CN106485182A - A kind of fuzzy Q R code restored method based on affine transformation - Google Patents

A kind of fuzzy Q R code restored method based on affine transformation Download PDF

Info

Publication number
CN106485182A
CN106485182A CN201610511613.0A CN201610511613A CN106485182A CN 106485182 A CN106485182 A CN 106485182A CN 201610511613 A CN201610511613 A CN 201610511613A CN 106485182 A CN106485182 A CN 106485182A
Authority
CN
China
Prior art keywords
step
image
qr code
roi
obtain
Prior art date
Application number
CN201610511613.0A
Other languages
Chinese (zh)
Other versions
CN106485182B (en
Inventor
唐颖奇
罗哉
王艳
胡晓峰
李冬
Original Assignee
中国计量大学
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 中国计量大学 filed Critical 中国计量大学
Priority to CN201610511613.0A priority Critical patent/CN106485182B/en
Publication of CN106485182A publication Critical patent/CN106485182A/en
Application granted granted Critical
Publication of CN106485182B publication Critical patent/CN106485182B/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/14Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
    • G06K7/1404Methods for optical code recognition
    • G06K7/1408Methods for optical code recognition the method being specifically adapted for the type of code
    • G06K7/14172D bar codes
    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/02Control of position or course in two dimensions
    • G05D1/021Control of position or course in two dimensions specially adapted to land vehicles
    • G05D1/0231Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
    • G05D1/0234Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using optical markers or beacons
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K7/00Methods or arrangements for sensing record carriers, e.g. for reading patterns
    • G06K7/10Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
    • G06K7/14Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
    • G06K7/1404Methods for optical code recognition
    • G06K7/146Methods for optical code recognition the method including quality enhancement steps

Abstract

The invention discloses providing a kind of method calibrating traditional image recovery method with QR code feature to be combined, traditional images restored method is make use of to obtain the preferable image of quality, the feature that again make use of QR code obtains the QR code of a standard, more quickness and high efficiency provide accurate QR code to position image for automatic guide vehicle, and obtain the fuzzy Q R code motion blur image restoration method based on affine transformation of the attitude information in automatic guide vehicle running.The present invention utilizes multi-resolution hierarchy, obtains ROI image, and image restoration time can be greatly lowered.Through calibration QR code, very high with the QR code similarity of standard it is ensured that identification.

Description

A kind of fuzzy Q R code restored method based on affine transformation

Technical field

The invention belongs to image processing field is and in particular to a kind of AGV locating module motion blur based on QR code restores Method.

Background technology

With the development of industrial automation technology, automatic guide vehicle (AGV) is widely adopted, and automatic guided vehicle can be by Route automatic running according to instruction, it is not necessary to artificial interfere, therefore, it is possible to effectively improve industrial automation, improves conevying efficiency.

Existing automatic guide vehicle air navigation aid is mainly the method such as magnetic stripe guiding, laser aiming, but these methods are all There is the limitation of itself, in recent years, with scientific and technological development, also occur in that some new location and navigation technologies, based on Quick Response Code The automatic guide vehicle air navigation aid of positioning, this visual guidance method can effectively improve the soft of the path design of automatic guide vehicle Property, but because two-dimension code structure is more complicated, image that automatic guided vehicle collects in motor process be fuzzy it is impossible to Direct Recognition, needs motion blur image restoration, conventional motion blur image restoration algorithm.

Image restoration is a hot spot technology in recent years, its purpose be restore from the observed image degrading real Image, its mathematical model can be described as formula (1):

G (x, y)=f (x, y) * h (x, y)+n (x, y) (1);

Wherein, x, y are image space coordinates, and the image upper left corner is initial point, and g (x, y) is the observed image degrading, f (x, y) It is true picture, h (x, y) is degradation model, n (x, y) is additive noise.By Fourier transformation, this model can be in frequency domain table It is shown as formula (2):

G (u, v)=F (u, v) H (u, v)+N (u, v) (2);

Wherein, u, v represent discrete frequency coordinate, G (u, v), F (u, v), H (u, v), N (u, v) be respectively g (x, y), f (x, Y), h (x, y), the Fourier transformation of n (x, y), H (u, v) be also called point spread function (point spread function, PSF), wherein in PSF, most important two parameters are blur direction and fuzzy distance respectively.Image restoration is exactly to find one again Former wave filter, obtains the estimation of f (x, y)

Restore to solve motion blur, also occur in that some image recovery methods in recent years, but automatic guided vehicle exists It is linear uniform motion in work process, that is, the direction of motion is all known with speed, such as China Patent No. Double iterative mixing blind restoration methods that CN201510445795.1 is announced, its iterative algorithm process time is long, recovery effect Bad, do not make full use of two parameters of known PSF it is impossible to ensure the absolutely identification of Quick Response Code, be not suitable for using Make the image recovery method of automatic guided vehicle, if the speed of automatic guided vehicle is slightly fast, Quick Response Code None- identified will be made, lead Cause can not effectively position, and the scheduling of impact automatic guided vehicle is realized, or even can produce derailing, causes inevitably to lose.

Content of the invention

For above-mentioned prior art exist defect, it is an object of the invention to provide a kind of by traditional image restoration side The method that method is calibrated with QR code feature is combined, that is, make use of traditional images restored method to obtain the preferable image of quality, and The feature that make use of QR code obtains the QR code of a standard, more quickness and high efficiency provide accurate QR code fixed for automatic guide vehicle Bit image, and obtain the fuzzy Q R code motion mould based on affine transformation of the attitude information in automatic guide vehicle running Paste image recovery method.

For achieving the above object, the technical scheme is that:A kind of fuzzy Q R code based on affine transformation restores Method, comprises the following steps:

Step (1), using the constant principle of moment of Hu, comprises the region of QR code in the motion blur image calculating parked ROI image;Step includes:

Step (1.1), using Gaussian image pyramid, downward to motion blur image three times down-sampled;

Step (1.2), detects the marginal information of down-sampled image downwards, calculates the Hu not bending moment of all marginal informations, root According to Hu, bending moment does not obtain the edge that size meets, as the edge of QR code;

Step (1.3), comprises the edge obtaining in step (1.2), obtains four angle points of minimum rectangle with minimum rectangle, It is multiplied by 3 respectively, the minimum obtaining QR code in motion blur image comprises rectangle, and segmentation figure picture obtains ROI image;

Step (2), carries out preliminary motion and obscures again with the ROI image that traditional image recovery method obtains to step (1) Former;Step includes:

Step (2.1), according to the direction of motion of automatic guided vehicle, obtains the motion blur direction of ROI image;

Step (2.2), according to the movement velocity of automatic guided vehicle, obtains the motion blur distance of ROI image;

Step (2.3), using vanishing moment number be 4 db4 wavelet transformation, obtain 4 sub- band diagrams of ROI image Picture, is that low frequency part LL of original image, vertical low frequency part HL of horizontal high-frequent of original image, the horizontal low frequencies of original image hang down respectively Straight HFS LH, the diagonal HFS HH of original image;Motion blur side is obtained in (2.2) according to step (2.1), step To with motion blur distance, using Wiener filtering to LL motion blur restore, using Laplace operator to LH, HL, HH edge Keep;

Step (2.4), using wavelet inverse transformation, 4 sub- band images is reconstructed the ROI image of recovery;

Step (3), using run-length encoding, obtains four angle points of QR code in the ROI image restoring, step includes:

Step (3.1), according to grey level histogram, using Otsu threshold method, the ROI image binaryzation that will restore;

Step (3.2), using median filter, dispels the salt-pepper noise in binaryzation ROI image and ringing effect;

Step (3.3), calculates the run-length encoding of binaryzation ROI image, according to the feature of QR code, is calculated binaryzation Four angle points of QR code in ROI image;

Step (4), using Homography matrix and Principle of Affine Transformation, is calculated the binaryzation just put and restores QR code Image, step includes:

Step (4.1), according to size in broad image for the QR code, obtains QR code-phase truer QR code in broad image Amplification, is calculated four angle points of amplification;

Step (4.2), using four obtained in four angle points of the QR code obtaining in step (3.3) and step (4.1) Angle point, is calculated the linear solution of Homography matrix;

Step (4.3), using Principle of Affine Transformation, is calculated the recovery QR code just put;

Step (5), the recovery just put QR code division is segmented into N*N module, according to the feature of QR code, travels through N*N mould Block, according to the feature of each module, current block is entered as black or white.

The invention has the beneficial effects as follows:

The present invention having the beneficial effects that compared with prior art:

1st, the present invention utilizes multi-resolution hierarchy, obtains ROI image, and image restoration time can be greatly lowered.

2nd, the present invention utilizes wavelet transformation, can improve the robustness of image restoration, improves image restoration effect.

3rd, through calibration QR code, very high with the QR code similarity of standard it is ensured that identification.

4th, algorithm is simple, it is easy to accomplish, real-time.

5 present invention is particularly suitable for motion blur do the automatic guided vehicle work process of locating module using QR code in Image restoration.

Term is explained:

QR code is one kind of two-dimensional bar code, and QR is derived from the abbreviation of English " Quick Response ", i.e. rapid-action meaning Think, wish that QR code can allow its content quickly be decoded from inventor.QR code can store additional information than common bar code, also need not Need adjusting to a line scanning device in scanning as common bar code.QR code can quickly read, compared with bar code before, QR code More rich information can be stored, including to word, URL address and other kinds of data encryption.

ROI, is region of interest abbreviation, refers to that sample determines for selecting in the data set of special-purpose Subset.The concept of one ROI is conventional in many applications, is exactly certain use of given image in image processing field Data in the region of special-purpose.

Brief description

When considered in conjunction with the accompanying drawings, by referring to detailed description below, can more completely more fully understand the present invention with And easily learn the adjoint advantage of many of which, but accompanying drawing described herein is used for providing a further understanding of the present invention, Constitute the part of the present invention, the schematic description and description of the present invention is used for explaining the present invention, does not constitute to this Bright improper restriction, wherein:

Fig. 1 is three gaussian pyramid down-sampled images downwards in embodiment step (1);

Fig. 2 is according to the Hu satisfactory profile that bending moment does not obtain in embodiment step (1);

Fig. 3 is the ROI image that in embodiment step (1), segmentation obtains;

Fig. 4 is that in embodiment step (2), motion blur restores ROI image;

Fig. 5 is affine transformation schematic diagram in embodiment step (4);

Fig. 6 is the recovery QR code just put in embodiment step (4);

Fig. 7 is the module map that in embodiment step (5), the Image Segmentation Methods Based on Features according to QR code becomes;

Fig. 8 is the restoration result in embodiment step (5) to broad image.

Specific embodiment

Below in conjunction with accompanying drawing to the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described.

The present invention comprises five steps, and concrete methods of realizing is as follows:

Step (1), utilize Hu constant principle of moment, in the motion blur image calculating parked, comprise the region of QR code ROI (region of interest) image.Concrete methods of realizing is as follows:

Step (1.1) utilizes Gaussian image pyramid, three samplings downward to motion blur image, down-sampled knot downwards Fruit as shown in figure 1, gaussian pyramid down-sampled formula (1) is downwards:

Wherein x, y are image space coordinates, and the image upper left corner is initial point (0,0), Gn+1(x, y) is (n+1)th grade of down-sampled figure Picture, Gn(x, y) is n-th grade of down-sampled image, and W (a, b)=W (a) * W (b) is the Gaussian convolution core that length is 5, and W (a) is x=a The value of Gaussian convolution core at point, W (b) points out the value of Gaussian convolution core for x=b.

Step (1.2) uses Canny operator to detect the marginal information of down-sampled image downwards, calculates the Hu at all edges not Bending moment, the formula (4) of Hu not bending moment is as follows:

Wherein p=0,1,2......, q=0,1,2......, m00Area for geometric moment.

Calculate the area at all edges that Canny operator edge detection obtains, obtain wherein satisfactory edge, regard It is the edge of QR code, the edge of QR code is as shown in Figure 2.

Step (1.3) minimum rectangle comprises the edge obtaining in step (1.2), obtains four angle points of minimum rectangle, It is multiplied by 3 respectively, the minimum obtaining QR code in motion blur image comprises rectangle, and it is as shown in Figure 3 that segmentation figure picture obtains ROI image.

Step (2), carries out preliminary motion smear restoration with traditional image recovery method to ROI image.Implemented Journey is as follows:

Step (2.1), automatic guided vehicle moves along y direction all the time, so the motion blur direction of ROI image is 90 °.

Step (2.2), the movement velocity of automatic guided vehicle is v, and the time of exposure of industrial camera is t, so ROI image Fuzzy distance be vt.

Step (2.3), using vanishing moment number be 4 " db4 " small echo, scale coefficient is:

{0.325803,1.010946,0.892200,-0.039575,-0.264507,0.043616,0.023252,- 0.014987}

Wavelet transformation is given by formula (5) and (6):

WhereinIt is the of original image respectively j0The value of x=m with y=n low frequency part (LL), the value of the vertical low frequency part of horizontal high-frequent (HL), level in four frequency sub-band of level The value of low frequency vertical HFS (LH), the value of diagonal HFS (HH), H, V, D are respectively intended to identify three high frequency frequencies Section.M and N is width and the height of image resolution ratio,

For jth0Level scaling function andValue at x=m, y=n for the j-th stage wavelet function, its Calculating process is as follows:

X, y are represented with t, whereinFor scaling function, For wavelet function.

H (n) is scale coefficient, and the number of vanishing moment is that 4 " db4 " wavelet scale coefficient is:

{0.325803,1.010946,0.892200,-0.039575,-0.264507,0.043616,0.023252,- 0.014987}

LL frequency sub-band is restored using Wiener filtering, Wiener filtering is given by:

WhereinIt isFourier transformation, H*(u, v) is the conjugate matrices of h (x, y) Fourier transformation, and K is Special constant, generally takes less than 1.

HL, LH, HH frequency sub-band is kept using Laplace operator edge respectively, Laplace operator is as follows:

According to process after LL, HL, LH, HH frequency sub-band, reconstruction image, as shown in Figure 4.

Step (3), using run-length encoding, obtains four angle points of QR code in the ROI image restoring.Implement process For:

Step (3.1), according to grey level histogram, using Otsu threshold method, the ROI image binaryzation that will restore.

Step (3.2), using median filter, dispels the salt-pepper noise in binaryzation ROI image and ringing effect.

Step (3.3), calculates the run-length encoding of binaryzation ROI image, according to the feature of QR code, is calculated binaryzation A in four angle points of QR code in ROI image, such as Fig. 51、A2、A3、A4, wherein XS-axis、YS- axis is binaryzation ROI image Coordinate system, zero is O (xs,ys).

Step (4), using Homography matrix and Principle of Affine Transformation, is calculated the binaryzation just put and restores QR code Image.Wherein affine transformation as shown in figure 5, wherein XN-axis、YN- axis is the coordinate system after conversion, and zero is O' (0,0), A1'、A'2、A'3、A'4Four angle points for the QR code after conversion.

Step (4.1), according to size in broad image for the QR code, obtains QR code-phase truer QR code in broad image Amplification, is calculated four angle points of amplification;

Step (4.2), using four obtained in four angle points of the QR code obtaining in step (3.3) and step (4.1) Angle point, is calculated the linear solution of Homography matrix;

Step (4.3), using Principle of Affine Transformation, is calculated the recovery QR code just put;

Affine transformation is given by:

Wherein H is to use for Homography homography matrix it is known that Homography homography matrix is the invertible matrix of 3*3 Come the projected position to calculate in the point on same three-dimensional planar in different two dimensional images, it is a man-to-man mapping, and H All there are correct mapping relations for each of input picture point.The linear solution calculating Homography needs 4 to not Conllinear point.4 pairs of not conllinear points are given by step (3.3) and step (4.1).

According to Principle of Affine Transformation, it is calculated the recovery QR code just put as shown in Figure 6.

Step (5), the recovery just put QR code division is segmented into N*N module, and the QR code of the Class1 that the present embodiment is selected is 21*21 module, as shown in Figure 7.Each module of traversal QR code, calculates institute's average gray a little in each module, if putting down All gray scale thinks that more than 150 this module is white, otherwise for black.Obtain fuzzy Q R code restoration result as shown in Figure 8.Can See it compared to other restoration algorithms, obtaining is a QR code close to standard.

Described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.Based on the present invention In embodiment, the every other enforcement that those of ordinary skill in the art are obtained under the premise of not making creative work Example, broadly falls into the scope of protection of the invention.

Claims (1)

1. a kind of fuzzy Q R code restored method based on affine transformation is it is characterised in that comprise the following steps:
Step(1), using the constant principle of moment of Hu, in the motion blur image calculating parked, comprise the ROI figure in the region of QR code Picture;Step includes:
Step(1.1), using Gaussian image pyramid, downward to motion blur image three times down-sampled;
Step(1.2), detect the marginal information of down-sampled image downwards, calculate the Hu not bending moment of all marginal informations, according to Hu Bending moment does not obtain the edge that size meets, as the edge of QR code;
Step(1.3), comprise step with minimum rectangle(1.2)In the edge that obtains, obtain four angle points of minimum rectangle, respectively It is multiplied by 3, the minimum obtaining QR code in motion blur image comprises rectangle, and segmentation figure picture obtains ROI image;
Step(2), with traditional image recovery method to step(1)The ROI image obtaining carries out preliminary motion smear restoration;Step Rapid inclusion:
Step(2.1), according to the direction of motion of automatic guided vehicle, obtain the motion blur direction of ROI image;
Step(2.2), according to the movement velocity of automatic guided vehicle, obtain the motion blur distance of ROI image;
Step(2.3), using vanishing moment number be 4 db4 wavelet transformation, obtain 4 sub- band images of ROI image, point Be not low frequency part LL of original image, vertical low frequency part HL of horizontal high-frequent of original image, original image horizontal low frequencies vertically high Frequency part LH, the diagonal HFS HH of original image;According to step(2.1), step(2.2)In obtain motion blur direction and fortune Dynamic fuzzy distance, is restored to LL motion blur using Wiener filtering, using Laplace operator, LH, HL, HH edge is kept;
Step(2.4), using wavelet inverse transformation, 4 sub- band images are reconstructed the ROI image of recovery;
Step(3), using run-length encoding, obtain four angle points of QR code in the ROI image restoring, step includes:
Step(3.1), according to grey level histogram, using Otsu threshold method, the ROI image binaryzation that will restore;
Step(3.2), using median filter, dispel the salt-pepper noise in binaryzation ROI image and ringing effect;
Step(3.3), calculate the run-length encoding of binaryzation ROI image, according to the feature of QR code, be calculated binaryzation ROI figure Four angle points of QR code in picture;
Step(4), using Homography matrix and Principle of Affine Transformation, it is calculated the binaryzation just put and restores QR code figure Picture, step includes:
Step(4.1), according to size in broad image for the QR code, obtain the amplification of QR code-phase truer QR code in broad image Multiple, is calculated four angle points of amplification;
Step(4.2), using step(3.3)In four angle points of QR code of obtaining and step(4.1)In four angle points obtaining, It is calculated the linear solution of Homography matrix;
Step(4.3), using Principle of Affine Transformation, it is calculated the recovery QR code just put;
Step(5), the recovery just put QR code division is segmented into N*N module, according to the feature of QR code, travels through N*N module, according to The feature of each module, current block is entered as black or white.
CN201610511613.0A 2016-06-27 2016-06-27 A kind of fuzzy Q R code restored methods based on affine transformation CN106485182B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610511613.0A CN106485182B (en) 2016-06-27 2016-06-27 A kind of fuzzy Q R code restored methods based on affine transformation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610511613.0A CN106485182B (en) 2016-06-27 2016-06-27 A kind of fuzzy Q R code restored methods based on affine transformation

Publications (2)

Publication Number Publication Date
CN106485182A true CN106485182A (en) 2017-03-08
CN106485182B CN106485182B (en) 2018-10-30

Family

ID=58238765

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610511613.0A CN106485182B (en) 2016-06-27 2016-06-27 A kind of fuzzy Q R code restored methods based on affine transformation

Country Status (1)

Country Link
CN (1) CN106485182B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3506227A1 (en) * 2017-12-28 2019-07-03 Honeywell International Inc. Systems and methods for intelligently recording video data streams

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060038891A1 (en) * 2003-01-31 2006-02-23 Masatoshi Okutomi Method for creating high resolution color image, system for creating high resolution color image and program creating high resolution color image
CN101079149A (en) * 2006-09-08 2007-11-28 浙江师范大学 Noise-possessing movement fuzzy image restoration method based on radial basis nerve network
CN102236887A (en) * 2011-03-11 2011-11-09 贵州大学 Motion-blurred image restoration method based on rotary difference and weighted total variation
US20120093399A1 (en) * 2010-10-15 2012-04-19 Chung-Ang University Industry-Academy Cooperation Foundation Apparatus and method for enhancing image quality of image captured by using multiple color-filter aperture
CN102708552A (en) * 2012-05-24 2012-10-03 中山大学 Rapid two-dimensional barcode image motion deblurring method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060038891A1 (en) * 2003-01-31 2006-02-23 Masatoshi Okutomi Method for creating high resolution color image, system for creating high resolution color image and program creating high resolution color image
CN101079149A (en) * 2006-09-08 2007-11-28 浙江师范大学 Noise-possessing movement fuzzy image restoration method based on radial basis nerve network
US20120093399A1 (en) * 2010-10-15 2012-04-19 Chung-Ang University Industry-Academy Cooperation Foundation Apparatus and method for enhancing image quality of image captured by using multiple color-filter aperture
CN102236887A (en) * 2011-03-11 2011-11-09 贵州大学 Motion-blurred image restoration method based on rotary difference and weighted total variation
CN102708552A (en) * 2012-05-24 2012-10-03 中山大学 Rapid two-dimensional barcode image motion deblurring method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
高敬礼: "多分辨率图像融合及仿射目标识别算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3506227A1 (en) * 2017-12-28 2019-07-03 Honeywell International Inc. Systems and methods for intelligently recording video data streams
US10582147B2 (en) 2017-12-28 2020-03-03 Ademco Inc. Systems and methods for intelligently recording video data streams

Also Published As

Publication number Publication date
CN106485182B (en) 2018-10-30

Similar Documents

Publication Publication Date Title
JP2017533482A (en) Lane data processing method, apparatus, storage medium and equipment
Min et al. Depth video enhancement based on weighted mode filtering
Pfeiffer et al. Exploiting the power of stereo confidences
Bradley et al. Adaptive thresholding using the integral image
Tombari et al. Classification and evaluation of cost aggregation methods for stereo correspondence
Luvizon et al. A video-based system for vehicle speed measurement in urban roadways
CN103426182B (en) The electronic image stabilization method of view-based access control model attention mechanism
Moghadam et al. Fast vanishing-point detection in unstructured environments
Barinova et al. Fast automatic single-view 3-d reconstruction of urban scenes
JP5178875B2 (en) Image processing method for corresponding point search
CN102354363B (en) Identification method of two-dimensional barcode image on high-reflect light cylindrical metal
Yu et al. A novel license plate location method based on wavelet transform and EMD analysis
CN100470452C (en) Method and system for implementing three-dimensional enhanced reality
US8457403B2 (en) Method of detecting and correcting digital images of books in the book spine area
CN101520904B (en) Reality augmenting method with real environment estimation and reality augmenting system
CN104766058B (en) A kind of method and apparatus for obtaining lane line
US20120275653A1 (en) Method for recognizing license plate image, and related computer program product, computer-readable recording medium, and image recognizing apparatus using the same
US8848978B2 (en) Fast obstacle detection
US8744168B2 (en) Target analysis apparatus, method and computer-readable medium
GB2520338A (en) Automatic scene parsing
CN101216941B (en) Motion estimation method under violent illumination variation based on corner matching and optic flow method
JPWO2012017650A1 (en) Object detection apparatus, object detection method, and program
CN103455797B (en) Detection and tracking method of moving small target in aerial shot video
TWI489418B (en) Parallax Estimation Depth Generation
US8401333B2 (en) Image processing method and apparatus for multi-resolution feature based image registration

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant