CN106447601A - Unmanned aerial vehicle remote image mosaicing method based on projection-similarity transformation - Google Patents
Unmanned aerial vehicle remote image mosaicing method based on projection-similarity transformation Download PDFInfo
- Publication number
- CN106447601A CN106447601A CN201610795444.8A CN201610795444A CN106447601A CN 106447601 A CN106447601 A CN 106447601A CN 201610795444 A CN201610795444 A CN 201610795444A CN 106447601 A CN106447601 A CN 106447601A
- Authority
- CN
- China
- Prior art keywords
- transformation
- image
- projection
- similarity
- similarity transformation
- 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.)
- Granted
Links
- 230000009466 transformation Effects 0.000 title claims abstract description 132
- 238000000034 method Methods 0.000 title claims abstract description 21
- 239000000284 extract Substances 0.000 claims abstract description 3
- 238000006243 chemical reaction Methods 0.000 claims description 23
- 230000000007 visual effect Effects 0.000 claims description 18
- 230000008859 change Effects 0.000 claims description 17
- 230000003139 buffering effect Effects 0.000 claims description 6
- 230000008878 coupling Effects 0.000 claims description 2
- 238000010168 coupling process Methods 0.000 claims description 2
- 238000005859 coupling reaction Methods 0.000 claims description 2
- 230000007704 transition Effects 0.000 claims description 2
- 230000004927 fusion Effects 0.000 claims 1
- 230000001131 transforming effect Effects 0.000 description 16
- 238000003384 imaging method Methods 0.000 description 10
- 239000011159 matrix material Substances 0.000 description 7
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 4
- 241001633663 Iris pseudacorus Species 0.000 description 2
- 231100000768 Toxicity label Toxicity 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000013213 extrapolation Methods 0.000 description 2
- 238000012886 linear function Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000005452 bending Methods 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/14—Transformations for image registration, e.g. adjusting or mapping for alignment of images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/60—Rotation of whole images or parts thereof
- G06T3/604—Rotation of whole images or parts thereof using coordinate rotation digital computer [CORDIC] devices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/32—Indexing scheme for image data processing or generation, in general involving image mosaicing
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an unmanned aerial vehicle remote sensing image mosaicing method, and in particular for unmanned aerial vehicle remote sensing images having small overlapping regions and under great differences of field of view. The method extracts feature points of an image, matches the feature points so as to determine projection, and smoothly changes projection transformation of the overlapping regions to similarity transformation of non-overlapping regions. The method includes the following detailed procedures: firstly changing coordinates of the acquired projection transformation; secondly, dividing the changed coordinate space into a plurality of different sub-spaces with a straight line which is parallel to a coordinate axis; then, determining a parameter U1 and U2 to acquire a new projection-similarity transformation; and finally using the acquired changing to conduct image mosaic and generate a mosic image. The changing takes registration precision of the overlapping regions and the consistency of the field of view of the non-overlapping regions at the same time, such that the mosaic image has high matching precision and maintains the field of view of the image to reduce image distortion.
Description
Technical field
The invention belongs to unmanned aerial vehicle remote sensing image automatic business processing field is and in particular to a kind of be based on projection-similarity transformation
Unmanned aerial vehicle remote sensing image split-joint method.
Background technology
Unmanned aerial vehicle remote sensing, as one of the development trend of remote sensing, has very strong ageing, pin in data acquisition
The advantages of to property and high flexibility, it is the important channel obtaining remotely-sensed data, has a wide range of applications in terms of emergency response, can
For the monitoring of forest fire, the emergency response of the natural calamity such as earthquake.Because areas imaging limits, single width unmanned plane image is past
Toward whole devastated cannot be covered, unmanned aerial vehicle remote sensing image need to be spliced to obtain large-scale observation.
Due to unmanned aerial vehicle platform stability problem so that between adjacent image degree of overlapping change greatly, the anglec of rotation change greatly,
So that traditional joining method is inapplicable for the splicing of unmanned aerial vehicle remote sensing image.Below consideration is needed in unmanned plane image mosaic
Some：First, the inclination angle of image is excessive, ship's control is little, gray scale is inconsistent etc. makes that images match difficulty is big, precision
Low, adversely affect to subsequent treatment；Second, image areas imaging is little, quantity is many, causes that workload is big, efficiency is low, needs
Research is efficient, the processing method of high degree of automation.Image mosaic is widely used in Photogrammetry and Remote Sensing field, therefore
Research for remote sensing images splicing has a lot, but the splicing for unmanned aerial vehicle remote sensing image yet suffers from many technology
A difficult problem.
Image mosaic be by combine a series of images obtain a width big there is wider array of view field image.Widely used
It is using wherein piece image as benchmark image that Method of Projection Change carries out two width image mosaic, by the overlapping region of two width images
Alignment, but this conversion can make Non-overlapping Domain produce serious distortion in terms of size and shape.Hartley etc. points out to adopt
Observation viewpoint can be changed with projective transformation so that after conversion on the basis of image visual field extension, it is single for therefore obtaining stitching image
After visual point image can make splicing, image produces larger shapes and sizes distortion.
The one-view image with wider wide field inevitably produces shapes and sizes distortion, and therefore good image is spelled
The scheme of connecing should produce the stitching image of multiple views.Basic thought is that the overlapping region to image to be obtained using projective transformation
Geometric alignment and the Non-overlapping Domain in image uses similarity transformation.Similarity transformation by translating, uniformly scaling and rotation form, because
This similarity transformation will not produce shape distortion and non-homogeneous scaling so that direction of visual lines does not change therefore can keep image
Viewpoint.
Some representative technology are had at present in terms of image mosaic.R.Marzotto etc. proposes a kind of automatic
The method building panoramic mosaic image, the method carries out global registration using affine transformation and generates panorama sketch.Substantial amounts of projective module
Type is suggested and target is so that the distortion minimization that projection produces, and such as Brown etc. just adopts in AutoStitch program
It is spherical projection, Zelnik-Manor etc. is replaced cylindrical surface projecting and executes projective transformation using multiple plane projections.Carroll
To reduce the distortion of spliced wide angle picture Deng the projective transformation proposing content preservation.Important literature includes：Marzotto
R., Fusiello A., Murino V.High resolution video mosaicing with global alignment
[C].Proceedings of the 2004IEEE Computer Society Conference on Computer
Vision and Pattern Recognition, 2004；M.Brown and D.G.Lowe.Automatic panoramic
image stitching using invariant features[J].International Journal of Computer
Vision, 74 (1)：59-73,2007 etc..
Image mosaic technology is frequently utilized that parameterised transform to carry out global or local image alignment.Gao proposes double lists should
Become and bring the scene processing containing two dominant plane, this transforming function transformation function is defined as the line that two spaces change the projective transformation of weight
Property combination, but this conversion is based on projective transformation, and the image of formation produces projection distortion.Lin etc. proposes the affine of smooth change
Conversion, this affine transformation allows local deformation for the overall situation is affine, and its important literature includes：J.Gao, S.J.Kim, and
M.S.Brown.Constructing image panoramas using dual-homography warping[C].In
Proceedings of IEEE CVPR 2011, pages 49-56,2011 etc..But these methods are not suitable for unmanned aerial vehicle remote sensing
Image, does not account for the image f iotaeld-of-view inconsistence problems that the unstability of remote sensing platform causes, is not suitable for unmanned aerial vehicle remote sensing figure
As splicing.The present invention provides a kind of unmanned aerial vehicle remote sensing image split-joint method based on projection-similarity transformation.
Content of the invention
The technical problem to be solved is little for unmanned aerial vehicle remote sensing image areas imaging, and parallax angle is big
Problem, provides one kind to carry out unmanned aerial vehicle remote sensing image split-joint method based on a kind of improved projection-similarity transformation.The method is first
Projective transformation between two width images is estimated according to the overlapping region between two width images.After obtaining image projection transformation, to throwing
By new extrapolation strategy, shadow conversion to determine that the similarity transformation of Non-overlapping Domain keeps the visual field of image.The change exchange the letters proposing
Number makes stitching image have high matching precision and the visual field of image can be kept to reduce image fault.
The thinking of the present invention is：By carrying out after coordinate transform to traditional projective transformation, and analyze the basis of its property
Go up the improvement it is proposed that to original projective transformation.By the projective transformation of overlapping region, it is gradually transitions non-by strategy of extrapolating
So that there is good geometrical registration precision overlapping region, the image of Non-overlapping Domain passes through simultaneously for the similarity transformation of overlapping region
Similarity transformation to reduce image fault to keep the visual field of image.
Technical scheme provides one kind and carries out unmanned aerial vehicle remote sensing image spelling based on improved projection-similarity transformation
Connect method it is characterised in that following implementation steps：
1) scale invariant feature-SIFT feature is extracted to image sequence；
2) carry out Image Feature Matching using the SIFT feature obtaining, picked using RANSAC algorithm during characteristic matching
Except Mismatching point pair；
3) after mutually corresponding characteristic point completes coupling between image, determine corresponding geometric transformation model, i.e. projection becomes
Change H；
4) coordinate system transformation is carried out to projective transformation, by rotation transformation to making original (x, y) coordinate system transformation to new
(u, v) coordinate system；
5) to projective transformation property analysis under new coordinate system, and with two lines U1 and U2, space is carried out division and be divided into
RH, RTAnd RSThree sub-spaces；
6) parameter U1 and U2 are solved, and determine final projection-similarity transformation w；
7) using transforming function transformation function w, conversion is executed to image, carry out geometry splicing；
8) carry out image co-registration using linear weighted function, obtain final stitching image.
Above-mentioned implementation steps are characterised by：After obtaining the projective transformation of image sequence, in the base that projective transformation is analyzed
On plinth, the projective transformation of overlapping region is gradually transformed to the similarity transformation of Non-overlapping Domain, produce new projection-similar change
Change.Projection-similarity transformation both can guarantee that the matching precision of overlapping region, also can guarantee that the visual field of Non-overlapping Domain is unanimously come simultaneously
Reduce image fault.
Little for unmanned aerial vehicle remote sensing image areas imaging, the big feature of parallax angle, the present invention provides a kind of new conversion
I.e. projection-similarity transformation carries out unmanned aerial vehicle remote sensing image mosaic.First two width are estimated according to the overlapping region between two width images
Projective transformation between image.After obtaining image projection transformation, projective transformation is determined non-overlapped by new extrapolation strategy
The similarity transformation in region is keeping the visual field of image.The transforming function transformation function proposing makes stitching image have high matching precision and can protect
Hold the visual field of image to reduce image fault.
Brief description
The unmanned aerial vehicle remote sensing image mosaic flow chart based on projection-similarity transformation for the Fig. 1
Fig. 2 projective transformation and the comparative result figure of the conversion proposing
Wherein Fig. 2 (A) is original image, and Fig. 2 (B) is projective transformation, and Fig. 2 (C) is the conversion proposing；
The splicing result figure of the different conversion of Fig. 3
Wherein Fig. 3 (A) is that projective transformation carries out splicing result figure, Fig. 3 (B) AutoStitch splicing result figure, Fig. 3 (C)
The conversion splicing result figure proposing；
Fig. 4 is little to overlapping region and splicing result figure that there is the inconsistent image in visual field
Overlapping region in Fig. 4 (A) and Fig. 3 (B) is red-label part, and the overlapping region in Fig. 4 (B) and Fig. 4 (C) is
Yellow flag part, in Fig. 4 (A), imaging process is located at the upper left corner imaging of image for camera, and Fig. 4 (B) camera is located at image
The lower left corner is imaged, and Fig. 4 (C) faces imaging contexts for ideal, and the conversion that Fig. 4 (D) proposes carries out splicing result to this three width image
Figure；
Fig. 5 is to image sequence splicing result figure
Fig. 5 (A) artwork, Fig. 5 (B) proposes conversion and carries out image sequence splicing result figure
Specific embodiment
Below in conjunction with the accompanying drawings the specific embodiment of the present invention is described as follows：
Image sequence extracts characteristic point and feature descriptor-SIFT feature, and by the Feature point correspondence extracting get up into
Row characteristic matching.
