CN104881841B - Aerial high-voltage power tower image splicing method based on edge features and point features - Google Patents
Aerial high-voltage power tower image splicing method based on edge features and point features Download PDFInfo
- Publication number
- CN104881841B CN104881841B CN201510263448.7A CN201510263448A CN104881841B CN 104881841 B CN104881841 B CN 104881841B CN 201510263448 A CN201510263448 A CN 201510263448A CN 104881841 B CN104881841 B CN 104881841B
- Authority
- CN
- China
- Prior art keywords
- image
- images
- point
- spliced
- feature
- 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 26
- 230000009466 transformation Effects 0.000 claims abstract description 23
- 230000004927 fusion Effects 0.000 claims abstract description 10
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims abstract description 5
- 238000007500 overflow downdraw method Methods 0.000 claims abstract description 4
- 230000007704 transition Effects 0.000 claims abstract description 4
- 230000005611 electricity Effects 0.000 claims description 16
- 238000000605 extraction Methods 0.000 claims description 12
- 239000000284 extract Substances 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 8
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000003709 image segmentation Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 description 4
- VMXUWOKSQNHOCA-UKTHLTGXSA-N ranitidine Chemical compound [O-][N+](=O)\C=C(/NC)NCCSCC1=CC=C(CN(C)C)O1 VMXUWOKSQNHOCA-UKTHLTGXSA-N 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 208000025274 Lightning injury Diseases 0.000 description 1
- 241000607479 Yersinia pestis Species 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910002056 binary alloy Inorganic materials 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 238000002790 cross-validation Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003647 oxidation Effects 0.000 description 1
- 238000007254 oxidation reaction Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention provides an aerial photography high-voltage power tower image splicing method based on edge features and point features, which comprises the following steps of: A. inputting two images to be spliced, and removing the backgrounds of the two images to be spliced; B. extracting edge characteristics, and extracting a multi-scale tower framework; C. roughly matching the images to obtain the approximate corresponding relation of the images to be spliced, and acquiring the approximate overlapping area of the two images on the basis of the roughly matched images; D. c, extracting surf characteristic points in the overlapping area of the step C; E. c, matching surf characteristic points under the constraint of the approximate corresponding relation to realize homonymy point matching; F. calculating by using surf homonymy points obtained in the step E through a least square algorithm to obtain image affine transformation parameters, and accurately registering the images; G. seamless splicing of the registered images is realized by using a linear fusion method, and the images are multiplied by corresponding weight functions respectively, so that the image overlapping areas realize smooth transition, and image fusion is realized; H. and generating a splicing map.
Description
Technical field
The present invention relates to a kind of high-tension electricity pylon image split-joint method of taking photo by plane based on edge feature and point feature.
Background technique
Power equipment is subjected to the work of normal mechanical load and electric load due to being chronically exposed in external environment
With, but also often encroached on by external force such as filth, lightning stroke, high wind, sleet and bird pests, multiple factors promote various elements on route
Aging, fatigue, oxidation and corrosion such as find and eliminate not in time, may develop into various failures, to the safety of electric system
It constitutes a threat to stabilization, and the main component of transmission line of electricity is all located on electric force pole tower, thus, the inspection and maintenance of shaft tower are electricity
One working contents of power equipment routing inspection.
With the fast development of aerial remote sens ing technique, electric power relevant departments have been widely used in Electric Power Patrol and maintenance,
But since the video camera for remote sensing monitoring is limited by technology at that time, the regional area of shaft tower can only be shot, shadow
The tour effect of image height high-tension electricity pylon, to obtain the panorama sketch of entire shaft tower, it is necessary to by image mosaic technology, will clap
Multiple images taken the photograph synthesize the complete panorama remote sensing images of a width by registration and fusion.
Currently, scholars have proposed a variety of image split-joint methods, most of method is the image mosaic based on point feature, king
State east etc. proposes a kind of fast algorithm of suitable aerial image splicing, using ORB characteristic point as matching characteristic, with binary system spy
It levies vector and carries out characteristic distance calculating, greatly improve feature extraction with characteristic matching speed.In process of image registration, use
Secondary neighbour's filter algorithm, cross validation algorithm and RANSAC algorithm for estimating, robustly calculate between stitching image sequence
Homography matrix.After image registration, the different image in same pixel position still has certain misalignment, by fusion
Picture position weighting, using improved α-hybrid algorithm, is included in calculating for image edge location information, enables image natural
Fusion, solves the problems, such as the edge slot of image mosaic.Gu takes bravely by utilizing Harris operator extraction feature, and utilizes feature
The main Gradient direction information of point peripheral region pixel, more accurately carries out Feature Points Matching, eliminates pseudo- matching double points, also to containing
Perspective and the image of scale transformation are analyzed, it is proposed that a kind of algorithm based on affine iteration is used for such characteristics of image
Point set is matched, but the algorithm accuracy is not enough, needs to be advanced optimized and improve.
Joining method development for high-tension electricity pylon image of taking photo by plane is started late, and is lack of pertinence, Chinese patent Shen
It 201210303832 please propose a kind of clear image panorama joining method of the high pressure stem tower height based on ORB characteristic point, read superelevation
High resolution compression bar tower image is sampled ultrahigh resolution image to be spliced using bilinear interpolation having a size of W × H
It reduces, obtains w × h image, wherein W, H, w, h are the integer greater than 0, and k is greater than 0 integer；Using ORB algorithm to all samplings
Image carries out feature extraction afterwards；Extracted ORB feature is slightly matched；The matching double points extracted using upper step, original
ORB feature is extracted in image block where the matching double points of ultrahigh resolution image again, is accurately matched；Pass through institute above
The matching double points asked calculate the transformation matrix H between adjacent image；The seamless spliced of ultrahigh resolution image is realized, is reduced
The time required to splicing, splicing efficiency is improved, especially has good beneficial effect to high-definition image.
The defect of above-mentioned technical proposal is, in traditional images stitching algorithm, in the process of image registration based on characteristic point
In the extraction and matching of characteristic point are directly carried out usually within the scope of entire image, increase algorithm calculation amount, and easily cause of the same name
Point error hiding, influences registration effect；In addition, existing high-tension electricity pylon image split-joint method does not consider image mesohigh power tower
The line feature and globality characteristic of frame, only from traditional sense to image carry out feature point extraction with match, pass through same place
It calculates transformation parameter and realizes image registration and splicing.
Summary of the invention
In view of the shortcomings of the prior art, the object of the present invention is to provide a kind of height of taking photo by plane based on edge feature and point feature
High-tension electricity pylon image split-joint method.
To achieve the goals above, the present invention provides a kind of high-voltage power tower of taking photo by plane based on edge feature and point feature
Frame image split-joint method, the image split-joint method include the following steps:
A, two images to be spliced are inputted, and remove two image backgrounds to be spliced；Based on high-tension electricity pylon and image
Middle major part atural object has obvious color difference (specific judgment basis) this feature, by setpoint color threshold value in image
Other background atural objects in addition to shaft tower are removed.
B, edge feature is extracted, multiple dimensioned shaft tower skeleton is extracted；
C, image is slightly matched, obtain image to be spliced is generally corresponding to relationship, the image basis after thick matching
The upper substantially overlapping region for obtaining two width images；
D, surf feature point extraction is carried out in the overlapping region of step C；
E, under being generally corresponding to the constraint of relationship described in the step C, the matching of surf characteristic point is carried out, realizes same place
Match；
F, image affine transformation parameter is calculated by least-squares algorithm in the surf same place obtained using step E, right
Image carries out accuracy registration；
G, image is seamless spliced after being registrated using the realization of linear fusion method, respectively by image multiplied by corresponding weight letter
Number makes image overlapping region realize gentle transition, realizes image co-registration；
H, spliced map is generated.
The present invention treats stitching image using edge feature and is slightly matched, and obtains substantially overlapping region and the correspondence of image
Relationship, subsequent characteristics point extracts and matching carries out under overlay region and its corresponding relationship constraint, can reduce same place error hiding
Situation.
The present invention considers the design feature of image mesohigh electric tower, by extracting shaft tower skeleton image, guarantees bar
The connectivity and topological relation integrality of tower are used for images match with better Shandong for skeleton extract result as edge feature
Stick.
Another specific embodiment according to the present invention in step B, extracts multiple dimensioned shaft tower skeleton based on wavelet transformation.
Another specific embodiment according to the present invention, step B include the following steps:
B1, wavelet transformation treat stitching image based on Harr wavelet function and carry out 3 layers of wavelet decomposition respectively；
B2, image segmentation, three layers of low-frequency approximation subgraph corresponding with its to raw video to be spliced carry out otsu threshold respectively
Value segmentation, obtains the bianry image comprising shaft tower；
B3, skeleton extract carry out the skeleton extract based on Mathematical Morphology to bianry image；
B4, the fusion of multiple dimensioned skeleton, carry out Multiscale Fusion to extracted skeleton image, obtain the complete of image to be spliced
Skeleton image；
B5, straight-line detection；
B6, shaft tower skeleton drawing is obtained.
Specifically, for example, being obtained to guarantee the connectivity of shaft tower extraction and the integrality of topological relation using wavelet transformation
To multiresolution image, ostu image segmentation is carried out to the low-frequency image of each scale, obtains shaft tower bianry image, and to two-value
Image carries out the skeleton extract based on mathematical morphology, and multiple dimensioned skeleton image is merged, and obtains final shaft tower skeleton
Image, the pixel in 3*3 neighborhood on scale s+1 are the associated domains of local model maximum value point (i, j) on scale s, are defined as
Fs,s+1(i, j), the collection of local model maximum value point is combined into M on scale s, uses Cs,s+1(i, j) indicates s spatial point (i, j) and s+1
The correlation in space, calculation formula are as follows:
Wherein,WithFor scale s, the gradient direction of the upper maximum point (i, j) of s+1, α is for direction difference
The threshold value of setting, if Cs,s+1(i, j)=1 then illustrates that maximum point (i, j) is related to maximum point on scale s+1 on scale
Connection, edge transfer need to be carried out, is not otherwise needed.
Another specific embodiment according to the present invention, in step C, for the linear feature of high-tension electricity pylon, to step B
Obtained in shaft tower skeleton image, carry out quick line segment detection using LSD algorithm, reject non-rectilinear sections in image, obtain
To shaft tower part as edge feature.
Another specific embodiment according to the present invention, step D specifically comprise the following steps:
D1, scale space building；
D2, extreme point detection；
D3, the building of feature point description.
Another specific embodiment according to the present invention, step E specifically comprise the following steps:
The first matched transformation parameter of E1, basis and wherein characteristic point coordinate in width image to be spelled, calculate characteristic point another
Corresponding position in one width image；
The matching same place of search this feature point in E2, certain contiguous range in the position；
E3, it is rejected using Ransac algorithm without match point, obtains accurate corresponding dot pair.
Compared with prior art, the present invention have it is following the utility model has the advantages that
Image split-joint method of the prior art based on point feature mostly carries out feature point extraction or benefit directly in whole picture image
Have geographic coordinate information with image and determine image overlay region, the present invention carries out preliminary matches to image using edge feature, obtains
Image overlay region and corresponding relationship are taken, suitable for the image mosaic of no geographical coordinate, and improves image mosaic precision.For high pressure
The distinguishing feature of electric tower, by obtaining the integral skeleton structure of shaft tower based on the multiple dimensioned skeletal extraction of wavelet transformation, and will
It is used for the image registration based on edge feature, and the joining method is made preferably to be suitable for shaft tower image mosaic, improves splicing
Precision.
The present invention is described in further detail with reference to the accompanying drawing.
Detailed description of the invention
Fig. 1 is the flow chart of the high-tension electricity pylon image mosaic of embodiment 1.
Specific embodiment
Embodiment 1
A kind of high-tension electricity pylon image split-joint method of taking photo by plane based on edge feature and point feature is present embodiments provided,
As shown in Figure 1, it includes the following steps:
A, two images to be spliced are inputted, and remove two image backgrounds to be spliced: the color based on high-tension electricity pylon
Feature, the gray threshold by setting each color component with shaft tower will there is the atural object of obvious color difference to go in image
It removes, extracts generated interference to reduce background atural object to subsequent shaft tower.
First, it will be seen that light image switchs to hsv color space by RGB color, utilizes the H color in hsv color space
Component given threshold carries out background rejecting, and specific threshold value setting is as follows:
Wherein, fRGB(x, y) is pixel value of the image at the position (x, y) under rgb space, gH(x, y) is under HSV space
H component value of the image at the position (x, y).
B, the multiple dimensioned shaft tower skeleton extract based on wavelet transformation: based on Harr wavelet function treat stitching image respectively into
3 layers of wavelet decomposition of row, wavelet transform is defined as:Wherein, m.n ∈ Z,For from
Wavelet function is dissipated, is metψ (k) is the wavelet basis letter for meeting wavelet transformation constraint condition
Number harr, is defined asThe wavelet transformation of image need to expand to one-dimensional wavelet transform
Two-dimensional discrete wavelet conversion.
Three layers of low-frequency approximation subgraph corresponding with its to raw video to be spliced carries out otsu Threshold segmentation respectively, is wrapped
Bianry image containing shaft tower carries out the skeleton extract based on Mathematical Morphology to bianry image, carries out to extracted skeleton image more
Scale fusion, obtains the complete skeleton image of image to be spliced, judges figure using the spatial coherence of spatial point on adjacent scale
As the relevance of maximum point, the pixel in 3*3 neighborhood on scale s+1 is the pass of local model maximum value point (i, j) on scale s
Join domain, is defined as Fs,s+1(i, j), the collection of local model maximum value point is combined into M on scale s, uses Cs,s+1(i, j) indicates s spatial point
The correlation of (i, j) and the space s+1, calculation formula are as follows:
Wherein,WithFor scale s, the gradient direction of the upper maximum point (i, j) of s+1, α is for direction difference
The threshold value of setting, if Cs,s+1(i, j)=1 then illustrates that maximum point (i, j) is related to maximum point on scale s+1 on scale
Connection, edge transfer need to be carried out, is not otherwise needed.
C, the image based on edge feature slightly matches side: for the linear feature of high-tension electricity pylon, to being obtained in step B
The shaft tower skeleton image obtained carries out quick line segment detection using LSD algorithm, rejects non-rectilinear sections in image, obtain shaft tower
Part is used as edge feature.Edge feature image preliminary matches are carried out using phase correlation method, edge image are carried out respectively fast
Fast Fourier transformation, obtains F1(u, v), F2(u, v) calculates crosspower spectrum
Coordinate at Fourier transformation of inverting amplitude maximum, as image translation amount are flat by calculating under polar coordinates by transforming to image
Shifting amount identical process calculates rotation parameter, by the way that image after Fourier transformation is converted to logarithmic form, calculates scale factor, obtains
To the transformation parameter of image, images match is carried out.
D, Surf feature point extraction: obtaining the substantially overlapping region of two width images on the basis of the image after slightly matching,
Surf feature point extraction is carried out in the overlapping region, mainly includes scale space building, extreme point detection, the sub- structure of feature point description
Build several processes.
E, the homotopy mapping under range constraint: the big of available image to be spliced is slightly matched by the image of step C
Corresponding relationship is caused, under this relation constraint, carries out the matching of surf characteristic point.Firstly, according to first matched transformation parameter and its
In characteristic point coordinate in width image to be spelled, calculate corresponding position of the characteristic point in another width image；Then, in the position
The matching same place of search this feature point in certain contiguous range；Finally, rejecting using Ransac algorithm without match point, standard is obtained
True corresponding dot pair.
F, image essence is registrated: using obtained surf same place, image affine transformation is calculated by least-squares algorithm
Parameter carries out accuracy registration to image.
G, visual fusion: image is seamless spliced after being registrated using the realization of linear fusion method, respectively by image multiplied by phase
The weighting function answered makes image overlapping region realize gentle transition.
H, spliced map is generated.
Although the present invention is disclosed above in the preferred embodiment, it is not intended to limit the invention the range of implementation.Any
The those of ordinary skill in field is not departing from invention scope of the invention, improves when can make a little, i.e., all according to this hair
Bright done same improvement, should be the scope of the present invention and is covered.
Claims (3)
1. a kind of high-tension electricity pylon image split-joint method of taking photo by plane based on edge feature and point feature, which is characterized in that described
Image split-joint method includes the following steps:
A, two images to be spliced are inputted, and remove the background of two images to be spliced；
B, edge feature is extracted, multiple dimensioned shaft tower skeleton is extracted based on wavelet transformation；The step B includes the following steps:
B1, wavelet transformation treat stitching image based on Harr wavelet function and carry out 3 layers of wavelet decomposition respectively；
B2, image segmentation, three layers of low-frequency approximation subgraph corresponding with its to raw video to be spliced carry out otsu threshold value point respectively
It cuts, obtains the bianry image comprising shaft tower；
B3, skeleton extract carry out the skeleton extract based on Mathematical Morphology to bianry image；
B4, the fusion of multiple dimensioned skeleton, carry out Multiscale Fusion to extracted skeleton image, obtain the complete skeleton of image to be spliced
Image；
B5, straight-line detection；
B6, shaft tower skeleton drawing is obtained；
C, just matching is carried out to image, obtain image to be spliced is generally corresponding to relationship, obtains on the basis of the image after just matching
Take the substantially overlapping region of two width images；Wherein, for the linear feature of high-tension electricity pylon, to bar obtained in step B
Tower skeleton image carries out quick line segment detection using LSD algorithm, rejects non-rectilinear sections in image, obtains shaft tower part work
For edge feature；
D, surf feature point extraction is carried out in the overlapping region of step C；
E, under being generally corresponding to the constraint of relationship described in the step C, the matching of surf characteristic point is carried out, realizes homotopy mapping；
F, image affine transformation parameter is calculated by least-squares algorithm, to image in the surf same place obtained using step E
Carry out accuracy registration；
G, the seamless spliced of image after registration is realized using linear fusion method, respectively by image multiplied by corresponding weighting function,
So that image overlapping region is realized gentle transition, realizes image co-registration；
H, spliced map is generated.
2. image split-joint method according to claim 1, which is characterized in that the step D specifically comprises the following steps:
D1, scale space building；
D2, extreme point detection；
D3, the building of feature point description.
3. image split-joint method according to claim 1, which is characterized in that the step E specifically comprises the following steps:
The first matched transformation parameter of E1, basis and wherein characteristic point coordinate in width image to be spelled, calculate characteristic point in another width
Corresponding position in image；
The matching same place of search this feature point in E2, certain contiguous range in the position；
E3, it is rejected using Ransac algorithm without match point, obtains accurate corresponding dot pair.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510263448.7A CN104881841B (en) | 2015-05-20 | 2015-05-20 | Aerial high-voltage power tower image splicing method based on edge features and point features |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510263448.7A CN104881841B (en) | 2015-05-20 | 2015-05-20 | Aerial high-voltage power tower image splicing method based on edge features and point features |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104881841A CN104881841A (en) | 2015-09-02 |
| CN104881841B true CN104881841B (en) | 2019-01-04 |
Family
ID=53949327
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510263448.7A Active CN104881841B (en) | 2015-05-20 | 2015-05-20 | Aerial high-voltage power tower image splicing method based on edge features and point features |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104881841B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105357433B (en) * | 2015-10-13 | 2018-12-07 | 哈尔滨工程大学 | A kind of adaptive method for panoramic imaging of high speed rotation focal length |
| CN106228601B (en) * | 2016-07-21 | 2019-08-06 | 山东大学 | Multiple dimensioned pyramidal CT image quick three-dimensional reconstructing method based on wavelet transformation |
| CN108230376B (en) * | 2016-12-30 | 2021-03-26 | 北京市商汤科技开发有限公司 | Remote sensing image processing method and device and electronic equipment |
| CN107767410A (en) * | 2017-10-27 | 2018-03-06 | 中国电子科技集团公司第三研究所 | The multi-band image method for registering of the multispectral system acquisition of polyphaser parallel optical axis |
| CN108537833B (en) * | 2018-04-18 | 2022-06-21 | 昆明物理研究所 | Infrared image rapid splicing method |
| CN109308715A (en) * | 2018-09-19 | 2019-02-05 | 电子科技大学 | A kind of optical imagery method for registering combined based on point feature and line feature |
| CN109584222B (en) * | 2018-11-19 | 2023-05-16 | 国网江西省电力有限公司电力科学研究院 | Unmanned aerial vehicle-based fault classification and identification method for photovoltaic module image |
| CN109993718B (en) * | 2019-03-05 | 2021-04-13 | 北京当红齐天国际文化科技发展集团有限公司 | Multi-channel projection image fusion method and device |
| CN110728176B (en) * | 2019-08-30 | 2022-11-11 | 长安大学 | Unmanned aerial vehicle visual image feature rapid matching and extracting method and device |
| WO2021230314A1 (en) * | 2020-05-14 | 2021-11-18 | 国立大学法人 東京大学 | Measurement system, vehicle, measurement device, measurement program, and measurement method |
| CN111861889B (en) * | 2020-07-31 | 2023-03-21 | 聚时科技(上海)有限公司 | Automatic splicing method and system for solar cell images based on semantic segmentation |
| CN113033508B (en) * | 2021-05-20 | 2021-08-10 | 成都奥伦达科技有限公司 | Point cloud-based fine routing inspection waypoint rapid generation method |
| CN115293971B (en) * | 2022-09-16 | 2023-02-28 | 荣耀终端有限公司 | Image splicing method and device |
| CN115661453B (en) * | 2022-10-25 | 2023-08-04 | 腾晖科技建筑智能(深圳)有限公司 | Tower crane object detection and segmentation method and system based on downward view camera |
| CN118134758A (en) * | 2024-05-08 | 2024-06-04 | 海南热带海洋学院 | Underwater ultrasonic image stitching method based on WT and SURF algorithm |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101794439A (en) * | 2010-03-04 | 2010-08-04 | 哈尔滨工程大学 | Image splicing method based on edge classification information |
| CN102855649A (en) * | 2012-08-23 | 2013-01-02 | 山东电力集团公司电力科学研究院 | Method for splicing high-definition image panorama of high-pressure rod tower on basis of ORB (Object Request Broker) feature point |
| US8411961B1 (en) * | 2007-08-22 | 2013-04-02 | Adobe Systems Incorporated | Method and apparatus for image feature matching in automatic image stitching |
| CN104361586A (en) * | 2014-10-31 | 2015-02-18 | 国网上海市电力公司 | Electric pole and tower maintenance and pre-warning system based on panoramas |
-
2015
- 2015-05-20 CN CN201510263448.7A patent/CN104881841B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8411961B1 (en) * | 2007-08-22 | 2013-04-02 | Adobe Systems Incorporated | Method and apparatus for image feature matching in automatic image stitching |
| CN101794439A (en) * | 2010-03-04 | 2010-08-04 | 哈尔滨工程大学 | Image splicing method based on edge classification information |
| CN102855649A (en) * | 2012-08-23 | 2013-01-02 | 山东电力集团公司电力科学研究院 | Method for splicing high-definition image panorama of high-pressure rod tower on basis of ORB (Object Request Broker) feature point |
| CN104361586A (en) * | 2014-10-31 | 2015-02-18 | 国网上海市电力公司 | Electric pole and tower maintenance and pre-warning system based on panoramas |
Non-Patent Citations (2)
| Title |
|---|
| 基于显著图的输电线路杆塔图像拼接方法;张旭 等;《计算机应用》;20150410;参见摘要，正文第一二部分 |
| 基于特征的遥感图像拼接技术研究;王慧玲;《中国优秀硕士学位论文全文数据库》;20140531;全文 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104881841A (en) | 2015-09-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104881841B (en) | Aerial high-voltage power tower image splicing method based on edge features and point features | |
| CN105741252B (en) | Video image grade reconstruction method based on rarefaction representation and dictionary learning | |
| CN104318569B (en) | Space salient region extraction method based on depth variation model | |
| CN104966270B (en) | A kind of more image split-joint methods | |
| CN111127318B (en) | Panoramic image splicing method in airport environment | |
| CN104200461B (en) | The remote sensing image registration method of block and sift features is selected based on mutual information image | |
| CN104751465A (en) | ORB (oriented brief) image feature registration method based on LK (Lucas-Kanade) optical flow constraint | |
| CN107818303B (en) | Unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method, system and software memory | |
| CN105069746A (en) | Video real-time human face substitution method and system based on partial affine and color transfer technology | |
| CN109558806A (en) | The detection method and system of high score Remote Sensing Imagery Change | |
| CN107680140B (en) | Depth image high-resolution reconstruction method based on Kinect camera | |
| CN106504192B (en) | A kind of power transmission line corridor geological disaster exploration image treatment method and system | |
| CN109376641B (en) | Moving vehicle detection method based on unmanned aerial vehicle aerial video | |
| CN105426825B (en) | A kind of power grid geographical wiring diagram method for drafting based on Aerial Images identification | |
| CN108665468B (en) | Device and method for extracting tangent tower insulator string | |
| CN109087245A (en) | Unmanned aerial vehicle remote sensing image mosaic system based on neighbouring relations model | |
| CN116681636B (en) | Light infrared and visible light image fusion method based on convolutional neural network | |
| CN103793894A (en) | Cloud model cellular automata corner detection-based substation remote viewing image splicing method | |
| CN103679740B (en) | ROI (Region of Interest) extraction method of ground target of unmanned aerial vehicle | |
| CN105550981A (en) | Image registration and splicing method on the basis of Lucas-Kanade algorithm | |
| CN108460833A (en) | A kind of information platform building traditional architecture digital protection and reparation based on BIM | |
| Zhang et al. | Preprocessing and fusion analysis of GF-2 satellite Remote-sensed spatial data | |
| CN111861866A (en) | Panoramic reconstruction method for substation equipment inspection image | |
| CN116228521A (en) | Heterologous image registration and fusion method | |
| CN111652240A (en) | Image local feature detection and description method based on CNN |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |