CN102855649B - Method for splicing high-definition image panorama of high-pressure rod tower on basis of ORB (Object Request Broker) feature point - Google Patents
Method for splicing high-definition image panorama of high-pressure rod tower on basis of ORB (Object Request Broker) feature point Download PDFInfo
- Publication number
- CN102855649B CN102855649B CN201210303832.1A CN201210303832A CN102855649B CN 102855649 B CN102855649 B CN 102855649B CN 201210303832 A CN201210303832 A CN 201210303832A CN 102855649 B CN102855649 B CN 102855649B
- Authority
- CN
- China
- Prior art keywords
- image
- orb
- double points
- matching double
- matching
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000005070 sampling Methods 0.000 claims abstract description 19
- 239000011159 matrix material Substances 0.000 claims abstract description 18
- 238000000605 extraction Methods 0.000 claims description 18
- 230000009466 transformation Effects 0.000 claims description 16
- 238000001514 detection method Methods 0.000 claims description 14
- 239000000284 extract Substances 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 5
- 238000012217 deletion Methods 0.000 claims description 3
- 230000037430 deletion Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 241000937378 Everettia interior Species 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 4
- 238000005562 fading Methods 0.000 abstract 1
- 230000000694 effects Effects 0.000 description 6
- 238000003384 imaging method Methods 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- 238000012360 testing method Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 2
- 238000010010 raising Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 244000291564 Allium cepa Species 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 230000002459 sustained effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses a method for splicing a high-definition image panorama of a high-pressure rod tower on the basis of an ORB (Object Request Broker) feature point. The method comprises the following steps: 1) reading ultrahigh resolution high-pressure rod tower images of which the sizes are W*H, and sampling and reducing the to-be-spliced ultrahigh resolution images by utilizing a bilinear interpolation method, thereby obtaining a w*h image, wherein W, H, w and h are integers greater than 0, W/w=H/h=k and k is the integer greater than 0; 2) utilizing an ORB algorithm to extract features of all the sampled images; 3) roughly matching the ORB features extracted in the step 2); 4) utilizing matching point pairs extracted in the previous step to extract the ORB features again in an image block in which the matching point pairs of an original ultrahigh resolution image are located, thereby finely matching; 5) calculating a transformational matrix H between adjacent images through the solved matching point pairs; and 6) utilizing a gradually fading method to fuse the adjacent ultrahigh resolution images. According to the method, the seamless splicing for the ultrahigh resolution images is realized, the time for splicing is reduced, the splicing efficiency is increased and the beneficial effect on the high-definition image is achieved.
Description
Technical field
The present invention relates to the clear image panorama joining method of a kind of high pressure stem tower height, particularly relate to the clear image panorama joining method of a kind of high pressure stem tower height based on ORB unique point.
Background technology
In recent years, the sustained and rapid development of Chinese national economy proposes more and more higher requirement to China's power industry.Because China territory is vast, power transmission line corridor is with a varied topography, the limitation of Traditional Man walking operation pattern highlights day by day, adopt unmanned vehicle lift-launch digital imaging apparatus to carry out detailed-oriented patrolling and examining to overhead transmission line and become possibility, although existing digital imaging apparatus resolution has reached careful requirement of seeing ultra-high-tension power transmission line gold utensil clearly, but because imaging device visual field is less, the high-definition image gathered can not comprise high pressure shaft tower armamentarium.
Panorama Mosaic technology has a wide range of applications in satellite remote sensing detection, meteorology, medical science, military affairs, Aero-Space, the protection of large area cultural heritage and virtual scene realize.Overhead transmission line high pressure shaft tower has the characteristics of image of large format, adopts common digital imaging apparatus cannot once pan and the image of ultrahigh resolution.Utilize image mosaic technology to solve the problem smoothly, successfully realize the synthesis of ultrahigh resolution high pressure shaft tower image.
Panorama Mosaic technology can by digital imaging apparatus gather multiple image and be spliced into the larger panoramic picture in a width visual field, and the panoramic picture distortion finally obtained is less, area-of-interest all centralized displaying on a Zhang Quanjing image.Panorama Mosaic technology relates generally to feature point extraction, Feature Points Matching and image fusion technology three aspect, and wherein the extraction effect of unique point directly affects later image splicing effect.
At present, SIFT and SURF is popular Feature Points Extraction, although above-mentioned two kinds of Feature Points Extraction, image mosaic and other a lot of in all have comparatively ripe application.But during feature point extraction to high-definition picture, just have a large amount of time for feature point extraction.
Summary of the invention
For solving produced problem in above-mentioned image mosaic, the present invention proposes the low and splicing effect of a kind of time complexity well based on the clear image panorama joining method of high pressure stem tower height of ORB unique point.It utilizes slightly mates the Feature Points Matching algorithm combined with exact matching, achieves the seamless spliced of ultrahigh resolution image, decreases splicing required time, improve splicing efficiency, particularly have good beneficial effect to high-definition image.
To achieve these goals, the present invention adopts following technical scheme:
The clear image panorama joining method of high pressure stem tower height based on ORB unique point, concrete steps are:
The first step: read ultrahigh resolution high pressure shaft tower image, and sampling is carried out to image reduce;
Second step: all imagery exploitation ORB algorithms after reducing sampling carry out feature extraction;
3rd step: utilize the ORB feature extracted to carry out most proximity matching, by RASANC algorithm, the matching double points obtained is screened, obtain thick matching double points;
4th step: the thick matching double points coordinate that in utilization, step is extracted, calculates the respective coordinates in original ultrahigh resolution image, and again extract ORB feature in the image block at the matching double points place of original ultrahigh resolution image, carry out exact matching;
5th step: calculate the transformation matrix H between adjacent image
0;
6th step: utilize to be fade-in and gradually go out method and merge ultrahigh resolution adjacent image, obtain ultrahigh resolution panoramic picture, splicing terminates.
In the described first step, the method that reduces of sampling is: utilize bilinear interpolation that ultrahigh resolution image to be spliced is carried out sampling and reduce, original image is of a size of W × H, and the picture size obtained is w × h, wherein W, H, w, h be greater than 0 integer,
k is for being greater than 0 integer.
The concrete steps of the feature extraction in described second step are:
(2-1) Oriented FAST feature point detection is carried out:
(2-2) Rotated BRIEF Feature Descriptor is generated;
Near unique point, the some points of random selecting are right, and these size groups putting right gray-scale value are synthesized a binary string, and using the Feature Descriptor of this binary string as this unique point.
The detailed process of described step (2-2) is:
A generates BRIEF Feature Descriptor;
B generates Rotated BRIEF Feature Descriptor;
The direction vector extracted in Oriented FAST algorithm is joined in BRIEF feature, rotates, obtain oriented BRIEF, be referred to as Steered BRIEF; Have high variance and the incoherent steered brief of height with greedy learning algorithm screening, result is referred to as rBRIEF; Calculate the distance of each SBRIEF and 0.5, and create container T; First SBRIEF is put into result container R, and removes from container T; From container T, take out next SBRIEF, and compare with all SBRIEF in result container R, if its correlativity is less than certain threshold value, then adds in result container R, otherwise abandon;
Repeat step b until have 256 SBRIEF in result container R, if be less than 256 SBRIEF in result container R, then change threshold value, and repeat above step.
The concrete steps of described 3rd step are as follows:
(3-1) LSH is selected as most proximity matching point to calculating;
(3-2) utilize RASANC algorithm to be screened by the thick matching double points that step (3-1) generates, select and meet the requirements of matching double points i.e. interior point, deletion error matching double points.
The detailed process of described step (3-2) is:
Point initialization in (a): randomly draw 4 pairs of matching double points in given matching double points;
B () calculates transformation matrix H by interior point
0;
C (), to matching double points remaining in matching double points, calculates they and transformation matrix H
0distance, if result is less than certain threshold value, then joined in interior set, and according to some set in new, use least square method to upgrade transformation matrix H
0, otherwise continue to judge remaining matching double points;
D () repeated execution of steps (c), until interior some number no longer increases.
The concrete steps of described 4th step are as follows:
(4-1) M is made
land M
rfor original two adjacent images of ultrahigh resolution, m
land m
rbe respectively two adjacent images after reducing of sampling, the thick matching double points coordinate that the 3rd step calculates is respectively (x
li, y
li) and (x
rj, y
rj) wherein 0≤i, j≤n, n be required match point logarithm;
(4-2) respectively with (X
li, Y
yli) and (X
rj, Y
rj) centered by, be that the range image block of radius is respectively I with γ
land I
r, wherein:
(4-3) at image block I
l, I
rthe middle ORB of extraction respectively feature;
(4-4) I is obtained
land I
rmatching double points;
(4-5) to all matching double points, repeat above step, generate the matching double points of exact matching.
The concrete steps of described 6th step are as follows:
(6-1) according to the transformation matrix H between image
0, corresponding image is converted, determines the overlapping region between image;
(6-2) I is made
land I
rbe respectively two adjacent images, I is the image after merging:
I(x，y)＝(1-τ(k))×I
l(x，y)+τ(k)×I
r(x，y)+d (1)
Wherein 0≤d≤1 is fine setting coefficient, and 0≤τ (k)≤1 is weighting function,
Wherein m is overlapping peak width, and k is leftmost pixel count from overlapping region, and such overlapping region is larger, and τ (k) will be milder, makes seamlessly transit between image.
Beneficial effect of the present invention:
1, utilize ORB algorithm institute extract minutiae, image mosaic application in have good effect, and its operation time two orders of magnitude faster than SIFT algorithm, an order of magnitude faster than SURF algorithm;
2, by carrying out ORB feature point extraction falling to adopt in image, the problem of the consuming time serious and low memory that unique point in ultrahigh resolution image too much causes is solved;
3, in ultrahigh resolution image, utilize ORB algorithm to carry out feature point extraction in the image block of matching double points place, and carry out exact matching, solve the error in the rear image mosaic of sampling;
4, the method combined with exact matching is slightly mated in utilization, had obvious minimizing its match time, and matching precision also has many raisings;
5, from whole ultrahigh resolution image splicing aspect, its splicing speed has had very large raising.
Accompanying drawing explanation
Fig. 1 is algorithm flow chart of the present invention;
Fig. 2,3 is image before splicing;
Fig. 4,5 is effect exploded view after splicing.
Embodiment
Below in conjunction with accompanying drawing and embodiment, the invention will be further described.
As shown in Figure 1, method step of the present invention is as follows:
The first step, read ultrahigh resolution high pressure shaft tower image, be of a size of W × H, utilize bilinear interpolation that ultrahigh resolution image to be spliced is carried out sampling and reduce, obtain w × h image, wherein W, H, w, h be greater than 0 integer,
k is for being greater than 0 integer;
Second step, ORB algorithm is utilized to carry out feature extraction to image after all samplings:
ORB feature have employed Oriented FAST feature point detection operator and Rotated BRIEF Feature Descriptor.ORB algorithm not only has the Detection results of SIFT feature, but also there is the characteristic of the aspects such as rotation, scaling, brightness change unchangeability, the most important thing is that its time complexity has had than SIFT to reduce greatly, make ORB algorithm have very large application prospect in high-definition image splicing and real time video image splicing.
Specifically comprise the following steps:
2-1) Oriented FAST feature point detection:
The present invention, by the basis of FAST feature point detection, adds direction vector, makes it have directivity.
A) FAST feature point detection algorithm is utilized to detect key point fast;
B) barycenter and the direction of key point place block is asked:
The extraction of block barycenter is as follows:
m
pq＝∑
x，yx
py
qI(x，y) (1)
Wherein m
pqfor the square of block, p, q ∈ (0,1), x, y ∈ block, C is required block barycenter; The extraction of Block direction:
θ＝atan2(m
01，m
10) (3)
The FAST key point in this belt transect direction just extracts.But FAST feature point detection can not process multi-scale image, but original image can be done pyramid, then carry out above step to each figure, such Oriented FAST just supports multiple dimensioned change.
2-2) generate Rotated BRIEF Feature Descriptor
The main thought of BRIEF be exactly near unique point the some points of random selecting right, these size groups putting right gray-scale value are synthesized a binary string, and using the Feature Descriptor of this binary string as this unique point.Its advantage is that arithmetic speed is very fast.
A) BRIEF Feature Descriptor is generated:
A given width figure;
To the smoothing process of image, reduce picture noise;
Image is selected the region unit p of one piece of SXS pixel, wherein 5≤S≤15, p extracts BRIEF feature:
Definition τ test, τ measuring and calculation formula is as follows:
X, y are two location of pixels in p, and namely x and y is the two-dimensional coordinate of shape as [u, v], and p (x) and p (y) is the brightness value of x and y pixel;
A BRIEF feature is exactly the binary string that several τ test composition, and structure specific [x, y] is right, according to following formula, obtains BRIEF feature
B) Rotated BRIEF Feature Descriptor is generated:
The direction vector extracted in Oriented FAST algorithm is joined in BRIEF feature, rotates, obtain oriented BRIEF, be referred to as Steered BRIEF;
Have high variance and the incoherent steered brief of height with the screening of one greedy learning algorithm, result is referred to as rBRIEF, and algorithm is as follows:
Calculate the distance of each SBRIEF and 0.5, and create container T;
Wherein, greedy algorithm:
First SBRIEF is put into result container R, and removes from container T;
From container T, take out next SBRIEF, and compare with all SBRIEF in result container R, if its correlativity is less than certain threshold value, then adds in result container R, otherwise abandon;
Repeat step b) until have 256 SBRIEF in result container R, if be less than 256 SBRIEF in result container R, then change (becoming large) threshold value, and repeat above step;
3rd step, slightly mate the ORB feature extracted in second step, concrete steps are as follows:
3-1) rBRIEF feature is binary mode feature, selects Locality Sensitive Hashing (LSH) as most proximity matching point to calculating;
RASANC algorithm 3-2) is utilized to be screened by the matching double points that upper step generates:
RANSAC is the abbreviation of " RANdom SAmple Consensus (random sampling is consistent) ", it is a kind of Robust estimation method, proposed in 1981 by Fischler and Bolles, it can estimate high-precision parameter from the data centralization of a large amount of exterior point.Its basic thought is, when carrying out parameter estimation, designing a search engine, utilizing this search engine iteration to filter out the input data consistent with estimated parameter, then utilizing these data to carry out parameter estimation.
The present invention uses RANSAC algorithm to screen all matching double points, selects the matching double points reaching parameter model requirement, point namely, deletion error matching double points, and specific algorithm is as follows:
Interior some initialization: randomly draw 4 pairs of matching double points in given matching double points;
Transformation matrix H is calculated by interior point
0;
To matching double points remaining in matching double points, calculate they and transformation matrix H
0distance, if result is less than certain threshold value, then joined in interior set, and according to some set in new, use least square method to upgrade transformation matrix H
0, otherwise continue to judge remaining matching double points;
Repeat previous step, until interior some number no longer increases.
4th step, by above-mentioned three steps, the transformation matrix H between adjacent image after sampling can be calculated
0, and the splicing of image is carried out by certain fusion method, but current transform matrix H
0calculate in image after sampling, if be directly used in ultrahigh resolution original image, after splicing, image overlapping region not necessarily can be very accurate, has the error of sampling quantity level pixel coordinate.The matching double points that the present invention utilizes step to extract, again extracts ORB feature, then carries out exact matching, solve produced problem above in the image block at the matching double points place of original ultrahigh resolution image.Concrete steps are as follows:
4-1) make M
land M
rfor original two adjacent images of ultrahigh resolution, m
land m
rbe respectively two adjacent images after reducing of sampling, the matching double points coordinate that the 3rd step calculates is respectively (x
li, y
li) and (x
rj, y
rj), wherein 0≤i, j≤n, n is required match point logarithm;
4-2) respectively with (X
li, Yy
li) and (X
rj, Y
rj) centered by, be that the range image block of radius is respectively I with γ
land I
r, γ is the integer of any 9≤γ≤100 scope here, wherein:
4-3) at image block I
l, I
rthe middle ORB of extraction respectively feature;
4-4) obtain I
land I
rmatching double points;
4-5) to all matching double points, repeat above step, generate the matching double points of exact matching.
5th step, by required matching double points, calculate the transformation matrix H between adjacent image above
0;
6th step, utilize to be fade-in and gradually go out method and merge ultrahigh resolution adjacent image, concrete steps are as follows:
6-1) according to the transformation matrix H between image
0, can convert corresponding image, determine the overlapping region between image;
6-2) make I
land I
rbe respectively two adjacent images, I is the image after merging:
I(x，y)＝(1-τ(k))×I
l(x，y)+τ(k)×I
r(x，y)+d (1)
Wherein 0≤d≤1 is fine setting coefficient, and 0≤τ (k)≤1 is weighting function,
Wherein m is overlapping peak width, and k is leftmost pixel count from overlapping region, and such overlapping region is larger, and τ (k) will be milder, makes seamlessly transit between image.
In order to check the beneficial effect of the present invention in ultrahigh resolution, in an experiment, the present invention carries out splicing effect test to 5 groups of ultrahigh resolution high pressure shaft tower images, and experiment condition is as follows: CPU is Intel Core i32.27GHz, 2G internal memory.Table 1 is that ORB compares on time complexity with feature point detecting method popular now, picture size size used is 5184X3456, as can be seen from Table 1, ORB feature point detection algorithm time complexity is obviously better than SURF and SIFT feature detection algorithm, a nearly order of magnitude faster than SURF, faster than SIFT nearly two orders of magnitude; Table 2 is on original ultrahigh resolution image, directly utilize ORB algorithm to carry out splicing piecemeal stitching algorithm the comparing on time complexity with first thick rear essence, as can be seen from the table, joining method of the present invention is than the method directly utilizing ORB algorithm to carry out splicing on original ultrahigh resolution image soon nearly 150%.In conjunction with above two experimental results, the present invention not only improves detection efficiency at unique point examination phase, and greatly reduce time complexity, and in entirety splicing, the present invention utilizes the block image joining method of first thick rear essence, equally greatly reduce the time complexity of splicing, for the splicing of ultrahigh resolution image improves splicing efficiency greatly.
Contrast detection time of table 1. each feature point detection algorithm
Table 2. direct splicing and piecemeal splice the time and contrast
By reference to the accompanying drawings the specific embodiment of the present invention is described although above-mentioned; but not limiting the scope of the invention; one of ordinary skill in the art should be understood that; on the basis of technical scheme of the present invention, those skilled in the art do not need to pay various amendment or distortion that creative work can make still within protection scope of the present invention.
Claims (5)
1., based on the clear image panorama joining method of high pressure stem tower height of ORB unique point, it is characterized in that, concrete steps are:
The first step: read ultrahigh resolution high pressure shaft tower image, and sampling is carried out to image reduce;
Second step: all imagery exploitation ORB algorithms after reducing sampling carry out feature extraction;
The concrete steps of the feature extraction in described second step are:
(2-1) Oriented FAST feature point detection is carried out:
(2-2) Rotated BRIEF Feature Descriptor is generated;
Near unique point, the some points of random selecting are right, and these size groups putting right gray-scale value are synthesized a binary string, and using the Feature Descriptor of this binary string as this unique point;
3rd step: utilize the ORB feature extracted to carry out most proximity matching, by RASANC algorithm, the matching double points obtained is screened, obtain thick matching double points;
The concrete steps of described 3rd step are as follows:
(3-1) LSH is selected to calculate most proximity matching point pair;
(3-2) utilize RASANC algorithm to be screened by the matching double points that step (3-1) generates, select and meet the requirements of matching double points i.e. interior point, deletion error matching double points;
4th step: the thick matching double points coordinate that in utilization, step is extracted, calculates the respective coordinates in original ultrahigh resolution image, and again extract ORB feature in the image block at the matching double points place of original ultrahigh resolution image, carry out exact matching;
The concrete steps of described 4th step are as follows:
(4-1) M is made
land M
rfor original two adjacent images of ultrahigh resolution, m
land m
rbe respectively two adjacent images after reducing of sampling, the thick matching double points coordinate that the 3rd step calculates is respectively (x
li, y
li) and (x
rj, y
rj) wherein 0≤i, j≤n, n be required match point logarithm;
(4-2) respectively with (X
li, Y
li) and (X
rj, Y
rj) centered by, be that the range image block of radius is respectively I with γ
land I
r, wherein:
(4-3) at image block I
l, I
rthe middle ORB of extraction respectively feature;
(4-4) I is obtained
land I
rmatching double points;
(4-5) to all matching double points, repeat above step, generate the matching double points of exact matching;
5th step: calculate the transformation matrix H between adjacent image
0;
6th step: utilize to be fade-in and gradually go out method and merge ultrahigh resolution adjacent image, obtain ultrahigh resolution panoramic picture, splicing terminates.
2. the clear image panorama joining method of a kind of high pressure stem tower height based on ORB unique point as claimed in claim 1, it is characterized in that, in the described first step, the method reduced of sampling is: utilize bilinear interpolation that ultrahigh resolution image to be spliced is carried out sampling and reduce, original image is of a size of W × H, the picture size obtained is w × h, wherein W, H, w, h be greater than 0 integer
q is for being greater than 0 integer.
3. the clear image panorama joining method of a kind of high pressure stem tower height based on ORB unique point as claimed in claim 1, it is characterized in that, the detailed process of described step (2-2) is:
A generates BRIEF Feature Descriptor;
B generates Rotated BRIEF Feature Descriptor;
The direction vector extracted in Oriented FAST algorithm is joined in BRIEF feature, rotates, obtain oriented BRIEF, be referred to as Steered BRIEF; Have high variance and the incoherent Steered BRIEF of height with greedy learning algorithm screening, result is referred to as rBRIEF; Calculate the distance of each SBRIEF and 0.5, and create container T; First SBRIEF is put into result container R, and removes from container T; From container T, take out next SBRIEF, and compare with all SBRIEF in result container R, if its correlativity is less than certain threshold value, then adds in result container R, otherwise abandon;
Repeat step b until have 256 SBRIEF in result container R, if be less than 256 SBRIEF in result container R, then change threshold value, and repeat above step.
4. the clear image panorama joining method of a kind of high pressure stem tower height based on ORB unique point as claimed in claim 1, it is characterized in that, the detailed process of described step (3-2) is:
Point initialization in (a): randomly draw 4 pairs of matching double points in given matching double points;
B () calculates transformation matrix H by interior point
0;
C (), to matching double points remaining in matching double points, calculates they and transformation matrix H
0distance, if result is less than certain threshold value, then joined in interior set, and according to some set in new, use least square method to upgrade transformation matrix H
0, otherwise continue to judge remaining matching double points;
D () repeated execution of steps (c), until interior some number no longer increases.
5. the clear image panorama joining method of a kind of high pressure stem tower height based on ORB unique point as claimed in claim 1, it is characterized in that, the concrete steps of described 6th step are as follows:
(6-1) according to the transformation matrix H between image
0, corresponding image is converted, determines the overlapping region between image;
(6-2) I is made
land I
rbe respectively two adjacent images, I is the image after merging:
I(x，y)＝(1-τ(k))×I
l(x，y)+τ(k)×I
r(x，y)+d (1)
Wherein 0≤d≤1 is fine setting coefficient, and 0≤τ (k)≤1 is weighting function,
Wherein m is overlapping peak width, and k is leftmost pixel count from overlapping region, and such overlapping region is larger, and τ (k) will be milder, makes seamlessly transit between image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210303832.1A CN102855649B (en) | 2012-08-23 | 2012-08-23 | Method for splicing high-definition image panorama of high-pressure rod tower on basis of ORB (Object Request Broker) feature point |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210303832.1A CN102855649B (en) | 2012-08-23 | 2012-08-23 | Method for splicing high-definition image panorama of high-pressure rod tower on basis of ORB (Object Request Broker) feature point |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102855649A CN102855649A (en) | 2013-01-02 |
| CN102855649B true CN102855649B (en) | 2015-07-15 |
Family
ID=47402210
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210303832.1A Active CN102855649B (en) | 2012-08-23 | 2012-08-23 | Method for splicing high-definition image panorama of high-pressure rod tower on basis of ORB (Object Request Broker) feature point |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102855649B (en) |
Families Citing this family (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103236050B (en) * | 2013-05-06 | 2015-08-12 | 电子科技大学 | A kind of auxiliary bank note worn coin method for reconstructing based on figure cluster |
| CN103336963B (en) * | 2013-07-08 | 2016-06-08 | 天脉聚源(北京)传媒科技有限公司 | A kind of method of image characteristics extraction and device |
| CN103761721B (en) * | 2013-12-30 | 2016-10-05 | 西北工业大学 | One is applicable to space rope system machine human stereo vision fast image splicing method |
| CN104200487A (en) * | 2014-08-01 | 2014-12-10 | 广州中大数字家庭工程技术研究中心有限公司 | Target tracking method based on ORB characteristics point matching |
| CN104462820B (en) * | 2014-12-10 | 2017-10-27 | 广东电网有限责任公司电力科学研究院 | A kind of method of power network shaft tower coordinate error detection |
| CN105007397B (en) * | 2015-04-30 | 2018-06-19 | 南方电网科学研究院有限责任公司 | Video compensation method for eliminating line segment cross mismatching points |
| CN104881841B (en) * | 2015-05-20 | 2019-01-04 | 南方电网科学研究院有限责任公司 | Aerial high-voltage power tower image splicing method based on edge features and point features |
| CN105223957B (en) * | 2015-09-24 | 2018-10-02 | 北京零零无限科技有限公司 | A kind of method and apparatus of gesture manipulation unmanned plane |
| CN106657816A (en) * | 2016-11-07 | 2017-05-10 | 湖南源信光电科技有限公司 | ORB algorithm based multipath rapid video splicing algorithm with image registering and image fusion in parallel |
| CN106909877B (en) * | 2016-12-13 | 2020-04-14 | 浙江大学 | Visual simultaneous mapping and positioning method based on dotted line comprehensive characteristics |
| CN107580175A (en) * | 2017-07-26 | 2018-01-12 | 济南中维世纪科技有限公司 | A kind of method of single-lens panoramic mosaic |
| CN107656613B (en) * | 2017-09-08 | 2020-12-18 | 国网智能科技股份有限公司 | Human-computer interaction system based on eye movement tracking and working method thereof |
| CN107703956A (en) * | 2017-09-28 | 2018-02-16 | 山东鲁能智能技术有限公司 | A kind of virtual interaction system and its method of work based on inertia capturing technology |
| CN107886100B (en) * | 2017-12-04 | 2020-10-30 | 西安思源学院 | BRIEF feature descriptor based on rank array |
| CN109961078B (en) * | 2017-12-22 | 2021-09-21 | 展讯通信（上海）有限公司 | Image matching and splicing method, device, system and readable medium |
| CN108455228B (en) * | 2017-12-29 | 2023-07-28 | 长春师范大学 | Automatic tire loading system |
| JP7244488B2 (en) * | 2018-03-15 | 2023-03-22 | 株式会社村上開明堂 | Composite video creation device, composite video creation method, and composite video creation program |
| CN109064385A (en) * | 2018-06-20 | 2018-12-21 | 何中 | 360 degree of panorama bandwagon effect Core Generators and delivery system |
| CN110866863A (en) * | 2018-08-27 | 2020-03-06 | 天津理工大学 | Automobile A-pillar perspective algorithm |
| CN109376773A (en) * | 2018-09-30 | 2019-02-22 | 福州大学 | Crack detecting method based on deep learning |
| CN110986889A (en) * | 2019-12-24 | 2020-04-10 | 国网河南省电力公司检修公司 | High-voltage substation panoramic monitoring method based on remote sensing image technology |
| CN113673283A (en) * | 2020-05-14 | 2021-11-19 | 惟亚(上海)数字科技有限公司 | Smooth tracking method based on augmented reality |
| CN113724177B (en) * | 2021-09-07 | 2023-12-15 | 北京大学深圳医院 | Lung nodule information fusion method, device, equipment and storage medium thereof |
| CN114373153B (en) * | 2022-01-12 | 2022-12-27 | 北京拙河科技有限公司 | Video imaging optimization system and method based on multi-scale array camera |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1901228A1 (en) * | 2005-06-08 | 2008-03-19 | Fujitsu Ltd. | Image processor |
| CN101345843A (en) * | 2008-08-28 | 2009-01-14 | 中兴通讯股份有限公司 | Method and system for implementing full view video of visible mobile terminal |
| CN101877140A (en) * | 2009-12-18 | 2010-11-03 | 北京邮电大学 | Panorama-based panoramic virtual tour method |
-
2012
- 2012-08-23 CN CN201210303832.1A patent/CN102855649B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1901228A1 (en) * | 2005-06-08 | 2008-03-19 | Fujitsu Ltd. | Image processor |
| CN101345843A (en) * | 2008-08-28 | 2009-01-14 | 中兴通讯股份有限公司 | Method and system for implementing full view video of visible mobile terminal |
| CN101877140A (en) * | 2009-12-18 | 2010-11-03 | 北京邮电大学 | Panorama-based panoramic virtual tour method |
Non-Patent Citations (2)
| Title |
|---|
| Orb: an efficient alternative to sift or surf;Ethan Rublee等;《Computer Vision-ICCV》;20111231;第1-8页 * |
| 基于特征点的全景图生成技术的研究;杨红喆;《中国优秀硕士学位论文全文数据库》;20090131;第9,10,23-29,37页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102855649A (en) | 2013-01-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102855649B (en) | Method for splicing high-definition image panorama of high-pressure rod tower on basis of ORB (Object Request Broker) feature point | |
| Yin et al. | FD-SSD: An improved SSD object detection algorithm based on feature fusion and dilated convolution | |
| CN104200461B (en) | The remote sensing image registration method of block and sift features is selected based on mutual information image | |
| US10621446B2 (en) | Handling perspective magnification in optical flow processing | |
| CN103034982B (en) | Image super-resolution rebuilding method based on variable focal length video sequence | |
| JP2019514123A (en) | Remote determination of the quantity stored in containers in geographical areas | |
| CN102859535A (en) | Daisy descriptor generation from precomputed scale-space | |
| CN106991695A (en) | A kind of method for registering images and device | |
| CN103167218A (en) | Super-resolution reconstruction method and equipment based on non-locality | |
| Li et al. | Bifnet: Bidirectional fusion network for road segmentation | |
| Wu et al. | Remote sensing image super-resolution via saliency-guided feedback GANs | |
| CN109242796A (en) | Character image processing method, device, electronic equipment and computer storage medium | |
| CN111105452A (en) | High-low resolution fusion stereo matching method based on binocular vision | |
| CN114612575A (en) | Camera parameter calibration and three-dimensional data generation method and system | |
| CN114677479A (en) | Natural landscape multi-view three-dimensional reconstruction method based on deep learning | |
| CN109242854A (en) | A kind of image significance detection method based on FLIC super-pixel segmentation | |
| CN107944471A (en) | A kind of ORB characteristic point matching methods based on Nonlinear Scale Space Theory | |
| CN105678307B (en) | Mobile terminal real-time characteristic based on FAST-SURF detects matching process | |
| CN117788296A (en) | Infrared remote sensing image super-resolution reconstruction method based on heterogeneous combined depth network | |
| Dehnavi et al. | Cost and power efficient FPGA based stereo vision system using directional graph transform | |
| CN105631849B (en) | The change detecting method and device of target polygon | |
| CN105894494A (en) | Three-line array stereo aerial surveying camera parallel spectral band registration method based on GPU technology | |
| Yang et al. | Remote Sensing Image Object Detection Based on Improved YOLOv3 in Deep Learning Environment | |
| CN105528772A (en) | Image fusion method based on guidance filtering | |
| CN112651351B (en) | Data processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 250002, No. 1, South Second Ring Road, Shizhong District, Shandong, Ji'nan Co-patentee after: State Grid Corporation of China Patentee after: Electric Power Research Institute of State Grid Shandong Electric Power Company Address before: 250002, No. 1, South Second Ring Road, Shizhong District, Shandong, Ji'nan Co-patentee before: State Grid Corporation of China Patentee before: Electric Power Research Institute of Shandong Electric Power Corporation |
|
| CP01 | Change in the name or title of a patent holder | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20130102 Assignee: National Network Intelligent Technology Co., Ltd. Assignor: Electric Power Research Institute of State Grid Shandong Electric Power Company Contract record no.: X2019370000006 Denomination of invention: Method for splicing high-definition image panorama of high-pressure rod tower on basis of ORB (Object Request Broker) feature point Granted publication date: 20150715 License type: Exclusive License Record date: 20191014 |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20201029 Address after: 250101 Electric Power Intelligent Robot Production Project 101 in Jinan City, Shandong Province, South of Feiyue Avenue and East of No. 26 Road (ICT Industrial Park) Patentee after: National Network Intelligent Technology Co.,Ltd. Address before: 250002, No. 1, South Second Ring Road, Shizhong District, Shandong, Ji'nan Patentee before: ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER Co. Patentee before: STATE GRID CORPORATION OF CHINA |
|
| EC01 | Cancellation of recordation of patent licensing contract | ||
| EC01 | Cancellation of recordation of patent licensing contract |
Assignee: National Network Intelligent Technology Co.,Ltd. Assignor: ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER Co. Contract record no.: X2019370000006 Date of cancellation: 20210324 |