Mismatching point in images match can have a strong impact on the precision of geometric transformation parameter, is carrying out projective transformation solution
Before, by Mismatching point to rejecting, Mismatching point pair need to be rejected using RANSAC algorithm.
To reject Mismatching point to after correct matching double points solve projective transformation matrix
Projective transformation matrix is analyzed
Projection functionIt is defined as the linear transformation under homogeneous coordinates
Then (x, y) arrives being mapped as of (x ', y ')：
Original coordinate system (x, y) is transformed to new coordinate system (u, v)：
After coordinate system transformation, new projective transformation matrix H is the mapping that under new coordinate system, point (u, v) arrives (x ', y ')
Wherein：
Then after coordinate system transformation, new coordinate mapping relations are expressed as
Mapping H is the function of u and v, i.e. [x ', y ']T=H (u, v)=[Hx(u, v), Hy(u, v)]T.The master of coordinate transform
Want advantage to be that in new projective transformation mapping, denominator term only has a coordinate u, be easy to analyze.
The property of the transformation matrix after coordinate transform：
(1) scaling change.Projective transformation H can be decomposed into an affine transformation matrix and a pure projective transformation.(u, v)
Regional area change by map H to measure in Jacobian J (u, v) of point (u, v)：
Wherein sAIt is one and u, the uncorrelated constant of v.The zoom factor of regional area change only relies only on parameter u, especially
It is that, when coordinate u becomes big, the local area of H becomes big, leads to big region deformation distortion.Fig. 2 (b) gives a projection distortion
Example, wherein region distortion become much larger along in u coordinate positive direction.
(2) linear relationship of H.After coordinate system transformation, find out from formula 5, if u=u0It is that a fixation is normal
Amount, then Hx(u0, v) and Hy(u0, v) linear function for v, has
(3) ruled surface.Ruled surface parametric form is described as s (u, v)=p (u)+v r (u), Hx(u, v) and Hy(u,
V) form of f (u) v+g (u) can be written as, their figure is ruled surface.
To the projective transformation matrix obtainingMake full use of the new projection of the attribute construction after coordinate transform-similar
Conversion.So that new conversion is projective transformation in the overlapping region of image, it is similarity transformation in Non-overlapping Domain.So that in overlap
There is high matching precision in region, keeps by similarity transformation the visual field of image to reduce image in the image of Non-overlapping Domain and loses
Very.
Build new projection-similarity transformation
First coordinate transform is carried out to original projective transformation matrix, make original coordinate system (x, y) transform to new coordinate system
(u, v).And with a line u=u1By plane space R2It is divided into two half spaces：RH=(u, v) | u≤u1And RL={ (u, v)
| u ＞ u1}.
To (u, v) ∈ RHPoint, u value is little, carries out projective transformation using original transform H.To (u, v) ∈ RL, because H is in RL
On can produce big deformation distortion, H is substituted using similarity transformation S.
Projection-similarity transformation function the w proposing must be continuous, otherwise, along RHAnd RLBorder u=u1Will produce
Raw obvious crack.In order to construct continuous transforming function transformation function, it is required that in cut-off rule u=u1Upper all of point (u, v) meets S
(u1, v)=H (u1, v).H(u1, v) be v linear function, then similarity transformation S be uniquely defined as：
Projection-similarity transformation the function obtaining is continuous (C0Continuously).But online u=u1Can there is bending in place.To make
Spliced image is smooth change it is desirable to the transforming function transformation function of conversion is C1Continuously.
Transforming function transformation function w is optimized so as to be continuously differentiable (C1) function makes information is smooth variation.By introducing
Buffering area, makes the projective transformation of overlapping region be smoothly transitted into similarity transformation so that transforming function transformation function w is continuous in this region
(C that can be micro-1Continuously).
By RLIt is further divided into transitional region RT=(u, v) | u1＜ u ＜ u2And similarity transformation region RS=(u, v) |
u2＜ u }.Transforming function transformation function is：
Wherein, u1And u2For the parameter in transforming function transformation function, T (u, v) is gradually to transform to similarity transformation S from projective transformation H (u, v)
The function of (u, v).
H (u, v) and S (u, v) can be expressed as the form of f (u) v+g (u), and the transforming function transformation function w in order to ensure to obtain is even
Continuous (C that can be micro-1) function, then T (u, v) form be：
Transforming function transformation function w.After H (u, v) and the determination of S (u, v) form, so that transforming function transformation function w is continuously differentiable (C1)
Function, then transforming function transformation function w form be：
Wherein
Determine parameter u1And u2.The purpose of the transforming function transformation function proposing is to keep every width figure as far as possible while aliging image
The viewpoint of picture, it is desirable that each image carries out similarity transformation as far as possible in splicing.
To each image I to be splicediOne cost function E of distributioniFor metric transformation function wiWith immediate similar change
The Frobenius norm of the difference changed.
When image is to splicing, total cost function is E (u1, u2)=E1(u1, u2)+E2(u1, u2), be parameter be u1With
u2Nonlinear function.
Parameter u1And u2Solution.By to (u1, u2) parameter space carry out canonical sampling total cost function is carried out
Optimization is processed, and estimates the cost of each sampled point and chooses the point (u with minimum cost1, u2) it is optimized parameter.
Using transforming function transformation function w, conversion is executed to image.To the n-th width image InLine translation is entered by the transforming function transformation function w obtaining, figure
As IiByEnter line translation, obtain geometrically splicing result.
Image co-registration is carried out using linear weighted function and obtains final image mosaic result
Fig. 3 gives and utilizes projective transformation, and AutoStitch and the projection-similarity transformation proposing to two width sizes are
The result that 6000*4000 unmanned aerial vehicle remote sensing image is spliced.Projection-similarity transformation that result display proposes is reducing projection mistake
The viewpoint of each image very can be kept simultaneously
Fig. 4 is splicing result that is little to overlapping region and there is the inconsistent image in visual field.Unmanned aerial vehicle remote sensing image is adjacent
Image can have that degree of overlapping is low, and to produce visual field inconsistent due to platform unstable.The weight of three images to be spliced of in figure
Folded region is less and visual field is inconsistent.Overlapping region in Fig. 4 (a) and Fig. 4 (b) is red-label part, Fig. 4 (b) and Fig. 4
C the overlapping region in () is yellow flag part.Fig. 4 (a) is located at the upper left corner imaging of image for camera, and Fig. 4 (b) phase seat in the plane
In the lower left corner imaging of image, Fig. 4 (c) faces imaging contexts for ideal.
Fig. 5 is that the conversion proposing carries out splicing result to image sequence.From experimental result it can be seen that propose projection-
Similarity transformation can be effectively little to image overlapping region to be spliced in unmanned aerial vehicle remote sensing image, and there is the inconsistent situation in visual field and carry out
Splicing, has high matching precision and good visual field consistent features.Go out from the experimental results, spliced using the method proposing
Image have preferable alignment accuracy and splice after image f iotaeld-of-view consistent so that spliced image more meets real scene.
Claims (4)
1. a kind of unmanned aerial vehicle remote sensing images joining method based on improved projection-similarity transformation, specifically includes following steps：
Step 1：Extract image sequence scale invariant feature-SIFT feature；
Step 2：Carry out Feature Matching using the SIFT feature obtaining, picked using RANSAC algorithm during characteristic matching
Except Mismatching point pair；
Step 3：After mutually corresponding characteristic point completes coupling between image, determine corresponding geometric transformation model, i.e. projection becomes
Change H；
Step 4：Coordinate system transformation is carried out to projective transformation, original (x, y) coordinate system transformation is made to new by rotation transformation
(u, v) coordinate system；
Step 5：The property of projective transformation under the coordinate system after conversion is analyzed, and with two lines U1 and U2, space is divided
For RH, RTAnd RSThree sub-spaces；
Step 6：Space dividing line parameter U1 and U2 are solved, and determines final projection-similarity transformation w；
Step 7：Using projection-similarity transformation w, conversion is executed to image, carry out geometry splicing；
Step 8：Carry out visual fusion using linear weighted function.
2. a kind of unmanned aerial vehicle remote sensing images joining method based on improved projection-similarity transformation according to claim 1,
It is characterized in that, described improved projection-similarity transformation is the combination of projective transformation and similarity transformation, to the characteristic point obtaining
Mated, determined projective transformation H；Coordinate transform is carried out to projective transformation H, is divided into different subspace RH, RTAnd RS,
It is respectively projective transformation, buffering area and similarity transformation space.Will be smoothed for the projective transformation of overlapping region by introducing buffering area
Cross similarity transformation so that the conversion obtaining has the property of projective transformation and similarity transformation simultaneously, and then make splicing image
There is high matching precision and the visual field of image can be kept to reduce image distortion.
3. a kind of unmanned aerial vehicle remote sensing images joining method based on improved projection-similarity transformation according to claim 2,
In order to ensure projective transformation seamlessly transitting to similarity transformation in buffering area, described projection-similarity transformation function is in buffering area
It is continuously differentiable.
4. projection-the similarity transformation according to claim 1 and 2 carries out unmanned aerial vehicle remote sensing images splicing, to different regions
Image mosaic is carried out using different conversion.Projective transformation space is mated using projective transformation, to similarity transformation space
Mated using projective transformation, buffering area is mated so that spliced image is using projection-similar transition conversion
Smooth change, reduce image fault.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610795444.8A CN106447601B (en) | 2016-08-31 | 2016-08-31 | Unmanned aerial vehicle remote sensing image splicing method based on projection-similarity transformation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610795444.8A CN106447601B (en) | 2016-08-31 | 2016-08-31 | Unmanned aerial vehicle remote sensing image splicing method based on projection-similarity transformation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106447601A true CN106447601A (en) | 2017-02-22 |
| CN106447601B CN106447601B (en) | 2020-07-24 |
Family
ID=58164973
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610795444.8A Active CN106447601B (en) | 2016-08-31 | 2016-08-31 | Unmanned aerial vehicle remote sensing image splicing method based on projection-similarity transformation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106447601B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018170857A1 (en) * | 2017-03-23 | 2018-09-27 | 深圳市大疆创新科技有限公司 | Method for image fusion and unmanned aerial vehicle |
| CN108648145A (en) * | 2018-04-28 | 2018-10-12 | 北京东软医疗设备有限公司 | Image split-joint method and device |
| CN108876861A (en) * | 2018-05-25 | 2018-11-23 | 中国空间技术研究院 | A kind of solid matching method of objects outside Earth rover |
| CN109118429A (en) * | 2018-08-02 | 2019-01-01 | 武汉大学 | A kind of medium-wave infrared-visible light multispectral image rapid generation |
| CN110689485A (en) * | 2019-10-14 | 2020-01-14 | 中国空气动力研究与发展中心超高速空气动力研究所 | SIFT image splicing method applied to infrared nondestructive testing of large pressure container |
| CN111860205A (en) * | 2020-06-29 | 2020-10-30 | 成都数之联科技有限公司 | Forest fire evaluation method based on multi-source remote sensing image and grid and storage medium |
| CN112508071A (en) * | 2020-11-30 | 2021-03-16 | 中国公路工程咨询集团有限公司 | BIM-based bridge disease marking method and device |
| CN112750075A (en) * | 2019-10-31 | 2021-05-04 | 中国科学院长春光学精密机械与物理研究所 | Low-altitude remote sensing image splicing method and device |
| CN113486697A (en) * | 2021-04-16 | 2021-10-08 | 成都思晗科技股份有限公司 | Forest smoke and fire monitoring method based on space-based multi-modal image fusion |
| CN113837270A (en) * | 2021-09-18 | 2021-12-24 | 广东人工智能与先进计算研究院 | Target identification method, device, equipment and storage medium |
| CN114742869A (en) * | 2022-06-15 | 2022-07-12 | 西安交通大学医学院第一附属医院 | Brain neurosurgery registration method based on pattern recognition and electronic equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020040208A1 (en) * | 2000-10-04 | 2002-04-04 | Flaherty J. Christopher | Data collection assembly for patient infusion system |
| CN101442619A (en) * | 2008-12-25 | 2009-05-27 | 武汉大学 | Method for splicing non-control point image |
| US20100141273A1 (en) * | 2008-12-09 | 2010-06-10 | General Electric Company | Systems and methods for using ferrite alignment keys in wireless remote sensors |
| CN103646424A (en) * | 2013-11-26 | 2014-03-19 | 北京空间机电研究所 | Aerial seamless virtual roaming system constructing method |
| CN104851094A (en) * | 2015-05-14 | 2015-08-19 | 西安电子科技大学 | Improved method of RGB-D-based SLAM algorithm |
| CN105243653A (en) * | 2015-11-05 | 2016-01-13 | 北京航天泰坦科技股份有限公司 | Fast mosaic technology of remote sensing image of unmanned aerial vehicle on the basis of dynamic matching |
| CN105931185A (en) * | 2016-04-20 | 2016-09-07 | 中国矿业大学 | Automatic splicing method of multiple view angle image |
-
2016
- 2016-08-31 CN CN201610795444.8A patent/CN106447601B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020040208A1 (en) * | 2000-10-04 | 2002-04-04 | Flaherty J. Christopher | Data collection assembly for patient infusion system |
| US20100141273A1 (en) * | 2008-12-09 | 2010-06-10 | General Electric Company | Systems and methods for using ferrite alignment keys in wireless remote sensors |
| CN101442619A (en) * | 2008-12-25 | 2009-05-27 | 武汉大学 | Method for splicing non-control point image |
| CN103646424A (en) * | 2013-11-26 | 2014-03-19 | 北京空间机电研究所 | Aerial seamless virtual roaming system constructing method |
| CN104851094A (en) * | 2015-05-14 | 2015-08-19 | 西安电子科技大学 | Improved method of RGB-D-based SLAM algorithm |
| CN105243653A (en) * | 2015-11-05 | 2016-01-13 | 北京航天泰坦科技股份有限公司 | Fast mosaic technology of remote sensing image of unmanned aerial vehicle on the basis of dynamic matching |
| CN105931185A (en) * | 2016-04-20 | 2016-09-07 | 中国矿业大学 | Automatic splicing method of multiple view angle image |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018170857A1 (en) * | 2017-03-23 | 2018-09-27 | 深圳市大疆创新科技有限公司 | Method for image fusion and unmanned aerial vehicle |
| CN108648145A (en) * | 2018-04-28 | 2018-10-12 | 北京东软医疗设备有限公司 | Image split-joint method and device |
| CN108876861A (en) * | 2018-05-25 | 2018-11-23 | 中国空间技术研究院 | A kind of solid matching method of objects outside Earth rover |
| CN109118429A (en) * | 2018-08-02 | 2019-01-01 | 武汉大学 | A kind of medium-wave infrared-visible light multispectral image rapid generation |
| CN109118429B (en) * | 2018-08-02 | 2023-04-25 | 武汉大学 | Method for rapidly generating intermediate wave infrared-visible light multispectral image |
| CN110689485A (en) * | 2019-10-14 | 2020-01-14 | 中国空气动力研究与发展中心超高速空气动力研究所 | SIFT image splicing method applied to infrared nondestructive testing of large pressure container |
| CN112750075A (en) * | 2019-10-31 | 2021-05-04 | 中国科学院长春光学精密机械与物理研究所 | Low-altitude remote sensing image splicing method and device |
| CN111860205A (en) * | 2020-06-29 | 2020-10-30 | 成都数之联科技有限公司 | Forest fire evaluation method based on multi-source remote sensing image and grid and storage medium |
| CN111860205B (en) * | 2020-06-29 | 2024-03-19 | 成都数之联科技股份有限公司 | Forest fire evaluation method based on multisource remote sensing images and grids and storage medium |
| CN112508071B (en) * | 2020-11-30 | 2023-04-18 | 中国公路工程咨询集团有限公司 | BIM-based bridge disease marking method and device |
| CN112508071A (en) * | 2020-11-30 | 2021-03-16 | 中国公路工程咨询集团有限公司 | BIM-based bridge disease marking method and device |
| CN113486697A (en) * | 2021-04-16 | 2021-10-08 | 成都思晗科技股份有限公司 | Forest smoke and fire monitoring method based on space-based multi-modal image fusion |
| CN113486697B (en) * | 2021-04-16 | 2024-02-13 | 成都思晗科技股份有限公司 | Forest smoke and fire monitoring method based on space-based multimode image fusion |
| CN113837270B (en) * | 2021-09-18 | 2022-08-30 | 广东人工智能与先进计算研究院 | Target identification method, device, equipment and storage medium |
| CN113837270A (en) * | 2021-09-18 | 2021-12-24 | 广东人工智能与先进计算研究院 | Target identification method, device, equipment and storage medium |
| CN114742869B (en) * | 2022-06-15 | 2022-08-16 | 西安交通大学医学院第一附属医院 | Brain neurosurgery registration method based on pattern recognition and electronic equipment |
| CN114742869A (en) * | 2022-06-15 | 2022-07-12 | 西安交通大学医学院第一附属医院 | Brain neurosurgery registration method based on pattern recognition and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106447601B (en) | 2020-07-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106447601A (en) | Unmanned aerial vehicle remote image mosaicing method based on projection-similarity transformation | |
| CN110211043B (en) | Registration method based on grid optimization for panoramic image stitching | |
| Liu et al. | Multiview geometry for texture mapping 2d images onto 3d range data | |
| CN104077760A (en) | Rapid splicing system for aerial photogrammetry and implementing method thereof | |
| Urban et al. | Finding a good feature detector-descriptor combination for the 2D keypoint-based registration of TLS point clouds | |
| CN107818598B (en) | Three-dimensional point cloud map fusion method based on visual correction | |
| CN104616247B (en) | A kind of method for map splicing of being taken photo by plane based on super-pixel SIFT | |
| Lin et al. | Cylindrical panoramic image stitching method based on multi-cameras | |
| CN116152068A (en) | Splicing method for solar panel images | |
| CN113012208A (en) | Multi-view remote sensing image registration method and system | |
| CN112862683A (en) | Adjacent image splicing method based on elastic registration and grid optimization | |
| Wan et al. | Drone image stitching using local mesh-based bundle adjustment and shape-preserving transform | |
| CN113706635B (en) | Long-focus camera calibration method based on point feature and line feature fusion | |
| Tian et al. | Aerial image mosaicking based on the 6-DoF imaging model | |
| Parmehr et al. | Automatic parameter selection for intensity-based registration of imagery to LiDAR data | |
| Lin et al. | A comprehensive review of image line segment detection and description: Taxonomies, comparisons, and challenges | |
| Wang et al. | Multi-features visual odometry for indoor mapping of UAV | |
| CN105184736B (en) | A kind of method of the image registration of narrow overlapping double-view field hyperspectral imager | |
| Aktar et al. | Robust mosaicking of maize fields from aerial imagery | |
| CN116128919A (en) | Multi-temporal image abnormal target detection method and system based on polar constraint | |
| Al-Ruzouq | Data fusion of multi-source imagery based on linear features registration | |
| CN107392878A (en) | A kind of image interfusion method of panorama camera | |
| Yu et al. | Multimodal urban remote sensing image registration via roadcross triangular feature | |
| Chen et al. | Medical image mosaic based on low-overlapping regions | |
| Zhang et al. | Image mosaic of bionic compound eye imaging system based on image overlap rate prior |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240515 Address after: 19 North Fourth Ring Road West, Haidian District, Beijing 100094 Patentee after: Aerospace Information Research Institute,Chinese Academy of Sciences Country or region after: China Address before: 100101 Institute of remote sensing earth, Chinese Academy of Sciences, Datun Road, Chaoyang District, Beijing Patentee before: Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences Country or region before: China |
|
| TR01 | Transfer of patent right |