CN106407983A - Image body identification, correction and registration method - Google Patents
Image body identification, correction and registration method Download PDFInfo
- Publication number
- CN106407983A CN106407983A CN201610818691.5A CN201610818691A CN106407983A CN 106407983 A CN106407983 A CN 106407983A CN 201610818691 A CN201610818691 A CN 201610818691A CN 106407983 A CN106407983 A CN 106407983A
- Authority
- CN
- China
- Prior art keywords
- image
- compartment
- identification
- background
- matrix
- 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 34
- 239000011159 matrix material Substances 0.000 claims abstract description 19
- 230000009466 transformation Effects 0.000 claims abstract description 13
- 230000011218 segmentation Effects 0.000 claims abstract description 7
- 238000001514 detection method Methods 0.000 claims abstract description 6
- 238000000605 extraction Methods 0.000 claims abstract description 6
- 230000001788 irregular Effects 0.000 claims abstract description 4
- 230000008569 process Effects 0.000 claims description 12
- 239000000284 extract Substances 0.000 claims description 4
- 230000003416 augmentation Effects 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims 1
- 230000003190 augmentative effect Effects 0.000 abstract 1
- 230000009286 beneficial effect Effects 0.000 abstract 1
- 235000013350 formula milk Nutrition 0.000 description 9
- 238000006243 chemical reaction Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 6
- 230000003044 adaptive effect Effects 0.000 description 4
- 238000003708 edge detection Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 238000003909 pattern recognition Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 239000012535 impurity Substances 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000004064 recycling Methods 0.000 description 2
- 206010044565 Tremor Diseases 0.000 description 1
- 239000000729 antidote Substances 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 238000003702 image correction Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/14—Transformations for image registration, e.g. adjusting or mapping for alignment of images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an image body identification, correction and registration method comprising the following steps: target detection: performing threshold segmentation by observing the features of an image, and roughly distinguishing between the foreground and the background; binarization: getting a black-and-white binary image, removing noise points through expansion, and retaining a compartment part in the original image; image feature extraction: positioning the four corners of the compartment based on the mathematical properties of trapezoid; perspective transformation: stretching an irregular convex quadrilateral to a regular rectangle through a transformation matrix; and registration and splicing: combining multiple box body images, and for the images with common image features, registering and splicing multiple box body images on the side by use of an Opencv open-source library Stitcher class or augmenting the images based on the properties of the matrix. The image body identification, correction and registration method is beneficial to eliminating background interference, reducing the sample set of box body identification and improving the efficiency of identification.
Description
Technical field
The invention belongs to image processing techniques, particularly a kind of image subject based on characteristics of image and pixel distribution pattern
Identification, rectification and method for registering.
Background technology
As mathematics, physics, geography is architectonic constantly complete and satellite technology, computer technology, internet
The progress in the multiple field such as technology, image document gets more and more, and is related to the process of image, the technology of storage and retrieval is to count greatly
Sharp weapon according to the epoch.The identification of image subject, correct with method for registering be widely used in geography, navigation etc. have scattered in a large number
In the field of image data.In traffic and transport field, the casing identification of harbour container can effectively improve conevying efficiency, reduce into
This, have great economic benefit.And the identification of image subject, rectification are conducive to excluding ambient interferences with method for registering, reduce
The sample set of casing identification, thus improve recognition efficiency.
The identification of image subject, rectification are by the body matter nothing to do with background separation in picture, rectification with method for registering
Lens distortion, the picture Registration and connection that multiple contain same characteristic features become pictures.The method relate generally to IMAQ and
The steps such as classification, image segmentation, image feature value extraction, image conversion, image registration and splicing.Wherein key issue is edge
Detection and image feature value extract part.
At present in industrial circle, pattern-recognition is based substantially on original image and is directly strengthened, clusters, identifies, but right
The less picture of target prospect ratio, it is impossible to foreground and background is distinguished in effective identification, increases the error processing in calculating process,
Reduce accuracy rate and the success rate of graphical analysis.Propose on the basis of herein first to carry out target prospect to original image to carry
Take, correct and project, eliminate the error impact that extraneous background is constituted on image procossing, effectively improve the accuracy rate of analysis identification,
Make the intermediate image in interactive process more visual and clear simultaneously, strengthen interactive experience.
There are statistical-simulation spectrometry, configuration mode identification, Fuzzy Pattern Recognition at present in terms of image recognition.Image segmentation is
A key technology in image procossing.The big Tianjin of Japanese scholars in 1979 proposes based on gradation of image characteristic, divides the image into
The two-part adaptive thresholding algorithm of target and background (Otsu N.A threshold selection method from
gray-level histograms[J].Automatica,1975,11(285-296):23-27.).California in 1986
Professor Canny of university proposes a kind of multistage edge detection algorithm, in the case of retaining original image attributes, substantially reduces
Data scale (Canny J.A computational approach to edge detection [J] .IEEE of image
Transactions on pattern analysis and machine intelligence,1986(6):679-698.).
Image flame detection aspect mainly applies to the interpolation technique in numerical analysis method, and the analysis geometric distortion of Zhao Qingpeng and Ma East China produces
The reason, propose a kind of adaptive geometric fault image correction solution in conjunction with numerical analysis method, experiment demonstrates its algorithm
Validity (Zhao Qingpeng, horse East China. adaptive geometric fault image antidote study [C] [the C] // 3rd man-machine ring of harmony
Border joint academic conference (HHME2007) collection of thesis .2007,2007.).In terms of image registration, Flussr is to fault image
Registration problems propose self organizing maps method, improve registration accuracy rate (Flusser J.An adaptive
method for image registration[J].Pattern Recognition,1992,25(1):45-54.).
Content of the invention
It is an object of the invention to provide a kind of identification of image subject, rectification and method for registering, the casing to container
It is identified, rectangle is corrected and Registration and connection, be the casing digital and damaged identification exclusion extraneous background interference in later stage, reduce sample
This collection, improves recognition efficiency.
The technical solution realizing the object of the invention is：A kind of identification of image subject, rectification and method for registering, step
As follows：
The first step, target detection, artwork extracts the region in compartment.For most of compartment be all red, blue,
Orange, image is transformed into HSV space, then using tone H feature, foreground and background is substantially made a distinction.Then utilize color
Adjust histogram, Otsu algorithm Threshold segmentation to separate prospect compartment and background, obtain bianry image, then removed by expansion process
Some noises.Retain car parts in artwork finally according to bianry image；
Second step, image characteristics extraction, orient the position at four angles in compartment；To black background part and white compartment portion
Divide and be entered as 0-1 matrix, then the distribution characteristics according to white car parts makes positioning to black and white border, then confirms four
The position at angle；
3rd step, perspective transform, irregular convex quadrangle is stretched to rectangle.Solve transformation matrix, its function is
Arbitrary quadrilateral and foursquare mutual conversion, then first can be converted into square by quadrangle, then be converted to square from square
Shape；
4th step, Registration and connection, multiple casing pictures are combined.Completed using Opencv storehouse Stitcher class of increasing income
The splicing Registration and connection of multiple casing figures of side.For the photo of not coplanar, that is, scaling is to highly identical, recycling matrix properties
By picture augmentation.
Present invention incorporates existing image processing techniques and algorithm, pointedly container body picture is carried out pre-
Process, advantage has：(1) speed ability is good.Rectification and the splicing of plurality of pictures can be processed rapidly.(2) picture number does not limit.Theoretical
On can realize any pictures rectification splicing.(3) algorithm accuracy rate is high.In to field of traffic, container body picture is rectified
Just splicing in experiment, 65% sample can accomplish the casing side after intercepting perpendicular to high and bottom, and inclination angle is about 90 °, 20%
Sample intercept rear box side and slanted floor angle and be about 85 °, only 5% sample can block former quadrangle, makes design sketch not
Completely.To sum up, this algorithm can effectively recover the linear processes distortion of picture, and picture can be carried out effective registration
Splicing.
Brief description
Fig. 1 is the identification based on characteristics of image and the image subject of pixel distribution pattern for the present invention, corrects and registering flow process
Figure
Fig. 2 be artwork and artwork through using hue histogram, Otsu algorithm Threshold segmentation by prospect compartment and background
Separately, the bianry image obtaining.
Fig. 3 is the binaryzation picture after expansion process, and a large amount of noises are eliminated.
Fig. 4 is that quadrangle transforms to foursquare transformation matrix principle schematic.
Fig. 5 is the result figure through matrixing rear box for the container picture.
Fig. 6 is that have the result figure that the picture Registration and connection of public characteristic obtains to two.
Specific embodiment
Below in conjunction with the accompanying drawings the present invention is described in further detail.
In conjunction with Fig. 1, the identification based on characteristics of image and the image subject of pixel distribution pattern for the present invention, correct and registering side
Method, step is as follows：
The first step, target detection, artwork extracts the region in compartment.For most of compartment be all red, blue,
Orange, image is transformed into HSV space, then using tone H feature, foreground and background is substantially made a distinction；
(1) image rgb space vector (r, g, b) is transformed into HSV space vector (h, s, v), then by the color of HSV space
Adjust H vector to be stripped out, form one-dimensional gray scale vector, using formula
(2) and then using hue histogram, Otsu algorithm Threshold segmentation prospect compartment and background are separated, obtain binary map
Picture；For image I (x, y), the segmentation threshold of foreground and background is denoted as T, and the pixel number belonging to prospect accounts for the ratio of entire image
Example is designated as ω0, its average gray μ0；The ratio that background pixel points account for entire image is ω1, its average gray is μ1；Image
Overall average gray scale is designated as μ, and inter-class variance is designated as g；Assume that the background of image is dark, and the size of image is M × N, in image
The number of pixels that the gray value of pixel is less than threshold value T is denoted as N0, pixel grey scale be more than threshold value T number of pixels be denoted as N1, then have：
N0+N1=M × N
ω0+ω1=1
μ=ω0×μ0+ω1×μ1
G=ω0(μ0-μ)2+ω1(μ1-μ)2
Penultimate formula is substituted into last formula, obtains equivalence formula：G=ω0ω1(μ0-μ1)2；Here it is
Inter-class variance；Obtain making maximum threshold value T of inter-class variance g using the method for traversal, as required；The binary image obtaining is such as
Fig. 2, it can be seen that white point is full of in black background, has a large amount of noises；
(3) again some noises are removed by expansion process, expansion algorithm is the structural element with 3 × 3, scan binary map
Each pixel of picture, does AND operation with structural element with the bianry image that it covers, if being all 0, structural images should
Pixel is 0；Otherwise 1, result：Bianry image is made to expand a circle；Picture such as Fig. 3 after expansion, a large amount of noises are eliminated；
Second step, image characteristics extraction, orient the position at four angles in compartment；
Compartment is similar to trapezoidal, but because the interference of yellow line, the shakiness at edge on the impure, track of background in picture
Qualitative so that in image compartment the border of redundancy occurs；
(1) obtain the 0-1 matrix of n row m row；Because the bianry image that Otsu algorithm obtains can only ensure prospect and background phase
Different to value, but it cannot be guaranteed that absolute value；Therefore take picture central region to be judged, allow compartment be located region in matrix
It is worth for 1, in other regions, matrix value is 0；Compartment extension occurs that impurity is designated as 1, have in background simultaneously the region of light tone also by
It is labeled as 1；
(2) determine the bound in compartment；Substantially up big and down small trapezoidal in compartment, often capable addition obtains accumulated value s to matrix.
According to the picture that actually photographed, find that the s value in the upper bound accounts for the 80～90% of m, but upper bound impurity appearing above, s value is about in m
10～30%；Some compartments can assume inclination.Therefore travel through half pictures from top to bottom, occur after s value increases to the 1/3 of m
During transition, that is, think arrival coboundary.Lower bound is obtained by similar method；
(3) determine the end points in compartment；Matrix each column is added and obtains accumulated value r, can be obtained by trapezoidal property, r value is in ideal
In the case of from left to right first increase to maximum from 0, keep constant, reduce to 0 afterwards again；Hardly result in equal difference in actual figure
Sequence, but constant partly can the sampling of middle holding obtains mean value, then by true from the middle position obtaining outward being mutated
Determine left and right circle, four end points can be determined successively；
3rd step, perspective transform, it is to a new view plane by picture projection, irregular convex quadrangle is stretched to
Rectangle；Solve transformation matrix, its function is the mutual conversion of arbitrary quadrilateral and rectangle, then first can be converted into quadrangle
Rectangle；
(1) reduction formula
U, v are original image coordinates, correspondence convert after Picture Coordinate x, y, simultaneously projection scaling multiple for w. its
In
X=x '/w ', y=y '/w '
Transformation matrix
4 parts can be splitted into,
Represent linear transformation, such as bi-directional scaling, cutting and rotation；
[a31a32]
For translating,
[a13a23]T
Produce perspective transform.Transformation for mula before rewriting can obtain：
So it is known that the corresponding several points of conversion can be to ask for transformation for mula.Conversely, specific transformation for mula also can obtain
Picture to after new conversion；Simply see a square to the conversion of quadrangle：
4 groups of corresponding points of conversion can be expressed as：
(0,0) → (x0, y0), (1,0) → (x1, y1), (1,1) → (x2, y2), (0,1) → (x3, y3)
Obtained according to transformation for mula：
a31=x0
a11+a31-a13x1=x1
a11+a21+a31-a13x2-a23x2=x2
a21+a31-a23x3=x3
a32=y0
a12+a32-a13y1=y1
a12+a22+a32-a13y2-a23y2=y2
a22+a32-a23y3=y3
Define several auxiliary variables：
Δx1=x1- x2Δx2=x3-x2Δx3=x0-x1+x2-x3
Δy1=y1-y2Δy2=y3-y2Δy3=y0-y1+y2- y3
Δx3, Δ y3Be all when 0 changing the plane with originally parallel, can obtain：
a11=x1-x0
a21=x2- x1
a31=x0
a12=y1- y0
a22=y2-y1
a32=y0
a13=0
a12=0
Δx3, Δ y3When being not 0, obtain：
a11=x1-x0+a12x1
a2l=x3-x0+a12x2
a31=x0
a12=y1-y0+a13y1
a22=y3-y0+a23y3
a32=y0
One square just can be transformed to quadrangle by the transformation matrix solving；Conversely, quadrangle transforms to pros
Shape is also the same；Then, we pass through to convert twice：Quadrangle transforms to square, then transforms to quadrangle from square
Just any one quadrangle can be transformed to another quadrangle, such as Fig. 4；Experiment effect figure such as Fig. 5；
4th step, Registration and connection, multiple casing pictures are combined；Completed using Opencv storehouse Stitcher class of increasing income
The splicing Registration and connection of multiple casing figures of side；For the photo of not coplanar, that is, scaling is to highly identical, recycling matrix properties
By picture augmentation；Opencv storehouse Stitcher class of increasing income is related to a lot of algorithms, such as：The extraction of characteristic point, characteristic point
Join, image co-registration etc.；These processes are encapsulated in Stitcher class；Program splicing effect such as Fig. 6.
Claims (3)
1. a kind of identification of image subject, correct with method for registering it is characterised in that step is as follows：
The first step, target detection, artwork extracts the region in compartment；For the color in compartment, image is transformed into HSV empty
Between, then using tone H feature, foreground and background is made a distinction；Then hue histogram, Otsu algorithm Threshold segmentation are utilized
Prospect compartment and background are separated, obtains bianry image, then some noises are removed by expansion process, finally according to bianry image
Retain car parts in artwork；
Second step, image characteristics extraction, orient the position at four angles in compartment；Black background part and white car parts are assigned
It is worth for 0-1 matrix, then the distribution characteristics according to white car parts makes positioning to black and white border, then confirms four angles
Position；
3rd step, perspective transform, irregular convex quadrangle is stretched to rectangle, solves transformation matrix, by matrix operation,
First quadrangle is converted into square, then is converted to rectangle from square；
4th step, Registration and connection, multiple casing pictures are combined.Side is completed using Opencv storehouse Stitcher class of increasing income
The splicing Registration and connection of multiple casing figures.For the photo of not coplanar, that is, scaling, to highly identical, recycles matrix properties will scheme
Piece augmentation.
2. the identification of image subject according to claim 1, correct with method for registering it is characterised in that target detection step
As follows：
The first step, image rgb space vector (r, g, b) is transformed into HSV space vector (h, s, v), then by the color of HSV space
Adjust H vector to be stripped out, form one-dimensional gray scale vector, foreground and background is distinguished, using formula
Prospect compartment and background are separated using hue histogram, Otsu algorithm Threshold segmentation, are obtained bianry image by second step,
Again some noises are removed by expansion process；Expansion algorithm is the structural element with N × N, scans bianry image each
Pixel, does AND operation with structural element with the bianry image that it covers, if being all 0, this pixel of structural images is 0；No
Then 1；Make bianry image expand a circle, ultimately form case face binary image.
3. the identification of image subject according to claim 1, correct with method for registering it is characterised in that：Shape after the first step
The casing becoming, formation class in compartment is trapezoidal, and that is, both sides are parallel up and down, the shape that the right and left tilts；According to this situation by every a line
Pixel value add up, then adjacent rows are made the difference line by line, obtain two difference mutated site, be i.e. two upper following positions；Will
The pixel value of each row adds up, and adjacent two row make the difference by column, obtains a difference by just becoming one difference of zero-sum and becomes positive by zero
Position, that is, following two end points；The pixel of each row is added up with given threshold T, less than T and from trapezoidal nearest position
I.e. two end points of top, finally give trapezoidal four end points.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610818691.5A CN106407983A (en) | 2016-09-12 | 2016-09-12 | Image body identification, correction and registration method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610818691.5A CN106407983A (en) | 2016-09-12 | 2016-09-12 | Image body identification, correction and registration method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106407983A true CN106407983A (en) | 2017-02-15 |
Family
ID=57999708
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610818691.5A Pending CN106407983A (en) | 2016-09-12 | 2016-09-12 | Image body identification, correction and registration method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106407983A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108171669A (en) * | 2017-12-29 | 2018-06-15 | 星阵（广州）基因科技有限公司 | A kind of image correction method based on OpenCV algorithms |
| CN109377455A (en) * | 2018-09-27 | 2019-02-22 | 浙江工业大学 | The improved multisequencing magnetic resonance image method for registering based on self-similarity |
| CN109840939A (en) * | 2019-01-08 | 2019-06-04 | 北京达佳互联信息技术有限公司 | Three-dimensional rebuilding method, device, electronic equipment and storage medium |
| CN110717865A (en) * | 2019-09-02 | 2020-01-21 | 苏宁云计算有限公司 | Picture detection method and device |
| CN111259177A (en) * | 2020-01-10 | 2020-06-09 | 深圳盒子信息科技有限公司 | Black-white binary signature picture storage method and system |
| CN111368889A (en) * | 2020-02-26 | 2020-07-03 | 腾讯科技（深圳）有限公司 | Image processing method and device |
| CN112183237A (en) * | 2020-09-10 | 2021-01-05 | 天津大学 | Automatic white blood cell classification method based on color space adaptive threshold segmentation |
| CN112257607A (en) * | 2020-10-23 | 2021-01-22 | 合肥工业大学 | Correction method for processing mobile phone image distortion acquired on production line |
| CN112381084A (en) * | 2020-10-12 | 2021-02-19 | 武汉沃亿生物有限公司 | Automatic contour recognition method for tomographic image |
| CN113377108A (en) * | 2021-06-09 | 2021-09-10 | 乐聚(深圳)机器人技术有限公司 | Method, device, equipment and storage medium for controlling biped robot |
| CN113421297A (en) * | 2021-07-02 | 2021-09-21 | 浙江德菲洛智能机械制造有限公司 | Strawberry shape symmetry analysis method |
| CN113580147A (en) * | 2021-09-02 | 2021-11-02 | 乐聚(深圳)机器人技术有限公司 | Robot control method, device, equipment and storage medium |
| CN114037558A (en) * | 2021-11-05 | 2022-02-11 | 国网上海市电力公司 | Transmission line anticollision early warning system based on multidimensional feature recognition |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040264806A1 (en) * | 2003-06-24 | 2004-12-30 | Microsoft Corporation | System and method for de-noising multiple copies of a signal |
| CN101776952A (en) * | 2010-01-29 | 2010-07-14 | 联动天下科技（大连）有限公司 | Novel interactive projection system |
| CN102902945A (en) * | 2012-09-28 | 2013-01-30 | 南京汇兴博业数字设备有限公司 | Distortion correction method of outer contour based on quick response matrix code |
| US20150379724A1 (en) * | 2014-06-26 | 2015-12-31 | The Governors Of The University Of Alberta | Cardiac segmentation with point correspondence |
-
2016
- 2016-09-12 CN CN201610818691.5A patent/CN106407983A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040264806A1 (en) * | 2003-06-24 | 2004-12-30 | Microsoft Corporation | System and method for de-noising multiple copies of a signal |
| CN101776952A (en) * | 2010-01-29 | 2010-07-14 | 联动天下科技（大连）有限公司 | Novel interactive projection system |
| CN102902945A (en) * | 2012-09-28 | 2013-01-30 | 南京汇兴博业数字设备有限公司 | Distortion correction method of outer contour based on quick response matrix code |
| US20150379724A1 (en) * | 2014-06-26 | 2015-12-31 | The Governors Of The University Of Alberta | Cardiac segmentation with point correspondence |
Non-Patent Citations (3)
| Title |
|---|
| XIAOWEI_CQU: ""【图像处理】透视变换Perspective Transformation"", 《CSDN》 * |
| 贺志明: ""数码相机拍摄的透视文本图像的校正"", 《上海工程技术大学学报》 * |
| 黄大坤等: ""多图无缝拼接的配准算法"", 《微型电脑应用》 * |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108171669A (en) * | 2017-12-29 | 2018-06-15 | 星阵（广州）基因科技有限公司 | A kind of image correction method based on OpenCV algorithms |
| CN109377455B (en) * | 2018-09-27 | 2021-08-03 | 浙江工业大学 | Improved multi-sequence magnetic resonance image registration method based on self-similarity |
| CN109377455A (en) * | 2018-09-27 | 2019-02-22 | 浙江工业大学 | The improved multisequencing magnetic resonance image method for registering based on self-similarity |
| CN109840939A (en) * | 2019-01-08 | 2019-06-04 | 北京达佳互联信息技术有限公司 | Three-dimensional rebuilding method, device, electronic equipment and storage medium |
| CN109840939B (en) * | 2019-01-08 | 2024-01-26 | 北京达佳互联信息技术有限公司 | Three-dimensional reconstruction method, three-dimensional reconstruction device, electronic equipment and storage medium |
| CN110717865B (en) * | 2019-09-02 | 2022-07-29 | 苏宁云计算有限公司 | Picture detection method and device |
| WO2021042823A1 (en) * | 2019-09-02 | 2021-03-11 | 苏宁云计算有限公司 | Picture test method and device |
| CN110717865A (en) * | 2019-09-02 | 2020-01-21 | 苏宁云计算有限公司 | Picture detection method and device |
| CN111259177A (en) * | 2020-01-10 | 2020-06-09 | 深圳盒子信息科技有限公司 | Black-white binary signature picture storage method and system |
| CN111368889A (en) * | 2020-02-26 | 2020-07-03 | 腾讯科技（深圳）有限公司 | Image processing method and device |
| CN111368889B (en) * | 2020-02-26 | 2023-10-17 | 腾讯科技（深圳）有限公司 | Image processing method and device |
| CN112183237A (en) * | 2020-09-10 | 2021-01-05 | 天津大学 | Automatic white blood cell classification method based on color space adaptive threshold segmentation |
| CN112381084A (en) * | 2020-10-12 | 2021-02-19 | 武汉沃亿生物有限公司 | Automatic contour recognition method for tomographic image |
| CN112381084B (en) * | 2020-10-12 | 2024-02-09 | 武汉沃亿生物有限公司 | Automatic contour recognition method for tomographic image |
| CN112257607B (en) * | 2020-10-23 | 2022-08-26 | 合肥工业大学 | Correction method for processing mobile phone image distortion acquired on production line |
| CN112257607A (en) * | 2020-10-23 | 2021-01-22 | 合肥工业大学 | Correction method for processing mobile phone image distortion acquired on production line |
| CN113377108A (en) * | 2021-06-09 | 2021-09-10 | 乐聚(深圳)机器人技术有限公司 | Method, device, equipment and storage medium for controlling biped robot |
| CN113377108B (en) * | 2021-06-09 | 2024-05-28 | 乐聚(深圳)机器人技术有限公司 | Control method, device and equipment of bipedal robot and storage medium |
| CN113421297A (en) * | 2021-07-02 | 2021-09-21 | 浙江德菲洛智能机械制造有限公司 | Strawberry shape symmetry analysis method |
| CN113580147A (en) * | 2021-09-02 | 2021-11-02 | 乐聚(深圳)机器人技术有限公司 | Robot control method, device, equipment and storage medium |
| CN114037558A (en) * | 2021-11-05 | 2022-02-11 | 国网上海市电力公司 | Transmission line anticollision early warning system based on multidimensional feature recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106407983A (en) | Image body identification, correction and registration method | |
| US11238274B2 (en) | Image feature extraction method for person re-identification | |
| CN109886896B (en) | Blue license plate segmentation and correction method | |
| CN104751142B (en) | A kind of natural scene Method for text detection based on stroke feature | |
| Mishra et al. | Segmenting “simple” objects using RGB-D | |
| US8355569B2 (en) | Object region extracting device | |
| CN111079545A (en) | Three-dimensional target detection method and system based on image restoration | |
| CN107818303B (en) | Unmanned aerial vehicle oil and gas pipeline image automatic contrast analysis method, system and software memory | |
| CN109409356B (en) | Multi-direction Chinese print font character detection method based on SWT | |
| CN104200461A (en) | Mutual information image selected block and sift (scale-invariant feature transform) characteristic based remote sensing image registration method | |
| CN102147867B (en) | Method for identifying traditional Chinese painting images and calligraphy images based on subject | |
| Partovi et al. | Roof Type Selection based on patch-based classsification using deep learning for high Resolution Satellite Imagery | |
| CN114529925A (en) | Method for identifying table structure of whole line table | |
| CN111161219B (en) | Robust monocular vision SLAM method suitable for shadow environment | |
| Mei et al. | A novel framework for container code-character recognition based on deep learning and template matching | |
| Fang et al. | He-slam: A stereo slam system based on histogram equalization and orb features | |
| CN112418210B (en) | Intelligent classification method for tower inspection information | |
| Samet et al. | An implementation of automatic contour line extraction from scanned digital topographic maps | |
| CN118658062A (en) | Occlusion environment pose estimation method based on foreground probability | |
| Chen | Superpixel segmentation based on Delaunay Triangulation | |
| CN104036494A (en) | Fast matching computation method used for fruit picture | |
| Tong et al. | Warped document image mosaicing method based on inflection point detection and registration | |
| Sun et al. | Complex building roof detection and strict description from LIDAR data and orthorectified aerial imagery | |
| CN110619331A (en) | Color distance-based color image field positioning method | |
| Nobuhara et al. | Multi-viewpoint silhouette extraction with 3D context-aware error detection, correction, and shadow suppression |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170215 |
|
| RJ01 | Rejection of invention patent application after publication |