CN109360145A - One kind is based on vortex pulsed infrared thermal image joining method - Google Patents
One kind is based on vortex pulsed infrared thermal image joining method Download PDFInfo
- Publication number
- CN109360145A CN109360145A CN201811276563.8A CN201811276563A CN109360145A CN 109360145 A CN109360145 A CN 109360145A CN 201811276563 A CN201811276563 A CN 201811276563A CN 109360145 A CN109360145 A CN 109360145A
- Authority
- CN
- China
- Prior art keywords
- thermal
- point
- induced imagery
- induced
- imagery
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 239000011159 matrix material Substances 0.000 claims abstract description 48
- 230000009466 transformation Effects 0.000 claims abstract description 35
- 230000004927 fusion Effects 0.000 claims abstract description 10
- 238000012163 sequencing technique Methods 0.000 claims abstract description 5
- 238000012545 processing Methods 0.000 claims description 17
- 238000012360 testing method Methods 0.000 claims description 14
- 238000009499 grossing Methods 0.000 claims description 4
- 230000010354 integration Effects 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000004064 recycling Methods 0.000 claims 1
- 230000003044 adaptive effect Effects 0.000 abstract description 4
- 230000000452 restraining effect Effects 0.000 abstract description 4
- 230000007547 defect Effects 0.000 description 9
- 238000011160 research Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000001757 thermogravimetry curve Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000001931 thermography Methods 0.000 description 1
Classifications
-
- G06T3/14—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4038—Scaling the whole image or part thereof for image mosaicing, i.e. plane images composed of plane sub-images
Abstract
The invention discloses one kind based on vortex pulsed infrared thermal image joining method, contain several width thermal-induced imageries of 30% or so overlay information first, in accordance with time sequencing acquisition, then the two width thermal-induced imageries acquired at first are subjected to pretreatment and extract key point, adaptive non-maximum restraining is carried out to the key point found again, reduce some unnecessary key points, extract characteristic point, then it is purified using RANSAC algorithm to characteristic point is extracted, seek out transformation matrix, and then obtain the Transformation Relation of Projection, finally according to the Transformation Relation of Projection, with being fade-in gradually, method successively carries out image mosaic fusion to thermal-induced imagery out.
Description
Technical field
The invention belongs to technical field of image processing, more specifically, are related to a kind of based on vortex pulsed infrared thermal map
As joining method.
Background technique
Currently, the research for being directed to infrared thermal imaging defects detection is the thermal-induced imagery or thermal image with single hot spot mostly
Sequence is as research object, to obtain the defect information of thermal-induced imagery.Since thermal infrared imager visual angle is narrow, collect
Single width thermal image scene it is also smaller, this method can only get the defect information of hot spot and surrounding, cannot be to research
Object carries out comprehensive defects detection, is easy to ignore existing association situation between multiple defects, can not be to entire research object
Defect information analyzed and judged.In view of the above-mentioned problems, being introduced under conditions of guaranteeing that thermal-induced imagery resolution ratio is constant
Multiple thermal-induced imageries are stitched together by thermal-induced imagery splicing, obtain the panorama thermal-induced imagery of big visual field, more
Facilitate the defect information for obtaining research object.
Requirement of the image mosaic technology common at present to image is relatively high, needs to refer to and deposits between image and registration image
The accurate matching of image can be completed in the characteristic point being largely mutually matched.But thermal image is imaged different from natural light, and
It is to illustrate the temperature information of surface of test piece with the form of image, different colors represents different temperature, due to test specimen table
The temperature of the face overwhelming majority does not have notable difference, this leads to difference existing for the color in thermal image between the overwhelming majority very
It is small, when splicing using traditional images joining method to thermal image, since characteristic point is less, the splicing of thermal image is caused to exist
It is difficult.Therefore, the accuracy for introducing new method raising thermal image splicing is very important.
Summary of the invention
It is an object of the invention to overcome the deficiencies of the prior art and provide one kind based on vortex pulsed infrared thermal image splicing
Method completes the accurate splicing of thermal-induced imagery under the premise of guaranteeing thermal-induced imagery resolution ratio.
For achieving the above object, the present invention is a kind of based on vortex pulsed infrared thermal image joining method, feature
It is, comprising the following steps:
(1), thermal-induced imagery is acquired
With thermal infrared imager according to the thermal-induced imagery of chronological order record test specimen, when record,
Test specimen is applied and is motivated, and keeps test specimen stationary, using thermal infrared imager according to certain translation
Distance gradually records several width thermal-induced imageries of test specimen, and makes left there are 30% between two adjacent width thermal-induced imageries
Right overlapping, several width thermal-induced imageries form infrared chart image set T, T=T according to the time sequencing of acquisition1,T2,…,Tm, Tm
Indicate m width thermal-induced imagery;
(2), thermal-induced imagery T is extracted1、T2In key point
(2.1), by thermal-induced imagery T1、T2Grayscale image is first converted to, then Gaussian Blur processing is carried out to grayscale image;
(2.2), with SOBEL verification Gaussian Blur, treated that two width thermal-induced imageries are filtered, using horizontal, vertical
Difference operator is filtered each pixel of filtered two width thermal-induced imagery, calculates on horizontal, vertical direction
Gradient value Ix, Iy;
(2.3), according to gradient value Ix, Iy, thermal-induced imagery T is constructed respectively1、T2In each pixel gradient matrix M
(x,y);
M (x, y)=[Ix2,Ix*Iy;Ix*Iy,Iy2]
Wherein, (x, y) indicates thermal-induced imagery T1Or T2The coordinate of middle pixel;
(2.4), Gaussian smoothing filter processing is carried out to the gradient matrix M (x, y) of each pixel, obtains each pixel
New gradient matrix
(2.5), the new gradient matrix of each pixel is utilizedCalculate the angle point receptance function value r of each pixel
(x,y);
(2.6), the angle point receptance function value r (x, y) of all pixels point is constructed into matrix R according to the position of preimage vegetarian refreshments,
In matrix R, it will be greater than threshold value threshold and be the pixel R (x, y) of the local maximum in certain field labeled as key
Point;
(3), characteristic point is extracted
(3.1), in thermal-induced imagery T1And T2In, the corresponding maximum pass of angle point receptance function value of pixel is found out respectively
Key point, and it is labeled as RT1Max and RT2Max, corresponding angle point receptance function value are denoted as rT1Max and rT2max;
(3.2), to thermal-induced imagery T1And T2It performs the following operation respectively: traversing all key points, if a certain key point
Angle point receptance function value meet: r (x, y) > rτmax*Crobust, then the radius of the key point is set as infinity, wherein τ
=1,2, τ=1 indicates to thermal-induced imagery T1Into processing, τ=1 is indicated to thermal-induced imagery T2Into processing;If a certain key point
Angle point receptance function value meets: r (x, y)≤rτmax*Crobust, then the key point is calculated to the key point nearest from it, and is counted
Calculate the distance L of this two key pointk;
Wherein, R (x, y) is this key point calculated,For the nearest key point of distance R (x, y), f (R (x,
y))、Be respectively R (x, y) andPoint set, CrobustFor constant;
Respectively by T1And T2In calculated all two key points distance LkIt is ranked up, and in T1And T2It is middle take respectively away from
From LkMaximum λ key point is as T1And T2Characteristic point;
(4), transformation matrix is purified and sought to characteristic point is extracted using RANSAC algorithm
(4.1), a transformation matrix H is constructed, transformation matrix H size is 3 × 3;
Wherein, (x, y) is thermal-induced imagery T1In characteristic point, (x', y') be thermal-induced imagery T2In characteristic point;h11
~h33For parameter to be determined;
Enable h33=1, then from thermal-induced imagery T1In randomly select 4 characteristic points, and any two points in this 4 characteristic points
It is not conllinear, from thermal-induced imagery T2In find out and thermal-induced imagery T1Corresponding 4 characteristic points, by the coordinate of this 4 pairs of characteristic points
It is successively updated in transformation matrix H, obtains transformation matrix H;
(4.2), by thermal-induced imagery T1In remaining characteristic point and thermal-induced imagery T2In the pairing of remaining characteristic point, so
The number of statistics pairing characteristic point afterwards updates transformation matrix H if it is more than last to find pairing feature point number;
(4.3), step (4.1)-(4.2) are repeated, finding one makes to match the most transformation matrix H of feature point number, and
As thermal-induced imagery T1、T2Transformation relation;
T2(x, y)=H*T1(x,y)
(5), according to thermal-induced imagery T1、T2Between transformation matrix H, determine stitching direction and integration region, recycle
It is fade-in and gradually goes out method to thermal-induced imagery T1、T2Carry out image co-registration;
Wherein,Indicate the thermal-induced imagery after the completion of i-th splicingIt is to be spliced with i+1
Thermal-induced imagery carries out the thermal-induced imagery obtained after splicing fusion;α is to be fade-in gradually to go out method fusion coefficients, and specially the right is wait spell
Connect the overlapping region of thermal-induced imagery point splice to the left side after the completion of thermal-induced imagery overlapping region left margin distance
With the ratio of the length of entire overlapping region;
(6), repeat the above steps (2)-(5), continues to splice thermal-induced imagery T according to the method described above3, and so on,
Splicing fusion until completing all thermal-induced imageries.
Goal of the invention of the invention is achieved in that
The present invention is a kind of based on vortex pulsed infrared thermal image joining method, contains 30% first, in accordance with time sequencing acquisition
Then the two width thermal-induced imageries acquired at first are carried out pretreatment extraction by several width thermal-induced imageries of the overlay information of left and right
Key point out, then adaptive non-maximum restraining is carried out to the key point found, some unnecessary key points are reduced, spy is extracted
Point is levied, is then purified using RANSAC algorithm to characteristic point is extracted, seeks out transformation matrix, and then obtain projective transformation
Relationship, finally according to the Transformation Relation of Projection, with being fade-in gradually, method successively carries out image mosaic fusion to thermal-induced imagery out.
Meanwhile a kind of vortex pulsed infrared thermal image joining method that is based on of the present invention also has the advantages that
(1), the characteristic point of thermal-induced imagery is few, and the characteristic point directly extracted with traditional method is very few,
It is difficult to accomplish the accurate matching of thermal-induced imagery.The present invention obtains meeting matched by extracting key point and screening key point
For key point as characteristic point, such obtained feature point number is more, and thermal-induced imagery splicing is more accurate, and effect is more preferable;
(2), several thermal-induced imageries are precisely spliced, obtains the thermal-induced imagery panorama sketch of big visual field, quickly obtains
Take the comprehensive information of research object, it is easier to which defect is judged and analyzed comprehensively;
(3), the thermal-induced imagery panorama sketch obtained with the present invention is complete relative to the thermal-induced imagery obtained with special installation
Jing Tu, more save the cost.
Detailed description of the invention
Fig. 1 is of the invention a kind of based on vortex pulsed infrared thermal image joining method flow chart;
Fig. 2 is T1、T2Two width Infrared Thermograms;
Fig. 3 is the key point schematic diagram for extracting two width Infrared Thermograms;
Fig. 4 is to carry out adaptive non-maximum restraining schematic diagram to key point;
Fig. 5 is the accurate matched feature point diagram extracted using RANSC algorithm;
Fig. 6 is that obtained result map is merged in method splicing out using being fade-in gradually.
Specific embodiment
A specific embodiment of the invention is described with reference to the accompanying drawing, preferably so as to those skilled in the art
Understand the present invention.Requiring particular attention is that in the following description, when known function and the detailed description of design perhaps
When can desalinate main contents of the invention, these descriptions will be ignored herein.
Embodiment
Fig. 1 is of the invention a kind of based on vortex pulsed infrared thermal image joining method flow chart.
In the present embodiment, as shown in Figure 1, the present invention is a kind of based on vortex pulsed infrared thermal image joining method, including
Following steps:
S1, acquisition thermal-induced imagery
With thermal infrared imager according to the thermal-induced imagery of chronological order record test specimen, when record,
Test specimen is applied and is motivated, and keeps test specimen stationary, using thermal infrared imager according to certain translation
Distance gradually records several width thermal-induced imageries of test specimen, and makes left there are 30% between two adjacent width thermal-induced imageries
Right overlapping, several width thermal-induced imageries form infrared chart image set T, T=T according to the time sequencing of acquisition1,T2,…,Tm, Tm
Indicate m width thermal-induced imagery;
S2, as shown in Fig. 2, extract thermal-induced imagery T1、T2In key point
S2.1, by thermal-induced imagery T1、T2Grayscale image is first converted to, then Gaussian Blur processing, purpose are carried out to grayscale image
It is some point fuzziness for being unlikely to be key point, it would be possible to be that the point of key point more highlights, be convenient to key
The extraction of point;
S2.2, with SOBEL verification Gaussian Blur, treated that two width thermal-induced imageries are filtered, using horizontal, vertical
Difference operator is filtered each pixel of filtered two width thermal-induced imagery, calculates on horizontal, vertical direction
Gradient value Ix, Iy is used for subsequent screening key point;SOBEL core is First-order Gradient operator, as Gaussian Blur processing operation
Core;
S2.3, according to gradient value Ix, Iy, construct thermal-induced imagery T respectively1、T2In each pixel gradient matrix M (x,
y);
M (x, y)=[Ix2,Ix*Iy;Ix*Iy,Iy2]
Wherein, (x, y) indicates thermal-induced imagery T1Or T2The coordinate of middle pixel;
S2.4, Gaussian smoothing filter processing is carried out to the gradient matrix M (x, y) of each pixel, obtains each pixel
New gradient matrix
In this way, the pixel gradient obtained by gaussian filtering smoothing processing includes compared with untreated pixel gradient
Noise it is less, keep matching result more accurate
S2.5, the new gradient matrix using each pixelCalculate the angle point receptance function value r of each pixel
(x,y);
S2.6, the angle point receptance function value r (x, y) of all pixels point is constructed into matrix R according to the position of preimage vegetarian refreshments,
In matrix R, it will be greater than threshold value threshold and be the pixel R (x, y) of the local maximum in certain field labeled as key
Point;
After carrying out Fuzzy Processing to the point on thermal-induced imagery by this step, then gradient calculating is carried out, is responded with angle point
The key point that functional value is extracted as judgment basis is as shown in figure 3, the available many key points of this process, need to lead to
It crosses following steps further to be screened, obtains the characteristic point that can be matched
S3, characteristic point is extracted
S3.1, in thermal-induced imagery T1And T2In, the corresponding maximum key of angle point receptance function value of pixel is found out respectively
Point, and it is labeled as RT1Max and RT2Max, corresponding angle point receptance function value are denoted as rT1Max and rT2max;
S3.2, to thermal-induced imagery T1And T2It performs the following operation respectively: traversing all key points, if a certain key point
Angle point receptance function value meet: r (x, y) > rτmax*Crobust, then the radius of the key point is set as infinity, wherein τ
=1,2, τ=1 indicates to thermal-induced imagery T1Into processing, τ=1 is indicated to thermal-induced imagery T2Into processing;If a certain key point
Angle point receptance function value meets: r (x, y)≤rτmax*Crobust, then the key point is calculated to the key point nearest from it, and is counted
Calculate the distance L of this two key pointk;
Wherein, R (x, y) is this key point calculated,For the nearest key point of distance R (x, y), f (R (x,
y))、Be respectively R (x, y) andPoint set, CrobustFor constant;
Respectively by T1And T2In calculated all two key points distance LkIt is ranked up, and in T1And T2It is middle take respectively away from
From LkMaximum λ key point is as T1And T2Characteristic point;
In this way, carrying out adaptive non-maximum restraining processing by the key point obtained to S2, remove some unnecessary passes
Key point, filters out roughly more accurate key point, and the key point extracted is as shown in Figure 4.
S4, transformation matrix is purified and sought to characteristic point is extracted using RANSAC algorithm (RANSAC)
S4.1, one transformation matrix H of building, transformation matrix H size are 3 × 3;
Wherein, (x, y) is thermal-induced imagery T1In characteristic point, (x', y') be thermal-induced imagery T2In characteristic point;h11
~h33For parameter to be determined;
Enable h33=1, then from thermal-induced imagery T1In randomly select 4 characteristic points, and any two points in this 4 characteristic points
It is not conllinear, from thermal-induced imagery T2In find out and thermal-induced imagery T1Corresponding 4 characteristic points, by the coordinate of this 4 pairs of characteristic points
It is successively updated in transformation matrix H, obtains transformation matrix H;
S4.2, by thermal-induced imagery T1In remaining characteristic point and thermal-induced imagery T2In the pairing of remaining characteristic point, then
The number of statistics pairing characteristic point updates transformation matrix H if it is more than last to find pairing feature point number;
S4.3, step S4.1-S4.2 is repeated, finding one makes to match the most transformation matrix H of feature point number, and conduct
Thermal-induced imagery T1、T2Transformation relation;
T2(x, y)=H*T1(x,y)
In this way, calculating the transformation between two images using 4 pairs of characteristic points to match not conllinear in two images
Matrix calculates transformation matrix by being repeated several times, choose make the most transformation matrix of feature point number of pairing as finally into
The transformation matrix of row splicing, and all pairing characteristic points that this transformation matrix will be used to be converted are as finally obtained spy
Point is levied, as a result as shown in Figure 5.
S5, according to thermal-induced imagery T1、T2Between transformation matrix H, determine stitching direction and integration region, recycle
It is fade-in and gradually goes out method to thermal-induced imagery T1、T2Carry out image co-registration;
Wherein,Indicate the thermal-induced imagery after the completion of i-th splicingWith to be spliced red of i+1
Outer thermal image carries out the thermal-induced imagery obtained after splicing fusion;α is to be fade-in gradually to go out method fusion coefficients, and specially the right is to be spliced
The point of the overlapping region of thermal-induced imagery splice to the left side after the completion of thermal-induced imagery overlapping region left margin distance with
The ratio of the length of entire overlapping region;
S6, repeat the above steps S2-S5, continues to splice thermal-induced imagery T according to the method described above3, and so on, until
Complete the splicing fusion of all thermal-induced imageries.
In the present embodiment, as shown in fig. 6, can be seen that use by the result map that approach described above and operation obtain
When the present invention carries out image mosaic, Feature Points Matching is more accurate, almost without there is the case where error hiding, and splices and obtains
Image effect is fine, defect can be judged and be analyzed comprehensively.
Although the illustrative specific embodiment of the present invention is described above, in order to the technology of the art
Personnel understand the present invention, it should be apparent that the present invention is not limited to the range of specific embodiment, to the common skill of the art
For art personnel, if various change the attached claims limit and determine the spirit and scope of the present invention in, these
Variation is it will be apparent that all utilize the innovation and creation of present inventive concept in the column of protection.
Claims (2)
1. one kind is based on vortex pulsed infrared thermal image joining method, which comprises the following steps:
(1), thermal-induced imagery is acquired
With thermal infrared imager according to the thermal-induced imagery of chronological order record test specimen, when record,
Test specimen is applied and is motivated, and keeps test specimen stationary, using thermal infrared imager according to certain translation distance
Several width thermal-induced imageries of test specimen are gradually recorded, and make between two adjacent width thermal-induced imageries that there are 30% or so
Overlapping, several width thermal-induced imageries form infrared chart image set T, T=T according to the time sequencing of acquisition1,T2,…,Tm, TmIt indicates
M width thermal-induced imagery;
(2), thermal-induced imagery T is extracted1、T2In key point
(2.1), by thermal-induced imagery T1、T2Grayscale image is first converted to, then Gaussian Blur processing is carried out to grayscale image;
(2.2), with SOBEL verification Gaussian Blur, treated that two width thermal-induced imageries are filtered, and utilizes horizontal, vertical difference
Operator is filtered each pixel of filtered two width thermal-induced imagery, calculates gradient horizontal, on vertical direction
Value Ix, Iy;
(2.3), according to gradient value Ix, Iy, thermal-induced imagery T is constructed respectively1、T2In each pixel gradient matrix M (x, y);
M (x, y)=[Ix2,Ix*Iy;Ix*Iy,Iy2]
Wherein, (x, y) indicates thermal-induced imagery T1Or T2The coordinate of middle pixel;
(2.4), Gaussian smoothing filter processing is carried out to the gradient gradient matrix M (x, y) of each pixel, obtains each pixel
New gradient matrix
(2.5), the new gradient matrix of each pixel is utilizedCalculate each pixel angle point receptance function value r (x,
y);
(2.6), the angle point receptance function value r (x, y) of all pixels point is constructed into matrix R according to the position of preimage vegetarian refreshments, in square
In battle array R, it will be greater than threshold value threshold and be the pixel R (x, y) of the local maximum in certain field labeled as key point;
(3), characteristic point is extracted
(3.1), in thermal-induced imagery T1And T2In, the corresponding maximum key point of angle point receptance function value of pixel is found out respectively,
And it is labeled as RT1Max and RT2Max, corresponding angle point receptance function value are denoted as rT1Max and rT2max;
(3.2), to thermal-induced imagery T1And T2It performs the following operation respectively: traversing all key points, if the angle of a certain key point
Point receptance function value meets: r (x, y) > rτmax*Crobust, then the radius of the key point is set as infinity, wherein and τ=1,
2, τ=1 indicates to thermal-induced imagery T1Into processing, τ=1 is indicated to thermal-induced imagery T2Into processing;If the angle point of a certain key point
Receptance function value meets: r (x, y)≤rτmax*Crobust, then the key point is calculated to the key point nearest from it, and is calculated
The distance L of this two key pointk;
Wherein, R (x, y) is this key point calculated,For the nearest key point of distance R (x, y), f (R (x, y)),Be respectively R (x, y) andPoint set;
Respectively by T1And T2In calculated all two key points distance LkIt is ranked up, and in T1And T2It is middle to take distance L respectivelyk
Maximum λ key point is as T1And T2Characteristic point;
(4), transformation matrix is purified and sought to characteristic point is extracted using RANSAC algorithm
(4.1), transformation matrix H is constructed, transformation matrix H size is 3 × 3;
Wherein, (x, y) is thermal-induced imagery T1In characteristic point, (x', y') be thermal-induced imagery T2In characteristic point;h11~h33
For parameter to be determined;
Enable h33=1, then from thermal-induced imagery T1In randomly select 4 characteristic points, and any two points in this 4 characteristic points are not
Collinearly, from thermal-induced imagery T2In find out and thermal-induced imagery T1Corresponding 4 characteristic points, successively by the coordinate of this 4 pairs of characteristic points
It is updated in transformation matrix H, obtains transformation matrix H;
(4.2), by thermal-induced imagery T1In remaining characteristic point and thermal-induced imagery T2In the pairing of remaining characteristic point, then unite
The number of meter pairing characteristic point updates transformation matrix H if it is more than last to find pairing feature point number;
(4.3), step (4.1)-(4.2) are repeated, finding one makes to match the most transformation matrix H of feature point number, and conduct
Thermal-induced imagery T1、T2Transformation relation;
T2(x, y)=H*T1(x,y)
(5), according to thermal-induced imagery T1、T2Between transformation matrix H, determine that stitching direction and integration region, recycling are fade-in
Gradually go out method to thermal-induced imagery T1、T2Carry out image co-registration;
Wherein,Indicate the thermal-induced imagery after the completion of i-th splicingWith to be spliced infrared of i+1
Thermal image carries out the thermal-induced imagery obtained after splicing fusion;α is to be fade-in gradually to go out method fusion coefficients, and specially the right is to be spliced red
The point of the overlapping region of outer thermal image splice to the left side after the completion of thermal-induced imagery overlapping region left margin distance with it is whole
The ratio of the length of a overlapping region;
(6), repeat the above steps (2)-(5), continues to splice thermal-induced imagery T according to the method described above3, and so on, until complete
It is merged at the splicing of all thermal-induced imageries.
2. according to claim 1 a kind of based on vortex pulsed infrared thermal image joining method, which is characterized in that described
The method for calculating the angle point receptance function value r (x, y) of each pixel are as follows:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811276563.8A CN109360145A (en) | 2018-10-30 | 2018-10-30 | One kind is based on vortex pulsed infrared thermal image joining method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811276563.8A CN109360145A (en) | 2018-10-30 | 2018-10-30 | One kind is based on vortex pulsed infrared thermal image joining method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109360145A true CN109360145A (en) | 2019-02-19 |
Family
ID=65347181
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811276563.8A Pending CN109360145A (en) | 2018-10-30 | 2018-10-30 | One kind is based on vortex pulsed infrared thermal image joining method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109360145A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110706191A (en) * | 2019-10-14 | 2020-01-17 | 中国空气动力研究与发展中心超高速空气动力研究所 | High-pressure container detection method based on infrared image splicing fusion algorithm |
CN111882520A (en) * | 2020-06-16 | 2020-11-03 | 歌尔股份有限公司 | Screen defect detection method and device and head-mounted display equipment |
CN112102307A (en) * | 2020-09-25 | 2020-12-18 | 杭州海康威视数字技术股份有限公司 | Method and device for determining heat data of global area and storage medium |
CN113532328A (en) * | 2021-07-16 | 2021-10-22 | 燕山大学 | Surface profile real-time measurement system and method in medium plate straightening process |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103487729A (en) * | 2013-09-06 | 2014-01-01 | 广东电网公司电力科学研究院 | Electrical equipment defect detection method based on fusion of ultraviolet video and infrared video |
CN104020177A (en) * | 2014-06-26 | 2014-09-03 | 重庆大学 | Dual-CCD (Charge Coupled Device) scanning imaging detection method for continuous casting slab surface defects |
CN107784632A (en) * | 2016-08-26 | 2018-03-09 | 南京理工大学 | A kind of infrared panorama map generalization method based on infra-red thermal imaging system |
CN108109128A (en) * | 2017-11-30 | 2018-06-01 | 电子科技大学 | A kind of thermal-induced imagery processing method based on spatial coherence |
CN108198181A (en) * | 2018-01-23 | 2018-06-22 | 电子科技大学 | A kind of thermal-induced imagery processing method based on region segmentation and image co-registration |
CN108195928A (en) * | 2017-12-19 | 2018-06-22 | 电子科技大学 | A kind of metallicl magnetic material defect detecting device based on image co-registration |
CN108226290A (en) * | 2018-01-08 | 2018-06-29 | 西安交通大学 | A kind of inside parts defect three-dimensional parameter extracting method based on ultrasonic phase array |
CN108682002A (en) * | 2018-04-03 | 2018-10-19 | 中国空气动力研究与发展中心超高速空气动力研究所 | Based on vortex pulse thermal imaging to enhance the method for Defects in Pressure Container feature |
-
2018
- 2018-10-30 CN CN201811276563.8A patent/CN109360145A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103487729A (en) * | 2013-09-06 | 2014-01-01 | 广东电网公司电力科学研究院 | Electrical equipment defect detection method based on fusion of ultraviolet video and infrared video |
CN104020177A (en) * | 2014-06-26 | 2014-09-03 | 重庆大学 | Dual-CCD (Charge Coupled Device) scanning imaging detection method for continuous casting slab surface defects |
CN107784632A (en) * | 2016-08-26 | 2018-03-09 | 南京理工大学 | A kind of infrared panorama map generalization method based on infra-red thermal imaging system |
CN108109128A (en) * | 2017-11-30 | 2018-06-01 | 电子科技大学 | A kind of thermal-induced imagery processing method based on spatial coherence |
CN108195928A (en) * | 2017-12-19 | 2018-06-22 | 电子科技大学 | A kind of metallicl magnetic material defect detecting device based on image co-registration |
CN108226290A (en) * | 2018-01-08 | 2018-06-29 | 西安交通大学 | A kind of inside parts defect three-dimensional parameter extracting method based on ultrasonic phase array |
CN108198181A (en) * | 2018-01-23 | 2018-06-22 | 电子科技大学 | A kind of thermal-induced imagery processing method based on region segmentation and image co-registration |
CN108682002A (en) * | 2018-04-03 | 2018-10-19 | 中国空气动力研究与发展中心超高速空气动力研究所 | Based on vortex pulse thermal imaging to enhance the method for Defects in Pressure Container feature |
Non-Patent Citations (1)
Title |
---|
徐鑫: "红外图像拼接技术研究", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110706191A (en) * | 2019-10-14 | 2020-01-17 | 中国空气动力研究与发展中心超高速空气动力研究所 | High-pressure container detection method based on infrared image splicing fusion algorithm |
CN110706191B (en) * | 2019-10-14 | 2022-09-20 | 中国空气动力研究与发展中心超高速空气动力研究所 | High-pressure container detection method based on infrared image splicing fusion algorithm |
CN111882520A (en) * | 2020-06-16 | 2020-11-03 | 歌尔股份有限公司 | Screen defect detection method and device and head-mounted display equipment |
CN111882520B (en) * | 2020-06-16 | 2023-10-17 | 歌尔光学科技有限公司 | Screen defect detection method and device and head-mounted display equipment |
CN112102307A (en) * | 2020-09-25 | 2020-12-18 | 杭州海康威视数字技术股份有限公司 | Method and device for determining heat data of global area and storage medium |
WO2022062862A1 (en) * | 2020-09-25 | 2022-03-31 | 杭州海康威视数字技术股份有限公司 | Method and apparatus for determining heat data of global region, and storage medium |
CN112102307B (en) * | 2020-09-25 | 2023-10-20 | 杭州海康威视数字技术股份有限公司 | Method and device for determining heat data of global area and storage medium |
CN113532328A (en) * | 2021-07-16 | 2021-10-22 | 燕山大学 | Surface profile real-time measurement system and method in medium plate straightening process |
CN113532328B (en) * | 2021-07-16 | 2023-07-25 | 燕山大学 | Real-time surface profile measuring system and method for medium plate straightening process |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109360145A (en) | One kind is based on vortex pulsed infrared thermal image joining method | |
CN105913093B (en) | A kind of template matching method for Text region processing | |
CN103679636B (en) | Based on point, the fast image splicing method of line double characteristic | |
CN105957007B (en) | Image split-joint method based on characteristic point plane similarity | |
EP1901228B1 (en) | Apparatus, method and program for image matching | |
CN102999750B (en) | A kind of fingerprint on site Enhancement Method removing background interference | |
CN109285194B (en) | Camera calibration board and camera calibration data acquisition method | |
CN102334141B (en) | Infrared resolution and contrast enhancement with fusion | |
CN104867123A (en) | Infrared Resolution And Contrast Enhancement With Fusion | |
CN102982520B (en) | Robustness face super-resolution processing method based on contour inspection | |
CN104134208B (en) | Using geometry feature from slightly to the infrared and visible light image registration method of essence | |
CN106709868A (en) | Image stitching method and apparatus | |
CN111583114B (en) | Automatic measuring device and measuring method for pipeline threads | |
CN102567979A (en) | Vehicle-mounted infrared night vision system and multi-source images fusing method thereof | |
CN103679672B (en) | Panorama image splicing method based on edge vertical distance matching | |
CN104392416A (en) | Video stitching method for sports scene | |
Ait-Aoudia et al. | Satellite and aerial image mosaicing-a comparative insight | |
CN106600653A (en) | Calibration method for optical center of zooming camera | |
Ma et al. | Remote sensing image registration based on multifeature and region division | |
CN109658366A (en) | Based on the real-time video joining method for improving RANSAC and dynamic fusion | |
CN104966283A (en) | Imaging layered registering method | |
CN109285118B (en) | Infrared thermal image splicing method with additional accessory layer | |
Liu et al. | Application of color filter adjustment and k-means clustering method in lane detection for self-driving cars | |
Chen et al. | Robust visible-infrared image matching by exploiting dominant edge orientations | |
Anzid et al. | A new SURF-based algorithm for robust registration of multimodal images data |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190219 |
|
RJ01 | Rejection of invention patent application after publication |