en) | SAR image sorting technique based on further feature study and watershed | |
CN105374033B (en) | SAR image segmentation method based on ridge ripple deconvolution network and sparse classification | |
CN109063594A (en) | Remote sensing images fast target detection method based on YOLOv2 | |
CN105930846B (en) | Classification of Polarimetric SAR Image method based on neighborhood information and SVGDL | |
CN105825502B (en) | A kind of Weakly supervised method for analyzing image of the dictionary study based on conspicuousness guidance | |
CN107403434B (en) | SAR image semantic segmentation method based on two-phase analyzing method | |
CN105930815A (en) | Underwater organism detection method and system | |
CN106846322B (en) | The SAR image segmentation method learnt based on curve wave filter and convolutional coding structure | |
CN109543688A (en) | A kind of novel meter reading detection and knowledge method for distinguishing based on multilayer convolutional neural networks | |
CN109409240A (en) | A kind of SegNet remote sensing images semantic segmentation method of combination random walk | |
Buddhiraju et al. | Comparison of CBF, ANN and SVM classifiers for object based classification of high resolution satellite images | |
CN106600595A (en) | Human body characteristic dimension automatic measuring method based on artificial intelligence algorithm | |
CN110503613A (en) | Based on the empty convolutional neural networks of cascade towards removing rain based on single image method | |
CN104408731B (en) | Region graph and statistic similarity coding-based SAR (synthetic aperture radar) image segmentation method |
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 |