CN110473205A - Remote sensing image information extracting method and system based on arrow bar phantom - Google Patents
Remote sensing image information extracting method and system based on arrow bar phantom Download PDFInfo
- Publication number
- CN110473205A CN110473205A CN201910617812.3A CN201910617812A CN110473205A CN 110473205 A CN110473205 A CN 110473205A CN 201910617812 A CN201910617812 A CN 201910617812A CN 110473205 A CN110473205 A CN 110473205A
- Authority
- CN
- China
- Prior art keywords
- arrow
- module
- remote sensing
- data
- bar phantom
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 69
- 238000004364 calculation method Methods 0.000 claims abstract description 16
- 230000002452 interceptive effect Effects 0.000 claims abstract description 15
- 238000012549 training Methods 0.000 claims abstract description 10
- 238000000605 extraction Methods 0.000 claims description 16
- 230000003595 spectral effect Effects 0.000 claims description 12
- 238000005267 amalgamation Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 5
- 230000015572 biosynthetic process Effects 0.000 claims description 4
- 238000003786 synthesis reaction Methods 0.000 claims description 4
- 241000239290 Araneae Species 0.000 claims description 3
- 230000002194 synthesizing effect Effects 0.000 claims description 3
- 230000008859 change Effects 0.000 abstract description 12
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 241000208340 Araliaceae Species 0.000 description 4
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 4
- 235000003140 Panax quinquefolius Nutrition 0.000 description 4
- 238000001514 detection method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 235000008434 ginseng Nutrition 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000001228 spectrum Methods 0.000 description 3
- 238000005520 cutting process Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 241000894007 species Species 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000000227 grinding Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Image Processing (AREA)
Abstract
The invention discloses remote sensing image information extracting methods and system based on arrow bar phantom, comprising the following steps: building arrow bar phantom；Object each in the arrow bar phantom is subjected to feature calculation；Training sample is acquired, the characteristic dimension after calculating is exercised supervision classification；The difference that two phase images of front and back divide figure spot in characteristic dimension is calculated, the big figure spot of difference is determined as doubtful region of variation, variation discovery information is extracted by arrow grid interactive editor.The beneficial effects of the present invention are: displaying live view checks arrow bar phantom information；Real-time update neighborhood topology index file realizes the automatic building of topological relation；It applies it to high spatial resolution remote sense image satellite images interpretation, in remote sensing image extracting change information, meets the application demand that different high score remotely-sensed data source information are extracted.
Description
Technical field
The present invention relates to remote sensing technology fields, it particularly relates to the remote sensing image information extraction side based on arrow bar phantom
Method and system.
Background technique
With the continuous transmitting of remote sensing satellite, remote sensing image and history land cover pattern data are increasing, in remote sensing image
Interactive editor is inevitable in information last handling process, and the strong interaction that traditional GIS platform is unable to satisfy information extraction post-processing needs
It asks, it is also difficult to the huge arrow grid data volume of support volume.
High spatial resolution remote sense image, compared to low resolution remote sensing image in tradition, spectral signature is less, space line
Reason, geological information are more abundant, and traditional using pixel as analytical unit, carry out the extraction of high score remote sensing information, can not be effectively sharp
The Spectral Properties of image can be completely combined using the object of homogeneity as analytical unit with the spatial texture of high score image, geometrical characteristic
Sign, textural characteristics, geometrical characteristic carry out the information extraction of high score data.
For the problems in the relevant technologies, currently no effective solution has been proposed.
Summary of the invention
For above-mentioned technical problem in the related technology, the present invention proposes that the remote sensing image information based on arrow bar phantom extracts
Method and system can not only avoid the weak problem of grid interactive editor's ability and utilize GIS means editor vector data mistake
Transfer problem caused by the topological issues and vector sum raster data type being easy to appear in journey are different.
To realize the above-mentioned technical purpose, the technical scheme of the present invention is realized as follows:
A kind of remote sensing image information extracting method based on arrow bar phantom, comprising the following steps:
Building arrow bar phantom；
Object each in the arrow bar phantom is subjected to feature calculation；
Training sample is acquired, the characteristic dimension after calculating is exercised supervision classification；
The difference that two phase images of front and back divide figure spot in characteristic dimension is calculated, the big figure spot of difference is determined as doubtful change
Change region, variation discovery information is extracted by arrow grid interactive editor.
Further, the building arrow bar phantom includes:
Obtain data, wherein data are raster data or input raster data and vector data；
Parameter is set, wherein the parameter includes region merging technique parameter and wave band weight parameter；
The image is subjected to piecemeal；
Merge block data according to similar area；
Segmented areas amalgamation result is synthesized, the arrow bar phantom is generated.
Further, include: by image progress piecemeal
The image is subjected to piecemeal；
Traverse the block number evidence of each image；
By every block number according to progress image first time region merging technique；
Generate the region merging technique result of every block number evidence.
Further, information includes raster data, vector data, region merging technique parameter setting letter in the arrow bar phantom
Breath, region merging technique result and topological Index file.
Further, the characteristic dimension includes vector characteristic, geometrical characteristic, spectral signature, textural characteristics and customized
Feature.
Another aspect of the present invention provides a kind of remote sensing image information extraction system based on multi-layer multi-scale division,
Include:
Module is constructed, for constructing arrow bar phantom；
Feature calculation module, for object each in the arrow bar phantom to be carried out feature calculation；
Supervised classification module exercises supervision the characteristic dimension after calculating classification for acquiring training sample；
Difference calculating module divides the difference of figure spot for calculating two phase images of front and back in characteristic dimension, and difference is big
Figure spot be determined as doubtful region of variation, pass through arrow grid interactive editor extract variation discovery information.
Further, the building module includes:
Module is obtained, for obtaining data, wherein data are raster data or input raster data and vector data；
Parameter setting module, for parameter to be arranged, wherein the parameter includes region merging technique parameter and wave band weight ginseng
Number；
Image block module, for the image to be carried out piecemeal；
Merging module, for merging block data according to similar area；
Synthesis module is generated the arrow bar phantom for synthesizing segmented areas amalgamation result.
Further, the image block module includes:
Piecemeal module, for the image to be carried out piecemeal；
Spider module, for traversing the block number evidence of each image；
First time region merging technique module is used for every block number according to progress image first time region merging technique；
Generation module, for generating the region merging technique result of every block number evidence.
Further, information includes raster data, vector data, region merging technique parameter setting letter in the arrow bar phantom
Breath, region merging technique result and topological Index file.
Further, the characteristic dimension includes vector characteristic, geometrical characteristic, spectral signature, textural characteristics and customized
Feature.
Beneficial effects of the present invention:
Displaying live view checks arrow bar phantom information, meanwhile, to swear that bar phantom for basic data, is supported from information extraction application
To interactive editor, roaming and fast browsing whole process, Data Format Transform consumes caused by saving because of data type difference
When；Arrow bar phantom records the topological relation between figure in real time, and figure spot object process cuts, digs a hole, concatenating, merging and assignment
A series of editors can realize the automatic building of topological relation with real-time update neighborhood topology index file；
To swear bar phantom as basic data, feature calculation, the supervised classification, variation discovery, arrow grid of arrow bar phantom are realized
The functions such as interactive editor apply it to high spatial resolution remote sense image satellite images interpretation, remote sensing image change information mentions
In taking, meet the application demand that different high score remotely-sensed data source information are extracted.
Detailed description of the invention
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without creative efforts, can also obtain according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the flow diagram of the building arrow bar phantom described according to embodiments of the present invention；
Fig. 2 is the arrow bar phantom display renderings described according to embodiments of the present invention；
Fig. 3 is that the arrow bar phantom described according to embodiments of the present invention is applied to the process that high score remote sensing image information interprets and shows
It is intended to；
Fig. 4 (a) is GF-1 remote sensing image according to an embodiment of the present invention+history road network effect data diagram；
Fig. 4 (b) is the effect diagram of SVM supervised classification result according to an embodiment of the present invention；
Fig. 5 is that the process of the remote sensing image information extracting method based on arrow bar phantom described according to embodiments of the present invention is shown
It is intended to；
Fig. 6 (a) is the remote sensing image of GF1 in 2014 according to embodiments of the present invention+change detection result diagram；
Fig. 6 (b) is the remote sensing image of GF1 in 2017 according to embodiments of the present invention+change detection result diagram；
Fig. 7 is that the structure of the remote sensing image information extraction system based on arrow bar phantom described according to embodiments of the present invention is shown
It is intended to.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art's every other embodiment obtained belong to what the present invention protected
Range.
As shown in Figure 3 and Figure 5, the remote sensing image information extraction side based on arrow bar phantom according to embodiments of the present invention
Method, comprising the following steps:
Building arrow bar phantom；
Specifically, there are two types of situations for creation arrow bar phantom: 1) using raster data as input, adjacent similar area being merged
For figure spot object, the index relative that each adjacent area merges is recorded, generates the data set of homogeneity figure spot object；2) with grid number
, as inputting, adjacent similar area is merged into figure spot object, figure spot object inherits vector in merging process according to vector data
Data shape and attributive character, while recording each adjacent area and merging index relative generate and cover grid and Vector Message
Figure spot object data set.
Object each in the arrow bar phantom is subjected to feature calculation；
Specifically, carrying out feature calculation, feature calculation to each object in arrow bar phantom to swear bar phantom as basic data
Have: vector characteristic, geometrical characteristic, spectral signature, textural characteristics and user-defined feature, wherein vector characteristic is that input data is
When grid and vector data, the attributive character of corresponding vector data；Geometrical characteristic has 5, mainly from figure spot object morphology into
Row description has area, perimeter, compact degree=area/perimeter, length-width ratio=length/width, the long and narrow degree=area/outer of figure spot
Envelope product；Spectral signature has 5, is mainly described from the spectral signature of segmentation figure spot, have the spectral value of figure spot, standard deviation,
Maximum value, minimum value, brightness value；Textural characteristics are mainly gray level co-occurrence matrixes feature, and gray level co-occurrence matrixes are a kind of by grinding
The spatial correlation characteristic of gray scale is studied carefully to describe the common method of texture；User-defined feature mainly provides Feature Calculator, for using
The customized new feature in family can carry out existing feature adding, subtract, the operation processings such as multiplication and division, power operation, obtain it is new from
Defined feature.
Training sample is acquired, the characteristic dimension after calculating is exercised supervision classification；
Specifically, providing sample management and supervised classification ability, wherein sample management have sample collection, sample delete and
Sample export function；Supervised classification method has Bayes, neural network, support vector machine, decision tree and random forest etc., according to
Ground mulching classification, locality species very originally, input the feature of different dimensions, using the classifier of supervised classification, realize distant
Feel image automatic interpretation.
The difference that two phase images of front and back divide figure spot in characteristic dimension is calculated, the big figure spot of difference is determined as doubtful change
Change region, variation discovery information is extracted by arrow grid interactive editor.
Specifically, providing remote sensing image variation detection method is differential technique, common method has Euclidean distance, manhatton distance
With Chebyshev's distance etc., divide the difference of figure spot in characteristic dimension by calculating two phase images of front and back, by the big figure of difference
Spot is determined as doubtful region of variation, obtains variation discovery information；Arrow grid interactive editor confirmation is provided, final variation discovery knot is obtained
The abilities such as fruit, including figure spot cuts, digs a hole, concatenating, merging, assignment, wherein figure spot cutting be a figure spot is cut into it is multiple
The process of figure spot；It is to carry out processing of digging a hole to existing figure spot that figure spot, which is dug a hole,；Figure spot concatenation is to carry out deburring to existing figure spot boundary
Processing；Figure spot merging is to merge processing to multiple adjacent figure spots；Figure spot assignment is to carry out assignment again to the attribute of figure spot
Process.
Specific embodiments of the present invention, the building arrow bar phantom include:
Obtain data, wherein data are raster data or input raster data and vector data；
Parameter is set, wherein the parameter includes region merging technique parameter and wave band weight parameter；
The image is subjected to piecemeal；
Merge block data according to similar area；
Segmented areas amalgamation result is synthesized, the arrow bar phantom is generated.
Specifically, S1 input data, wherein input data includes input raster data or input raster data and arrow
Measure data；Parameter is arranged in S2, wherein setting area merges parameter and wave band weight parameter, design parameter are shown in Table 1.
Table 1 is arranged parameter and describes
S3 homogeneous region merges, and carries out piecemeal to the image of input, traverses the block number evidence of each image, to every block number according into
Row image first time region merging technique merges and stops being controlled by scale parameter, generates the region merging technique result of every block number evidence.
S4 segmented areas amalgamation result merges, and carries out the region merging technique result between different masses to inlay synthesis.
S5 output arrow bar phantom, arrow bar phantom display effect are shown in that Fig. 2, the arrow bar phantom file of output include: 1) to input number
According to information, including raster data and vector data information, raster data information has geographical coordinate, wave band number, locating depth, data class
The essential informations such as type, vector data information have the essential informations such as geographical coordinate, vector attributive character；2) region merging technique parameter setting
Information；3) region merging technique is as a result, be the figure spot object result of grid format；4) topological Index file records adjacent figure spot object
Neighborhood relationships.
Specific embodiments of the present invention, the image, which is carried out piecemeal, includes:
The image is subjected to piecemeal；
Traverse the block number evidence of each image；
By every block number according to progress image first time region merging technique；
Generate the region merging technique result of every block number evidence.
Specific embodiments of the present invention, information includes raster data, vector data, region merging technique ginseng in the arrow bar phantom
Number setting information, region merging technique result and topological Index file.
Specific embodiments of the present invention, the characteristic dimension include vector characteristic, geometrical characteristic, spectral signature, texture spy
It seeks peace user-defined feature.
Another aspect of the present invention provides the remote sensing image information extraction system based on multi-layer multi-scale division, comprising:
Module is constructed, for constructing arrow bar phantom；
Feature calculation module, for object each in the arrow bar phantom to be carried out feature calculation；
Supervised classification module exercises supervision the characteristic dimension after calculating classification for acquiring training sample；
Difference calculating module divides the difference of figure spot for calculating two phase images of front and back in characteristic dimension, and difference is big
Figure spot be determined as doubtful region of variation, pass through arrow grid interactive editor extract variation discovery information.
Specific embodiments of the present invention, the building module include:
Module is obtained, for obtaining data, wherein data are raster data or input raster data and vector data；
Parameter setting module, for parameter to be arranged, wherein the parameter includes region merging technique parameter and wave band weight ginseng
Number；
Image block module, for the image to be carried out piecemeal；
Merging module, for merging block data according to similar area；
Synthesis module is generated the arrow bar phantom for synthesizing segmented areas amalgamation result.
Specific embodiments of the present invention, the image block module include:
Piecemeal module, for the image to be carried out piecemeal；
Spider module, for traversing the block number evidence of each image；
First time region merging technique module is used for every block number according to progress image first time region merging technique；
Generation module, for generating the region merging technique result of every block number evidence.
Specific embodiments of the present invention, information includes raster data, vector data, region merging technique ginseng in the arrow bar phantom
Number setting information, region merging technique result and topological Index file.
Specific embodiments of the present invention, the characteristic dimension include vector characteristic, geometrical characteristic, spectral signature, texture spy
It seeks peace user-defined feature.
Two embodiments of detection are explained and changed with supervised classification remote sensing, carry out arrow bar phantom and its are applied to high score remote sensing
Image information extracts explanation.
In order to facilitate understanding above-mentioned technical proposal of the invention, below by way of in specifically used mode to of the invention above-mentioned
Technical solution is described in detail.
Swear that bar phantom is interpreted applied to high score remote sensing image information
Arrow bar phantom is applied to the interpretation of high score remote sensing image information, detailed process is shown in Fig. 3, firstly, choosing Erhai area
GF1 multi-spectrum remote sensing image and history road network data are shown in Fig. 4 (a) as input data, secondly, creation arrow bar phantom, so
Afterwards, to swear that bar phantom carries out feature calculation for basic data, the feature of each figure spot object is calculated, the specific feature that calculates has:
(1) vector characteristic: the category feature of history road network data；
(2) mean value: figure layer average value is calculated in the figure layer value by constituting all n pixels an of imaged object；
(3) standard deviation: standard deviation is calculated in the figure layer value by constituting all n pixels an of imaged object；
(4) brightness: the average value of the spectrum mean value of an imaged object；
(5) maximum value: the maximum value that the figure layer value by constituting all n pixels an of imaged object sorts；
(6) minimum value: the minimum value that the figure layer value by constituting all n pixels an of imaged object sorts；
(7) NDWI:NDWI (Normalized Difference Water Index normalizes aqua index), formula are as follows:
NDWI=(p (Green)-p (NIR))/(p (Green)+p (NIR))；
(8) NDVI:NDVI (Normalized Difference Vegetation Index, normalized differential vegetation index),
Formula are as follows: NDVI=(p (NIR)-p (R))/(p (NIR)+p (R))；
On the basis of features described above calculates, existing category of roads is inherited；Then, it acquires forest land, arable land, water body, build
Four kinds of ground species training samples are built, various dimensions feature is inputted, is supported vector machine (abbreviation SVM, Support Vector
Machine) supervised classification.
Wherein, the SVM supervised classification of use is a kind of two points of disaggregated models, and basic model is defined as on feature space
It is spaced maximum linear classifier, learning strategy is margin maximization, can finally be converted into a convex quadratic programming problem
Solution.Principle is as follows:
Assuming that input training sample { (x1, y1) ..., (xl, yl)}(xi∈Rn, yi∈ { -1,1 }, i=1,2 ..., l) may be used
By hyperplane<w, x>+b=0 (b ∈ R) linear partition is two classes, and hyperplane must meet yi(<w, x>+b)>=1 (i=1,
2 ..., l),<*, *>expression inner product of vector, if the class interval of optimal hyperlane required by SVM is maximum, this is equivalent to solve
Quadratic programming problem:
s.t.yi(< w, xi>+b) >=1 i=1,2 ..., l
Decision function based on optimal separating hyper plane are as follows:
Wherein,For the unique solution of above-mentioned quadratic programming problem.
When the training data linearly inseparable in original feature space, slack variable ζ is introducedi>=0 (i=1,2 ...,
L), it allows there are error sample, corresponding optimization problem are as follows:
s.t.yi(< w, xi>+b)≥1-ζi ζi≥0；I=1,2 ..., l
Wherein, C > 0 is penalty factor, for controlling the degree to error sample punishment.
On the basis of supervised classification result, is cut, dug a hole, concatenated, merged, assignment function using arrow grid interactive editor, it is right
Supervised classification result carries out precise modification, obtains final information extraction achievement, sees Fig. 4 (b).
Swear that bar phantom is applied to remote sensing image extracting change information
In the present embodiment, arrow bar phantom is applied to remote sensing image extracting change information, detailed process is shown in Fig. 5.Choose Qi
Lianshan Mountain 2014 and GF1 multi-spectrum remote sensing image in 2017, are shown in Fig. 6.Firstly, two phase remote sensing image datas are carried out band overlapping
Arrow bar phantom is created as input data；Secondly, to swear that bar phantom for basic data, carries out feature calculation, calculating feature has:
(1) homogeney: reflecting the homogeney of image texture, measurement image texture localized variation number.Its value is then said greatly
Lack variation, highly uniform, the formula in part between the different zones of bright image texture are as follows:Wherein, i, j distinguish table
Show that row and column, N indicate the number of row or column, PI, jFor the value after pixel (i, j) standardization.
(2) mean value: the regular degree of mean value reflection texture.Texture is rambling, it is difficult to description, it is worth smaller；Rule
Property is strong, and the value easily described is bigger.Formula are as follows:Wherein, i, j respectively indicate row and column, and N indicates row or column
Number, PI, jFor the value after pixel (i, j) standardization.
(3) standard deviation: the measurement of variance and standard deviation reflection pixel value and mean bias, when the grey scale change of image compares
When big, variance criterion difference is larger.Formula are as follows:Wherein, i, j respectively indicate row and column, and N is indicated
The number of row or column, PI, jFor the value after pixel (i, j) standardization, uI, jFor gray level co-occurrence matrixes mean value.
(4) angular second moment: being that gray level co-occurrence matrixes element value obtains quadratic sum, so also referred to as energy, reflects image ash
Spend be evenly distributed degree and texture fineness.Texture is thick, and energy is big, and texture is thin, and energy is small.Formula are as follows:Its
In, i, j respectively indicate row and column, and N indicates the number of row or column, PI, jFor the value after pixel (i, j) standardization.
(5) contrast: the value of metric matrix is the number how being distributed with localized variation in image, has reacted the clear of image
The rill depth of clear degree and texture.The rill of texture is deeper, and contrast is bigger, and effect is more clear；Conversely, reduced value is small, then rill
Shallowly, effect is fuzzy.Formula are as follows:Wherein, i, j respectively indicate row and column, and N indicates the number of row or column,
PI, jFor the value after pixel (i, j) standardization, uI, jFor gray level co-occurrence matrixes mean value, δI, jFor gray level co-occurrence matrixes standard deviation.
(6) correlation: for measure the gray level of image be expert at or column direction on similarity degree, therefore be worth size it is anti-
Local gray level correlation is answered, value is bigger, and correlation is also bigger.Formula are as follows:Wherein, i, j distinguish table
Show that row and column, N indicate the number of row or column, PI, jFor the value after pixel (i, j) standardization.
(7) non-similarity: it is similar to contrast, but be linearly increasing.If local contrast is higher, non-similarity
Also higher.Formula are as follows:Wherein, i, j respectively indicate row and column, and N indicates the number of row or column, PI, jFor picture
Value after plain (i, j) standardization.
(8) entropy: image includes the randomness metrics of information content, when all values are equal or pixel value table in co-occurrence matrix
When revealing maximum randomness, entropy is maximum；Therefore entropy shows the complexity of image grayscale distribution, and entropy is bigger, image
It is more complicated.Formula are as follows:Wherein, i, j respectively indicate row and column, and N indicates the number of row or column, PI, jFor
Value after pixel (i, j) standardization.
On the basis of features described above calculates, it is changed discovery, the method for use is direct differential technique, in the present embodiment
Two phase image feature differences are calculated using Euclidean distance, the Euclidean distance principle of use is as follows:
Euclidean distance indicates that, in the actual distance in n-dimensional space between two points, n is the formula in space:
Wherein, n indicates characteristic dimension, X={ x1, x2..., xnIndicate first point feature, Y={ y1, y2, ...,
Yn } indicate second point feature.
Euclidean distance calculated result is ranked up, the big area output of difference is doubtful region of variation, by manually handing over
Mutual edit validation obtains final variation discovery as a result, seeing Fig. 6 a and Fig. 6 b, and black lines are rendered into variation discovery as a result, discovery
Qilianshan Area 2014-2017 increases a road newly.
In conclusion displaying live view checks arrow bar phantom information by means of above-mentioned technical proposal of the invention, meanwhile, with
Swear that bar phantom is basic data, support is applied to interactive editor, roaming and fast browsing whole process from information extraction, saves
Because Data Format Transform is time-consuming caused by data type is different；Arrow bar phantom records the topological relation between figure in real time, schemes
Spot object by cutting, digging a hole, concatenating, merging and a series of editors of assignment, can with real-time update neighborhood topology index file,
Realize the automatic building of topological relation；To swear that bar phantom for basic data, realizes feature calculation, the supervision point of arrow bar phantom
Class, variation discovery, the arrow functions such as grid interactive editor apply it to high spatial resolution remote sense image satellite images interpretation, distant
Feel in image extracting change information, meets the application demand that different high score remotely-sensed data source information are extracted.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Within mind and principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of remote sensing image information extracting method based on arrow bar phantom, which comprises the following steps:
Building arrow bar phantom；
Object each in the arrow bar phantom is subjected to feature calculation；
Training sample is acquired, the characteristic dimension after calculating is exercised supervision classification；
The difference that two phase images of front and back divide figure spot in characteristic dimension is calculated, the big figure spot of difference is determined as doubtful variation zone
Variation discovery information is extracted by arrow grid interactive editor in domain.
2. the remote sensing image information extracting method according to claim 1 based on arrow bar phantom, which is characterized in that the building
Swear that bar phantom includes:
Obtain data, wherein the data are raster data or input raster data and vector data；
Parameter is set, wherein the parameter includes region merging technique parameter and wave band weight parameter；
The image is subjected to piecemeal；
Merge block data according to similar area；
Segmented areas amalgamation result is synthesized, the arrow bar phantom is generated.
3. the remote sensing image information extracting method according to claim 1 based on arrow bar phantom, which is characterized in that by the shadow
Include: as carrying out piecemeal
The image is subjected to piecemeal；
Traverse the block number evidence of each image；
By every block number according to progress image first time region merging technique；
Generate the region merging technique result of every block number evidence.
4. the remote sensing image information extracting method according to claim 1 based on arrow bar phantom, which is characterized in that the arrow grid
Information includes raster data, vector data, region merging technique parameter setting information, region merging technique result and topological Index text in model
Part.
5. the remote sensing image information extracting method based on arrow bar phantom described in -4 any one according to claim 1, which is characterized in that
The characteristic dimension includes vector characteristic, geometrical characteristic, spectral signature, textural characteristics and user-defined feature.
6. a kind of remote sensing image information extraction system based on multi-layer multi-scale division characterized by comprising
Module is constructed, for constructing arrow bar phantom；
Feature calculation module, for object each in the arrow bar phantom to be carried out feature calculation；
Supervised classification module exercises supervision the characteristic dimension after calculating classification for acquiring training sample；
Difference calculating module divides the difference of figure spot for calculating two phase images of front and back, by the big figure of difference in characteristic dimension
Spot is determined as doubtful region of variation, extracts variation discovery information by arrow grid interactive editor.
7. according to claim 6 based on the remote sensing image information extraction system of multi-layer multi-scale division, which is characterized in that
The building module includes:
Module is obtained, for obtaining data, wherein data are raster data or input raster data and vector data；
Parameter setting module, for parameter to be arranged, wherein the parameter includes region merging technique parameter and wave band weight parameter；
Image block module, for the image to be carried out piecemeal；
Merging module, for merging block data according to similar area；
Synthesis module is generated the arrow bar phantom for synthesizing segmented areas amalgamation result.
8. according to claim 6 based on the remote sensing image information extraction system of multi-layer multi-scale division, which is characterized in that
The image block module includes:
Piecemeal module, for the image to be carried out piecemeal；
Spider module, for traversing the block number evidence of each image；
First time region merging technique module is used for every block number according to progress image first time region merging technique；
Generation module, for generating the region merging technique result of every block number evidence.
9. according to claim 6 based on the remote sensing image information extraction system of multi-layer multi-scale division, which is characterized in that
Information includes raster data, vector data, region merging technique parameter setting information, region merging technique result and opens up in the arrow bar phantom
Flutter index file.
10. based on the remote sensing image information extraction system of multi-layer multi-scale division according to claim any one of 6-9,
It is characterized in that, the characteristic dimension includes vector characteristic, geometrical characteristic, spectral signature, textural characteristics and user-defined feature.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910617812.3A CN110473205A (en) | 2019-07-10 | 2019-07-10 | Remote sensing image information extracting method and system based on arrow bar phantom |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910617812.3A CN110473205A (en) | 2019-07-10 | 2019-07-10 | Remote sensing image information extracting method and system based on arrow bar phantom |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110473205A true CN110473205A (en) | 2019-11-19 |
Family
ID=68507496
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910617812.3A Pending CN110473205A (en) | 2019-07-10 | 2019-07-10 | Remote sensing image information extracting method and system based on arrow bar phantom |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110473205A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112800992A (en) * | 2021-02-03 | 2021-05-14 | 成都鼎安华智慧物联网股份有限公司 | Extraction method based on remote sensing data high-temperature abnormal information |
| CN112945352A (en) * | 2021-02-03 | 2021-06-11 | 成都鼎安华智慧物联网股份有限公司 | Extraction method based on remote sensing data water level abnormal information |
| CN113240026A (en) * | 2021-05-24 | 2021-08-10 | 中国科学院重庆绿色智能技术研究院 | Vector-grid combined inland water surface floater batch identification and extraction method |
| CN113822900A (en) * | 2021-07-09 | 2021-12-21 | 武汉大学 | Vector constraint object-oriented new image sample automatic selection method and system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080009966A1 (en) * | 2006-07-05 | 2008-01-10 | Battelle Energy Alliance, Llc | Occupancy Change Detection System and Method |
| CN101510374A (en) * | 2009-03-12 | 2009-08-19 | 中国科学院遥感应用研究所 | Method for automatically updating land-use map based on historical data and remote sensing data |
| CN101644572A (en) * | 2009-06-24 | 2010-02-10 | 中国科学院地理科学与资源研究所 | Detection method of ocean eddy variation based on historical similarity cases |
| CN103500344A (en) * | 2013-09-02 | 2014-01-08 | 中国测绘科学研究院 | Method and module for extracting and interpreting information of remote-sensing image |
| CN104751478A (en) * | 2015-04-20 | 2015-07-01 | 武汉大学 | Object-oriented building change detection method based on multi-feature fusion |
| CN104834006A (en) * | 2015-04-28 | 2015-08-12 | 中国地质大学（北京） | Processing method and apparatus of structural cause deposition parameter diagram |
| CN105225227A (en) * | 2015-09-07 | 2016-01-06 | 中国测绘科学研究院 | The method and system that remote sensing image change detects |
| CN108846832A (en) * | 2018-05-30 | 2018-11-20 | 理大产学研基地(深圳)有限公司 | A kind of change detecting method and system based on multi-temporal remote sensing image and GIS data |
| CN108986113A (en) * | 2018-07-06 | 2018-12-11 | 航天星图科技（北京）有限公司 | A kind of block parallel multi-scale division algorithm based on LLTS frame |
-
2019
- 2019-07-10 CN CN201910617812.3A patent/CN110473205A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080009966A1 (en) * | 2006-07-05 | 2008-01-10 | Battelle Energy Alliance, Llc | Occupancy Change Detection System and Method |
| CN101510374A (en) * | 2009-03-12 | 2009-08-19 | 中国科学院遥感应用研究所 | Method for automatically updating land-use map based on historical data and remote sensing data |
| CN101644572A (en) * | 2009-06-24 | 2010-02-10 | 中国科学院地理科学与资源研究所 | Detection method of ocean eddy variation based on historical similarity cases |
| CN103500344A (en) * | 2013-09-02 | 2014-01-08 | 中国测绘科学研究院 | Method and module for extracting and interpreting information of remote-sensing image |
| CN104751478A (en) * | 2015-04-20 | 2015-07-01 | 武汉大学 | Object-oriented building change detection method based on multi-feature fusion |
| CN104834006A (en) * | 2015-04-28 | 2015-08-12 | 中国地质大学（北京） | Processing method and apparatus of structural cause deposition parameter diagram |
| CN105225227A (en) * | 2015-09-07 | 2016-01-06 | 中国测绘科学研究院 | The method and system that remote sensing image change detects |
| CN108846832A (en) * | 2018-05-30 | 2018-11-20 | 理大产学研基地(深圳)有限公司 | A kind of change detecting method and system based on multi-temporal remote sensing image and GIS data |
| CN108986113A (en) * | 2018-07-06 | 2018-12-11 | 航天星图科技（北京）有限公司 | A kind of block parallel multi-scale division algorithm based on LLTS frame |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112800992A (en) * | 2021-02-03 | 2021-05-14 | 成都鼎安华智慧物联网股份有限公司 | Extraction method based on remote sensing data high-temperature abnormal information |
| CN112945352A (en) * | 2021-02-03 | 2021-06-11 | 成都鼎安华智慧物联网股份有限公司 | Extraction method based on remote sensing data water level abnormal information |
| CN113240026A (en) * | 2021-05-24 | 2021-08-10 | 中国科学院重庆绿色智能技术研究院 | Vector-grid combined inland water surface floater batch identification and extraction method |
| CN113240026B (en) * | 2021-05-24 | 2022-03-25 | 中国科学院重庆绿色智能技术研究院 | Vector-grid combined inland water surface floater batch identification and extraction method |
| CN113822900A (en) * | 2021-07-09 | 2021-12-21 | 武汉大学 | Vector constraint object-oriented new image sample automatic selection method and system |
| CN113822900B (en) * | 2021-07-09 | 2023-09-22 | 武汉大学 | Method and system for automatically selecting new image sample based on vector constraint object-oriented |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110473205A (en) | Remote sensing image information extracting method and system based on arrow bar phantom | |
| Bruzzone et al. | A multilevel context-based system for classification of very high spatial resolution images | |
| Darwish et al. | Image segmentation for the purpose of object-based classification | |
| de Pinho et al. | Land-cover classification of an intra-urban environment using high-resolution images and object-based image analysis | |
| Cousty et al. | Hierarchical segmentations with graphs: quasi-flat zones, minimum spanning trees, and saliency maps | |
| Lodha et al. | Aerial LiDAR data classification using support vector machines (SVM) | |
| Malinverni et al. | Hybrid object-based approach for land use/land cover mapping using high spatial resolution imagery | |
| CN101840581B (en) | Method for extracting profile of building from satellite remote sensing image | |
| CN107330875A (en) | Based on the forward and reverse heterogeneous water body surrounding enviroment change detecting method of remote sensing images | |
| dos Santos et al. | Interactive multiscale classification of high-resolution remote sensing images | |
| Sukhia et al. | Content-based remote sensing image retrieval using multi-scale local ternary pattern | |
| Ge et al. | Hyperspectral and LiDAR data classification using kernel collaborative representation based residual fusion | |
| Ziaei et al. | A rule-based parameter aided with object-based classification approach for extraction of building and roads from WorldView-2 images | |
| CN107992856A (en) | High score remote sensing building effects detection method under City scenarios | |
| Zhao et al. | Integration of ZiYuan-3 multispectral and stereo imagery for mapping urban vegetation using the hierarchy-based classifier | |
| Ver Hoef et al. | A primer on topological data analysis to support image analysis tasks in environmental science | |
| Chattaraj et al. | Semi-automated object-based landform classification modelling in a part of the Deccan Plateau of central India | |
| Wu et al. | Fusing optical and synthetic aperture radar images based on shearlet transform to improve urban impervious surface extraction | |
| Wang et al. | Simultaneous extracting area and quantity of agricultural greenhouses in large scale with deep learning method and high-resolution remote sensing images | |
| Lizarazo et al. | Fuzzy image segmentation for urban land-cover classification | |
| Su | Superpixel-based principal component analysis for high resolution remote sensing image classification | |
| Engstrom et al. | Evaluating the Relationship between Contextual Features Derived from Very High Spatial Resolution Imagery and Urban Attributes: A Case Study in Sri Lanka | |
| Zhao et al. | Improving object-oriented land use/cover classification from high resolution imagery by spectral similarity-based post-classification | |
| CN106570124B (en) | Remote sensing images semantic retrieving method and system based on object level correlation rule | |
| Shen et al. | Cropland extraction from very high spatial resolution satellite imagery by object-based classification using improved mean shift and one-class support vector machines |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191119 |
|
| RJ01 | Rejection of invention patent application after publication |