CN106339985A - Method for aviation image mosaic by selecting mosaic lines from vector building data - Google Patents

Abstract

The invention provides a method for aviation image mosaic by selecting mosaic lines from vector building data. The method is characterized by comprising the following steps that 1) an overlap area of adjacent images within each air strip is calculated according to boundaries of an effective area of each image in the air strip; 2) an overlap area of adjacent air strips among the air strips is calculated according to the obtained boundaries of the effective area of each air strip; 3) on the basis of existing vector building data, mosaic lines are searched from the overlap areas of the adjacent images in the air strips and the overlap areas of the adjacent air strips among the air strips; 4) all the mosaic lines in and among the air strips are cut mutually, and effective mosaic polygons are constructed for the images respectively; and 5) data in the effective mosaic polygon of each image is merged to form a final mosaic image which includes all the single images. According to the method for aviation image mosaic, the vector data is used to construct the initial mosaic polygons, grid images themselves are used for local optimization, time for constructing the mosaic polygons can be shortened, crossing protruded objects as buildings can be avoided, and the amount of post manual editing is reduced substantially.

Description

Choose a kind of data from vector house and inlay the method that line is inlayed to aviation image

Technical field

The present invention relates to the processing technology field of aviation image, choose from vector house data particularly to one kind and inlay The method that line is inlayed to aviation image.

Background technology

" National Program for Medium-to Long-term Scientific and Technological Development (2006～the year two thousand twenty) " clearly will develop high-resolution pair Ground observation system (referred to as " high score is special ") is classified as one of 16 key special subjects.To the year two thousand twenty, this system is built up comprehensively And come into operation, by being obviously improved, China is round-the-clock, round-the-clock, Global coverage earth observation ability.However, with remote sensing shadow The resolution more and more higher of picture, application in all trades and professions for the remotely-sensed data is more and more extensive, covers given area (as Wuhan City) remote sensing image data amount assume explosive growth, promptly increase tb number till now from the gb order of magnitude of 20 beginnings of the century Magnitude (Chen Jiechen, 2012；2008；Xu Difeng, 2009；Primary outstanding person, 2013) it is intended that, obtaining now compared with the past with ratio The high resolution image of identical ground region needs to inlay more remote sensing images.This inlays to quick, the intelligence of remote sensing image Put forward higher requirement.Its difficult point is how to avoid passing through what the digital elevation models such as house can not be corrected so that inlaying line Object (wang et al., 2012).

However, the theory inlayed of remote sensing image and method but develop relative backward, lead to a lot of surveying and mapping units and the department must More people must be employed, or quality is inlayed in reduction, the remote sensing image to meet increasingly increase is inlayed.As Wuhan City makes once The main city zone image of 1:2000 scale, needs to shoot and inlay about 10000 images, runs and know in the world on single pc machine Tessellation software orthovista (inpho, 2014) of name, to inlay these images, needs day and night ceaselessly computing 10 days about, Single technician changes to inlay line and be set into figure and also needs to 6 months about (wang et al., 2012).Such efficiency makes Obtain relevant departments to have to put into substantial amounts of high-performance computer and manpower, inlay work to complete in one month.With This simultaneously, Surveying and Mapping Industry have accumulated the very strong vector data of a large amount of actualities within the past few decades, such as house and road, these Vector data can serve the production of a lot of mapping products as priori, such as image mosaic.wang et al. (2013) proposition is a kind of inlays line selection selection method based on global optimum's aviation image of vector road data, so that inlaying line Walk along road axis, thus avoiding house as far as possible.However, also rarely having scholar's research how by vector house market demand so far Inlay in aviation image, and research is applied to efficiency and the effect of image mosaic.

Content of the invention

For the problems referred to above, the goal of the invention of the present invention is to provide one kind to choose from vector house and road data to inlay The method that polygon is inlayed to aviation image, it is built using vector data initially inlays polygon, then with Raster Images itself These inlay polygon by what vector data generated to carry out local optimum, inlay the polygon structure time to shorten, and guarantee any One section final to inlay line and can avoid passing through the prominent object such as house, thus significantly reducing human-edited's amount in later stage.

The concrete technical scheme of the present invention is to choose a kind of data from vector house to inlay line aviation image is inlayed Method is it is characterised in that comprise the following steps:

1) it is based on left-turn algorithm, the side to every Extraction of Image effective coverage in every air strips of the aviation image shooting Boundary line, in every air strips, adjacent image overlap in the border line computation air strips according to every image effective coverage in this air strips Area；

2) merge the effective coverage in every air strips, obtain the boundary line of the effective coverage of every air strips, according to obtain Adjacent air strips overlay region between the border line computation air strips of the effective coverage of every air strips；

3) it is based on existing vector house data, adjacent air strips overlay region between adjacent image overlap area and air strips in air strips Line is inlayed in interior searching；

4) inlay the mutual cutting of line by between in all air strips and air strips, be then based on left-turn algorithm, by inlaying after cutting Line joins end to end, and is that polygon individually effectively inlayed by every image structure in every air strips, finally inlays image and only use The data effectively inlayed in polygon of every image, remainder is given up；

5) data effectively inlayed in polygon of every image is merged, formed and include all single images Inlay image eventually.

Further, described step 3) in find that to inlay the method for line be one by one in adjacent image overlap area in air strips Extract and inlay line between all adjacent images, find in adjacent air strips overlay region between air strips when inlaying line, adjacent air strips are worked as Make adjacent image, using with adjacent image overlap area find inlay line identical method extract adjacent air strips between inlay line, If two adjacent images are m degree superimposed image, that is, in addition to this two images itself, also open with m-2 other images have overlapping, m >= 2, the method inlaying line between two adjacent images of extraction comprises the following steps:

(1) in two adjacent image overlap areas based on constraint delaunay triangulation network algorithm extraction, between vector house vacant lot Intermediate line and overlay region skeleton line, then by overlay region between vector house the intermediate line in vacant lot and overlay region skeleton line Merge, composition candidate inlays line storehouse；

(2) effective width inlaying line according to candidate inlays line tax by cost to candidate, is sought based on dijkstra algorithm The candidate looking for the cost between origin-to-destination minimum inlays line；

(3) the minimum candidate of the cost that step (2) extracted using m degree superimposed image data is inlayed line and enters in overlay region Row optimization processing, forms and final inlays line it is ensured that any one section final inlayed line and can avoid passing through in image data Exist, and the prominent object not comprised in vector house data.

Further, inlay line and assign to candidate according to the effective width that candidate inlays line in described step (2) and pass through into This, finding the minimum candidate of cost between origin-to-destination based on dijkstra algorithm and inlaying the concrete grammar of line is, if l_iFor Candidate in i-th two adjacent image overlap areas inlays line, and this candidate inlays line l_iEffective width vadw (l_i) as the following formula I () calculates:

Wherein, d (l_i) represent and inlay line l with candidate_iMinimum range between adjacent room, projd (h_i1,h_i2) it is two phases Height displacement's sum in next door room, (ii) calculating as the following formula:

projd(h_i1,h_i2)=tan θ_j1*h_i1(l_i)+tanθ_i2*h_i2(l_i) (ii)

Wherein, θ_i1And θ_i2Inlaying line l for two adjacent images with candidate respectively_iThe downwards angle of visibility in two adjacent houses； h_i1(l_i)、h_i2(l_i) be this two houses height,

Candidate inlays line l_iBy cost tcost (l_i) (iii) calculating as the following formula:

tcost(l_i)=κ (l_i)×len(l_i) (iii)

Wherein, len (l_i) represent that candidate inlays line l_iLength；κ(l_i) represent that candidate inlays line l_iWidth to cost Contribution, κ (l_i) (iv) calculating as the following formula:

Wherein, c is set to constant 0.1, and minw is the minimum widith of the intermediate line in vacant lot between house in overlay region,

If all candidates of forming of the intermediate line in vacant lot and overlay region skeleton line inlay line set, cs between sl represents by house Represent that connecting candidate in some sl inlays line and the path by inlaying line beginning and end, inlaying line beginning and end is phase Two intersection points of adjacent image effective coverage, as the following formula (v) calculate cs in either path cs by cost f (cs):

$f\left(cs\right)=\mathrm{\σ}t\cos t\left(l_i\right),l_i\&element;cs\&element;sl---\left(v\right)$

Based on dijkstra algorithm, the minimum candidate of cost in cs can be found and inlay line ms, it is min [f by cost (cs)],cs∈cs∈sl.

Further, in described step (3), the minimum candidate of cost inlays line and is optimized process in overlay region Method is, it is first determined the minimum candidate of cost inlays the optimization processing scope of line, and in distance costs, minimum candidate inlays line In scope in optimization distance bufdis (ms), (vi) optimizes to it as the following formula:

If the line of inlaying in optimizing is rs, it passes through cost tcost (rs), and (vii) calculates as the following formula:

$t\cos t\left(rs\right)=\mathrm{\σ}t\cos t(u,v),(u,v)\&element;rs\&element;d---\left(vii\right)$

In formula, d refers to the hunting zone being defined by above formula (vi), and the line rs that inlays during tcost (u, v) refers to optimize passes through pixel The cost of (u, v), tcost (u, v) is the maximum in pixel (u, v) for the overlay region of m image and the difference of minima, presses Formula (viii) calculates:

Tcost (u, v)=max [l_j(u,v)]-min[l_j(u, v)], j=1 ..., m (viii)

In formula, l_j(u, v) refers to j-th image brightness value in pixel (u, v), and brightness value is by the red, green, blue three of image Wave band (ix) calculating gained as the following formula:

$l_j(u,v)=0.3p_j^r(u,v)+0.59p_j^g(u,v)+0.11p_j^b(u,v)---\left(ix\right)$

Wherein,WithRefer to the red, green, blue three in j-th image for the pixel (u, v) The pixel value of wave band,

Based on grid dijkstra algorithm, can find pass through in rs cost minimum inlay line, as finally inlay line.

Due to taking above technical scheme, it has the advantage that 1 to the present invention) method of the present invention can be by history Mapping big data significantly improves inlays efficiency.Experiment shows, and generally acknowledges the best edge based on dijkstra algorithm of effect at present Rule system of selection is compared, and can save 80-90% operation time.This is because the node of vector data is generally far fewer than grid number According to pixel although the later stage is also required to inlay line based on raster data optimization, but optimization range not enough original 1/10, thus efficiency Significantly higher.2) can significantly improve by mapping big data and inlay quality, pass through less house.Based on three groups of different buildings The experiment of inlaying of density shows: with based on vector road inlay line selection selection method compared with, the present invention can pass through 10-40% less House；And with present generally acknowledge effect best based on dijkstra algorithm inlay line selection selection method compared with, the present invention can wear less The more house of 1-15%.The completed region of the city domain that the method that aviation image is inlayed is particularly suitable for house data rich of the present invention Aviation image inlay.

Brief description

Fig. 1 is multiple boats of a plurality of air strips based on vector house data in the method that aviation image is inlayed of the present invention Empty image mosaic schematic flow sheet；

Fig. 2 is that the method flow inlaying line between two images of extraction in the method that aviation image is inlayed of the present invention shows It is intended to；

Fig. 3 be in the method that aviation image is inlayed of the present invention by two houses the candidate in vacant lot inlay line and have Effect width indication figure；

Fig. 4 is the optimization processing scope schematic diagram that minimum candidate is inlayed with line in the method for the present invention, gray area For inlaying the optimization processing scope of line.

Specific embodiment

With reference to Figure of description, technical scheme is further described.

As accompanying drawing 1 shows, choose a kind of data from vector house and inlay the method that line is inlayed to aviation image, its feature exists In comprising the following steps:

1) it is based on left-turn algorithm, the side to every Extraction of Image effective coverage in every air strips of the aviation image shooting Boundary line, effective coverage refers to the region of non-null value in image.In every air strips, according to every image effective coverage in this air strips Adjacent image overlap area in the line computation air strips of border；

2) merge the effective coverage in every air strips, obtain the boundary line of the effective coverage of every air strips, according to obtain Adjacent air strips overlay region between the border line computation air strips of the effective coverage of every air strips；

3) it is based on existing vector house data, adjacent air strips overlay region between adjacent image overlap area and air strips in air strips Line is inlayed in interior searching；

As shown in Figure 2, the method that in adjacent image overlap area in air strips, line is inlayed in searching is to extract all phases one by one Inlay line between adjacent image, find when inlaying line in adjacent air strips overlay region between air strips, by adjacent air strips as adjacent image, Using with adjacent image overlap area find inlay line identical method extract adjacent air strips between inlay line, if two adjacent shadows Picture is m degree superimposed image, and that is, in addition to this two images itself, also opening other images with m-2 has overlapping, m >=2, extracts two phases The method inlaying line between adjacent image comprises the following steps:

(1) in two adjacent image overlap areas based on constraint delaunay triangulation network algorithm extraction, between vector house vacant lot Intermediate line and overlay region skeleton line, then by overlay region between vector house the intermediate line in vacant lot and overlay region skeleton line Merge, composition candidate inlays line storehouse；

(2) effective width inlaying line according to candidate inlays line tax by cost to candidate, is sought based on dijkstra algorithm The candidate looking for the cost between origin-to-destination minimum inlays line；

(3) the minimum candidate of the cost that step (2) extracted using m degree superimposed image data is inlayed line and enters in overlay region Row optimization processing, forms and final inlays line it is ensured that any one section final inlayed line and can avoid passing through in image data Exist, and the prominent object not comprised in vector house data, prominent object includes house, temporary construction stacks thing etc..

4) inlay the mutual cutting of line by between in all air strips and air strips, be then based on left-turn algorithm, by inlaying after cutting Line joins end to end, and is that polygon individually effectively inlayed by every image structure in every air strips, finally inlays image and only use The data effectively inlayed in polygon of every image, remainder is given up；

5) data effectively inlayed in polygon of every image is merged, form the final edge including all single images Embedding image.

The effective width inlaying line in described step (2) according to candidate inlays line tax by cost to candidate, is based on The concrete grammar that the minimum candidate of cost that dijkstra algorithm is found between origin-to-destination inlays line is, if l_iFor i-th two Candidate in Zhang Xianglin image overlap area inlays line, and this candidate inlays line l_iEffective width vadw (l_i) (i) calculating as the following formula:

Wherein, d (l_i) represent and inlay line l with candidate_iMinimum range between adjacent room, as shown in Figure 3, projd (h_i1,h_i2) be two adjacent room height displacement's sum, as the following formula (ii) calculate:

projd(h_i1,h_i2)=tan θ_i1*h_i1(l_i)+tanθ_i2*h_i2(l_i) (ii)

Wherein, θ_i1And θ_i2Inlaying line l for two adjacent images with candidate respectively_iThe downwards angle of visibility in two adjacent houses； h_i1(l_i)、h_i2(l_i) be this two houses height,

Candidate inlays line l_iBy cost tcost (l_i) (iii) calculating as the following formula:

tcost(l_i)=κ (l_i)×len(l_i) (iii)

Wherein, len (l_i) represent that candidate inlays line l_iLength；κ(l_i) represent that candidate inlays line l_iWidth to cost Contribution, κ (l_i) (iv) calculating as the following formula:

Wherein, in order to ensure the skeleton line of overlay region, the intermediate line in vacant lot between house is far above in overlay region by cost Minimum widith by cost so that make least cost inlays the intermediate line that vacant lot between house preferentially followed the tracks of by line, c is set to often Number 0.1, minw is the minimum widith of the intermediate line in vacant lot between house in overlay region, if sl represent by house between vacant lot intermediate line Inlay line set with all candidates of overlay region skeleton line composition, cs represents that in some sl of connection, candidate inlays line and by edge The path of rule beginning and end, inlays two intersection points that line beginning and end is adjacent image effective coverage, in prior art If in number of hits more than two, two farthest intersection points of selected distance.(v) calculates passing through of the either path cs in cs as the following formula Cost f (cs):

$f\left(cs\right)=\mathrm{\σ}t\cos t\left(l_i\right),l_i\&element;cs\&element;sl---\left(v\right)$

Based on dijkstra algorithm, the minimum candidate of cost in cs can be found and inlay line ms, it is min [f by cost (cs)],cs∈cs∈sl.

In described step (3), the minimum candidate of cost inlays line and is optimized the method for process in overlay region and is, first Determine that the minimum candidate of cost inlays the optimization processing scope of line, as shown in Figure 4, minimum candidate inlays line in distance costs In scope in optimization distance bufdis (ms), (vi) optimizes to it as the following formula:

I.e. when the minimum candidate of cost inlay the effective width of line between 10m (rice) -60m (rice) when it is believed that two rooms It is likely to road between room, the scope being optimized process using image data is defined as 5m (rice), otherwise, when candidate inlays The effective width of line segment be less than 10m (rice) or more than 60m (rice) when it is believed that there will more than likely be between two houses other buildings or Prominent object, the scope being optimized process using image data is defined as 50m (rice).

If the line of inlaying in optimizing is rs, it passes through cost tcost (rs), and (vii) calculates as the following formula:

$t\cos t\left(rs\right)=\mathrm{\σ}t\cos t(u,v),(u,v)\&element;rs\&element;d---\left(vii\right)$

In formula, d refers to the hunting zone being defined by above formula (vi), and the line rs that inlays during tcost (u, v) refers to optimize passes through pixel The cost of (u, v), tcost (u, v) is the maximum in pixel (u, v) for the overlay region of m image and the difference of minima.Optimize In inlay the initial value that line is rs and can be set to the minimum candidate of cost in cs and inlay line ms by the cost sum of all pixels, Also can set in other arbitrary d positioned at hunting zone and by inlaying the path of line beginning and end, inlay the beginning and end of line It is respectively two intersection points of adjacent image effective coverage.By cost, (viii) calculates as the following formula:

Tcost (u, v)=max [l_j(u,v)]-min[l_j(u, v)], j=1 ..., m (viii)

In formula, l_j(u, v) refers to j-th image brightness value in pixel (u, v), and brightness value is by the red, green, blue three of image Wave band (ix) calculating gained as the following formula:

$l_j(u,v)=0.3p_j^r(u,v)+0.59p_j^g(u,v)+0.11p_j^b(u,v)---\left(ix\right)$

Wherein,WithRefer to the red, green, blue three in j-th image for the pixel (u, v) The pixel value of wave band, is 0.3,0.59 and 0.11 by prior art setting three wave band weights of red, green, blue.

Based on grid dijkstra algorithm, can find pass through in rs cost minimum inlay line, as finally inlay line.

Below with three groups of aviation orthography data, the present invention is chosen from vector house data and inlay line pair The efficiency of the method that aviation image is inlayed and effect are tested, and and calculate based on vector road with based on dijkstra The method of method is contrasted.The relevant parameter of three group image data is as shown in table 1.The present invention with based on vector road be based on The method accuracy comparison of dijkstra algorithm is as shown in Table 2-4.

The relevant parameter of 1 three groups of aviation images of table

As shown in Table 1, three groups of aviation images are respectively from three kinds of different regions of site coverage, including city, suburb And rural area, the data precision in vector house is from 0.5-5m.

Three kinds of method contrasts based on 36 city images for the table 2

^*Indicate: dijkstra algorithm is independent of vector data.Candidate inlays between the intersection point that line is two image overlay regions Straight line, the time that therefore calculates has leveled off to 0s.

Three kinds of method contrasts based on 6 suburb images for the table 3

Three kinds of method contrasts based on 110 rural images for the table 4

From table 2-4, choosing from vector house and road data that the present invention provides inlays polygon to aviation shadow The method relatively based on vector road for the method that picture is inlayed has been compared substantially with the line selection selection method of inlaying based on dijkstra algorithm Advantage: 1, with present generally acknowledge effect best based on dijkstra algorithm inlay line selection selection method compared with, 80-90% can be saved Operation time.2nd, the present invention can significantly improve by mapping big data and inlay quality, pass through less house.With based on vector The line selection selection method of inlaying of road is compared, and the present invention can pass through the house of 10-40% less；And the base best with generally acknowledged effect at present Line selection selection method of inlaying in dijkstra algorithm is compared, and the present invention can pass through the house of 1-15% less.As can be seen here, the method Greater advantages are inlaid with to the urban district aviation image of house data rich, are avoided that and pass through almost all of house.3rd, it is based on arrow The maximum memory that the method in amount house needs is close with the method based on vector road, but is significantly less than based on dijkstra algorithm Method (the only 26-30% of the latter about).

The various embodiments described above are merely to illustrate the present invention, and the connection of each part and structure all can be varied from, On the basis of technical solution of the present invention, all improvement connection and the structure of individual part being carried out according to the principle of the invention and equivalent Conversion, all should not exclude outside protection scope of the present invention.

Claims (4)

1. choose a kind of data from vector house below inlaying the method that line inlays to aviation image it is characterised in that including Step:

1) it is based on left-turn algorithm, the border to every Extraction of Image effective coverage in every air strips of the aviation image shooting Line, in every air strips, adjacent image overlap area in the border line computation air strips according to every image effective coverage in this air strips；

2) merge the effective coverage in every air strips, obtain the boundary line of the effective coverage of every air strips, according to every obtained Adjacent air strips overlay region between the border line computation air strips of the effective coverage of air strips；

3) it is based on existing vector house data, seek in adjacent air strips overlay region between adjacent image overlap area and air strips in air strips Look for and inlay line；

4) inlay the mutual cutting of line by between in all air strips and air strips, be then based on left-turn algorithm, by after cutting to inlay line first Tail is connected, and is that polygon individually effectively inlayed by every image structure in every air strips, finally inlays image only using every The data effectively inlayed in polygon of image, remainder is given up；

5) data effectively inlayed in polygon of every image is merged, form the final edge including all single images Embedding image.

2. choose a kind of data from vector house as claimed in claim 1 and inlay the method that line is inlayed to aviation image, its Be characterised by, described step 3) in find that to inlay the method for line be to extract all phases one by one in adjacent image overlap area in air strips Inlay line between adjacent image, find when inlaying line in adjacent air strips overlay region between air strips, by adjacent air strips as adjacent image, Using with adjacent image overlap area find inlay line identical method extract adjacent air strips between inlay line, if two adjacent shadows Picture is m degree superimposed image, and that is, in addition to this two images itself, also opening other images with m-2 has overlapping, m >=2, extracts two phases The method inlaying line between adjacent image comprises the following steps:

(1) in vacant lot between vector house in two adjacent image overlap areas based on constraint delaunay triangulation network algorithm extraction Top-stitching and the skeleton line of overlay region, then by overlay region between vector house the intermediate line in vacant lot and overlay region skeleton line close And, composition candidate inlays line storehouse；

(2) effective width inlaying line according to candidate inlays line tax by cost to candidate, is found based on dijkstra algorithm Point inlays line to the minimum candidate of the cost between terminal；

(3) the minimum candidate of the cost that step (2) extracted using m degree superimposed image data inlay line carry out in overlay region excellent Change is processed, formed final inlay line it is ensured that any one section final inlay line and can avoid passing through in image data exist , and the prominent object not comprised in vector house data.

3. choose a kind of data from vector house as claimed in claim 2 and inlay the method that line is inlayed to aviation image, its It is characterised by: the effective width inlaying line in described step (2) according to candidate inlays line tax by cost to candidate, is based on The concrete grammar that the minimum candidate of cost that dijkstra algorithm is found between origin-to-destination inlays line is, if l_iFor i-th two Candidate in Zhang Xianglin image overlap area inlays line, and this candidate inlays line l_iEffective width vadw (l_i) (i) calculating as the following formula:

Wherein, d (l_i) represent and inlay line l with candidate_iMinimum range between adjacent room, projd (h_i1,h_i2) it is two phase next doors Height displacement's sum in room, (ii) calculating as the following formula:

projd(h_i1,h_i2)=tan θ_i1*h_i1(l_i)+tanθ_i2*h_i2(l_i) (ii)

Wherein, θ_i1And θ_i2Inlaying line l for two adjacent images with candidate respectively_iThe downwards angle of visibility in two adjacent houses；h_i1 (l_i)、h_i2(l_i) be this two houses height,

Candidate inlays line l_iBy cost tcost (l_i) (iii) calculating as the following formula:

tcost(l_i)=κ (l_i)×len(l_i) (iii)

Wherein, len (l_i) represent that candidate inlays line l_iLength；κ(l_i) represent that candidate inlays line l_iThe contribution to cost for the width, κ(l_i) (iv) calculating as the following formula:

Wherein, c is set to constant 0.1, and minw is the minimum widith of the intermediate line in vacant lot between house in overlay region, if sl represents by room Between room, all candidates of the intermediate line in vacant lot and overlay region skeleton line composition inlay line set, and cs represents candidate in some sl of connection Inlay line and the path by inlaying line beginning and end, inlay two that line beginning and end is adjacent image effective coverage Intersection point, as the following formula (v) calculate cs in either path cs by cost f (cs):

F (cs)=∑ tcost (l_i),l_i∈cs∈sl (v)

Based on dijkstra algorithm, the minimum candidate of cost in cs can be found and inlay line ms, it is min [f by cost (cs)],cs∈cs∈sl.

4. choose a kind of data from vector house as claimed in claim 3 and inlay the method that line is inlayed to aviation image, its It is characterised by: in described step (3), the minimum candidate of cost inlays line and is optimized the method for process in overlay region and is, first Determine that the minimum candidate of cost inlays the optimization processing scope of line, minimum candidate inlays line optimization distance in distance costs In scope in bufdis (ms), (vi) optimizes to it as the following formula:

If the line of inlaying in optimizing is rs, it passes through cost tcost (rs), and (vii) calculates as the following formula:

Tcost (rs)=∑ tcost (u, v), (u, v) ∈ rs ∈ d (vii)

In formula, d refers to the hunting zone being defined by above formula (vi), tcost (u, v) refer to optimize in inlay line rs pass through pixel (u, V) cost, tcost (u, v) is the maximum in pixel (u, v) for the overlay region of m image and the difference of minima, as the following formula (viii) calculate:

Tcost (u, v)=max [l_j(u,v)]-min[l_j(u, v)], j=1 ..., m (viii)

In formula, l_j(u, v) refers to j-th image brightness value in pixel (u, v), and brightness value is by three wave bands of red, green, blue of image (ix) calculating gained as the following formula:

$l_j(u,v)=0.3p_j^r(u,v)+0.59p_j^g(u,v)+0.11p_j^b(u,v)---\left(ix\right)$

Wherein,WithRefer to pixel (u, v) three wave bands of red, green, blue in j-th image Pixel value,

Based on grid dijkstra algorithm, can find pass through in rs cost minimum inlay line, as finally inlay line.