CN103699883B - A kind of method utilizing village landmark group identification to locate buildings - Google Patents
A kind of method utilizing village landmark group identification to locate buildings Download PDFInfo
- Publication number
- CN103699883B CN103699883B CN201310693089.XA CN201310693089A CN103699883B CN 103699883 B CN103699883 B CN 103699883B CN 201310693089 A CN201310693089 A CN 201310693089A CN 103699883 B CN103699883 B CN 103699883B
- Authority
- CN
- China
- Prior art keywords
- village
- candidate
- real
- landmark group
- target structures
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a kind of method utilizing village landmark group identification to locate buildings, it is characterized in that, comprise the steps: (1) choose from satellite photo comprise target structures thing and village group face striograph, according to the air route of having planned, choose candidate village; (2) generate village landmark group reference diagram, set up the restriction relation between region, candidate village and target structures thing; (3) obtain a frame to scheme in real time, perspective transform is carried out to reference diagram, set up the restriction relation storehouse in candidate village; (4) according to the restriction relation storehouse that step (3) obtains, in real-time figure, selected village is determined; (5) extrapolate the position of target structures thing in real-time figure, in the real-time figure of Primary Location, comprise the local region of interest of target structures thing; (6) the building target identification based on template matches is carried out at local region of interest.The method location is fast, false alarm rate is low, reliability is high, for catching when aircraft forward sight is navigated and locating above ground structure and provide gordian technique.
Description
Technical field
The invention belongs to imaging Target Recognition field, more specifically, relate to a kind of method utilizing village landmark group identification to locate buildings.
Background technology
Any aircraft, before flight, all will do routeing in the ground preparatory stage, and this is a basic precursor conditions, then just according to the waypoint navigation flight of regulation.In the flight course of reality, due to the deviation of aircraft inertia navigator fix, cause the deviation of the line of flight, and then the deviation causing air craft carried imaging sensor optical axis to point to.Thus target image may not at the center of visual field, and also possibility to the left relative to center possibility is to the right, and possibility is on the upper side also may be on the lower side, causes real target location may have angular displacement with the navigation position of expectation.If just by means of inertial navigation unit by aircraft navigation to predetermined target location, will become abnormal difficult due to the existence of above-mentioned deviation.
After images match airmanship is born, navigated by Downward-looking scene matching and be once widely adopted, but due to the requirement of this technology in the matching process for scene uniqueness higher, that is, if similar scene repeats, navigating to predetermined target will have any problem.In order to solve this difficulty, be born novel airmanship, i.e. forword-looking imaging airmanship, corrects navigation deviation by the forword-looking imaging target identification of optical sensor.
Many have the culture of strategic importance in rural area, such as airport, missile emplacements etc., these buildings all can cover up measure in actual combat, or due to the factor such as distant, illumination condition is bad, weather environment is severe, cause it relative to the rural area scene at its place, feature is not significantly not enough to as identical criterion, and therefore, direct-detection fixation and recognition is infeasible.And general forward sight target navigation by recognition method is all take Direct Recognition target to correct practical flight air route and the deviation planning air route, therefore, it is impossible or insecure for locating three-dimensional buildings in this way.In order to solve this navigator fix difficult problem, new positioning identifying method must be invented.
Summary of the invention
For above defect or the Improvement requirement of prior art, the invention provides a kind of method utilizing village landmark group identification to locate buildings, under effectively solving moving platform condition, when buildings optical imagery feature to be identified is not remarkable, identify the problem of reliability of positioning difference, location is fast, and false alarm rate is low, reliability is high, for catching when aircraft forward sight is navigated and locating above ground structure and provide gordian technique.
For achieving the above object, the invention provides a kind of method utilizing village landmark group identification to locate buildings, it is characterized in that, comprise the steps: (1) choose from satellite photo comprise target structures thing and village group face striograph, according to the air route of having planned, choose wherein several alternatively village, village having complete area; (2) generate village landmark group reference diagram, in village landmark group reference diagram, set up the restriction relation between region, candidate village and target structures thing; (3) obtain a frame to scheme in real time, perspective transform is carried out to village landmark group reference diagram, in perspective transform result images, set up the restriction relation storehouse in candidate village; (4) according to the restriction relation storehouse that step (3) obtains, in the obtained real-time figure of each frame, the selected village belonging to candidate village is determined; (5) the selected village that the restriction relation between the region, candidate village obtained according to step (2) and target structures thing and step (4) obtain, extrapolate the position of target structures thing in real-time figure, in the real-time figure of Primary Location, comprise the local region of interest of target structures thing; (6) the local region of interest obtained in step (5) carries out the building target identification based on template matches, finally completes the fixation and recognition to buildings.
Preferably, described candidate village is 4 ~ 7.
Preferably, described step (2) comprises the steps: that (2-1) generates village landmark group reference diagram further; (2-2) in reference diagram, the angle β of calculated line VT and ON and the length L of line segment VT
vt, wherein, V is candidate's village landmark group central point, and T is target structures thing central point, and O is satellite hub coordinate points, and ON is for taking O as the straight line that starting point is done towards direct north, and candidate's village landmark group central point is the mean value of candidate village regional centroid coordinate.
Preferably, described step (3) comprises the steps: that (3-1) obtains a frame and scheme in real time further, according to the flight support parameter of the real time imagery line number of real-time figure, real time imagery columns and moving platform, calculate the position coordinates of landmark group constraint central point in front view, perspective transform is carried out to village landmark group reference diagram; (3-2) in perspective transform design sketch, calculate each village regional centroid coordinate, therefrom a triangle is determined as a combination in optional three villages, total
individual triangle, calculates the long L of each leg-of-mutton longest edge
mxkwith diagonal angle ∠ Ω
k, set up the restriction relation storehouse in candidate village
wherein, n is the number in candidate village.
Preferably, described step (4) comprises the steps: that (4-1) carries out binary segmentation to the real-time figure of a described frame further, obtains the bianry image comprising region, m village, calculates each village regional centroid coordinate, therefrom a triangle is determined as a combination in optional three villages, total
individual triangle, calculates the long L of each leg-of-mutton longest edge
mxf' and the size ∠ Ω at diagonal angle
f', wherein,
(4-2) according to the restriction relation storehouse that step (3-2) obtains
to (L
mxf', ∠ Ω
f') screen, choose and L
mxkerror ± 8 pixels and with ∠ Ω
kerror the combination of ± 3 °, as the centre of form triangle in selected village, determine selected village centre of form coordinate V
g' (x
vg', y
vg'), g=1,2,3...n
sfor village sequence number, n
sfor selected village sum, thus obtain the center of the selected village landmark group of figure in real time.
In general, the above technical scheme conceived by the present invention compared with prior art, has following beneficial effect:
1, because employing step (2) establishes the restriction relation between region, candidate village and target structures thing, and extract selected village by step (4), and then comprise the region of interest of target structures thing by step (5) Primary Location, efficiently solve target structures thing feature not obvious or be blocked hide cause cannot Direct Recognition location problem.
2, owing to adopting step (6) to identify location building target in local region of interest, not only considerably reduce the time overhead performed needed for building recognition location tasks, also greatly reduce the false alarm rate of building recognition location.
Accompanying drawing explanation
Fig. 1 is the village regional architecture thing recognition positioning method schematic flow sheet of the embodiment of the present invention;
Fig. 2 be comprise target structures thing and village group face striograph;
Fig. 3 is the candidate village regional satellite front elevation according to planning air route;
Fig. 4 is village landmark group reference diagram;
Fig. 5 is that a frame is schemed in real time;
Fig. 6 is perspective transform schematic diagram;
Fig. 7 is design sketch village landmark group reference diagram being carried out to perspective transform;
Fig. 8 is the design sketch real-time figure of a frame being carried out to binary segmentation;
Fig. 9 is the confirmation result figure in candidate village in the real-time figure of a frame;
Figure 10 is the preliminary recognition result figure of village landmark group indirect addressing;
Figure 11 is target structures thing template;
Figure 12 is Primary Location and final recognition result figure.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.In addition, if below in described each embodiment of the present invention involved technical characteristic do not form conflict each other and just can mutually combine.
