CN105447452A - Remote sensing sub-pixel mapping method based on spatial distribution characteristics of features - Google Patents

Remote sensing sub-pixel mapping method based on spatial distribution characteristics of features Download PDF

Info

Publication number
CN105447452A
CN105447452A CN201510781816.7A CN201510781816A CN105447452A CN 105447452 A CN105447452 A CN 105447452A CN 201510781816 A CN201510781816 A CN 201510781816A CN 105447452 A CN105447452 A CN 105447452A
Authority
CN
China
Prior art keywords
sub
pixel
pixed mapping
linear
objects
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
Application number
CN201510781816.7A
Other languages
Chinese (zh)
Inventor
葛咏
陈跃红
贾远信
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Institute of Geographic Sciences and Natural Resources of CAS
Original Assignee
Institute of Geographic Sciences and Natural Resources of CAS
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Institute of Geographic Sciences and Natural Resources of CAS filed Critical Institute of Geographic Sciences and Natural Resources of CAS
Priority to CN201510781816.7A priority Critical patent/CN105447452A/en
Publication of CN105447452A publication Critical patent/CN105447452A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/13—Satellite images
    • 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
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/176—Urban or other man-made structures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/182—Network patterns, e.g. roads or rivers
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes
    • G06V20/188—Vegetation
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Data Mining & Analysis (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Astronomy & Astrophysics (AREA)
  • Remote Sensing (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a remote sensing sub-pixel mapping method based on the spatial distribution characteristics of features, comprising the following steps: first, classifying remote sensing images into an area pattern, a line pattern and a point pattern according to the spatial geometric characteristics of features; then, under the hypothesis of spatial dependence, using a sub-pixel mapping method based on vector boundary to process area features, using a line feature template sub-pixel mapping method to process line features, and using a sub-pixel mapping method based on space pattern consistency matching to process point features; and finally, embedding the sub-pixel mapping results of the three spatial patterns to get a sub-pixel map of the images. According to the invention, a sub-pixel mapping theoretical model capable of processing point, line and area features at the same time and based on the spatial distribution characteristics of features is built theoretically, and the simulation precision is high.

Description

A kind of remote sensing sub-pixed mapping drafting method based on atural object spatial distribution characteristic
Technical field
The present invention relates to a kind of sub-pixed mapping drafting method, belong to Geo_spatial Information Technology field.
Background technology
Remote sensing science and technology are applied to land resources survey extensively and profoundly, hydrologic process is simulated, landscape ecological, agricultural assesss and field (plum Anxin, 2001 such as disaster monitoring, forest resources management, Global climate change; Zhou Chenghu etc., 1999), the remote sensing image and products thereof of high spatial, time and spectral resolution adapts to it to study the development trend (Zhao's inch, 2003) applied further in each field.But in remote sensing image acquisition process, due to the impact of external environment condition and the limitation of internal sensor self, cause ubiquity mixed pixel (MixedPixel) in image, thus limit the application of remote sensing image and constrain extraction (DeJongandvanderMeer, 2004 of sensor information; Fisher, 1997).Especially, in Classification in Remote Sensing Image process, mixed pixel can make traditional Hard clustering (HardClassification) result produce error and uncertain (Atkinson, 2009; Bai Yanchen and Wang Jingfeng, 2003; Ge Yong and Wang Jingfeng, 2003; Ge Yong and Li Sanping, 2008).Mixed pixel is only appointed as reducing Hard clustering the error that single classification brings to result, soft classification (SoftClassification) method is suggested and expresses by the more scientific and reasonable mode of one namely classification results describes pixel shared by each atural object classification than the form of (also claiming Soft Inform ation, mark image) area ratio with hundred.Although softer point of soft classification results gives expression to how rational information, the locus of each atural object classification in mixed pixel is but unknown, and this brings again new uncertainty (Tatemetal, 2002a; Ibrahimetal, 2005), namely sub-pixed mapping category attribute locus is uncertain.For take into account neural classifier can express scientifically and rationally classification results and hard class methods can clear and definite not concrete locus class methods can the locus advantage separately of clear and definite not concrete locus, sub-pixed mapping drawing (or claim super-resolution system, Sub-pixelMappingorSuper-resolutionMapping) concept is formally proposed and is enjoyed concern and the attention (Atkinson, 1997) of numerous scholars.
The drawing of remote sensing sub-pixed mapping is the soft class result utilizing low spatial resolution, orients the atural object spatial distribution map under given any high spatial resolution in mixed pixel.The main object of sub-pixed mapping drawing research is mixed pixel, and its Producing reason complexity is various, and the mixed pixel type that different image-forming conditions produces is also different.According to the atural object spatial distribution characteristic that mixed pixel is inner different, relatively generally acknowledge that a kind of class standard is divided into two class (WoodcockandStrahler, 1987): (1) H type mixed pixel (H-TypeorH-Resolution) refers to that atural object patch is greater than the area of a pixel and the mixed pixel that produces at multiple atural object intersection, atural object is wherein the H type atural object with spatial coherence distribution characteristics, can be similar to and think that it is the atural object (AreaPattern, AP) of planar pattern; (2) L-type mixed pixel (L-TypeorL-Resolution) refers to the mixed pixel that the size that atural object patch is less than a pixel area produces in pixel inside with scattered independently formal distribution, atural object is wherein the L-type atural object with special heterogeneity distribution characteristics, can be similar to and think that it is the atural object (PointPattern, PP) of point-like pattern.But, three large geographic elements not only comprise the area feature in H type mixed pixel and the punctual geo-objects in L-type mixed pixel, also comprise and planar and the visibly different linear ground object (LinePattern, LP) with elongated connectedness of punctual geo-objects spatial distribution characteristic.Therefore, the sub-pixed mapping drawing of linear ground object is also that numerous scholar urgently pays close attention to and one of emphasis studied.
Through the developmental research in more than ten years, Chinese scholars proposes many outstanding sub-pixed mapping drafting algorithms in succession, wherein most sub-pixed mapping drafting method is mainly based on the maximum space distribution of locating point, line and area feature in view picture image in mixed pixel of spatial coherence simultaneously, and research separately for the point-like in L-type mixed pixel and linear ground object drawing is very micro-.Although the sub-pixed mapping drafting algorithm maximum based on spatial coherence has good treatment effect to area feature, but it can be made to produce the aggregation of punctual geo-objects when processing and having the punctual geo-objects of heterogeneous distribution characteristics, and be often difficult to when processing linear ground object ensure the connective spatial distribution characteristic of linear ground object and easily produce " sawtooth effect ".Thus, sub-pixed mapping draughtsmanship is developed so far that to impel people to recognize be only can restrict its further development and application based on the sub-pixed mapping drafting method that spatial coherence is maximum, also makes people pay attention to and studies the drawing problem of the inner point-like of mixed pixel and wire spatial distribution characteristic atural object.Therefore, while considering the area feature sub-pixed mapping drawing in H type mixed pixel, must pay much attention to and study the drawing problem of the inner point-like of mixed pixel and wire spatial distribution characteristic atural object, dotted line area feature of analyzing and researching theoretically produces the mechanism of dissimilar mixed pixel, character and type, the treatment technology that simultaneously can process dotted line area feature sub-pixed mapping drawing problem in mixed pixel is sought from method, thus make up the deficiency of current sub-pixed mapping drawing research, improve the precision of the meticulous land cover classification figure under sub-pixed mapping yardstick further.
Summary of the invention
The technical matters that the present invention solves: overcome the deficiencies in the prior art, a kind of remote sensing sub-pixed mapping drafting method of atural object spatial distribution characteristic is provided, build one theoretically and can locate dotted line area feature, sub-pixed mapping Map compilation theory model based on atural object spatial distribution characteristic simultaneously, the result good visual effect of sub-pixed mapping drawing and precision is high.
First, for mark image C, utilize region growing model and shape index, judge the Spatial Distribution Pattern (planar distribution, wire distribution and spot distribution) of atural object.Secondly, the area feature sub-pixed mapping chartography based on boundary polygon is utilized to carry out sub-pixed mapping drawing to the area feature of mark image; Linear die matching algorithm is utilized to carry out sub-pixed mapping drawing to the linear ground object of mark image; Atural object spatial model consistance matching algorithm is utilized to carry out sub-pixed mapping drawing to the punctual geo-objects of mark image.The last sub-pixed mapping charting results inlaying area feature, punctual geo-objects and linear ground object successively obtains the sub-pixed mapping charting results of view picture mark image C, is the value of null pixel with the type of ground objects replacement attribute that neighborhood frequency is maximum.
Technical scheme of the present invention: a kind of remote sensing sub-pixed mapping drafting method based on atural object spatial distribution characteristic, comprises the steps:
Step 1, pre-service is carried out to input remote sensing image, and then the component ratio utilizing neural classifier to obtain in remote sensing image each pixel shared by often kind of atural object classification is called abundance or mark image; Described often kind of atural object classification is three kinds of atural objects, i.e. punctual geo-objects, linear ground object and area feature;
Step 2, utilize region growing model, by the zones of different that mark Image Segmentation becomes, each region after adopting ordered data storage organization to store segmentation, is convenient to calculate the area in each region, girth and border;
The type of ground objects of step 3, zoning: the shape density index in the region of mark image after computed segmentation, according to corresponding discrimination threshold, realizes the division of punctual geo-objects, linear ground object and area feature;
Step 4, three kinds of atural object division results according to step 3, sub-pixed mapping drafting method based on linear ground object template matches is adopted to linear ground object mixed pixel, sub-pixed mapping drafting method based on vector border is adopted to area feature mixed pixel, sub-pixed mapping drafting method based on spatial coherence and Model Matching is adopted to punctual geo-objects mixed pixel, obtains punctual geo-objects, linear ground object, area feature sub-pixed mapping charting results;
Step 5, inlay the sub-pixed mapping charting results of punctual geo-objects, linear ground object, area feature, finally obtain sub-pixed mapping charting results.
Described step 2 utilizes region growing model to split mark image, the basic thought of region growing model to have similar quality, comprise gray level, texture, gradient information set of pixels form region altogether, finally reach the object of Image Segmentation, specific implementation process is as follows:
(1) a random selecting n pixel in mark image, as seed;
(2) region at seed place is expanded, until all pixels of mark image are all divided into specific region, for the neighborhood pixel of seed, if the threshold epsilon that the difference of the pixel value of its pixel value and seed is specifying, then expand this neighborhood pixel in the region at current seed place.
In order to ensure the connectedness of Linear feature, described threshold epsilon value is 1/S, sub-pixed mapping drawing is soft class result from low spatial resolution remote sensing image, obtains the process of remote sensing land classification figure under given high spatial resolution, S be in sub-pixed mapping drawing by low resolution to high-resolution amplification factor.
In described step 3, the shape density index in the region of mark image after computed segmentation, according to corresponding discrimination threshold, the division realizing punctual geo-objects, linear ground object and area feature is implemented as;
I＝ρ 1×S shape+ρ 2×D density
S s h a p e = L 4 × A , D d e n s i t y = N M
Wherein, I represents shape density index; S shaperepresent with the shape index in region, L represents the pixel number of zone boundary, and A represents the area in region, D densityrepresent the dnesity index in region, N represents the pixel number of region of out bound rectangle, and M represents the number of region of out bound square boundary pixel, ρ 1, ρ 2represent weighted index, wherein ρ 1+ ρ 2=1; If S>=2.3, D≤1.1, I>=1.6, this region is linear ground object; Otherwise this Region dividing is area feature; For remaining region, fractional value is greater than 1/S 2pixel be divided into punctual geo-objects.
Described step 4, adopts linear ground object template matching method to the method for linear ground object sub-pixed mapping drawing, and the sub-pixed mapping comprising definition linear die, coupling linear die and linear ground object charts three steps, specific as follows:
(1) linear die is defined: in order to indicate Linear feature complete as far as possible compactly, the present invention utilizes the linear die of 20 kind of 3 × 3 pixel, and linear die comprises straight line and curve type;
(2) linear die is mated: in order to find out the Linear feature template in the most applicable mixed pixel, the present invention utilizes the related coefficient of 3 × 3 local windows of 3 × 3 template pixels and mark image to represent their correlativity, and computing formula is as follows.
r j k = Σ m = - 1 1 Σ m = - 1 1 T k ( m , n ) × F j c ( x + m , y + n ) Σ m = - 1 1 Σ m = - 1 1 T k ( m , n ) 2 × Σ m = - 1 1 Σ n = - 1 1 F j c ( x + m , y + n )
Wherein, F jcrepresent a jth pixel of C class atural object mark image, x, y represent the coordinate of this pixel, T krepresent a kth linear die, m, n represent the position of pixel in 3 × 3 pixel linear die, and specify that the coordinate of top left corner pixel is-1 ,-1, and the present invention adopts and makes r jkmaximum linear die is as the optimal Template of this mixed pixel; (x+m, y+n) represents the neighborhood pixel of a jth pixel, at the fractional value of C class atural object mark image;
(3) the sub-pixed mapping drawing of Linear feature: in F jcabove-mentioned two steps determine the best trend of linear ground object in this mixed pixel, the i.e. trend of best linear template, definition pixel j top left co-ordinate is (0,0), lower right corner coordinate (1,1), make the size of linear die equal with mixed pixel j, then in linear die, the coordinate of pixel passes through following formulae discovery:
m ′ = 2 × ( m + 2 ) - 1 6 n ′ = 2 × ( n + 2 ) - 1 6 , m , n - 1 , 0 , 1
The pixel of each sub-pixed mapping in mixed pixel can represent with following formula:
x ′ = 2 × x - 1 2 × S y ′ = 2 × y - 1 2 × S , x , y = 1 , 2 , ... , S
Wherein, x, y are the ranks number of sub-pixed mapping, and S is the multiple of amplification factor and image augmentation, calculate new pixel i.e. each sub-pixed mapping to optimum linear template T kintermediate value is the minimum Eustachian distance of 1 pixel, carries out ascending order arrangement to these minimum euclidean distances, front F cj× S 2the type of ground objects of sub-pixed mapping be C class atural object.
The method of described step 4 pair area feature sub-pixed mapping drawing adopts the sub-pixed mapping drafting method based on vector border to comprise following four steps:
(1) for each mixed pixel with vector border extraction model, obtain the length and location of every bar line segment on mixed pixel inner boundary;
(2) clockwise every bar line segment of generating of Connection Step (1) successively, obtains the initial Polygonal Boundary of atural object C in mixed pixel;
(3) every bar line segment intersection of the initial Polygonal Boundary that obtains of delete step (2) and step (1), namely obtains final vector border;
(4) use ray diagnostic method, carry out assignment to the atural object in vector border.
The sub-pixed mapping drafting method based on spatial coherence and Model Matching that described step 4 pair punctual geo-objects adopts, utilize the Spatial Distribution Pattern of not blue index analysis punctual geo-objects, distribution pattern comprises dispersion, Stochastic sum is assembled, because the spatial model of punctual geo-objects changes along with the size of spatial dimension, the present invention adopts the not blue index of 3 × 3 pixel window calculation level shape atural objects, and detailed process is as follows;
(1) to the mark image of C class type of ground objects, area feature sub-pixed mapping chartography is first adopted to obtain the image of its sub-pixed mapping drawing;
(2) to the punctual geo-objects pixel j in the mark image of C class type of ground objects, the not blue exponential quantity I of its place local window is calculated; For the image that this mixed pixel is obtained by sub-pixed mapping drafting method, calculate 3 × 3 × S of mixed pixel j 2not blue index I';
(3) if | I'-I|≤θ, θ distinguish the threshold value whether pixel level punctual geo-objects mate with sub-pixed mapping ground shape atural object space distribution, then the image that this sub-pixed mapping charts is the net result of punctual geo-objects; If | I'-I| >=θ, the property value of two pixels in the sub-pixed mapping drawing image of random adjustment mixed pixel j, recalculates not blue exponential quantity, until meet | I'-I|≤θ, obtains the punctual geo-objects sub-pixed mapping charting results of the mark image of C class type of ground objects.
Punctual geo-objects, linear ground object, the integrated process obtained in final sub-pixed mapping charting results of inlaying of area feature sub-pixed mapping charting results are by described step 5: first by the sub-pixed mapping of area feature drawing image layer as a setting, superpose the sub-pixed mapping charting results of punctual geo-objects and the sub-pixed mapping drawing image of linear ground object successively; In the process of inlaying, if the pixel value of sub-pixed mapping drawing image is null, then use 3 × 3 window traversal neighborhoods, the classification that occurrence number is maximum is the category attribute of this pixel.
The present invention's advantage is compared with prior art: the present invention is the sub-pixed mapping drafting method considering spatial distribution pattern (area feature, linear ground object and area feature).Sub-pixed mapping for area feature charts, and it can ensure the maximization of area feature spatial coherence, effectively can improve the sub-pixed mapping cartographic accuracy of area feature; Sub-pixed mapping for linear ground object charts, and it can ensure the connectedness of linear ground object, and can produce the Linear feature border of relative smooth; Sub-pixed mapping for punctual geo-objects charts, and it can ensure that the punctual geo-objects distribution pattern in sub-pixed mapping charting results is similar to reference to image, and then improves cartographic accuracy.Compared with classic method, this algorithm from ground angle, take into full account atural object Spatial Distribution Pattern (planar distribution, wire distribution and spot distribution), for different spaces pattern terrain classification and control, by force explanatory, the result good visual effect precision of sub-pixed mapping drawing is high.
Accompanying drawing explanation
Fig. 1 is main flow chart of the present invention;
Fig. 2 is the linear die figure in the present invention, comprises straight line and curve type, represents the trend of linear ground object in real world.
Fig. 3 is experimental data of the present invention, and (a) is the standard False color comp osite image of 400*400 pixel TM image, and be a kind of common, in the remote sensing image processing such as vegetation, crops and Land_use change, this is the most frequently used band combination.(b) for spatial resolution higher, the Google Earth image consistent with (a) spatial dimension;
Fig. 4 is the mark image (be followed successively by vegetation, waters, road and be buildings) of four kinds of atural objects that the soft classification of TM image obtains, and the pixel value of mark image is the ratio in this mixed pixel shared by type of ground objects C, and mxm. is 1, and minimum is 0;
Fig. 5 is in four class atural object mark images, obtains the spatial distribution pattern in region after over-segmentation, and spatial distribution pattern comprises spot distribution, wire distribution and planar distribution.Wherein (a) is vegetation, and (b) is waters, and (c) is road, and (d) is buildings;
Fig. 6 is the contrast of different sub-pixed mapping drafting algorithm, a SVM Hard clustering result that () is Google Earth image data, b sub-pixed mapping charting results that () is SAM algorithm, c sub-pixed mapping charting results that () is HIIPD algorithm, d sub-pixed mapping charting results that () is PSA algorithm, e sub-pixed mapping charting results that () is LPSA algorithm, f sub-pixed mapping charting results that () is MRF algorithm, g sub-pixed mapping charting results that () is MAP algorithm, h sub-pixed mapping charting results that () is SPMv algorithm, (i) sub-pixed mapping drafting method that adopt to by the present invention based on space atural object distribution characteristics.
Embodiment
The present invention is further illustrated below in conjunction with accompanying drawing and specific embodiment.
As shown in Figure 1, specific embodiment of the invention process is as follows:
Step 1, the pre-service such as geometry correction, atmospheric correction and denoising are carried out to remote sensing image, the ratio that real remote sensing image obtains in each pixel shared by various atural object classification through soft classification by the training sample then obtained according to on-the-spot investigation, i.e. mark image, in this, as input data of the present invention.Because soft assorting process often exists error, therefore for avoiding the error effect sub-pixed mapping of soft assorting process to chart, the present invention utilizes high resolution image to describe the specific implementation process of sub-pixed mapping drawing.The present invention utilizes the high resolution image of Google Maps, in this, as reference diagram for verifying the validity of sub-pixed mapping drafting method.
Step 2, utilize region growing model to split mark image, detailed process is as follows:
(1) a random selecting n pixel in mark image, as seed;
(2) region at seed place is expanded, until all pixels of mark image are all divided into specific region, for the neighborhood pixel of seed, if the threshold epsilon that the difference of the pixel value of its pixel value and seed is specifying, then expand this neighborhood pixel in the region at current seed place.In order to ensure the connectedness of Linear feature, in be 1/S, S the be sub-pixed mapping drawing of described threshold epsilon value by low resolution to high-resolution amplification factor.
The shape density index of step 3, calculating mark Image Segmentation rear region, computing formula is as follows:
I＝ρ 1×S shape+ρ 2×D density
S s h a p e = L 4 × A , D d e n s i t y = N M
Wherein, I represents shape density index; S shaperepresent with the shape index in region, L represents the pixel number of zone boundary, and A represents the area in region, D densityrepresent the dnesity index in region, N represents the pixel number of region of out bound rectangle, and M represents the number of region of out bound square boundary pixel, ρ 1, ρ 2represent weighted index, wherein ρ 1+ ρ 2=1; If S>=2.3, D≤1.1, I>=1.6, this region is linear ground object; Otherwise this Region dividing is area feature; For remaining region, fractional value is greater than 1/S 2pixel be divided into punctual geo-objects.
Step 4, corresponding sub-pixed mapping drafting method is adopted to three kinds of atural objects.
First for linear ground object, the sub-pixed mapping drafting method of linear ground object template matches is adopted.Detailed process is as follows:
(1) linear die is defined: in order to indicate Linear feature complete as far as possible compactly, the present invention utilizes the linear die of 20 kind of 3 × 3 pixel, and linear die comprises straight line and curve type;
(2) linear die is mated: in order to find out the Linear feature template in the most applicable mixed pixel, the present invention utilizes the related coefficient of 3 × 3 local windows of 3 × 3 template pixels and mark image to represent their correlativity, and computing formula is as follows.
r j k = Σ m = - 1 1 Σ m = - 1 1 T k ( m , n ) × F j c ( x + m , y + n ) Σ m = - 1 1 Σ m = - 1 1 T k ( m , n ) 2 × Σ m = - 1 1 Σ n = - 1 1 F j c ( x + m , y + n )
Wherein, F jcrepresent a jth pixel of C class atural object mark image, x, y represent the coordinate of this pixel, T krepresent a kth linear die, m, n represent the position of pixel in 3 × 3 pixel linear die, and specify that the coordinate of top left corner pixel is-1 ,-1, and the present invention adopts and makes r jkmaximum linear die is as the optimal Template of this mixed pixel; (x+m, y+n) represents the neighborhood pixel of a jth pixel, at the fractional value of C class atural object mark image;
(3) the sub-pixed mapping drawing of Linear feature: for F jcabove-mentioned two steps determine the best trend of linear ground object in this mixed pixel, the i.e. trend of best linear template, definition pixel j top left co-ordinate is (0,0), lower right corner coordinate (1,1), make the size of linear die equal with mixed pixel j, then in linear die, the coordinate of pixel passes through following formulae discovery:
m ′ = 2 × ( m + 2 ) - 1 6 n ′ = 2 × ( n + 2 ) - 1 6 , m , n - 1 , 0 , 1
The pixel of each sub-pixed mapping in mixed pixel can represent with following formula:
x ′ = 2 × x - 1 2 × S y ′ = 2 × y - 1 2 × S , x , y = 1 , 2 , ... , S
Wherein, x, y are the ranks number of sub-pixed mapping, and S is the multiple of amplification factor and image augmentation, calculate new pixel i.e. each sub-pixed mapping to optimum linear template T kintermediate value is the minimum Eustachian distance of 1 pixel, carries out ascending order arrangement to these minimum euclidean distances, front F cj× S 2the type of ground objects of sub-pixed mapping be C class atural object.
Secondly the sub-pixed mapping drafting method based on vector border is adopted to comprise following Four processes for area feature:
(1) for each mixed pixel with vector border extraction model, obtain the length and location of every bar line segment on mixed pixel inner boundary; (2) clockwise every bar line segment of generating of Connection Step (1) successively, obtains the initial Polygonal Boundary of C class atural object in mixed pixel; (3) every bar line segment intersection of the initial Polygonal Boundary that obtains of delete step (2) and step (1), namely obtains final vector border; (4) use ray diagnostic method, carry out assignment to the atural object in vector border.
Finally punctual geo-objects is adopted to the sub-pixed mapping drafting method of pattern match.(1) the not blue index I' of not blue exponential quantity I and 3 × 3 × S sub-pixed mapping of 3 × 3 local windows at mark image mixed pixel j place is calculated respectively.(2) two not blue indexes are compared, if their difference is less than the threshold value of regulation, current sub-pixed mapping image is the sub-pixed mapping charting results of punctual geo-objects, otherwise the property value of two pixels in the sub-pixed mapping drawing image of random adjustment pixel j, recalculate not blue exponential quantity, until the difference meeting two kinds of not blue indexes is less than threshold value, obtain the sub-pixed mapping drawing image of punctual geo-objects.
Step 5 punctual geo-objects, linear ground object, area feature sub-pixed mapping charting results are integrated.Inlaying the process obtained in final sub-pixed mapping charting results is: first by the sub-pixed mapping of area feature drawing image layer as a setting, superposes the sub-pixed mapping charting results of punctual geo-objects and the sub-pixed mapping drawing image of linear ground object successively; In the process of inlaying, if the pixel value of sub-pixed mapping drawing image is null, then use 3 × 3 window traversal neighborhoods, the classification that occurrence number is maximum is the category attribute of this pixel.
The SPMs method proposed for contrast the present invention and traditional sub-pixed mapping drafting method performance in an experiment, this process has carried out the experiment of 8 kinds of sub-pixed mapping drafting methods.Experimental result is respectively: in Fig. 6, b sub-pixed mapping charting results that () is SAM algorithm, c sub-pixed mapping charting results that () is HIIPD algorithm, d sub-pixed mapping charting results that () is PSA algorithm, e sub-pixed mapping charting results that () is LPSA algorithm, f sub-pixed mapping charting results that () is MRF algorithm, g sub-pixed mapping charting results that () is MAP algorithm, h sub-pixed mapping charting results that () is SPMv algorithm, (i) sub-pixed mapping drafting method that adopt to by the present invention based on space atural object distribution characteristics.Compared with other are invented, the present invention is more accurate to result during process linear ground object, because the present invention has taken into full account the connectedness of linear ground object.And the present invention has also fully shown the details of area feature, and ensure that higher accuracy.
Non-elaborated part of the present invention belongs to the known technology of those skilled in the art.
The foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, for the technician of this neighborhood, the present invention can have various modifications and variations.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within right of the present invention.

Claims (8)

1., based on a remote sensing sub-pixed mapping drafting method for atural object spatial distribution characteristic, it is characterized in that performing step is as follows:
Step 1, pre-service is carried out to input remote sensing image, and then the component ratio utilizing neural classifier to obtain in remote sensing image each pixel shared by often kind of atural object classification is called abundance or mark image; Described often kind of atural object classification is three kinds of atural objects, i.e. punctual geo-objects, linear ground object and area feature;
Step 2, utilize region growing model, by the zones of different that mark Image Segmentation becomes, each region after adopting ordered data storage organization to store segmentation, is convenient to calculate the area in each region, girth and border;
The type of ground objects of step 3, zoning: the shape density index in the region of mark image after computed segmentation, according to corresponding discrimination threshold, realizes the division of punctual geo-objects, linear ground object and area feature;
Step 4, three kinds of atural object division results according to step 3, sub-pixed mapping drafting method based on linear ground object template matches is adopted to linear ground object mixed pixel, sub-pixed mapping drafting method based on vector border is adopted to area feature mixed pixel, sub-pixed mapping drafting method based on spatial coherence and Model Matching is adopted to punctual geo-objects mixed pixel, obtains punctual geo-objects, linear ground object, area feature sub-pixed mapping charting results;
Step 5, inlay the sub-pixed mapping charting results of punctual geo-objects, linear ground object, area feature, finally obtain sub-pixed mapping charting results.
2. a kind of remote sensing sub-pixed mapping drafting method based on atural object spatial distribution characteristic according to claim 1, is characterized in that: described step 2 utilizes region growing model, and the zones of different specific implementation process become by mark Image Segmentation is as follows:
(1) a random selecting n pixel in mark image, as seed;
(2) region at seed place is expanded, until all pixels of mark image are all divided into specific region, for the neighborhood pixel of seed, if the threshold epsilon that the difference of the pixel value of its pixel value and seed is specifying, then expand this neighborhood pixel in the region at current seed place.
3. a kind of remote sensing sub-pixed mapping drafting method based on atural object spatial distribution characteristic according to claim 2, is characterized in that: in be 1/S, S the be sub-pixed mapping drawing of described threshold epsilon value by low resolution to high-resolution amplification factor.
4. a kind of remote sensing sub-pixed mapping drafting method based on atural object spatial distribution characteristic according to claim 1, it is characterized in that: in described step 3, the shape density index in the region of mark image after computed segmentation, according to corresponding discrimination threshold, the division realizing punctual geo-objects, linear ground object and area feature is implemented as;
I＝ρ 1×S shape+ρ 2×D density
S s h a p e = L 4 × A , D d e n s i t y = N M
Wherein, I represents shape density index, S shaperepresent with the shape index in region, L represents the pixel number of zone boundary, and A represents the area in region, D densityrepresent the dnesity index in region, N represents the pixel number of region of out bound rectangle, and M represents the number of region of out bound square boundary pixel, ρ 1, ρ 2represent weighted index, wherein ρ 1+ ρ 2=1;
If S>=2.3, D≤1.1, I>=1.6, this region is linear ground object; Otherwise this Region dividing is area feature; For remaining region, fractional value is greater than 1/S 2pixel be divided into punctual geo-objects.
5. a kind of remote sensing sub-pixed mapping drafting method based on atural object spatial distribution characteristic according to claim 1, it is characterized in that: described step 4, linear ground object template matching method is adopted to the method for linear ground object sub-pixed mapping drawing, comprise definition linear die, coupling linear die and the sub-pixed mapping of linear ground object to chart three steps, specific as follows:
(1) linear die is defined: utilize the linear die of 20 kind of 3 × 3 pixel to carry out template matches, linear die comprises straight line and curve type;
(2) matching template is linear: utilize the related coefficient of 3 × 3 local windows of 3 × 3 template pixels and mark image to represent their correlativity, computing formula is as follows:
r j k = Σ m = - 1 1 Σ m = - 1 1 T k ( m , n ) × F j c ( x + m , y + n ) Σ m = - 1 1 Σ m = - 1 1 T k ( m , n ) 2 × Σ m = - 1 1 Σ m = - 1 1 F j c ( x + m , y + n )
Wherein, F jcrepresent a jth pixel of C class atural object mark image, x, y represent the coordinate of this pixel, T krepresent a kth linear die, m, n represent the position of pixel in 3 × 3 pixel linear die, and specify that the coordinate of top left corner pixel is-1 ,-1, and the present invention adopts and makes r jkmaximum linear die is as the optimal Template of this mixed pixel; (x+m, y+n) represents the neighborhood pixel of a jth pixel, at the fractional value of C class atural object mark image;
(3) the sub-pixed mapping drawing of Linear feature: for F jcabove-mentioned two steps determine the best trend of linear ground object in this mixed pixel, the i.e. trend of best linear template, definition pixel j top left co-ordinate is (0,0), lower right corner coordinate (1,1), make the size of linear die equal with mixed pixel j, then in linear die, the coordinate of pixel passes through following formulae discovery:
m ′ = 2 × ( n + 2 ) - 1 6 n ′ = 2 × ( n + 2 ) - 1 6 , m , n = - 1 , 0 , 1
The pixel of each sub-pixed mapping in mixed pixel can represent with following formula:
x ′ = 2 × x - 1 2 × S y ′ = 2 × y - 1 2 × S , x , y = 1 , 2 , ... , S
Wherein, x, y are the ranks number of sub-pixed mapping, and S is the multiple of amplification factor and image augmentation, calculate new pixel i.e. each sub-pixed mapping to optimum linear template T kintermediate value is the minimum Eustachian distance of 1 pixel, carries out ascending order arrangement to these minimum euclidean distances, front F cj× S 2the type of ground objects of sub-pixed mapping be C class atural object.
6. a kind of remote sensing sub-pixed mapping drafting method based on atural object spatial distribution characteristic according to claim 1, is characterized in that: the method for described step 4 pair area feature sub-pixed mapping drawing adopts the sub-pixed mapping drafting method based on vector border to comprise following four steps:
(1) for each mixed pixel with vector border extraction model, obtain the length and location of every bar line segment on mixed pixel inner boundary;
(2) clockwise every bar line segment of generating of Connection Step (1) successively, obtains the initial Polygonal Boundary of C class atural object in mixed pixel;
(3) every bar line segment intersection of the initial Polygonal Boundary that obtains of delete step (2) and step (1), namely obtains final vector border;
(4) use ray diagnostic method, carry out assignment to the atural object in vector border.
7. a kind of remote sensing sub-pixed mapping drafting method based on atural object spatial distribution characteristic according to claim 1, is characterized in that: the sub-pixed mapping drafting method based on spatial coherence and Model Matching that described step 4 pair punctual geo-objects adopts, and detailed process is as follows;
(1) to the mark image of C class type of ground objects, area feature sub-pixed mapping chartography is first adopted to obtain the image of its sub-pixed mapping drawing;
(2) to the punctual geo-objects pixel j in the mark image of C class type of ground objects, the not blue exponential quantity I of its place local window is calculated; For the image that this mixed pixel is obtained by sub-pixed mapping drafting method, calculate 3 × 3 × S of mixed pixel j 2not blue index I';
(3) if | I'-I|≤θ, θ distinguish the threshold value whether pixel level punctual geo-objects mate with sub-pixed mapping ground shape atural object space distribution, then the image that this sub-pixed mapping charts is the net result of punctual geo-objects; If | I'-I| >=θ, the property value of two pixels in the sub-pixed mapping drawing image of random adjustment mixed pixel j, recalculating not blue exponential quantity, until meet | I'-I|≤θ, obtains the punctual geo-objects sub-pixed mapping charting results of the mark image of C class type of ground objects.
8. a kind of remote sensing sub-pixed mapping drafting method based on atural object spatial distribution characteristic according to claim 1, it is characterized in that: punctual geo-objects, linear ground object, the integrated process obtained in final sub-pixed mapping charting results of inlaying of area feature sub-pixed mapping charting results are by described step 5: first by the sub-pixed mapping of area feature drawing image layer as a setting, superpose the sub-pixed mapping charting results of punctual geo-objects and the sub-pixed mapping drawing image of linear ground object successively; In the process of inlaying, if the pixel value of sub-pixed mapping drawing image is null, then use 3 × 3 window traversal neighborhoods, the classification that occurrence number is maximum is the category attribute of this pixel.
CN201510781816.7A 2015-11-13 2015-11-13 Remote sensing sub-pixel mapping method based on spatial distribution characteristics of features Pending CN105447452A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510781816.7A CN105447452A (en) 2015-11-13 2015-11-13 Remote sensing sub-pixel mapping method based on spatial distribution characteristics of features

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510781816.7A CN105447452A (en) 2015-11-13 2015-11-13 Remote sensing sub-pixel mapping method based on spatial distribution characteristics of features

Publications (1)

Publication Number Publication Date
CN105447452A true CN105447452A (en) 2016-03-30

Family

ID=55557612

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510781816.7A Pending CN105447452A (en) 2015-11-13 2015-11-13 Remote sensing sub-pixel mapping method based on spatial distribution characteristics of features

Country Status (1)

Country Link
CN (1) CN105447452A (en)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106846246A (en) * 2016-12-16 2017-06-13 河海大学 A kind of object-based Super-resolution Mapping for Remote Sensing Images
CN107220615A (en) * 2017-05-24 2017-09-29 河海大学 A kind of city impervious surface information extracting method for merging point of interest big data
CN109871884A (en) * 2019-01-25 2019-06-11 曲阜师范大学 A kind of support vector machines object-oriented Remote Image Classification merging multiple features
CN110009066A (en) * 2019-03-15 2019-07-12 北京航空航天大学 The spectrum picture sub-pixed mapping space-location method and device that logic-based returns
CN111899300A (en) * 2020-07-30 2020-11-06 北京航空航天大学 Abundance correction method and device for light field spectral data subpixel positioning
CN112949699A (en) * 2021-02-09 2021-06-11 国家卫星气象中心(国家空间天气监测预警中心) Remote sensing image classification model establishing and verifying method, system and electronic equipment
CN113127485A (en) * 2021-03-24 2021-07-16 中国电力工程顾问集团西南电力设计院有限公司 Method for processing linear ground object in power transmission line engineering measurement software

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103559506A (en) * 2013-11-19 2014-02-05 中国科学院地理科学与资源研究所 Sub-pixel drawing method based on vector boundaries

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103559506A (en) * 2013-11-19 2014-02-05 中国科学院地理科学与资源研究所 Sub-pixel drawing method based on vector boundaries

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
MAT-ISA N A等: "Seeded region growing features extraction algorithm; its potential use in improving screening for cervical cancer", 《INTERNATIONAL JOURNAL OF THE COMPUTER, THE INTERNET AND MANAGEMENT》 *
陈跃红: "基于地物空间分布特征的亚像元制图", 《道客巴巴》 *

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106846246A (en) * 2016-12-16 2017-06-13 河海大学 A kind of object-based Super-resolution Mapping for Remote Sensing Images
CN106846246B (en) * 2016-12-16 2019-07-05 河海大学 A kind of object-based Super-resolution Mapping for Remote Sensing Images
CN107220615A (en) * 2017-05-24 2017-09-29 河海大学 A kind of city impervious surface information extracting method for merging point of interest big data
CN107220615B (en) * 2017-05-24 2020-08-14 河海大学 Urban impervious surface information extraction method fusing interest point big data
CN109871884A (en) * 2019-01-25 2019-06-11 曲阜师范大学 A kind of support vector machines object-oriented Remote Image Classification merging multiple features
CN110009066A (en) * 2019-03-15 2019-07-12 北京航空航天大学 The spectrum picture sub-pixed mapping space-location method and device that logic-based returns
CN111899300A (en) * 2020-07-30 2020-11-06 北京航空航天大学 Abundance correction method and device for light field spectral data subpixel positioning
CN111899300B (en) * 2020-07-30 2022-05-31 北京航空航天大学 Abundance correction method and device for light field spectral data subpixel positioning
CN112949699A (en) * 2021-02-09 2021-06-11 国家卫星气象中心(国家空间天气监测预警中心) Remote sensing image classification model establishing and verifying method, system and electronic equipment
CN112949699B (en) * 2021-02-09 2024-05-14 国家卫星气象中心(国家空间天气监测预警中心) Remote sensing image classification model building and verifying method and system and electronic equipment
CN113127485A (en) * 2021-03-24 2021-07-16 中国电力工程顾问集团西南电力设计院有限公司 Method for processing linear ground object in power transmission line engineering measurement software
CN113127485B (en) * 2021-03-24 2023-02-03 中国电力工程顾问集团西南电力设计院有限公司 Method for processing linear ground object in power transmission line engineering measurement software

Similar Documents

Publication Publication Date Title
Chen et al. Automatic building information model reconstruction in high-density urban areas: Augmenting multi-source data with architectural knowledge
Yang et al. An individual tree segmentation method based on watershed algorithm and three-dimensional spatial distribution analysis from airborne LiDAR point clouds
CN105447452A (en) Remote sensing sub-pixel mapping method based on spatial distribution characteristics of features
Butenuth et al. Integration of heterogeneous geospatial data in a federated database
Shi Principles of modeling uncertainties in spatial data and spatial analyses
Pepe et al. A novel method based on deep learning, GIS and geomatics software for building a 3D city model from VHR satellite stereo imagery
Li Remotely sensed images and GIS data fusion for automatic change detection
Xiao et al. Building segmentation and modeling from airborne LiDAR data
Peeters et al. Automated recognition of urban objects for morphological urban analysis
CN110827302A (en) Point cloud target extraction method and device based on depth map convolutional network
CN110992366A (en) Image semantic segmentation method and device and storage medium
CN109308451A (en) A kind of high score data information extraction system and method
CN103035006A (en) High-resolution aerial image partition method based on LEGION and under assisting of LiDAR
CN104751479A (en) Building extraction method and device based on TIN data
Zhao et al. Combining ICESat-2 photons and Google Earth Satellite images for building height extraction
CN102945550B (en) A kind of method building remote sensing image semanteme based on Gaussian scale-space
Zhang et al. Impervious surface extraction from high-resolution satellite image using pixel-and object-based hybrid analysis
CN102609721B (en) Remote sensing image clustering method
Miky et al. A combined contour lines iteration algorithm and Delaunay triangulation for terrain modeling enhancement
Shirowzhan et al. Developing metrics for quantifying buildings’ 3D compactness and visualizing point cloud data on a web-based app and dashboard
Zhou 3D urban modeling from city-scale aerial LiDAR data
CN105631849A (en) Polygon object change detection method and device
Wolf et al. Applicability of neural networks for image classification on object detection in mobile mapping 3d point clouds
CN115131571A (en) Building local feature point identification method based on six fields of point cloud pretreatment
Mahphood et al. Virtual first and last pulse method for building detection from dense LiDAR point clouds

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20160330

WD01 Invention patent application deemed withdrawn after publication