CN109118525A - A kind of dual-band infrared image airspace method for registering - Google Patents
A kind of dual-band infrared image airspace method for registering Download PDFInfo
- Publication number
- CN109118525A CN109118525A CN201710488664.0A CN201710488664A CN109118525A CN 109118525 A CN109118525 A CN 109118525A CN 201710488664 A CN201710488664 A CN 201710488664A CN 109118525 A CN109118525 A CN 109118525A
- Authority
- CN
- China
- Prior art keywords
- image
- model
- dual
- module
- registration
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- 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
- G06T2207/10048—Infrared image
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Closed-Circuit Television Systems (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Transforming Light Signals Into Electric Signals (AREA)
- Photometry And Measurement Of Optical Pulse Characteristics (AREA)
Abstract
The invention discloses a kind of dual-band infrared image airspace method for registering, the method of the present invention constructs dual-band infrared image airspace registration arrangement first, constructed system includes: model construction module, region is chosen and characteristic point is to demarcating module, model parameter determining module, model authentication module and dual-band infrared image registration module, airspace, which is constructed, by model construction module is registrated affine Transform Model, region is chosen and characteristic point carries out target area selection and characteristic point to calibration to demarcating module, model parameter determining module determines the parameter in RST affine Transform Model, model authentication module verifies registration model, dual-band infrared image registration module is registrated dual-band infrared image.The method of the present invention can be realized the airspace registration of static dual-band infrared image, and two band image spatial features is made to be consistent, and solve the problems, such as that concern target position is inconsistent when fusing image data.This method parameter is few, and calculating process is succinct.
Description
Technical field
The present invention relates to a kind of image airspace method for registering, especially a kind of dual-band infrared image airspace method for registering.
Background technique
Domestic dual-band infrared camera or thermal imaging system mostly use two independent single band infrared sensors, in optical frames
Using the form of aperture light splitting altogether on head.It is Polaroid by Cassegrain's microscope group and scene progress, it then will using dichronic mirror
Optical path light splitting, and there is subsequent correction microscope group to carry out secondary imaging.It can be seen that Same Scene in this way on two wave bands.However this
Kind of optical lens requirement on machining accuracy is high, and adjustment technique is difficult, the result and design value that optical lens is realized often visual field size,
There are deviations on observation angle, axle center base position so as to Same Scene imaging results on airspace size, angle, phase
The features such as relationship are not quite identical, bring puzzlement to image analysis and Data Fusion.
Image registration be multiple sensors or single sensor at different conditions to Same Scene obtain image alignment,
The process of superposition.Including the alignment of geometric space global or local, the alignment of corresponding points pixel grey scale etc..Spatial alternation mathematical model has
The diversified forms such as affine transformation, polynomial transformation, transmission variation model.For the image of dual-band infrared camera, due to obtaining
The image spectrum taken is all in infrared band, and the same set of optical lens of use observes scene, time consistency, and the purpose of registration is
In order to correct optical lens bring deviation on airspace on two wave bands.
Summary of the invention
It is an object of that present invention to provide a kind of dual-band infrared image airspace method for registering, solve dual-band infrared camera light
Learn camera lens visual field difference in size, optical axis deviation, observation benchmark difference bring dual-band image spatial feature on two wave bands
Inconsistent problem.
A kind of specific steps of dual-band infrared image airspace method for registering are as follows:
The first step constructs dual-band infrared image airspace registration arrangement
Dual-band infrared image airspace registration arrangement, comprising: model construction module, region choose and characteristic point is to calibration mold
Block, model parameter determining module, model authentication module and dual-band infrared image registration module.It is described:
The function of model construction module are as follows: building airspace is registrated affine Transform Model；
Region is chosen and characteristic point is to the function of demarcating module are as follows: carries out target area selection and characteristic point to calibration；
The function of model parameter determining module are as follows: determine the parameter in RST affine Transform Model；
The function of model authentication module are as follows: registration model is verified；
The function of dual-band infrared image registration module are as follows: dual-band infrared image is registrated.
Second step model construction module constructs airspace and is registrated affine Transform Model
After two-waveband infrared optical camera lens is Polaroid, subsequent microscope group carries out spherical aberration and chromatic aberration correction, figure to infrared image
As having the preferable linearity, therefore airspace registration affine Transform Model building is suitable for the RST affine transformation mould of the linearity
Type, i.e. rotation-scaling-translation affine Transform Model carry out RST affine transformation, the table of RST affine Transform Model to infrared image
Up to formula are as follows:
In formula:
For the image pixel coordinates after registration；
S is image scaling ratio, S=[SX,SY]T, SXFor orientation pantograph ratio, SYFor pitching pantograph ratio；
For image rotation matrix；
θ is rotation angle；
For image pixel coordinates subject to registration；
For pixel coordinate offset.
Third step region is chosen and characteristic point carries out target area selection and characteristic point to calibration to demarcating module
Region choose and characteristic point to demarcating module to the infrared image of two wave bands, be that origin is established directly using picture centre
Angular coordinate system judges that target area falls into the quadrant position in coordinate system, so that it is determined that the area image is in optical lens visual field
In angle position, complete regional choice.
Region is chosen and characteristic point obtains characteristic point pair: P and P', Q and Q' to demarcating module.
4th step model parameter determining module determines the parameter in RST affine Transform Model
Model parameter determining module determines the parameter in RST affine Transform Model: scaling S, rotation angle, θ and translation
It measures [Δ X, Δ Y]T。
Scaling S is determined by two wave band visual field ratios of optical lens, if benchmark image wave band optical lens visual field is
Orientation εX1, pitching εY1；Image band visual field subject to registration is orientation εX2, pitching εY2；The then scaling in orientation and pitching point
It Wei not SX=εX1/εX2,SY=εY1/εY2。
Rotation angle, θ determines that coordinate is pixel to (P, P') and (Q, Q') coordinate in the picture by characteristic point
Ranks value in the picture.Model parameter determining module obtains the slope of straight line PQ and P'Q' respectively:
In formula:
X1、Y1It is characterized point P pixel coordinate；
X1'、Y1' it is characterized point P' pixel coordinate；
X2、Y2It is characterized point Q pixel coordinate；
X2'、Y2' it is characterized point Q' pixel coordinate.
Rotate angle, θ=θ2-θ1, wherein being counterclockwise positive.
In dual-band optical lens, input path is divided into two optical paths after Polaroid, by spectroscope of turning back, point
Not carry out secondary imaging, there is a certain error for 45 ° of established angles of spectroscope when adjustment, the verticality of subsequent reflection mirror and datum level
Or the depth of parallelism, there is also certain error range, these errors bring two optical paths not coaxial, so that the image of two wave bands exists
Relative displacement, i.e., the translational movement [Δ X, Δ Y] in modelT.Translational movement is determined by the coordinate difference mean value of characteristic point pair:
In formula:
Δ X' is characterized a little to orientation translational movement mean value；
Δ Y' is characterized a little to pitching translating amount mean value.
It is influenced by rotation angle and pantograph ratio, RST affine Transform Model translational movement are as follows:
Δ X=Δ X'cos θ SX, Δ Y=Δ Y'sin θ SY (4)
5th step model authentication module verifies registration model
Model authentication module carries out corresponding dual-band infrared image to registration model parameter using two characteristic points
Verifying:
Selecting first characteristic point is benchmark image to the wherein piece image in two wave bands of dual-band infrared image, separately
One band image is image subject to registration.Have determined three model parameters are substituted into model, are to sit with image center
Origin is marked, rectangular coordinate system is established, the ranks number of each pixel are the coordinate (X, Y) of the pixel in the picture in image, will
Each pixel coordinate is substituted into RST affine Transform Model and is calculated in image, and is rounded up to calculated result, is obtained
Coordinate (X', Y') of each pixel in new images is the image after being registrated according to the image that new coordinate is formed.Determine registration
Whether first characteristic point is overlapped with the coordinate of the corresponding characteristic point in its benchmark image in image afterwards.Using identical side
Formula carries out identical judgement to the dual-band infrared image of second characteristic point pair.
When the coordinate determined twice is all overlapped, decision model parameter is correct.When the coordinate that any time determines is not overlapped,
Then re-scale characteristic point pair.
6th step dual-band infrared image registration module is registrated dual-band infrared image
For needing to be registrated image, dual-band infrared image registration module determines benchmark image and image subject to registration first,
Then the pixel coordinate of image attention region subject to registration or entire image is substituted into RST affine Transform Model and is calculated, obtained
To coordinate of each pixel in new images, thus the image after being registrated.
So far, the airspace registration of dual-band infrared image is completed.
More preferably, wherein region is chosen in third step and characteristic point obtains demarcating module by the way of optical imagery calibration
It takes characteristic point pair: point source target being imaged on testing stand with same set of dual-band infrared camera, and point source target is made to mark on a map
As falling into target area, the dual-band infrared image is recorded, first characteristic point is obtained to P and P', then adjusts optical lens
Angle makes point source target be marked on the imaging of another angle position of the region, records corresponding dual-band infrared image, obtains second
Characteristic point is to Q and Q'.
A kind of dual-band infrared image airspace registration arrangement, comprising: model construction module, region choose and characteristic point is to mark
Cover half block, model parameter determining module, model authentication module and dual-band infrared image registration module；Wherein model construction module
It constructs airspace and is registrated affine Transform Model；Region is chosen and characteristic point carries out target area selection and characteristic point pair to demarcating module
Calibration；Model parameter determining module determines three parameters in RST affine Transform Model: scaling S, rotation angle, θ peace
Shifting amount [Δ X, Δ Y]T；Model authentication module verifies registration model；Dual-band infrared image registration module is to two waveband
Infrared image is registrated.
The method of the present invention can be realized the airspace registration of static dual-band infrared image, keep two band image airspaces special
Sign is consistent, and solves the problems, such as that concern target position is inconsistent when fusing image data.This method parameter is few, calculating process letter
It is clean.
Specific embodiment
A kind of specific steps of dual-band infrared image airspace method for registering are as follows:
The first step constructs dual-band infrared image airspace registration arrangement
Dual-band infrared image airspace registration arrangement, comprising: model construction module, region choose and characteristic point is to calibration mold
Block, model parameter determining module, model authentication module and dual-band infrared image registration module.It is described:
The function of model construction module are as follows: building airspace is registrated affine Transform Model；
Region is chosen and characteristic point is to the function of demarcating module are as follows: carries out target area selection and characteristic point to calibration；
The function of model parameter determining module are as follows: determine the parameter in RST affine Transform Model；
The function of model authentication module are as follows: registration model is verified；
The function of dual-band infrared image registration module are as follows: dual-band infrared image is registrated.
Second step model construction module constructs airspace and is registrated affine Transform Model
After two-waveband infrared optical camera lens is Polaroid, subsequent microscope group carries out spherical aberration and chromatic aberration correction, figure to infrared image
As having the preferable linearity, therefore airspace registration affine Transform Model building is suitable for the RST affine transformation mould of the linearity
Type, i.e. rotation-scaling-translation affine Transform Model carry out RST affine transformation, the table of RST affine Transform Model to infrared image
Up to formula are as follows:
In formula:
For the image pixel coordinates after registration；
S is image scaling ratio, S=[SX,SY]T, SXFor orientation pantograph ratio, SYFor pitching pantograph ratio；
For image rotation matrix；
θ is rotation angle；
For image pixel coordinates subject to registration；
For pixel coordinate offset.
Third step region is chosen and characteristic point carries out target area selection and characteristic point to calibration to demarcating module
Region choose and characteristic point to demarcating module to the infrared image of two wave bands, be that origin is established directly using picture centre
Angular coordinate system judges that target area falls into the quadrant position in coordinate system, so that it is determined that the area image is in optical lens visual field
In angle position, complete regional choice.
Region is chosen and characteristic point demarcating module is demarcated using optical imagery by the way of obtain characteristic point pair: with same set of
Dual-band infrared camera is imaged point source target on testing stand, and point source target image is made to fall into target area, record
The dual-band infrared image obtains first characteristic point to P and P', then adjusts optical lens angle, point source target is made to be marked on this
The imaging of another angle position of region, records corresponding dual-band infrared image, obtains second characteristic point to Q and Q'.
4th step model parameter determining module determines the parameter in RST affine Transform Model
Model parameter determining module determines the parameter in RST affine Transform Model: scaling S, rotation angle, θ and translation
It measures [Δ X, Δ Y]T。
Scaling S is determined by two wave band visual field ratios of optical lens, if benchmark image wave band optical lens visual field is
Orientation εX1, pitching εY1；Image band visual field subject to registration is orientation εX2, pitching εY2；The then scaling in orientation and pitching point
It Wei not SX=εX1/εX2,SY=εY1/εY2。
Rotation angle, θ determines that coordinate is pixel to (P, P') and (Q, Q') coordinate in the picture by characteristic point
Ranks value in the picture.Model parameter determining module obtains the slope of straight line PQ and P'Q' respectively:
In formula:
X1、Y1It is characterized point P pixel coordinate；
X1'、Y1' it is characterized point P' pixel coordinate；
X2、Y2It is characterized point Q pixel coordinate；
X2'、Y2' it is characterized point Q' pixel coordinate.
Rotate angle, θ=θ2-θ1, wherein being counterclockwise positive.
In dual-band optical lens, input path is divided into two optical paths after Polaroid, by spectroscope of turning back, point
Not carry out secondary imaging, there is a certain error for 45 ° of established angles of spectroscope when adjustment, the verticality of subsequent reflection mirror and datum level
Or the depth of parallelism, there is also certain error range, these errors bring two optical paths not coaxial, so that the image of two wave bands exists
Relative displacement, i.e., the translational movement [Δ X, Δ Y] in modelT.Translational movement is determined by the coordinate difference mean value of characteristic point pair:
In formula:
Δ X' is characterized a little to orientation translational movement mean value；
Δ Y' is characterized a little to pitching translating amount mean value.
It is influenced by rotation angle and pantograph ratio, RST affine Transform Model translational movement are as follows:
Δ X=Δ X'cos θ SX, Δ Y=Δ Y'sin θ SY (4)
5th step model authentication module verifies registration model
Model authentication module carries out corresponding dual-band infrared image to registration model parameter using two characteristic points
Verifying:
Selecting first characteristic point is benchmark image to the wherein piece image in two wave bands of dual-band infrared image, separately
One band image is image subject to registration.Have determined three model parameters are substituted into model, are to sit with image center
Origin is marked, rectangular coordinate system is established, the ranks number of each pixel are the coordinate (X, Y) of the pixel in the picture in image, will
Each pixel coordinate is substituted into RST affine Transform Model and is calculated in image, and is rounded up to calculated result, is obtained
Coordinate (X', Y') of each pixel in new images is the image after being registrated according to the image that new coordinate is formed.Determine registration
Whether first characteristic point is overlapped with the coordinate of the corresponding characteristic point in its benchmark image in image afterwards.Using identical side
Formula carries out identical judgement to the dual-band infrared image of second characteristic point pair.
When the coordinate determined twice is all overlapped, decision model parameter is correct.When the coordinate that any time determines is not overlapped,
Then re-scale characteristic point pair.
6th step dual-band infrared image registration module is registrated dual-band infrared image
For needing to be registrated image, dual-band infrared image registration module determines benchmark image and image subject to registration first,
Then the pixel coordinate of image attention region subject to registration or entire image is substituted into RST affine Transform Model and is calculated, obtained
To coordinate of each pixel in new images, thus the image after being registrated.
So far, the airspace registration of dual-band infrared image is completed.
Claims (3)
1. a kind of dual-band infrared image airspace method for registering, it is characterised in that specific steps are as follows:
The first step constructs dual-band infrared image airspace registration arrangement
Dual-band infrared image airspace registration arrangement, comprising: model construction module, region choose and characteristic point to demarcating module,
Model parameter determining module, model authentication module and dual-band infrared image registration module；It is described:
The function of model construction module are as follows: building airspace is registrated affine Transform Model；
Region is chosen and characteristic point is to the function of demarcating module are as follows: carries out target area selection and characteristic point to calibration；
The function of model parameter determining module are as follows: determine the parameter in RST affine Transform Model；
The function of model authentication module are as follows: registration model is verified；
The function of dual-band infrared image registration module are as follows: dual-band infrared image is registrated；
Second step model construction module constructs airspace and is registrated affine Transform Model
After two-waveband infrared optical camera lens is Polaroid, subsequent microscope group carries out spherical aberration and chromatic aberration correction, image tool to infrared image
There is the preferable linearity, therefore airspace registration affine Transform Model building is suitable for the RST affine Transform Model of the linearity, i.e.,
Rotation-scaling-translation affine Transform Model carries out RST affine transformation, the expression formula of RST affine Transform Model to infrared image
Are as follows:
In formula:
For the image pixel coordinates after registration；
S is image scaling ratio, S=[SX,SY]T, SXFor orientation pantograph ratio, SYFor pitching pantograph ratio；
For image rotation matrix；
θ is rotation angle；
For image pixel coordinates subject to registration；
For pixel coordinate offset；
Third step region is chosen and characteristic point carries out target area selection and characteristic point to calibration to demarcating module
Region choose and characteristic point to demarcating module to the infrared image of two wave bands, be that origin establishes right angle seat using picture centre
Mark system, judges that target area falls into the quadrant position in coordinate system, so that it is determined that the area image is in optical lens visual field
Regional choice is completed in angle position；
Region is chosen and characteristic point obtains characteristic point pair: P and P', Q and Q' to demarcating module；
4th step model parameter determining module determines the parameter in RST affine Transform Model
Model parameter determining module determines the parameter in RST affine Transform Model: scaling S, rotation angle, θ and translational movement
[ΔX,ΔY]T；
Scaling S is determined by two wave band visual field ratios of optical lens, if benchmark image wave band optical lens visual field is orientation
εX1, pitching εY1；Image band visual field subject to registration is orientation εX2, pitching εY2；Then the scaling in orientation and pitching is respectively
SX=εX1/εX2,SY=εY1/εY2；
Rotation angle, θ determines that coordinate is being schemed for pixel by characteristic point to (P, P') and (Q, Q') coordinate in the picture
Ranks value as in；Model parameter determining module obtains the slope of straight line PQ and P'Q' respectively:
In formula:
X1、Y1It is characterized point P pixel coordinate；
X1'、Y1' it is characterized point P' pixel coordinate；
X2、Y2It is characterized point Q pixel coordinate；
X2'、Y2' it is characterized point Q' pixel coordinate；
Rotate angle, θ=θ2-θ1, wherein being counterclockwise positive；
In dual-band optical lens, input path is divided into two optical paths after Polaroid, by spectroscope of turning back, respectively into
Row secondary imaging, there is a certain error for 45 ° of established angles of spectroscope when adjustment, the verticality or flat of subsequent reflection mirror and datum level
For row degree there is also certain error range, these errors bring two optical paths not coaxial, so that the image of two wave bands exists relatively
Displacement, i.e., the translational movement [Δ X, Δ Y] in modelT.Translational movement is determined by the coordinate difference mean value of characteristic point pair:
In formula:
Δ X' is characterized a little to orientation translational movement mean value；
Δ Y' is characterized a little to pitching translating amount mean value；
It is influenced by rotation angle and pantograph ratio, RST affine Transform Model translational movement are as follows:
Δ X=Δ X'cos θ SX, Δ Y=Δ Y'sin θ SY (4)
5th step model authentication module verifies registration model
Model authentication module verifies corresponding dual-band infrared image to registration model parameter using two characteristic points:
Selecting first characteristic point is benchmark image to the wherein piece image in two wave bands of dual-band infrared image, another
Band image is image subject to registration.Have determined three model parameters are substituted into model, using image center as coordinate original
Point establishes rectangular coordinate system, and the ranks number of each pixel are the coordinate (X, Y) of the pixel in the picture in image, by image
In each pixel coordinate substitute into RST affine Transform Model and calculated, and round up, obtain each to calculated result
Coordinate (X', Y') of the pixel in new images is the image after being registrated according to the image that new coordinate is formed；After determining registration
Whether first characteristic point is overlapped with the coordinate of the corresponding characteristic point in its benchmark image in image.It is right in a like fashion to adopt
The dual-band infrared image of second characteristic point pair carries out identical judgement；
When the coordinate determined twice is all overlapped, decision model parameter is correct；When the coordinate that any time determines is not overlapped, then weigh
New feature point for calibration pair.
6th step dual-band infrared image registration module is registrated dual-band infrared image
For needing to be registrated image, dual-band infrared image registration module determines benchmark image and image subject to registration first, then
The pixel coordinate of image attention region subject to registration or entire image is substituted into RST affine Transform Model and is calculated, is obtained every
Coordinate of a pixel in new images, thus the image after being registrated；
So far, the airspace registration of dual-band infrared image is completed.
2. dual-band infrared image as described in claim 1 airspace method for registering, it is characterised in that: region in the third step
It chooses and characteristic point obtains characteristic point pair by the way of optical imagery calibration to demarcating module: with same set of dual-band infrared phase
Machine is imaged point source target on testing stand, and point source target image is made to fall into target area, records the dual-band infrared
Image obtains first characteristic point to P and P', then adjusts optical lens angle, point source target is made to be marked on another angle of the region
Position imaging is spent, corresponding dual-band infrared image is recorded, obtains second characteristic point to Q and Q'.
3. a kind of dual-band infrared image airspace registration arrangement, characterized by comprising: model construction module, region choose and spy
Sign point is to demarcating module, model parameter determining module, model authentication module and dual-band infrared image registration module；Wherein model
It constructs module building airspace and is registrated affine Transform Model；Region choose and characteristic point to demarcating module carry out target area selection and
Characteristic point is to calibration；Model parameter determining module determines three parameters in RST affine Transform Model: scaling S, rotation angle
Spend θ and translational movement [Δ X, Δ Y]T；Model authentication module verifies registration model；Dual-band infrared image registration module pair
Dual-band infrared image is registrated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710488664.0A CN109118525B (en) | 2017-06-23 | 2017-06-23 | Dual-waveband infrared image spatial domain registration method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710488664.0A CN109118525B (en) | 2017-06-23 | 2017-06-23 | Dual-waveband infrared image spatial domain registration method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109118525A true CN109118525A (en) | 2019-01-01 |
| CN109118525B CN109118525B (en) | 2021-08-13 |
Family
ID=64732326
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710488664.0A Active CN109118525B (en) | 2017-06-23 | 2017-06-23 | Dual-waveband infrared image spatial domain registration method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109118525B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111047699A (en) * | 2019-12-18 | 2020-04-21 | 中国石油工程建设有限公司华北分公司 | Seabed detection image display method, device, equipment and storage medium |
| CN111122447A (en) * | 2019-11-25 | 2020-05-08 | 北华航天工业学院 | Multispectral wave band correction method for unmanned aerial vehicle |
| CN111311658A (en) * | 2020-04-02 | 2020-06-19 | 烟台艾睿光电科技有限公司 | Image registration method of dual-light imaging system and related device |
| CN112509035A (en) * | 2020-11-26 | 2021-03-16 | 江苏集萃未来城市应用技术研究所有限公司 | Double-lens image pixel point matching method for optical lens and thermal imaging lens |
| CN113409372A (en) * | 2021-06-25 | 2021-09-17 | 浙江商汤科技开发有限公司 | Image registration method, related device, equipment and storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016030305A1 (en) * | 2014-08-29 | 2016-03-03 | Thomson Licensing | Method and device for registering an image to a model |
-
2017
- 2017-06-23 CN CN201710488664.0A patent/CN109118525B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016030305A1 (en) * | 2014-08-29 | 2016-03-03 | Thomson Licensing | Method and device for registering an image to a model |
Non-Patent Citations (1)
| Title |
|---|
| 胡津翔 等: "《基于仿射变换的双波段红外图像配准算法研究》", 《科技广场》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111122447A (en) * | 2019-11-25 | 2020-05-08 | 北华航天工业学院 | Multispectral wave band correction method for unmanned aerial vehicle |
| CN111122447B (en) * | 2019-11-25 | 2022-03-25 | 北华航天工业学院 | Multispectral wave band correction method for unmanned aerial vehicle |
| CN111047699A (en) * | 2019-12-18 | 2020-04-21 | 中国石油工程建设有限公司华北分公司 | Seabed detection image display method, device, equipment and storage medium |
| CN111311658A (en) * | 2020-04-02 | 2020-06-19 | 烟台艾睿光电科技有限公司 | Image registration method of dual-light imaging system and related device |
| CN111311658B (en) * | 2020-04-02 | 2023-11-07 | 烟台艾睿光电科技有限公司 | Image registration method and related device for dual-light imaging system |
| CN112509035A (en) * | 2020-11-26 | 2021-03-16 | 江苏集萃未来城市应用技术研究所有限公司 | Double-lens image pixel point matching method for optical lens and thermal imaging lens |
| CN113409372A (en) * | 2021-06-25 | 2021-09-17 | 浙江商汤科技开发有限公司 | Image registration method, related device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109118525B (en) | 2021-08-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109118525A (en) | A kind of dual-band infrared image airspace method for registering | |
| US7479982B2 (en) | Device and method of measuring data for calibration, program for measuring data for calibration, program recording medium readable with computer, and image data processing device | |
| CN110057295B (en) | Monocular vision plane distance measuring method without image control | |
| Luhmann et al. | Sensor modelling and camera calibration for close-range photogrammetry | |
| CN105118055B (en) | Camera position amendment scaling method and system | |
| CN101136192B (en) | System and method for automated calibration and correction of display geometry and color | |
| CN104392435B (en) | Fisheye camera scaling method and caliberating device | |
| JP4224260B2 (en) | Calibration apparatus, method, result diagnosis apparatus, and calibration chart | |
| US8098958B2 (en) | Processing architecture for automatic image registration | |
| US20180122099A1 (en) | Image processing apparatus having automatic compensation function for image obtained from camera, and method thereof | |
| Zhang et al. | A universal and flexible theodolite-camera system for making accurate measurements over large volumes | |
| CN105379264A (en) | System and method for imaging device modelling and calibration | |
| CN103813151A (en) | Image processing apparatus and method, image processing system and program | |
| US9563952B2 (en) | Determination of a measurement error | |
| CN106056620A (en) | Calibration board for line laser position calibration and line laser camera measurement system calibration method | |
| JP4270949B2 (en) | Calibration chart image display device, calibration device, and calibration method | |
| US20210120194A1 (en) | Temperature measurement processing method and apparatus, and thermal imaging device | |
| CN101354796B (en) | Omnidirectional stereo vision three-dimensional rebuilding method based on Taylor series model | |
| CN111220129A (en) | Focusing measurement method with rotating holder and terminal | |
| CN109242779A (en) | A kind of construction method, device and the automatic vehicle control system of camera imaging model | |
| CN115830103A (en) | Monocular color-based transparent object positioning method and device and storage medium | |
| WO2023201578A1 (en) | Extrinsic parameter calibration method and device for monocular laser speckle projection system | |
| CN113724337A (en) | Camera dynamic external parameter calibration method and device without depending on holder angle | |
| CN113963065A (en) | Lens internal reference calibration method and device based on external reference known and electronic equipment | |
| CN112258581A (en) | On-site calibration method for multi-fish glasses head panoramic camera |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |