CN102682266B - Cylindrical surface bidimensional bar code reading method based on image splicing - Google Patents

Abstract

The invention provides a cylindrical surface bidimensional bar code reading method based on image splicing. The method comprises the following steps of: acquiring a group of bidimensional bar code pictures in a rotatable manner, correcting non-uniform image illumination, enhancing edge information of bar codes, identifying bar code images and bar code positions, classifying bar code modules, roughly registering the bar code images, finely registering the bar code images, splicing the images and identifying the bar codes.. The identification theory based on an image of the conventional bidimensional bar code identification system is changed, so that cylindrical surface bar code information can be completely acquired, and the problems of distortion, non-uniform illumination and the like of the cylindrical surface bar codes can be solved. According to the characteristic of a Data Matrix bidimensional bar code image, the outstanding problems of low correctness, low efficiency and the like during splicing of bidimensional bar code images can be well solved by the designed splicing fusion algorithm; and the cylindrical surface bidimensional bar code information can be read quickly and accurately.

Description

A kind of cylinder two-dimensional bar code reading method based on Image Mosaics

Technical field

The present invention relates to cylinder two-dimensional bar code recognition technical field, be specially a kind of cylinder two-dimensional bar code reading method based on Image Mosaics.

Background technology

The direct marking two-dimensional bar code of product surface is quick and precisely distinguished, and is the basis of product lifecycle management and tracking of information, is to become the key that improves stock control efficiency, realizes production process information collection and real-time tracing.Current multiselect is the two-dimensional bar code permanent identification as product by two-dimentional Data Matrix bar code, and this is because Data Matrix barcode encoding capacity is large, density is high, error correcting capability is strong.

The image that current two-dimensional bar code reading method all adopts CCD camera collection one width to contain two-dimensional bar code, then image is carried out to a series of processing, remove bar code region, background location, then extract barcode data information.That existing two-dimensional bar code recognition system can only be processed is complete under flat state, deformation is less and the even situation of illumination under two-dimensional barcode image.The closed code reader MATRIX2000 that the hand-held Dataman 7500 that for example U.S.'s Cognex is produced and Germany produce.But in reality, often run into columniform product, because the two-dimensional bar code and the planar bar code that are identified on cylinder have very large difference, cause recognition difficulty, specific as follows:

1, be identified at the two-dimensional bar code on cylinder, because cylinder carrying can cause cylinder distortion, cause the dimension scale of bar code to change.

, generally all can there is uneven illumination to a certain degree in the two-dimensional barcode image 2, gathering on cylinder.If cylinder smoother, particularly when metal cylinder, adopts and will form serious highlighted reflectively, cause information thoroughly to lose.

3, the width two-dimensional barcode image gathering on cylinder, due to cylinder block or acquisition angles improper, easily cause the information of collection imperfect.

4, these above-mentioned typical differences, less at cylinder product diameter, and marking is in the time that the two-dimensional strip yardstick of cylinder is larger, performance more obvious.

These typical differences make the two-dimensional bar code recognition on cylinder more difficult, even cannot distinguish.The two-dimensional bar code reading device of existing maturation, in the time of recognition cylinder two-dimensional bar code, generally recognition rate is lower, has had a strong impact on recognition efficiency, thereby has limited the application of 2D bar code technology in cylinder Product Identifying is followed the trail of.In existing technical research, in order to increase recognition efficiency and the accuracy of cylinder two-dimensional bar code, generally take the measure of two aspects: the one, take the safeguard measure in enhancing processing or the use procedure in marking process.The article that what Xie Zhifeng as upper in China Mechanical Engineering 05 phase in 2011 etc. delivered be entitled as " the process parameter optimizing research of the direct marking two-dimensional bar code of piece surface laser ", introduce how Optimal Parameters improves the quality of marking, these measures are conceived to improve and safeguard contrast and the quality of metal surface two-dimensional bar code, not from solving in essence cylinder bar code reading difficulty, so the effect obtaining bad.The 2nd, take the auxiliary method with image co-registration of hardware, but existing image interfusion method, do not consider a series of features of two-dimensional barcode image, be directly used in cylinder two-dimensional barcode image and merge, efficiency is low, and error rate is high.For example application number is " 201110100489.6 ", denomination of invention is the patent of " reading device and the reading method of the direct marking two-dimensional bar code of a kind of metal cylinder ", has realized rotating acquisition cylindrical picture from hardware aspect, eliminates highlighted reflective, but it is idealized that software is processed, can not be practical.

Summary of the invention

The technical matters solving

For the problem that solves prior art existence the present invention proposes a kind of cylinder two-dimensional bar code reading method based on Image Mosaics, by the Data Matrix bar code image feature of marking on research cylinder, design the difficult problem that a kind of brand-new image split-joint method solves the cylinder two-dimensional bar code aspects such as uneven illumination, cylinder distort in the time of recognition, Information Monitoring is imperfect, obtain fast and accurately high-quality two-dimensional barcode image to reach, realize cylinder information acquisition, improve the object of cylinder two-dimensional bar code recognition efficiency.

Technical scheme

Technical scheme of the present invention is:

Described a kind of cylinder two-dimensional bar code reading method based on Image Mosaics, is characterized in that: comprise the following steps:

Step 1: the N width image of continuous acquisition two-dimensional bar code

the full detail that described N width image has comprised two-dimensional bar code; The width of every width image is w, is highly h, pix ⁿ _{i, j}represent n width image M vtemp _nin i be listed as the pixel value of the capable pixel of j;

Step 2: correcting image uneven illumination:

Step (2-1): choose arbitrarily in comprise bar code information piece image Mvtemp, upwards travel through and ask for longitudinal gradient from h/2 in the middle of image M vtemp:

${\mathrm{grad}}_j^{\mathrm{Mvtemp}}=\underset{i=0}{\overset{w}{\mathrm{\Σ}}}({\mathrm{pix}}_{i,j}^{\mathrm{Mvtemp}}-{\mathrm{pix}}_{i,j+1}^{\mathrm{Mvtemp}})j\∈(h/2,h)$

Wherein,

the capable longitudinal Grad of j in presentation video Mvtemp, and at y _uprow is got maximum longitudinally Grad;

Step (2-2): the illuminance array of background area in computed image MVtemp:

$I_i^{\mathrm{Mvtemp}}=\frac{1}{20}\underset{j=y_{\mathrm{up}}}{\overset{y_{\mathrm{up}}+\mathrm{\β}}{\mathrm{\Σ}}}{\mathrm{pix}}_{i,j}^{\mathrm{Mvtemp}}i\∈[0,w)$

Wherein, in image M vtemp, background area refers to that the 0th row are to w-1 row, y _upwalk to y _upthe region that+β is capable, β gets 10～h-y _up;

the illuminance of presentation video Mvtemp background area i row;