Indirect searching order calibration method more meets the vision guided navigation mechanism of the mankind, that is, when we are when being intended to catch a target, if when this target is not remarkable relative to the scene characteristic at its place, we take the ground object structure in the scene of first target acquisition place with certain significant characteristics usually, indirect addressing is to our interested target again, and then the indirect addressing reached object of interest and identification.
In general the house of rural area is all exist in the mode of cluster, piecemeal village is all easily seen in the scene of rural area, and the general house in rural area all can not be very high, it is a kind of obviously characteristic in FLIR (Forward-Looking Infrared) figure, even if severe in weather condition, or target is when being subject to hiding, village landmark group is utilized also to be feasible as ground marker navigation.How to utilize village landmark group to carry out the detection and location of the Fabricated structures of rural areas, constitute core content of the present invention.
As shown in Figure 1, the method utilizing village landmark group identification to locate buildings of the embodiment of the present invention comprises the steps:
(1) choose from satellite photo comprise target structures thing and village group face striograph, according to the air route of having planned, choose wherein several alternatively village, village having complete area.
As shown in Figure 2, the resolution facing striograph comprising target structures thing and village group is 1.2 meters, and size is 5960 × 3450 pixels.The flying height of aircraft is for being 1500 meters, and the angle of entry angle and direct north is 243 °.As shown in Figure 3, according to the air route of having planned, choose wherein 4 alternatively villages, village having a complete area.
(2) generate village landmark group reference diagram, in village landmark group reference diagram, set up the restriction relation between region, candidate village and target structures thing.
(2-1) village landmark group reference diagram is generated as shown in Figure 4.
(2-2) for candidate village set of regions, feature is calculated as follows: centre of form coordinate is VC
i(x
vci, y
vci), wherein, i=1,2,3...n are village sequence number, and n is village sum, and four selected village centre of form coordinates are respectively village VC
1(175,129), village VC
2(91,347), village VC
3(168,381) and village VC
4(81,396).Target structures thing central point T (x
t, y
t)=(189,264), the mean value that village landmark group central point V (104,313) is village regional centroid coordinate.Satellite hub coordinate is O (ROW/2, COL/2), and wherein, ROW is the real time imagery line number of real-time figure, and COL is the real time imagery columns of real-time figure.Be that starting point makes straight line towards direct north with O, the note angle β of ON, calculated line VT and ON and the length L of line segment VT
vt.
(3) obtain a frame to scheme in real time, as shown in Figure 5, according to real time imagery line number ROW, the real time imagery columns COL of real-time figure and the flight support parameter of moving platform, perspective transform is carried out to village landmark group reference diagram, perspective transform schematic diagram as shown in Figure 6, sets up the restriction relation storehouse in candidate village in perspective transform result images.
(3-1) according to real time imagery line number ROW, the real time imagery columns COL of real-time figure and the flight support parameter of moving platform, the position coordinates (X of terrestrial reference constraint central point in front view is calculated
r, Y
r), perspective transform is carried out to village landmark reference diagram, the following formula of concrete employing:
X
r=ROW/2+(∠OMP-θ)*ROW/φ
Wherein,
OT
0=h/tanθ
OM=OT
0+(y
c-y
0)×cosα+(x
c-x
0)×sinα
tan(∠OMP)=h/OM
Wherein, (x
0, y
0) be the beam axis sight point of moving platform under earth coordinates, OT
0optical axis for moving platform points to the intersection point T with the earth surface level
0and the distance between the horizontal projection point O of moving platform imager on the earth surface level, M is the central point T of airfield runway constraint
1at projection and the OT of optical axis longitudinal direction
0the intersection point of straight line, OM is the distance between horizontal projection point O and some M, and φ is longitudinal imaging viewing field angle of moving platform imager,
for the horizontal angle of image of imager, α is the imaging side parallactic angle of moving platform imager, and θ is the imaging angle of pitch of moving platform imager, and h is the image height of moving platform imager.The design sketch of perspective transform as shown in Figure 7.
(3-2) in perspective transform design sketch, each village regional centroid coordinate V is calculated
j(x
vj, y
vj), j=1,2,3...n are village sequence number.The centre of form coordinate in four villages is respectively V
1(24,91), V
2(32,153), V
3(214,151) and V
4(63,164), village landmark group central point V
othe mean value that (83,137) are village regional centroid coordinate.From wherein optional three villages, then have
individual combination, a triangle is determined in each combination, total
individual triangle.Calculate the long L of each leg-of-mutton longest edge
mxkwith the size ∠ Ω at diagonal angle
k, set up the restriction relation storehouse in candidate village
(4) according to the restriction relation storehouse that step (3) obtains, in the obtained real-time figure of each frame, the selected village belonging to candidate village is determined.
(4-1) binary segmentation is carried out to the real-time figure of a frame, obtain the bianry image comprising region, village.The design sketch of binary segmentation is carried out as shown in Figure 8 to the real-time figure of a frame, wherein has region, eight villages, calculate its centre of form coordinate VS respectively
l(x
vsl, y
vsl), l=1,2,3...m are village sequence number, and m is village sum.From wherein optional three villages, then have
individual combination, a triangle is determined in each combination, total
individual triangle.Calculate the long L of each leg-of-mutton longest edge
mxf' and the size ∠ Ω at diagonal angle
f', wherein,
(4-2) according to the restriction relation storehouse that step (3) obtains
to (L
mxf', ∠ Ω
f') screen, choose and L
mxkerror ± 8 pixels and with ∠ Ω
kerror the combination of ± 3 °, as the centre of form triangle in selected village, thus determine selected village centre of form coordinate V
g' (x
vg', y
vg'), g=1,2,3...n
sfor village sequence number, n
sfor selected village sum.Finally obtain 4 selected village centre of form coordinates and be respectively V
1' (37,78), V
2' (64,126), V
3' (97,135) and V
4' (178,124), as shown in Figure 9.Thus obtain the center V ' (94,116) of the selected village landmark group of figure in real time.
(5) the selected village in the real-time figure that the region, candidate village in the reference diagram obtained according to step (2) and the restriction relation between target structures thing and step (4) obtain, extrapolate the position of target structures thing in real-time figure, in the real-time figure of Primary Location, comprise the local region of interest of target structures thing.
Particularly, with reference to the angle β of figure cathetus VT and ON and the length L of line segment VT
vtcorrespond in real-time figure, obtain the angle β ' of V ' T ' with O ' N ' and the length L of V ' T '
vt', the center V ' of the selected village landmark group that integrating step (4) obtains, calculates the central point T0 ' of target structures thing in real-time figure (x ', y ').In the present embodiment, T0 ' (x ', y ')=(165,91), getting region S{ (x, y) | x '-50<x<x '+50, y '-50<y<y '+50} just locates local region of interest, as shown in Figure 10, wherein T0 ' is first positioning result to the preliminary recognition result figure of village landmark group indirect addressing.
(6) the local region of interest obtained in step (5) carries out the building target identification based on template matches, finally completes the fixation and recognition to buildings.
(6-1) Figure 11 is the "T"-shaped target structures thing template made, and target structures thing is that airfield runway connects road.
(6-2) in local, the first location region of interest that step (5) obtains, carry out the building target identification based on template matches, recognition result as shown in figure 12, wherein larger frame representative just positioning result, less frame represents the final recognition result in region of interest.
Those skilled in the art will readily understand; the foregoing is only preferred embodiment of the present invention; not in order to limit the present invention, all any amendments done within the spirit and principles in the present invention, equivalent replacement and improvement etc., all should be included within protection scope of the present invention.
Claims (3)
1. utilize village landmark group identification to locate a method for buildings, it is characterized in that, comprise the steps:
(1) choose from satellite photo comprise target structures thing and village group face striograph, according to the air route of having planned, choose wherein several alternatively village, village having complete area;
(2) generate village landmark group reference diagram, in village landmark group reference diagram, set up the restriction relation between region, candidate village and target structures thing;
(3) obtain a frame to scheme in real time, perspective transform is carried out to village landmark group reference diagram, in perspective transform result images, set up the restriction relation storehouse in candidate village;
Comprise the steps: further
(3-1) obtain a frame to scheme in real time, according to the flight support parameter of the real time imagery line number of real-time figure, real time imagery columns and moving platform, calculate the position coordinates of landmark group constraint central point in front view, perspective transform is carried out to village landmark group reference diagram;
(3-2) in perspective transform design sketch, calculate each village regional centroid coordinate, therefrom a triangle is determined as a combination in optional three villages, total
individual triangle, calculates the long L of each leg-of-mutton longest edge
mxkwith diagonal angle ∠ Ω
k, set up the restriction relation storehouse in candidate village
wherein, n is the number in candidate village;
(4) according to the restriction relation storehouse that step (3) obtains, in the obtained real-time figure of each frame, the selected village belonging to candidate village is determined;
Comprise the steps: further
(4-1) carry out binary segmentation to the real-time figure of a described frame, obtain the bianry image comprising region, m village, calculate each village regional centroid coordinate, therefrom a triangle is determined as a combination in optional three villages, total
individual triangle, calculates the long L of each leg-of-mutton longest edge
mxf' and the size ∠ Ω at diagonal angle
f', wherein,
(4-2) according to the restriction relation storehouse that step (3-2) obtains
to (L
mxf', ∠ Ω
f') screen, choose and L
mxkerror ± 8 pixels and with ∠ Ω
kerror the combination of ± 3 °, as the centre of form triangle in selected village, determine selected village centre of form coordinate V
g' (x
vg', y
vg'), g=1,2,3...n
sfor village sequence number, n
sfor selected village sum, thus obtain the center of the selected village landmark group of figure in real time;
(5) the selected village that the restriction relation between the region, candidate village obtained according to step (2) and target structures thing and step (4) obtain, extrapolate the position of target structures thing in real-time figure, in the real-time figure of Primary Location, comprise the local region of interest of target structures thing;
(6) the local region of interest obtained in step (5) carries out the building target identification based on template matches, finally completes the fixation and recognition to buildings.
2. utilize village landmark group identification to locate the method for buildings as claimed in claim 1, it is characterized in that, in described step (1), described candidate village is 4 ~ 7.
3. utilize village landmark group identification to locate the method for buildings as claimed in claim 1 or 2, it is characterized in that, described step (2) comprises the steps: further
(2-1) village landmark group reference diagram is generated;
(2-2) in reference diagram, the angle β of calculated line VT and ON and the length L of line segment VT
vt, wherein, V is candidate's village landmark group central point, and T is target structures thing central point, and O is satellite hub coordinate points, and ON is for taking O as the straight line that starting point is done towards direct north, and candidate's village landmark group central point is the mean value of candidate village regional centroid coordinate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310693089.XA CN103699883B (en) | 2013-12-17 | 2013-12-17 | A kind of method utilizing village landmark group identification to locate buildings |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310693089.XA CN103699883B (en) | 2013-12-17 | 2013-12-17 | A kind of method utilizing village landmark group identification to locate buildings |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103699883A CN103699883A (en) | 2014-04-02 |
| CN103699883B true CN103699883B (en) | 2016-04-13 |
Family
ID=50361406
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310693089.XA Active CN103699883B (en) | 2013-12-17 | 2013-12-17 | A kind of method utilizing village landmark group identification to locate buildings |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103699883B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018098731A1 (en) * | 2016-11-30 | 2018-06-07 | 深圳益强信息科技有限公司 | Aerial reconnaissance target determination method and device based on flight vehicle |
| WO2018223355A1 (en) * | 2017-06-09 | 2018-12-13 | 深圳市乃斯网络科技有限公司 | Game map positioning implementation method and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101509782A (en) * | 2009-03-09 | 2009-08-19 | 华中科技大学 | Small-sized ground marker capturing and positioning method |
| CN101620671A (en) * | 2009-08-14 | 2010-01-06 | 华中科技大学 | Method for indirectly positioning and identifying three-dimensional buildings by using riverway landmarks |
| CN101726297A (en) * | 2009-12-18 | 2010-06-09 | 华中科技大学 | Plane landmark selection and reference map preparation method for front-view navigation guidance |
-
2013
- 2013-12-17 CN CN201310693089.XA patent/CN103699883B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101509782A (en) * | 2009-03-09 | 2009-08-19 | 华中科技大学 | Small-sized ground marker capturing and positioning method |
| CN101620671A (en) * | 2009-08-14 | 2010-01-06 | 华中科技大学 | Method for indirectly positioning and identifying three-dimensional buildings by using riverway landmarks |
| CN101726297A (en) * | 2009-12-18 | 2010-06-09 | 华中科技大学 | Plane landmark selection and reference map preparation method for front-view navigation guidance |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103699883A (en) | 2014-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10274958B2 (en) | Method for vision-aided navigation for unmanned vehicles | |
| Yol et al. | Vision-based absolute localization for unmanned aerial vehicles | |
| CN103822635B (en) | The unmanned plane during flying spatial location real-time computing technique of view-based access control model information | |
| CN107251055B (en) | Corridor capture | |
| Yahyanejad et al. | Incremental mosaicking of images from autonomous, small-scale uavs | |
| CN107727079A (en) | The object localization method of camera is regarded under a kind of full strapdown of Small and micro-satellite | |
| Brenner | Extraction of features from mobile laser scanning data for future driver assistance systems | |
| Chen et al. | Drogue tracking using 3D flash lidar for autonomous aerial refueling | |
| US20120314032A1 (en) | Method for pilot assistance for the landing of an aircraft in restricted visibility | |
| US8649917B1 (en) | Apparatus for measurement of vertical obstructions | |
| CN101620671B (en) | Method for indirectly positioning and identifying three-dimensional buildings by using riverway landmarks | |
| Chiabrando et al. | Direct photogrammetry using UAV: tests and first results | |
| US20100097460A1 (en) | Apparatus for measurement of vertical obstructions | |
| KR102195179B1 (en) | Orthophoto building methods using aerial photographs | |
| JP4978615B2 (en) | Target identification device | |
| CN106468547A (en) | Utilize multiple optical pickocffs is independent of global positioning system for self-conductance aircraft（“GPS”）Navigation system | |
| ES2854944T3 (en) | Procedure and device for georeferencing aerial image data using SAR image data | |
| CN108955685A (en) | A kind of tanker aircraft tapered sleeve pose measuring method based on stereoscopic vision | |
| KR101771492B1 (en) | Method and system for mapping using UAV and multi-sensor | |
| US20110246069A1 (en) | Method for determining the trajectory of a ballistic missile | |
| US11460302B2 (en) | Terrestrial observation device having location determination functionality | |
| CN103456027B (en) | Time sensitivity target detection positioning method under airport space relation constraint | |
| Andert et al. | On the safe navigation problem for unmanned aircraft: Visual odometry and alignment optimizations for UAV positioning | |
| CN103699883B (en) | A kind of method utilizing village landmark group identification to locate buildings | |
| JP5769149B2 (en) | Mobile mapping system, roadside object measurement method using the same, and position specifying program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |