CN106355572A - Method for automatic positioning of tire mold image - Google Patents

Method for automatic positioning of tire mold image Download PDF

Info

Publication number
CN106355572A
CN106355572A CN201610698016.3A CN201610698016A CN106355572A CN 106355572 A CN106355572 A CN 106355572A CN 201610698016 A CN201610698016 A CN 201610698016A CN 106355572 A CN106355572 A CN 106355572A
Authority
CN
China
Prior art keywords
image
roi
roi image
search window
flat type
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.)
Granted
Application number
CN201610698016.3A
Other languages
Chinese (zh)
Other versions
CN106355572B (en
Inventor
蔡念
陈裕潮
岑冠东
丁鹏
陈新度
王晗
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangdong University of Technology
Original Assignee
Guangdong University of Technology
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 Guangdong University of Technology filed Critical Guangdong University of Technology
Priority to CN201610698016.3A priority Critical patent/CN106355572B/en
Publication of CN106355572A publication Critical patent/CN106355572A/en
Application granted granted Critical
Publication of CN106355572B publication Critical patent/CN106355572B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20092—Interactive image processing based on input by user
    • G06T2207/20104—Interactive definition of region of interest [ROI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection

Abstract

The invention discloses a method for automatic positioning of tire mold image, which comprises the steps of scanning a tire mold to be detected and acquiring an original image, and generating a plurality of ROI images to be measured according to the original image; obtaining the straight-type image corresponding to the CAD drawing of the tire mold and positioning a reference ROI image on the straight-type image; generating a search window corresponding to the reference ROI image on the straight-type image;moving the search window to the preset vector, and positioning a contrast ROI image in the moved search window; obtaining an overlapping area of the reference ROI image and the contrast ROI image on the straight-type image;continuing to move the search window in units of the preset vector when the difference degree of the partial image in the overlapping area coincides with the preset conditions, and positioning in turn ROI images which are to be measures and have a ordinal relation with the contrast ROI image each time the search window is moved to a new position. The invention has the advantages of rapid and accurate positioning and high reliability.

Description

The automatic positioning method of tire-mold image
Technical field
The present invention relates to image processing field, more particularly, to a kind of automatic positioning method of tire-mold image.
Background technology
Framing is the important image processing step of image recognition front end, is the importance of graph image research, Computer vision is used widely, such as in the industrial production, framing is used for automatic segmentation, intelligent blanking, automatically Assembling and defects detection etc..
Character defect in tire-mold product, including the biting of character, wrongly typed and Duo Yin.Character defects detection it is simply that Requirement detects undesirable character, if any stroke defect, bite or wrongly typed character.Tire-mold framing conduct The front end applications of character defects detection, different from traditional images positioning, by mating tire-mold material picture, position cad Relevant position on (computer aided design, computer-aided design) figure.The positioning side of existing tire-mold image Method needs to roi to be measured (region of interest, area-of-interest) image classification, and preprocessing process is more complicated, needs Will too many priori, be unfavorable for promoting, meanwhile, be all global search to the positioning of roi image to be measured every time, efficiency is not Height, and only by Similarity Measure, robustness is not strong.
Content of the invention
The embodiment of the present invention provides a kind of automatic positioning method of tire-mold image, to solve existing localization method effect The problem that rate is not high, robustness is not strong.
Embodiments provide a kind of automatic positioning method of tire-mold image, comprising:
Successively tire-mold to be detected is scanned and gathers with several original images of acquisition, and original image is carried out pre- Process and generate several roi images to be measured；
Obtain the cad design drawing corresponding flat type image of tire-mold to be detected, described flat type image positions Benchmark roi image, described benchmark roi image is the piece image in roi image to be measured；
In described flat type image, generate search window corresponding with described benchmark roi image-region, described search window Mouth is not less than described benchmark roi image-region；
By mobile for described search window default vector, and in search window after movement, positioning compares roi image, described Comparing roi image is the piece image in addition to described benchmark roi image in described roi image to be measured；
Obtain the described benchmark roi image and described comparison roi image overlapping region on described flat type image；
Diversity factor when the described benchmark roi image and described comparison roi image topography in described overlapping region When meeting pre-conditioned, continue to move to described search window in units of described default vector, and often move in described search window When moving to new position, by the acquisition orders of described original image, there is ordering relation in positioning and the described roi image that compares successively Roi image to be measured.
Further, described successively tire-mold to be detected be scanned and gather several original images of acquisition, and to former Beginning image carries out the step that pretreatment generates several roi images to be measured, comprising:
Successively tire-mold to be detected is scanned according to the default anglec of rotation and gathers with several original images of acquisition, and Tire outer arc shape profile is obtained after respectively the every original image being gathered being processed；
Behind the center of circle of matching tire outer arc shape profile and radius, will be round outside tire to be measured by polar coordinate transform Arc image is converted to flat type testing image, and described flat type testing image is entered after row threshold division, positions tyre mould Tool image-region, generates several roi images to be measured.
Further, the described cad design drawing corresponding flat type image obtaining tire-mold to be detected, in described flat type Positioning datum roi image on image, described benchmark roi image is the step of the piece image in roi image to be measured, comprising:
Obtain the cad design drawing corresponding flat type image of tire-mold to be detected；
Choose default one roi image to be measured as described benchmark roi image；
Using normalized crosscorrelation method (ncc, normalized cross correlation) in described flat type figure As the described benchmark roi image of upper positioning.
Further, described employing normalized crosscorrelation method positions described benchmark roi image on described flat type image Step, particularly as follows:
Ncc (u, v) is obtained by below equation, and takes correlation coefficient maximum (u, v) to exist as described benchmark roi image Positioning result on described flat type image:
n c c ( u , v ) = 1 n σ ( d , φ ) &element; r f ( d , φ ) - m f s f 2 · g ( d + u , φ + v ) - m g ( u , v ) s g 2 ( u , v )
Wherein, f (d, φ) and g (d, φ) is the pixel value of roi image to be measured and flat type image respectively, and n is search window The sum of all pixels of flat type image in mouthful, r is the area-of-interest of flat type image in search window, mfIt is flat in search window The meansigma methodss of straight type gradation of image,It is the variance of flat type image intensity value in search window, mg(u, v) is search window position Move (u, v) corresponding meansigma methodss in flat type gradation of image afterwards,It is that search window displacement (u, v) corresponds to afterwards in flat type The variance of image intensity value, in above formula:
m f = 1 n σ ( d , φ ) &element; r f ( d , φ )
s f 2 = 1 n σ ( d , φ ) &element; r ( f ( d , φ ) - m f ) 2
m g ( u , v ) = 1 n σ ( d , φ ) &element; r g ( d + u , φ + v )
s g 2 ( u , v ) = 1 n σ ( d , φ ) &element; r ( g ( d + u , φ + v ) - m g ( u , v ) ) 2 .
Further, described move described search window presets vector, and in search window after movement, positioning compares Roi image, described comparison roi image is the step of the piece image in described roi image to be measured in addition to described benchmark roi image Suddenly, comprising:
According to the acquisition orders of described original image, obtain the order of corresponding roi image to be measured；
By mobile for described search window default vector, described default vector is become by polar coordinate by the described default anglec of rotation Change acquisition；
Choose the to be measured roi image adjacent with described benchmark roi image sequence as comparing roi image.
Further, the described Local map when described benchmark roi image and described comparison roi image in described overlapping region When the diversity factor of picture meets pre-conditioned, continue to move to described search window in units of described default vector, and search described When rope window movement is to new position, by the acquisition orders of described original image, positioning compares roi image presence order with described The step of the roi image to be measured of relation, comprising:
Judge the difference of the described benchmark roi image and described comparison roi image topography in described overlapping region It is pre-conditioned whether degree meets；
If so, continue to move to described search window in units of described default vector, and mobile extremely in described search window During new position, positioning with mobile before the adjacent roi image to be measured of roi image sequence to be measured in search window, repeat Step is until position all roi images to be measured on described flat type image.
Further, the described judgement described benchmark roi image and described comparison roi image local in described overlapping region After whether the diversity factor of image meets pre-conditioned step, also include:
If it is not, then again choosing benchmark roi image and being positioned on described flat type image, then re-execute described In described flat type image, generate search window corresponding with described benchmark roi image-region, described search window is not less than The step of described benchmark roi image-region.
Further, judge the described benchmark roi image and described comparison roi image topography in described overlapping region Diversity factor whether meet pre-conditioned step, comprising:
Defining adjacent two roi images to be measured is itAnd it+1, the local in described overlapping region is obtained by below equation The forward error fe of image:
Fe=1-st,t+1
Whereinot,t+1It is itHog (the histograms-of- of the topography in overlapping region Gradients histograms of oriented gradients) feature；ot+1,tIt is it+1Hog feature in overlapping region；
Judge fe ＜ th, then the diversity factor of the topography in described overlapping region meets pre-conditioned, and wherein th is poor Different degree threshold value.
Further, judge the described benchmark roi image and described comparison roi image topography in described overlapping region Diversity factor whether meet pre-conditioned step, comprising:
Defining adjacent three roi images to be measured is it-1、itAnd it+1, wherein, it-1And itBetween form the first overlapping region, itAnd it+1Between form the second overlapping region；The forward direction obtaining the topography in described overlapping region by below equation is by mistake Difference fe:
Fe=1-st,t+1
Whereinot,t+1It is itThe hog feature of the topography in the first overlapping region；ot+1,tIt is it+1Hog feature in the first overlapping region；
Obtain the backward error be of the topography in described overlapping region by below equation；
Be=1-st-1,t
Whereinot,t-1It is itThe hog feature of the topography in the second overlapping region；ot-1,t It is it-1The hog feature of the topography in the second overlapping region；
Judge fbe ＜ th, then the diversity factor of the topography in described overlapping region meets pre-conditioned, wherein fbe= Max (fe, be), th are diversity factor threshold values.
Further, described diversity factor threshold value value is 0.1.
The beneficial effect of the embodiment of the present invention is: by positioning datum roi image on flat type image, regeneration and base The corresponding search window of quasi- roi image-region, by mobile search window, positioning and benchmark roi image presence order are closed successively Other roi images to be measured including comparison roi image of system, the embodiment of the present invention takes full advantage of the time between image Order and spatial relationship, it is not necessary to be classified and Global localization to all roi images to be measured, both ensure that the accurate of positioning Property, in turn ensure that the speed of positioning.Meanwhile, the embodiment of the present invention passes through judgment standard roi image and described comparison roi image exists It is pre-conditioned whether the diversity factor of the topography in described overlapping region meets, and then judges Global localization and default vector Whether accurately, there is higher reliability, strong robustness.
Brief description
In order to be illustrated more clearly that the technical scheme of the embodiment of the present invention, below will be to required in embodiment of the present invention description Accompanying drawing to be used is briefly described, and drawings in the following description are only some embodiments of the present invention, for this area For those of ordinary skill, on the premise of not paying creative work, other accompanying drawings can also be obtained according to these accompanying drawings.
Fig. 1 is the flow chart of the automatic positioning method of tire-mold image of first embodiment of the invention；
Fig. 2 is the benchmark roi image of the present invention and the acquisition methods comparing topography in overlapping region for the roi image Schematic diagram；
Fig. 3 is the flow chart of the automatic positioning method of tire-mold image of second embodiment of the invention.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete Site preparation description is it is clear that described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.It is based on Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of not making creative work Embodiment, broadly falls into the scope of protection of the invention.
First embodiment
With reference to Fig. 1, it is the flow chart of the first embodiment of the automatic positioning method of tire-mold image of the present invention, the party Method includes:
Step 101, is scanned and gathers several original images of acquisition successively to tire-mold to be detected, and to original graph Generate several roi images to be measured as carrying out pretreatment.
In the present embodiment, above-mentioned original image has ordering relation, and the adjacent original image of order comprises to tyre mould The repeated acquisition region of tool, each original image can generate one or more roi image, exemplary, and roi image to be measured can be right Answer the pattern of tire-mold and the image of word segment.
Step 102, obtains the cad design drawing corresponding flat type image of tire-mold to be detected, in described flat type figure As upper positioning datum roi image, described benchmark roi image is the piece image in roi image to be measured.
Benchmark roi image can be in any piece image or roi image to be measured in roi image to be measured The width roi image obtaining by default selection rule, in the present embodiment, for ease of subsequent calculations and positioning, preferably uses pre- If the width roi image that selection rule obtains is as roi image to be measured.
Step 103, in described flat type image, generates search window corresponding with described benchmark roi image-region, institute State search window and be not less than described benchmark roi image-region.
Refer to Fig. 2, definition adjacent reference roi image is it, it is i that definition compares roi imaget+1.In flat type image In, generate search window (in figure is represented with dotted line frame) so that described benchmark roi image region is corresponding, in the present embodiment, search Rope window can be slightly larger than benchmark roi image region.
Step 104, by mobile for described search window default vector, and in search window after movement, positioning compares roi Image, described comparison roi image is the piece image in described roi image to be measured in addition to described benchmark roi image.
Above-mentioned default vector can be obtained by priori or preset formula, by presetting vector by mobile for search window Afterwards, positioned in comparison roi image search window after movement, specifically, comparing roi image can be and benchmark The adjacent piece image of roi image sequence.
Step 105, obtains the described benchmark roi image and described comparison roi image overlap on described flat type image Region.
In the present embodiment, the overlapping region in this step can correspond to the repeated acquisition region of above-mentioned tire-mold.? The topography obtaining benchmark roi image in overlapping region respectively and comparing roi image.
Step 106, when the described benchmark roi image and described comparison roi image topography in described overlapping region Diversity factor when meeting pre-conditioned, continue to move to described search window in units of described default vector, and in described search When window often moves to new position, by the acquisition orders of described original image, positioning compares roi image presence order with described The roi image to be measured of relation.
This step is compared to the topography of said reference roi image and comparison roi image, and judges its difference Whether degree, judge the diversity factor of the described benchmark roi image and described comparison roi image topography in described overlapping region Meet pre-conditioned method, particularly as follows: obtain the forward error of the topography in described overlapping region by below equation Fe:
Fe=1-st,t+1
Whereinot,t+1It is itThe hog feature of the topography in overlapping region；ot+1,tIt is it+1? The hog feature of overlapping region；
When judging fe ＜ th, then the diversity factor of the topography in described overlapping region meets pre-conditioned, and wherein th is Diversity factor threshold value, value can be 0.1.Now the selection of judgment standard roi image and default vector value is accurate.Continue with Described default vector moves described search window for unit, and when search window moves to new position for the second time, positioning is to be measured Roi image it+2；When search window third time moves to new position, position roi image i to be measuredt+3, the like until all Roi image to be measured positions all on flat type image.Wherein, it、it+1、it+2、it+3The acquisition orders of corresponding original image, and Sequentially adjoin successively.
The beneficial effect of the embodiment of the present invention is: by positioning datum roi image on flat type image, regeneration and base The corresponding search window of quasi- roi image-region, by mobile search window, positioning and benchmark roi image presence order are closed successively Other roi images to be measured including comparison roi image of system, the embodiment of the present invention takes full advantage of the time between image Order and spatial relationship, it is not necessary to be classified and Global localization to all roi images to be measured, both ensure that the accurate of positioning Property, in turn ensure that the speed of positioning.Meanwhile, the embodiment of the present invention passes through judgment standard roi image and described comparison roi image exists It is pre-conditioned whether the diversity factor of the topography in described overlapping region meets, and then judges Global localization and default vector Whether accurately, there is higher reliability, strong robustness.
Second embodiment
With reference to Fig. 2, it is the flow chart of the automatic positioning method second embodiment of the tire-mold image of the present invention, the method Including:
Step 201, successively tire-mold to be detected is scanned according to the default anglec of rotation and gathers with acquisition, and several are former Beginning image, and obtain tire outer arc shape profile after respectively the every original image being gathered being processed.
This step is non-limiting as a kind of specific embodiment of original image, specifically, obtains tire outer arc The mode of shape profile includes: successively tire-mold to be detected is scanned and gathers with several original images of acquisition, and right respectively After the every original image being gathered carries out image denoising and Threshold segmentation process, obtain tire-mold profile, and then according to wheel Wide curvature disconnects profile, thus according to the direction of every section of profile, length and curvature, obtaining tire outer arc shape profile.
Step 202, behind the center of circle of matching tire outer arc shape profile and radius, by polar coordinate transform by wheel to be measured Tire outer arc shape image is converted to flat type testing image, and described flat type testing image is entered after row threshold division, fixed Position tire-mold image-region, generates several roi images to be measured.
Above-mentioned roi image to be measured can be with the image of the pattern of corresponding tire mould and word segment.
Step 203, obtains the cad design drawing corresponding flat type image of tire-mold to be detected.
The implementation of this step is similar with step 201 and step 202, repeats no more here.
Step 204, chooses default one roi image to be measured as described benchmark roi image.
Step 205, positions described benchmark roi image using normalized crosscorrelation method on described flat type image.
In the present embodiment, ncc (u, v) can be obtained by below equation, and take maximum (u, the v) conduct of correlation coefficient Described benchmark roi image positioning result on described flat type image:
n c c ( u , v ) = 1 n σ ( d , φ ) &element; r f ( d , φ ) - m f s f 2 · g ( d + u , φ + v ) - m g ( u , v ) s g 2 ( u , v )
Wherein, f (d, φ) and g (d, φ) is the pixel value of roi image to be measured and flat type image respectively, and n is search window The sum of all pixels of flat type image in mouthful, r is the area-of-interest of flat type image in search window, mfIt is flat in search window The meansigma methodss of straight type gradation of image,It is the variance of flat type image intensity value in search window, mg(u, v) is search window position Move (u, v) corresponding meansigma methodss in flat type gradation of image afterwards,It is that search window displacement (u, v) corresponds to afterwards in flat type The variance of image intensity value, in above formula:
m f = 1 n σ ( d , φ ) &element; r f ( d , φ )
s f 2 = 1 n σ ( d , φ ) &element; r ( f ( d , φ ) - m f ) 2
m g ( u , v ) = 1 n σ ( d , φ ) &element; r g ( d + u , φ + v )
s g 2 ( u , v ) = 1 n σ ( d , φ ) &element; r ( g ( d + u , φ + v ) - m g ( u , v ) ) 2 .
Step 206, in described flat type image, generates search window corresponding with described benchmark roi image-region, institute State search window and be not less than described benchmark roi image-region.
This step is identical with the corresponding step of first embodiment, repeats no more here.
Step 207, according to the acquisition orders of described original image, obtains the order of corresponding roi image to be measured.
When a width original image multiple roi image to be measured of corresponding generation, should be multiple based on a width original image generation The order of roi image to be measured is corresponding with the collection direction of original image.
Step 208, by mobile for described search window default vector, described default vector is passed through by the described default anglec of rotation Polar coordinate transform obtains.
In the present embodiment, default vector can be brought in preset formula by the described default anglec of rotation and draw.Default arrow Amount can be corresponding with above-mentioned collection direction in the same direction or corresponding reverse.Exemplary, roi image to be measured suitable Sequence be image 1, image 2, image 3 ..., image n.
Step 209, chooses the to be measured roi image adjacent with described benchmark roi image sequence as comparing roi image.
Step 210, judges the described benchmark roi image and described comparison roi image Local map in described overlapping region It is pre-conditioned whether the diversity factor of picture meets.
In this step, judge the described benchmark roi image and described comparison roi image local in described overlapping region Whether the diversity factor of image meets pre-conditioned method, comprising:
Defining adjacent three roi images to be measured is it-1、itAnd it+1, wherein, it-1And itBetween form the first overlapping region, itAnd it+1Between form the second overlapping region；The forward direction obtaining the topography in described overlapping region by below equation is by mistake Difference fe:
Fe=1-st,t+1
Whereinot,t+1It is itThe hog feature of the topography in the first overlapping region；ot+1,tIt is it+1Hog feature in the first overlapping region；
Obtain the backward error be of the topography in described overlapping region by below equation；
Be=1-st-1,t
Whereinot,t-1It is itThe hog feature of the topography in the second overlapping region；ot-1,tIt is it-1The hog feature of the topography in the second overlapping region；
Judge fbe ＜ th, then the diversity factor of the topography in described overlapping region meets pre-conditioned, wherein fbe= Max (fe, be), th are diversity factor threshold values, and value can be 0.1.
Step 211, if so, continues to move to described search window in units of described default vector, and in described search window Mouthful mobile to new position when, positioning with mobile before the adjacent roi image to be measured of roi image sequence to be measured in search window, heavy Execute this step again until all roi images to be measured are positioned on described flat type image.
Step 212, if it is not, then again choosing benchmark roi image and being positioned on described flat type image, more again Execution step s205.
In the embodiment of the present invention, judge that roi image to be measured and default vector take by forward error and backward error simultaneously The accuracy of value, further increases the reliability of the embodiment of the present invention.Predictably, in the present invention forward error and after Can be also used for the positioning of other roi images to be measured in addition to benchmark roi image to the determination methods of error, to carry further The Position location accuracy of the high embodiment of the present invention.Meanwhile, when judging Wrong localization, the embodiment of the present invention can choose base automatically again Quasi- roi image, and positioned, without personnel's error correction, automatic intelligent level is higher.
It should be understood that each step can in each embodiment of the present invention in several embodiments provided herein To be realized by corresponding virtual functional units.Each functional unit can be integrated in a processing unit or each Unit is individually physically present it is also possible to two or more units are integrated in a unit.Above-mentioned integrated unit both may be used To be realized in the form of hardware, it would however also be possible to employ the form of SFU software functional unit is realized.
If described integrated unit is realized and as independent production marketing or use using in the form of SFU software functional unit When, can be stored in a computer read/write memory medium.Based on such understanding, technical scheme is substantially The part in other words prior art being contributed or all or part of this technical scheme can be in the form of software products Embody, this computer software product is stored in a storage medium, including some instructions with so that a computer Equipment (can be personal computer, server, or network equipment etc.) or processor (processor) execution the present invention each The all or part of step of embodiment methods described.And aforesaid storage medium includes: u disk, portable hard drive, read only memory (rom, read-only memory), random access memory (ram, random access memory), magnetic disc or CD Etc. various can be with the medium of store program codes.
The foregoing is only presently preferred embodiments of the present invention, not in order to limit the present invention, all essences in the present invention Any modification, equivalent and improvement made within god and principle etc., should be included within the scope of the present invention.

Claims (10)

1. a kind of automatic positioning method of tire-mold image is it is characterised in that include:
Successively tire-mold to be detected is scanned and gathers with several original images of acquisition, and pretreatment is carried out to original image Generate several roi images to be measured；
Obtain the cad design drawing corresponding flat type image of tire-mold to be detected, positioning datum on described flat type image Roi image, described benchmark roi image is the piece image in roi image to be measured；
In described flat type image, generate search window corresponding with described benchmark roi image-region, described search window is not Less than described benchmark roi image-region；
By mobile for described search window default vector, and in search window after movement, positioning compares roi image, described comparison Roi image is the piece image in described roi image to be measured in addition to described benchmark roi image；
Obtain the described benchmark roi image and described comparison roi image overlapping region on described flat type image；
When the diversity factor of the described benchmark roi image and described comparison roi image topography in described overlapping region meets When pre-conditioned, in units of described default vector, continue to move to described search window, and often move extremely in described search window During new position, by the acquisition orders of described original image, there is treating of ordering relation with the described roi image that compares in positioning successively Survey roi image.
2. method according to claim 1 is it is characterised in that described be scanned to tire-mold to be detected successively and adopt Collection obtains several original images, and original image is carried out with the step that pretreatment generates several roi images to be measured, comprising:
Successively tire-mold to be detected is scanned according to the default anglec of rotation and gathers with several original images of acquisition, and respectively Tire outer arc shape profile is obtained after the every original image being gathered is processed；
Behind the center of circle of matching tire outer arc shape profile and radius, by polar coordinate transform by tire outer arc shape to be measured Image is converted to flat type testing image, and described flat type testing image is entered after row threshold division, positions tire-mold figure As region, generate several roi images to be measured.
3. method according to claim 2 is it is characterised in that the cad design drawing pair of described acquisition tire-mold to be detected The flat type image answered, positioning datum roi image on described flat type image, described benchmark roi image is roi image to be measured In piece image step, comprising:
Obtain the cad design drawing corresponding flat type image of tire-mold to be detected；
Choose default one roi image to be measured as described benchmark roi image；
Described benchmark roi image is positioned on described flat type image using normalized crosscorrelation method.
4. method according to claim 3 is it is characterised in that described employing normalized crosscorrelation method is in described flat type The step that described benchmark roi image is positioned on image, particularly as follows:
Ncc (u, v) is obtained by below equation, and takes correlation coefficient maximum (u, v) as described benchmark roi image described Positioning result on flat type image:
n c c ( u , v ) = 1 n σ ( d , φ ) &element; r f ( d , φ ) - m f s f 2 · g ( d + u , φ + v ) - m g ( u , v ) s g 2 ( u , v )
Wherein, f (d, φ) and g (d, φ) is the pixel value of roi image to be measured and flat type image respectively, and n is in search window The sum of all pixels of flat type image, r is the area-of-interest of flat type image in search window, mfIt is flat type in search window The meansigma methodss of gradation of image,It is the variance of flat type image intensity value in search window, mg(u, v) be search window displacement (u, V) meansigma methodss in flat type gradation of image are corresponded to afterwards,It is that search window displacement (u, v) corresponds to afterwards in flat type image The variance of gray value, in above formula:
m f = 1 n σ ( d , φ ) &element; r f ( d , φ )
s f 2 = 1 n σ ( d , φ ) &element; r ( f ( d , φ ) - m f ) 2
m g ( u , v ) = 1 n σ ( d , φ ) &element; r g ( d + u , φ + v )
s g 2 ( u , v ) = 1 n σ ( d , φ ) &element; r ( g ( d + u , φ + v ) - m g ( u , v ) ) 2 .
5. method according to claim 2 it is characterised in that described by mobile default for described search window vector, and In search window after movement, positioning compares roi image, and described comparison roi image is to remove described base in described roi image to be measured The step of the piece image outside quasi- roi image, comprising:
According to the acquisition orders of described original image, obtain the order of corresponding roi image to be measured；
By mobile for described search window default vector, described default vector is obtained by polar coordinate transform by the described default anglec of rotation Take；
Choose the to be measured roi image adjacent with described benchmark roi image sequence as comparing roi image.
6. method according to claim 5 it is characterised in that described when described benchmark roi image and described comparison roi figure When meeting pre-conditioned as the diversity factor of the topography in described overlapping region, continue to move in units of described default vector Move described search window, and when described search window movement is to new position, by the acquisition orders of described original image, position Compare the step that roi image has the roi image to be measured of ordering relation with described, comprising:
The diversity factor judging the described benchmark roi image and described comparison roi image topography in described overlapping region is No meet pre-conditioned；
If so, continue to move to described search window in units of described default vector, and in described search window movement to new During position, positioning and the adjacent roi image to be measured of roi image sequence to be measured in search window before movement, repeat this step Until all roi images to be measured are positioned on described flat type image.
7. method according to claim 6 is it is characterised in that described judgement described benchmark roi image and described comparison roi Image, after whether the diversity factor of the topography in described overlapping region meets pre-conditioned step, also includes:
If it is not, then again benchmark roi image being positioned is chosen on described flat type image, then re-execute described in institute State in flat type image, generate search window corresponding with described benchmark roi image-region, described search window is not less than described The step of benchmark roi image-region.
8. the method according to any one of claim 1 to 7 is it is characterised in that judge described benchmark roi image and described ratio Whether pre-conditioned step is met to the diversity factor of topography in described overlapping region for the roi image, comprising:
Defining adjacent two roi images to be measured is itAnd it+1, the topography in described overlapping region is obtained by below equation Forward error fe:
Fe=1-st,t+1
Whereinot,t+1It is itThe hog feature of the topography in overlapping region；ot+1,tIt is it+1In overlap The hog feature in region；
Judge fe ＜ th, then the diversity factor of the topography in described overlapping region meets pre-conditioned, and wherein th is diversity factor Threshold value.
9. the method according to any one of claim 1 to 7 is it is characterised in that judge described benchmark roi image and described ratio Whether pre-conditioned step is met to the diversity factor of topography in described overlapping region for the roi image, comprising:
Defining adjacent three roi images to be measured is it-1、itAnd it+1, wherein, it-1And itBetween form the first overlapping region, itWith it+1Between form the second overlapping region；Obtain the forward error fe of the topography in described overlapping region by below equation:
Fe=1-st,t+1
Whereinot,t+1It is itThe hog feature of the topography in the first overlapping region；ot+1,tIt is it+1? The hog feature of the first overlapping region；
Obtain the backward error be of the topography in described overlapping region by below equation；
Be=1-st-1,t
Whereinot,t-1It is itThe hog feature of the topography in the second overlapping region；ot-1,tIt is it-1 The hog feature of the topography in the second overlapping region；
Judge fbe ＜ th, then the diversity factor of the topography in described overlapping region meets pre-conditioned, wherein fbe=max (fe, be), th is diversity factor threshold value.
10. method according to claim 8 or claim 9 is it is characterised in that described diversity factor threshold value value is 0.1.
CN201610698016.3A 2016-08-19 2016-08-19 The automatic positioning method of tire-mold image Active CN106355572B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610698016.3A CN106355572B (en) 2016-08-19 2016-08-19 The automatic positioning method of tire-mold image

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610698016.3A CN106355572B (en) 2016-08-19 2016-08-19 The automatic positioning method of tire-mold image

Publications (2)

Publication Number Publication Date
CN106355572A true CN106355572A (en) 2017-01-25
CN106355572B CN106355572B (en) 2019-05-21

Family

ID=57844330

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610698016.3A Active CN106355572B (en) 2016-08-19 2016-08-19 The automatic positioning method of tire-mold image

Country Status (1)

Country Link
CN (1) CN106355572B (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105067638A (en) * 2015-07-22 2015-11-18 广东工业大学 Tire fetal-membrane surface character defect detection method based on machine vision
CN105069749A (en) * 2015-07-22 2015-11-18 广东工业大学 Splicing method for tire mold images
CN105675626A (en) * 2016-02-26 2016-06-15 广东工业大学 Character defect detecting method of tire mold

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105067638A (en) * 2015-07-22 2015-11-18 广东工业大学 Tire fetal-membrane surface character defect detection method based on machine vision
CN105069749A (en) * 2015-07-22 2015-11-18 广东工业大学 Splicing method for tire mold images
CN105675626A (en) * 2016-02-26 2016-06-15 广东工业大学 Character defect detecting method of tire mold

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
KALAL Z ET AL.: "Forward-backward error: Automatic detection of tracking failures", 《2010 20TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION》 *
杨延竹,盛佳伟: "一种基于双目视觉的轴类锻件几何尺寸测量方法", 《锻压技术》 *

Also Published As

Publication number Publication date
CN106355572B (en) 2019-05-21

Similar Documents

Publication Publication Date Title
US8873837B2 (en) Image-based crack detection
CN109118473B (en) Angular point detection method based on neural network, storage medium and image processing system
CN106897990B (en) The character defect inspection method of tire-mold
EP3596449A1 (en) Structure defect detection using machine learning algorithms
CN103048329B (en) A kind of road surface crack detection method based on active contour model
Pascoe et al. Robust direct visual localisation using normalised information distance.
CN101014977A (en) Lesion boundary detection
CN103914827A (en) Method for visual inspection of shortages of automobile sealing strip profile
CN112488046B (en) Lane line extraction method based on high-resolution images of unmanned aerial vehicle
CN107192716A (en) A kind of workpiece, defect quick determination method based on contour feature
CN106778551A (en) A kind of fastlink and urban road Lane detection method
CN110503638B (en) Spiral adhesive quality online detection method
CN109858438B (en) Lane line detection method based on model fitting
CN108491786A (en) A kind of method for detecting human face based on hierarchical network and Cluster merging
Bae et al. COP: A new corner detector
CN103150725B (en) Based on SUSAN edge detection method and the system of non-local mean
CN106296587A (en) The joining method of tire-mold image
CN109544513A (en) A kind of steel pipe end surface defect extraction knowledge method for distinguishing
Pavlidis Image analysis
CN106355572A (en) Method for automatic positioning of tire mold image
CN116091987A (en) Industrial scene-oriented multi-strategy image anomaly sample generation method
CN112991327B (en) Steel grid welding system, method and terminal equipment based on machine vision
CN110717471B (en) B-ultrasonic image target detection method based on support vector machine model and B-ultrasonic scanner
CN114170202A (en) Weld segmentation and milling discrimination method and device based on area array structured light 3D vision
Mason et al. Unsupervised discovery of object classes with a mobile robot

Legal Events

Date Code Title Description
C06 Publication
PB01 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