Step (2-3): the average light illumination of background area in computed image Mvtemp

${\stackrel{\‾}{I}}^{\mathrm{Mvtemp}}=\frac{1}{(\mathrm{\β}+1)w}\underset{i=0}{\overset{w-1}{\mathrm{\Σ}}}\underset{j=y_{\mathrm{up}}}{\overset{y_{\mathrm{up}}+\mathrm{\β}}{\mathrm{\Σ}}}{\mathrm{pix}}_{i,j}^{\mathrm{Mvtemp}}$

Step (2-4): reverse correcting image sequence

in the uneven illumination of every piece image:

${\mathrm{pix}}_{i,j}^n={\mathrm{pix}}_{i,j}^n\·{\stackrel{\‾}{I}}^{\mathrm{Mvtemp}}/I_i^{\mathrm{Mvtemp}}n\∈[0,N-1],i\∈[0,w-1],j\∈[0,h-1],$

Step 3: use Roberts operator to image sequence do edge contour information extraction, then edge strength information is increased in original image, realize bar edges information and strengthen;

Step 4: identification bar code image and barcode position:

Step (4-1): sequence of computed images

in the transverse projection data of every width image

${\mathrm{avg}}_j^n=\frac{1}{w}\underset{i=0}{\overset{w}{\mathrm{\Σ}}}{\mathrm{pix}}_{i,j}^{n}j\∈[0,h)$

For n width image, obtain one group of transverse projection data

calculate

in minimum value

Step (4-2): by transverse projection data

entirety translation downwards

transverse projection data after calculating translation

maximal value

and mean value

to the transverse projection data after translation

be weighted mean filter and medium filtering, wherein weighted mean Filtering Template is:

$\frac{1}{9}\×\left[\begin{array}{ccccc}1& 2& 3& 2& 1\end{array}\right]$

Medium filtering adopts 5 × 1 moving window; To filtered transverse projection data

carry out Threshold segmentation, segmentation function is:

Step (4-3): to process step (4-2) transverse projection data after treatment

carry out data fitting, the form of fitting function is:

$y=\left\{\begin{array}{c}{a}_1(0\&le;x<x_1)\\ b_1(x_1\&le;x\&le;h/2)\\ a_2(h/2<x\&le;x_2)\\ b_2(x_2<x<h)\end{array}\right.$

Wherein, a ₁, b ₁, a ₂, b ₂for the matching variable of linear fit function, x ₁, x ₂for the waypoint of linear fit function; Adopt least square fitting, obtain error of fitting and be:

${\hat{S}}_{\min }^2\left(x_1^n\right)=\underset{i=0}{\overset{h/2}{\mathrm{\&Sigma;}}}{\left({\mathrm{avg}}_i^n\right)}^2-\frac{{\left(\underset{i=0}{\overset{x_1^n-1}{\mathrm{\&Sigma;}}}{\mathrm{avg}}_i^n\right)}^2}{x_1^n}-\frac{{\left(\underset{i=x_1^n}{\overset{h/2}{\mathrm{\&Sigma;}}}{\mathrm{avg}}_i^n\right)}^2}{h/2-x_1^n+1}$

${\hat{S}}_{\min }^2\left(x_2^n\right)=\underset{i=h/2+1}{\overset{h-1}{\mathrm{\&Sigma;}}}{\left({\mathrm{avg}}_i^n\right)}^2-\frac{{\left(\underset{i=h/2+1}{\overset{x_2^n}{\mathrm{\&Sigma;}}}{\mathrm{avg}}_i^n\right)}^2}{x_2^n-h/2}-\frac{{\left(\underset{i=x_2^n+1}{\overset{h-1}{\mathrm{\&Sigma;}}}{\mathrm{avg}}_i^n\right)}^2}{h-x_2^n-1}$

In n width image, digital simulation error

corresponding while getting minimum value

order

digital simulation error

corresponding while getting minimum value

order

Step (4-4): repeating step (4-1)～step (4-3), obtains image sequence

in every width image with

thereby obtain array

Step (4-5): sequence of computed images in every width image separately

longitudinal Grad of position:

${\mathrm{grad}}_{y_d^n}^n=\underset{i=0}{\overset{w-1}{\mathrm{\&Sigma;}}}|{\mathrm{pix}}_{i,y_d^n}^n-{\mathrm{pix}}_{i,y_d^n+1}^n|$

Thereby obtain array

calculate array

mean value from array

in the data of n=0 start backward successively with

relatively, when being relatively greater than to the data of n=η

till; From array

in the data of n=N-1 start forward successively with

relatively, when being relatively greater than to the data of n=κ

till; Image sequence

in, Mvtemp _η～Mvtemp _κimage be bar code image;

Step (4-6): calculate in all bar code images

the mean value y of value _s; Appoint the piece image of getting in bar code image, calculate this image

position and

longitudinal Grad of position; When

longitudinal Grad at place is greater than

longitudinal Grad at place, by all bar code images from separately

position is cutting upwards, retains

as new bar code image; When

longitudinal Grad at place is greater than

longitudinal Grad at place, by all bar code images from separately

the downward cutting in position, retains

as new bar code image; The bar code image newly obtaining is saved as again by original order

picture traverse w is constant, is highly y _s;

Step 5: the dividing mode of determining bar code module:

Step (5-1): will

transversely arranged in order, longitudinally alignment is merged into piece image NMvtemp, and image NMvtemp is highly y _s, width is Mw; Image NMvtemp is done to longitudinal gradient projection:

${y\mathrm{grad}}_j^{\mathrm{NMvtemp}}=\underset{i=0}{\overset{\mathrm{Mw}-1}{\mathrm{\&Sigma;}}}|{\mathrm{pix}}_{i,j}^{\mathrm{NMvtemp}}-{\mathrm{pix}}_{i,j+1}^{\mathrm{NMvtemp}}|j\&Element;[0,y_s)$

Wherein

the capable longitudinal gradient of j in presentation video NMvtemp;

Step (5-2): choose dividing mode l × l ∈ C from two-dimensional bar code Module Division mode set C={L × L}, obtain the longitudinal cut-point of a pack module:

$H={\left\{h_m\right|h_m=\frac{y_s}{l}\&times;m\}}_{m=1}^{l-1}$

Step (5-3): computed image NMvtemp is the longitudinal cut-point h of each module in set H _mlongitudinal Grad at place

and calculate all longitudinal Grad

mean value, as the Grad of dividing mode l × l ∈ C;

Step (5-4): repeat step (5-2)～step (5-3), the Grad of all dividing mode in set of computations C, gets the dividing mode p × p of Grad maximum as two-dimensional bar code transverse module dividing mode;

Step 6: the thick registration of bar code image:

Step (6-1): will

in every piece image NMvtemp _nall be converted into the data matrix of a p × w, conversion calculations mode is:

$X^n={\left[{(x_{k,i}^n=\frac{1}{M_s}\underset{j=k{M}_s}{\overset{(k+1)M_s}{\mathrm{\&Sigma;}}}{\mathrm{pix}}_{i,j}^n)}_{i=0}^{w-1}\right]}_{k=0}^{p-1}$

Wherein X ⁿpresentation video NMvemp _ncorresponding data matrix,

the capable i column element of k in representing matrix, M _srepresent the longitudinal size of bar code module, M _s=y _s/ p;

Step (6-2): will

in data matrix X corresponding to adjacent two width images ⁿ, X ⁿ⁺¹stepping is overlapping, stepping columns δ _gscope be 1 row～5 row; Calculate each mean square deviation S of overlapping region element when overlapping ²(g ⁿ):

$S^2\left(g^n\right)=\frac{1}{p{g}^n}\left(\underset{k=0}{\overset{p-1}{\mathrm{\&Sigma;}}}\underset{i=0}{\overset{g^n}{\mathrm{\&Sigma;}}}{(X_{k,i}^{n+1}-X_{k,w-i-g^n}^n)}^2\right)(1\&le;g^n\&le;w)$

Wherein S ²(g ⁿ) the overlapping g of expression data matrix ⁿmean square deviation when column element;

Step (6-3): computational data matrix X ⁿ, X ⁿ⁺¹all S in the overlapping process of stepping ²(g ⁿ), get the wherein overlapping columns corresponding to three mean square deviations of value minimum

as adjacent two width image NMvtemp _nand NMvtemp _n+1between thick registration position, and note

Step (6-4): repeating step (6-2)～step (6-3), to image sequence

in every two width adjacent images carry out thick registration, obtain thick registration position sequence

Step 7: bar code image essence registration:

Step (7-1): the adjacent two width image NMvtemp of calculating that adopt classical similarity measure method _nand NMvemp _n+1at matched position g ⁿthe matching degree R at place ⁿ(g ⁿ):

$R^n\left(g^n\right)=\frac{\underset{i=0}{\overset{g^n-1}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{y_s-1}{\mathrm{\&Sigma;}}}({\mathrm{pix}}_{w-i-g^n,j}^n\&times;{\mathrm{pix}}_{i,j}^{n+1})}{\sqrt{\underset{i=0}{\overset{g^n-1}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{y_s-1}{\mathrm{\&Sigma;}}}{\left({\mathrm{pix}}_{w-i-g^n,j}^n\right)}^2}\sqrt{\underset{i=0}{\overset{g^n-1}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{y_s-1}{\mathrm{\&Sigma;}}}{\left({\mathrm{pix}}_{i,j}^{n+1}\right)}^2}}{g}^n\&Element;U(g_1^n,\mathrm{\&delta;})\&cap;U(g_2^n,\mathrm{\&delta;})\&cap;U(g_3^n,\mathrm{\&delta;})$

The columns δ of fine setting gets δ _g+ 1;

Step (7-2): get matching degree R ⁿ(g ⁿ) maximal value be adjacent two width image NMvtemp _nand NMvtemp _n+1between optimum matching degree, be designated as R ⁿ=max{R ⁿ(g ⁿ), and by max{R ⁿ(g ⁿ) corresponding position g ⁿas adjacent two width image NMvtemp _nand NMvtemp _n+1between smart registration position, and remember smart registration position C ⁿ=g ⁿ;

Step (7-3): repeating step (7-1)～step (7-2), sequence of computed images

in optimum matching degree and the smart registration position of every two width adjacent images, obtain optimum matching number of degrees group

with smart registration position array

Step 8: Image Mosaics merges and bar-code identification:

Step (8-1): traversal optimum matching number of degrees group

with two location point n of matching degree numerical value minimum ₁and n ₂as waypoint, by image sequence

be divided into three parts;

Step (8-2): the image of each part is fade-in to the method for weighted mean gradually going out according to the smart registration position employing between image and splices fusion, obtain the composograph Part of three parts ₀, Part ₁, Part ₂, three width picture traverses are respectively w ₀, w ₁, w ₂, be highly y _s, the smart registration position between three width images is respectively PC ₀, PC ₁,

${\mathrm{PC}}_1=C^{n_2};$

Step (8-3): calculate

middle image NMvtemp ₀transverse gradients:

$x{\mathrm{grad}}_j^{\mathrm{NMvtem}{p}_0}=\underset{j=0}{\overset{y_s-1}{\mathrm{\&Sigma;}}}({\mathrm{pix}}_{i+1,j}^{\mathrm{NMvtem}{p}_0}-{\mathrm{pix}}_{i,j}^{\mathrm{NMvtem}{p}_0})i\&Element;[0,w)$

representative image NMvtemp ₀the transverse gradients of middle i row; Obtain array and calculate

in maximal value

position x _lbe image NMvtemp ₀middle bar code region and white space boundary position;

Step (8-4): calculate

middle image NMvtemp _m-1transverse gradients:

$x{\mathrm{grad}}_j^{\mathrm{NMvtem}{p}_{M-1}}=\underset{j=0}{\overset{y_s-1}{\mathrm{\&Sigma;}}}({\mathrm{pix}}_{i,j}^{\mathrm{NMvtem}{p}_{M-1}}-{\mathrm{pix}}_{i+1,j}^{\mathrm{NMvtem}{p}_{M-1}})i\&Element;[0,w)$

representative image NMvtemp _m-1the transverse gradients of middle i row; Obtain array and calculate in maximal value

position x _rbe image NMvtemp _m-1middle bar code region and white space boundary position;

Step (8-5): by Part ₀, Part ₁, Part ₂width after three width Image Mosaics merge is w _m=(x _l+ y _s+ w-x _r), be highly y _s; Set up image memory buffer zone, size is w _m× y _s; By image Part ₀put into left side, buffer zone, by image Part ₂put into right side, buffer zone; Judgement | (w _m-w ₀-w ₂)-(w ₁-PC ₀-PC ₁) |≤10, if meet Rule of judgment by image Part ₁according to Part ₀, Part ₂smart registration position PC ₀, PC ₁put into buffer zone, overlapping region employing is fade-in the method for weighted mean gradually going out and completes Image Mosaics fusion, enters step (8-7), enters step (8-6) if do not meet Rule of judgment;

Step (8-6): in image memory buffer zone by Part ₁be placed between two width images starting position and Part ₀coincidence w row, then Part ₁stepping moves right, until Part ₁with Part ₂coincidence w only classifies as; By Part in step motion process ₁with Part ₀, Part ₂associating overlapping region is merged in the region overlapping respectively, calculates associating overlapping region matching degree, the same step of matching degree computing method (7-1), and the position of getting the maximum matching degree in associating overlapping region in step motion process is as registration position, by image Part ₁put into buffer zone according to registration position, overlapping region employing is fade-in the method for weighted mean gradually going out and completes Image Mosaics fusion;

Step (8-7): splicing in image memory buffer zone is merged to the two-dimensional barcode image obtaining, carry out cutting according to width, get two-dimensional barcode image x _l～x _l+ y _spart, obtains the y that a width is new _s× y _sdata Matrix two-dimensional barcode image, barcode block size is M _s× M _s; Use decode system to read the bar code information in new Data Matrix two-dimensional barcode image, decode system is decoded and error correction to it according to decoding principle and Reed-Solomon error correction algorithm.

Beneficial effect

A kind of cylinder two-dimensional bar code recognition technology based on Image Mosaics that the present invention proposes, change the recognition principle of existing two-dimensional bar code recognition system based on piece image, can complete collection cylinder bar code information, eliminate the problem such as distortion and uneven illumination of cylinder bar code.According to the feature of Data Matrix two-dimensional barcode image, the splicing blending algorithm of design can run into the outstanding problems such as accuracy is not high, efficiency is low in fine solution two-dimensional barcode image splicing, realizes and distinguishes fast and accurately cylinder two-dimensional barcode information.According to inventor's rough estimates, in the time not using reading method of the present invention, reading a cylinder two-dimensional bar code on average needs 2～4 seconds, and only needs 0.4～0.8 second after using reading method of the present invention, has improved the efficiency of 3～5 times, and the recognition rate greatly improving.

Accompanying drawing explanation

Fig. 1: process flow diagram of the present invention;

Fig. 2: actual convex data plot;

Fig. 3: the convex data plot after standardization;

Fig. 4: the processing procedure figure in embodiment;

Embodiment

Below in conjunction with specific embodiment, the present invention is described:

This example is taken as the cylinder metal of Φ 6, under recognition frock, with MV1300, gathers the sequence image of a group.It is as follows that this example is chosen camera parameter: shutter speed 10us, and gain-adjusted 60, picking rate is at a high speed.For achieving the above object, the total process of technical scheme of the present invention as shown in Figure 1.

Step 1: under recognition frock, regulate and reduce baffle plate gap, the bar code image that makes to gather do not exist highlighted reflective till, then rotate the N width image of continuous acquisition two-dimensional bar code

the full detail that described N width image has comprised two-dimensional bar code; The width of every width image is w=97, is highly h=384,

represent n width image M vtemp _nin i be listed as the pixel value of the capable pixel of j.8 width images in the present embodiment, are gathered, as shown in (a) in accompanying drawing 4.

Step 2: correcting image uneven illumination:

This step is utilized the uneven illumination in bar code region in the illuminance Changing Pattern correcting image of background area in image, and step is as follows:

Step (2-1): choose arbitrarily in comprise bar code information piece image Mvtemp, upwards travel through and ask for longitudinal gradient from h/2 in the middle of image M vtemp:

${\mathrm{grad}}_j^{\mathrm{Mvtemp}}=\underset{i=0}{\overset{w}{\mathrm{\&Sigma;}}}({\mathrm{pix}}_{i,j}^{\mathrm{Mvtemp}}-{\mathrm{pix}}_{i,j+1}^{\mathrm{Mvtemp}})j\&Element;(h/2,h)$

Wherein,

the capable longitudinal Grad of j in presentation video Mvtemp, and at y _uprow is got maximum longitudinally Grad;

Step (2-2): the illuminance array of background area in computed image Mvtemp:

$I_i^{\mathrm{Mvtemp}}=\frac{1}{20}\underset{j=y_{\mathrm{up}}}{\overset{y_{\mathrm{up}}+\mathrm{\&beta;}}{\mathrm{\&Sigma;}}}{\mathrm{pix}}_{i,j}^{\mathrm{Mvtemp}}i\&Element;[0,w)$

Wherein, in image M vtemp, background area refers to that the 0th row are to w-1 row, y _upwalk to y _upthe region that+β is capable, β gets 10～h-y _up;

the illuminance of presentation video Mvtemp background area i row;

Step (2-3): the average light illumination of background area in computed image Mvtemp

${\stackrel{\&OverBar;}{I}}^{\mathrm{Mvtemp}}=\frac{1}{(\mathrm{\&beta;}+1)w}\underset{i=0}{\overset{w-1}{\mathrm{\&Sigma;}}}\underset{j=y_{\mathrm{up}}}{\overset{y_{\mathrm{up}}+\mathrm{\&beta;}}{\mathrm{\&Sigma;}}}{\mathrm{pix}}_{i,j}^{\mathrm{Mvtemp}}$

Step (2-4): reverse correcting image sequence in the uneven illumination of every piece image:

${\mathrm{pix}}_{i,j}^n={\mathrm{pix}}_{i,j}^n\&CenterDot;{\stackrel{\&OverBar;}{I}}^{\mathrm{Mvtemp}}/I_i^{\mathrm{Mvtemp}}n\&Element;[0,N-1],i\&Element;[0,w-1],j\&Element;[0,h-1],$

In the present embodiment, the longitudinal pixel of image belongs to the same radian of cylinder, under the parallel bar shaped white light source in both sides, image be longitudinally that illumination is uniform, horizontal illuminance diminishes in the middle of side direction gradually from two.Based on this rule, the present embodiment utilizes the uneven illumination in bar code region in the illuminance Changing Pattern correcting image in image empty region.Calculate the rectangle white space in the 4th width image in sequence image: longitudinal 15～35, horizontal 0～233, average light illumination

image after reverse correction is as shown in (b) in Fig. 4.

Step 3: use Roberts operator to image sequence

do edge contour information extraction, then edge strength information is increased in original image, realize bar edges information and strengthen; Image after enhancing is as shown in (c) in Fig. 4.

Step 4: identification bar code image and barcode position:

Which image is this step, by the method for Gray Projection in fitted figure picture, identify and contain bar code, and which image is not containing bar code, and the lengthwise position at the bar code place in the image that contains bar code, realizes the identification of bar code image and barcode position.Step is as follows:

Step (4-1): sequence of computed images

in the transverse projection data of every width image

${\mathrm{avg}}_j^n=\frac{1}{w}\underset{i=0}{\overset{w}{\mathrm{\&Sigma;}}}{\mathrm{pix}}_{i,j}^{n}j\&Element;[0,h)$

For n width image, obtain one group of transverse projection data

calculate

in minimum value

due to bar code region data for projection different from background, so

data can present convex shape, are called convex data, as shown in Figure 2.

Step (4-2): for simplifying the operand calculating, by transverse projection data

entirety translation downwards

transverse projection data after calculating translation maximal value

and mean value

to the transverse projection data after translation

be weighted mean filter and medium filtering, further eliminated the impacts such as various noises, pollution, wherein weighted mean Filtering Template is:

$\frac{1}{9}\&times;\left[\begin{array}{ccccc}1& 2& 3& 2& 1\end{array}\right]$

Medium filtering adopts 5 × 1 moving window; To filtered transverse projection data

carry out Threshold segmentation, eliminate convex data and have crenellated phenomena, segmentation function is:

By the standardized operation of step (4-2), make that convex data are irregular has obtained very large improvement, as shown in Figure 3.

Step (4-3): to process step (4-2) transverse projection data after treatment

carry out data fitting, the form of fitting function is:

$y=\left\{\begin{array}{c}{a}_1(0\&le;x<x_1)\\ b_1(x_1\&le;x\&le;h/2)\\ a_2(h/2<x\&le;x_2)\\ b_2(x_2<x<h)\end{array}\right.$

Wherein, a ₁, b ₁, a ₂, b ₂for the matching variable of linear fit function, x ₁, x ₂for the waypoint of linear fit function; Adopt least square fitting, obtain error of fitting and be:

${\hat{S}}_{\min }^2\left(x_1^n\right)=\underset{i=0}{\overset{h/2}{\mathrm{\&Sigma;}}}{\left({\mathrm{avg}}_i^n\right)}^2-\frac{{\left(\underset{i=0}{\overset{x_1^n-1}{\mathrm{\&Sigma;}}}{\mathrm{avg}}_i^n\right)}^2}{x_1^n}-\frac{{\left(\underset{i=x_1^n}{\overset{h/2}{\mathrm{\&Sigma;}}}{\mathrm{avg}}_i^n\right)}^2}{h/2-x_1^n+1}$

${\hat{S}}_{\min }^2\left(x_2^n\right)=\underset{i=h/2+1}{\overset{h-1}{\mathrm{\&Sigma;}}}{\left({\mathrm{avg}}_i^n\right)}^2-\frac{{\left(\underset{i=h/2+1}{\overset{x_2^n}{\mathrm{\&Sigma;}}}{\mathrm{avg}}_i^n\right)}^2}{x_2^n-h/2}-\frac{{\left(\underset{i=x_2^n+1}{\overset{h-1}{\mathrm{\&Sigma;}}}{\mathrm{avg}}_i^n\right)}^2}{h-x_2^n-1}$

In n width image, digital simulation error

corresponding while getting minimum value

order

digital simulation error

corresponding while getting minimum value

order

Step (4-4): repeating step (4-1)～step (4-3), obtains image sequence

in every width image

with

thereby obtain array

Step (4-5): sequence of computed images

in every width image separately

longitudinal Grad of position:

${\mathrm{grad}}_{y_d^n}^n=\underset{i=0}{\overset{w-1}{\mathrm{\&Sigma;}}}|{\mathrm{pix}}_{i,y_d^n}^n-{\mathrm{pix}}_{i,y_d^n+1}^n|$

Thereby obtain array

calculate array

mean value

from array

in the data of n=0 start backward successively with

relatively, when being relatively greater than to the data of n=η

till; From array in the data of n=N-1 start forward successively with

relatively, when being relatively greater than to the data of n=κ

till; Image sequence

in, Mvtemp _η～Mvtemp _κimage be bar code image;

Step (4-6): calculate in all bar code images

the mean value y of value _s; Appoint the piece image of getting in bar code image, calculate this image

position and

longitudinal Grad of position; When

longitudinal Grad at place is greater than longitudinal Grad at place, by all bar code images from separately

position is cutting upwards, retains

as new bar code image; When

longitudinal Grad at place is greater than

longitudinal Grad at place, by all bar code images from separately the downward cutting in position, retains as new bar code image; The bar code image newly obtaining is saved as again by original order

picture traverse w is constant, is highly y _s.

In the present embodiment, step 4 by the every width image of matching in the method for Gray Projection, identify which image and contain bar code, which image is containing bar code, the lengthwise position at the bar code place in the image that contains bar code, realizes the identification of bar code image and barcode position.Obtain minimum error of fitting position in every width image by calculating, standardization, matching projection convex data $\{y_d^0=153,y_u^0=196\},$ $\{y_d^1=34,y_u^1=345\},$ $\{y_d^2=34,y_u^2=345\},$ $\{y_d^3=35,y_u^3=346\},$ $\{y_d^4=31,y_u^4=342\},$ $\{y_d^5=29,y_u^5=341\},$ $\{y_d^6=31,y_u^d=343\},$ $\{y_d^7=150,y_u^7=200\},$ Judge in known the first width and last piece image and do not contain two-dimensional bar code by gradient.According to matching position cutting image, picture altitude y _s=311, width w is constant, has eliminated like this vertical misalignment between image, and the image after cutting is as shown in (d) in Fig. 4.

Step 5: the dividing mode of determining bar code module:

This step is according to the projection of image vertical gradient, and by judging that best pre-mode of dividing gradient realizes the horizontal Module Division of bar code, process is as follows:

Step (5-1): will transversely arranged in order, longitudinally alignment is merged into piece image NMvtemp, and image NMvtemp is highly y _s, width is Mw; Image NMvtemp is done to longitudinal gradient projection:

${y\mathrm{grad}}_j^{\mathrm{NMvtemp}}=\underset{i=0}{\overset{\mathrm{Mw}-1}{\mathrm{\&Sigma;}}}|{\mathrm{pix}}_{i,j}^{\mathrm{NMvtemp}}-{\mathrm{pix}}_{i,j+1}^{\mathrm{NMvtemp}}|j\&Element;[0,y_s)$

Wherein

the capable longitudinal gradient of j in presentation video NMvtemp;

in presentation video NMvtemp, i is listed as the pixel value of the capable pixel of j;

Step (5-2): choose dividing mode l × l ∈ C from two-dimensional bar code Module Division mode set C={L × L}, obtain the longitudinal cut-point of a pack module:

$H={\left\{h_m\right|h_m=\frac{y_s}{l}\&times;m\}}_{m=1}^{l-1}$

Step (5-3): computed image NMvtemp is the longitudinal cut-point h of each module in set H _mlongitudinal Grad at place and calculate all longitudinal Grad

mean value, as the Grad of dividing mode l × l ∈ C;

Step (5-4): repeat step (5-2)～step (5-3), the Grad of all dividing mode in set of computations C, gets the dividing mode p × p of Grad maximum as two-dimensional bar code transverse module dividing mode.

The two-dimensional bar code Module Division mode set of selecting in the present embodiment

according to image vertical gradient projection waveform, by judging that best pre-mode of dividing gradient realizes the horizontal Module Division of bar code.The Grad of various pre-dividing mode is:

Dividing mode	8*8	10*10	12*12	14*14	16*16	18*18	20*20	22*22	24*24
										Grad	489.42	328.22	1079.0	333.15	367.33	446.29	349.10	343.33	640.69

Known, 12*12 is the optimum level Module Division mode of two-dimensional bar code, divides effect as shown in (e) in Fig. 4.

Step 6: the thick registration of bar code image:

Through above-mentioned steps, the skew between image is only present in laterally, also needs the horizontal registration between image to be divided into thick registration and two stages of smart registration.This step is converted into data matrix by image according to divided horizontal bar code module, then mates by data matrix the thick registration of realizing between adjacent image, and process is as follows:

Step (6-1): will

in every piece image NMvtemp _nall be converted into the data matrix of a p × w, conversion calculations mode is:

$X^n={\left[{(x_{k,i}^n=\frac{1}{M_s}\underset{j=k{M}_s}{\overset{(k+1)M_s}{\mathrm{\&Sigma;}}}{\mathrm{pix}}_{i,j}^n)}_{i=0}^{w-1}\right]}_{k=0}^{p-1}$

Wherein X ⁿpresentation video NMvtemp _ncorresponding data matrix,

the capable i column element of k in representing matrix, M _srepresent the longitudinal size of bar code module, M _s=y _s/ p; In step 6 and later step

represent n width image NMvtemp _nin i be listed as the pixel value of the capable pixel of j;

Step (6-2): will

in data matrix X corresponding to adjacent two width images ⁿ, X ⁿ⁺¹stepping is overlapping, stepping columns δ _gscope be 1 row～5 row; Calculate each mean square deviation S of overlapping region element when overlapping ²(g ⁿ):

$S^2\left(g^n\right)=\frac{1}{p{g}^n}\left(\underset{k=0}{\overset{p-1}{\mathrm{\&Sigma;}}}\underset{i=0}{\overset{g^n}{\mathrm{\&Sigma;}}}{(X_{k,i}^{n+1}-X_{k,w-i-g^n}^n)}^2\right)(1\&le;g^n\&le;w)$

Wherein S ²(g ⁿ) the overlapping g of expression data matrix ⁿmean square deviation when column element;

Step (6-3): computational data matrix X ⁿ, X ⁿ⁺¹all S in the overlapping process of stepping ²(g ⁿ), get the wherein overlapping columns corresponding to three mean square deviations of value minimum as adjacent two width image NMvtemp _nand NMvtemp _n+1between thick registration position, and note

Step (6-4): repeating step (6-2)～step (6-3), to image sequence

in every two width adjacent images carry out thick registration, obtain thick registration position sequence

Step 7: bar code image essence registration:

This step, by each thick registration position fine setting, is calculated best matching degree, finally determines smart registration between adjacent two width images, and detailed process is as follows:

Step (7-1): the adjacent two width image NMvtemp of calculating that adopt classical similarity measure method _nand NMvtemp _n+1at matched position g ⁿthe matching degree R at place ⁿ(g ⁿ):

$R^n\left(g^n\right)=\frac{\underset{i=0}{\overset{g^n-1}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{y_s-1}{\mathrm{\&Sigma;}}}({\mathrm{pix}}_{w-i-g^n,j}^n\&times;{\mathrm{pix}}_{i,j}^{n+1})}{\sqrt{\underset{i=0}{\overset{g^n-1}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{y_s-1}{\mathrm{\&Sigma;}}}{\left({\mathrm{pix}}_{w-i-g^n,j}^n\right)}^2}\sqrt{\underset{i=0}{\overset{g^n-1}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{y_s-1}{\mathrm{\&Sigma;}}}{\left({\mathrm{pix}}_{i,j}^{n+1}\right)}^2}}{g}^n\&Element;U(g_1^n,\mathrm{\&delta;})\&cap;U(g_2^n,\mathrm{\&delta;})\&cap;U(g_3^n,\mathrm{\&delta;})$

The columns δ of fine setting gets δ _g+ 1;

Step (7-2): get matching degree R ⁿ(g ⁿ) maximal value be adjacent two width image NMvtemp _nand NMvtemp _n+1between optimum matching degree, be designated as R ⁿ=max{R ⁿ(g ⁿ), and by max{R ⁿ(g ⁿ) corresponding position g ⁿas adjacent two width image NMvtemp _nand NMvtemp _n+1between smart registration position, and remember smart registration position C ⁿ=g ⁿ;

Step (7-3): repeating step (7-1)～step (7-2), sequence of computed images

in optimum matching degree and the smart registration position of every two width adjacent images, obtain optimum matching number of degrees group

with smart registration position array

In the present embodiment, step 6 and step 7 are set up image data matrix, calculate Minimum Mean Square Error and determine the thick registration between image, then determine optimum matching degree and smart registration position between image according to classical similarity measure method, as follows:

Step 8: Image Mosaics merges and bar-code identification:

Smart registration position and the matching degree of this step based between image, is first fused into three fragments by picture splicing, then inserts predetermined two-dimensional barcode image memory buffer region, completes Image Mosaics and merges, and then realizes two-dimensional barcode information recognition.Process is as follows:

Step (8-1): traversal optimum matching number of degrees group

with two location point n of matching degree numerical value minimum ₁and n ₂as waypoint, by image sequence

be divided into three parts;

Optimum matching degrees of data as shown above, three parts are respectively { NMvtemp ₀, NMvtemp ₁, { NMvtemp ₂and { NMvtemp ₃, NMvtemp ₄, NMvtemp ₅.

Step (8-2): the image of each part is fade-in to the method for weighted mean gradually going out according to the smart registration position employing between image and splices fusion, obtain the composograph Part of three parts ₀, Part ₁, Part ₂, as shown in (f1), (f2), (f3) in Fig. 4.Three width picture traverses are respectively w ₀=172, w ₁=97, w ₂=166, height y _ssmart registration position between=311, three width images is respectively PC ₀=41, PC ₁=43;

Step (8-3): calculate middle image NMvtemp ₀transverse gradients:

$x{\mathrm{grad}}_j^{\mathrm{NMvtem}{p}_0}=\underset{j=0}{\overset{y_s-1}{\mathrm{\&Sigma;}}}({\mathrm{pix}}_{i+1,j}^{\mathrm{NMvtem}{p}_0}-{\mathrm{pix}}_{i,j}^{\mathrm{NMvtem}{p}_0})i\&Element;[0,w)$

representative image NMvtemp ₀the transverse gradients of middle i row; Obtain array

and calculate

in maximal value

position x _lbe image NMvtemp ₀middle bar code region and white space boundary position; The present embodiment x _l=12.

Step (8-4): calculate

middle image NMvtemp _m-1transverse gradients:

$x{\mathrm{grad}}_i^{\mathrm{NMvtem}{p}_{M-1}}=\underset{j=0}{\overset{y_s-1}{\mathrm{\&Sigma;}}}({\mathrm{pix}}_{i,j}^{\mathrm{NMvtem}{p}_{M-1}}-{\mathrm{pix}}_{i+1,j}^{\mathrm{NMvtem}{p}_{M-1}})i\&Element;[0,w)$

representative image NMvtemp _m-1the transverse gradients of middle i row; Obtain array and calculate

in maximal value

position x _rbe image NMvtemp _m-1middle bar code region and white space boundary position; The present embodiment x _r=76.

Step (8-5): by Part ₀, Part ₁, Part ₂width after three width Image Mosaics merge is w _m=(x _l+ y _s+ w-x _r)=344 are highly y _s; Set up image memory buffer zone, size is w _m× y _s; By image Part ₀put into left side, buffer zone, by image Part ₂put into right side, buffer zone; Judgement | (w _m-w ₀-w ₂)-(w ₁-PC ₀-PC ₁) |≤10, if meet Rule of judgment by image Part ₁according to Part ₀, Part ₂smart registration position PC ₀, PC ₁put into buffer zone, overlapping region employing is fade-in the method for weighted mean gradually going out and completes Image Mosaics fusion, enters step (8-7), enters step (8-6) if do not meet Rule of judgment;

In the present embodiment | (w _m-w ₀-w ₂)-(w ₁-PC ₀-PC ₁) |=7≤10.Meet Rule of judgment, by image Part ₁according to Part ₀, Part ₂smart registration position PC ₀, PC ₁put into buffer zone, complete Image Mosaics and merge, result is as shown in (g) in Fig. 4.

Step (8-6): in image memory buffer zone by Part ₁be placed between two width images starting position and Part ₀coincidence w row, then Part ₁stepping moves right, until Part ₁with Part ₂coincidence w only classifies as; By Part in step motion process ₁with Part ₀, Part ₂associating overlapping region is merged in the region overlapping respectively, calculates associating overlapping region matching degree, the same step of matching degree computing method (7-1), and the position of getting the maximum matching degree in associating overlapping region in step motion process is as registration position, by image Part ₁put into buffer zone according to registration position, overlapping region employing is fade-in the method for weighted mean gradually going out and completes Image Mosaics fusion;

Step (8-7): splicing in image memory buffer zone is merged to the two-dimensional barcode image obtaining, carry out cutting according to width, get two-dimensional barcode image x _l～x _l+ y _spart, obtains the y that a width is new _s× y _sdata Matrix two-dimensional barcode image, barcode block size is M _s× M _s, as shown in (h) in Fig. 4; Use decode system to read the bar code information in new Data Matrix two-dimensional barcode image, decode system is decoded and error correction to it according to decoding principle and Reed-Solomon error correction algorithm.

Claims (1)

1. the cylinder two-dimensional bar code reading method based on Image Mosaics, is characterized in that: comprise the following steps:

Step 1: the N width image of continuous acquisition two-dimensional bar code

the full detail that described N width image has comprised two-dimensional bar code; The width of every width image is w, is highly h,

represent n width image M vtemp _nin i be listed as the pixel value of the capable pixel of j;

Step 2: correcting image uneven illumination:

Step (2-1): choose arbitrarily

in comprise bar code information piece image Mvtemp, upwards travel through and ask for longitudinal gradient from h/2 in the middle of image M vtemp:

${\mathrm{grad}}_j^{\mathrm{Mvtemp}}=\underset{i=0}{\overset{w}{\mathrm{\&Sigma;}}}({\mathrm{pix}}_{i,j}^{\mathrm{Mvtemp}}-{\mathrm{pix}}_{i,j+1}^{\mathrm{Mvtemp}}),j\&Element;(h/2,h)$

Wherein,

the capable longitudinal Grad of j in presentation video Mvtemp, and at y _uprow is got maximum longitudinally Grad;

Step (2-2): the illuminance array of background area in computed image Mvtemp:

$I_i^{\mathrm{Mvtemp}}=\frac{1}{20}\underset{j=y_{\mathrm{up}}}{\overset{y_{\mathrm{up}}+\mathrm{\&beta;}}{\mathrm{\&Sigma;}}}{\mathrm{pix}}_{i,j}^{\mathrm{Mvtemp}},i\&Element;[0,w)$

Wherein, in image M vtemp, background area refers to that the 0th row are to w-1 row, y _upwalk to y _upthe region that+β is capable, β gets 10～h-y _up;

the illuminance of presentation video Mvtemp background area i row;

Step (2-3): the average light illumination of background area in computed image Mvtemp

${\stackrel{\&OverBar;}{I}}_{\mathrm{Mvtemp}}=\frac{1}{(\mathrm{\&beta;}+1)w}\underset{i=0}{\overset{w-1}{\mathrm{\&Sigma;}}}\underset{j=y_{\mathrm{up}}}{\overset{y_{\mathrm{up}}+\mathrm{\&beta;}}{\mathrm{\&Sigma;}}}{\mathrm{pix}}_{i,j}^{\mathrm{Mvtemp}}$

Step (2-4): reverse correcting image sequence in the uneven illumination of every piece image:

${\mathrm{pix}}_{i,j}^n={\mathrm{pix}}_{i,j}^n\&CenterDot;{\stackrel{\&OverBar;}{I}}_{\mathrm{Mvtemp}}/I_i^{\mathrm{Mvtemp}},n\&Element;[0,N-1],i\&Element;[0,w-1],j\&Element;[0,h-1],$

Step 3: use Roberts operator to image sequence

do edge strength information extraction, then edge strength information is increased in original image, realize edge strength information and strengthen;

Step 4: identification bar code image and barcode position:

Step (4-1): sequence of computed images

in the transverse projection data of every width image

${\mathrm{avg}}_j^n=\frac{1}{w}\underset{i=0}{\overset{w}{\mathrm{\&Sigma;}}}{\mathrm{pix}}_{i,j}^n,j\&Element;[0,h)$

For n width image, obtain one group of transverse projection data

calculate in minimum value

Step (4-2): by transverse projection data

entirety translation downwards

transverse projection data after calculating translation

maximal value and mean value

to the transverse projection data after translation

be weighted mean filter and medium filtering, wherein weighted mean Filtering Template is:

$\begin{array}{ccccc}\frac{1}{9}\&times;[1& 2& 3& 2& 1]\end{array}$

Medium filtering adopts 5 × 1 moving window; To filtered transverse projection data carry out Threshold segmentation, segmentation function is:

Step (4-3): to process step (4-2) transverse projection data after treatment

carry out data fitting, the form of fitting function is:

$y=\left\{\begin{array}{c}{a}_1(0\&le;x<x_1)\\ b_1(x_1\&le;x\&le;h/2)\\ a_2(h/2<x\&le;x_2)\\ b_2(x_2<x<h)\end{array}\right.$

Wherein, a ₁, b ₁, a ₂, b ₂for the matching variable of linear fit function, x ₁, x ₂for the waypoint of linear fit function; Adopt least square fitting, obtain error of fitting and be:

${\hat{S}}_{\min }^2\left(x_1^n\right)=\underset{i=0}{\overset{h/2}{\mathrm{\&Sigma;}}}{\left({\mathrm{avg}}_i^n\right)}^2-\frac{{\left(\underset{i=0}{\overset{x_1^n-1}{\mathrm{\&Sigma;}}}{\mathrm{avg}}_i^n\right)}^2}{x_1^n}-\frac{{\left(\underset{{i=x}_1^n}{\overset{h/2}{\mathrm{\&Sigma;}}}{\mathrm{avg}}_i^n\right)}^2}{h/2-x_1^n+1}$

${\hat{S}}_{\min }^2\left(x_2^n\right)=\underset{i=h/2+1}{\overset{h-1}{\mathrm{\&Sigma;}}}{\left({\mathrm{avg}}_i^n\right)}^2-\frac{{\left(\underset{i=h/2+1}{\overset{x_2^n}{\mathrm{\&Sigma;}}}{\mathrm{avg}}_i^n\right)}^2}{x_2^n-h/2}-\frac{{\left(\underset{{i=x}_2^n+1}{\overset{h-2}{\mathrm{\&Sigma;}}}{\mathrm{avg}}_i^n\right)}^2}{h-x_2^{n-}1}$

In n width image, digital simulation error

corresponding while getting minimum value

order

digital simulation error

corresponding while getting minimum value

order

Step (4-4): repeating step (4-1)～step (4-3), obtains image sequence

in every width image

with thereby obtain array

Step (4-5): sequence of computed images

in every width image separately

longitudinal Grad of position:

${\mathrm{grad}}_{y_d^n}^n=\underset{i=0}{\overset{w-1}{\mathrm{\&Sigma;}}}|{\mathrm{pix}}_{i,y_d^n}^n-{\mathrm{pix}}_{i,y_d^n+1}^n|$

Thereby obtain array

calculate array mean value

from array

in the data of n=0 start backward successively with

relatively, when being relatively greater than to the data of n=η

till; From array

in the data of n=N-1 start forward successively with

relatively, when being relatively greater than to the data of n=κ

till; Image sequence

in, the image of Mvtemp η～Mvtemp κ is bar code image;

Step (4-6): calculate in all bar code images

the mean value y of value _s; Appoint the piece image of getting in bar code image, calculate this image

position and

longitudinal Grad of position; When longitudinal Grad at place is greater than longitudinal Grad at place, by all bar code images from separately

position is cutting upwards, retains

as new bar code image; When

longitudinal Grad at place is greater than

longitudinal Grad at place, by all bar code images from separately

the downward cutting in position, retains

as new bar code image; The bar code image newly obtaining is saved as again by original order

picture traverse w is constant, is highly ys;

Step 5: the dividing mode of determining bar code module:

Step (5-1): will

transversely arranged in order, longitudinally alignment is merged into piece image NMvtemp, and image NMvtemp is highly y _s, width is Mw; Image NMvtemp is done to longitudinal gradient projection:

${y\mathrm{grad}}_j^{\mathrm{NMvemp}}=\underset{i=0}{\overset{\mathrm{Mw}-1}{\mathrm{\&Sigma;}}}|{\mathrm{pix}}_{i,j}^{\mathrm{NMvtemp}}-{\mathrm{pix}}_{i,j+1}^{\mathrm{NMvtemp}}|,j\&Element;[0,y_s)$

Wherein the capable longitudinal gradient of j in presentation video NMvtemp;

in presentation video NMvtemp, i is listed as the pixel value of the capable pixel of j;

Step (5-2): choose dividing mode l × l ∈ C from two-dimensional bar code Module Division mode set C={L × L}, obtain the longitudinal cut-point of a pack module:

$H=\left\{h_m\right|h_m=\frac{y_s}{l}\&times;m{\}}_{m=1}^{l-1}$

Step (5-3): computed image NMvtemp is the longitudinal cut-point h of each module in set H _mlongitudinal Grad at place

and calculate all longitudinal Grad

mean value, as the Grad of dividing mode l × l ∈ C;

Step (5-4): repeat step (5-2)～step (5-3), the Grad of all dividing mode in set of computations C, gets the dividing mode p × p of Grad maximum as two-dimensional bar code transverse module dividing mode;

Step 6: the thick registration of bar code image:

Step (6-1): will

in every piece image NMvtemp _nall be converted into the data matrix of a p × w, conversion calculations mode is:

$X^n={\left[{(x_{k,i}^n=\frac{1}{M_s}\underset{j={\mathrm{kM}}_s}{\overset{(k+1)M_s}{\mathrm{\&Sigma;}}}{\mathrm{pix}}_{i,j}^n)}_{i=0}^{w-1}\right]}_{k=0}^{p-1}$

Wherein Xn presentation video NMvtemp _ncorresponding data matrix,

the capable i column element of k in representing matrix, M _srepresent the longitudinal size of bar code module, M _s=y _s/ p; In step 6 and later step

represent n width image NMvtemp _nin i be listed as the pixel value of the capable pixel of j;

Step (6-2): will

in data matrix X corresponding to adjacent two width images ⁿ, X ⁿ⁺¹stepping is overlapping, stepping columns δ _gscope be 1 row～5 row; Calculate each mean square deviation S of overlapping region element when overlapping ²(g ⁿ):

$S^2\left(g^n\right)=\frac{1}{{\mathrm{pg}}^n}\left(\underset{k=0}{\overset{p-1}{\mathrm{\&Sigma;}}}\underset{i=0}{\overset{g^n}{\mathrm{\&Sigma;}}}{(X_{k,i}^{n+1}-X_{k,w-i-g^n}^n)}^2\right),(1\&le;g^n\&le;w)$

Wherein S ²(g ⁿ) the overlapping g of expression data matrix ⁿmean square deviation when column element;

Step (6-3): computational data matrix X ⁿ, X ⁿ⁺¹all S in the overlapping process of stepping ²(g ⁿ), get the wherein overlapping columns corresponding to three mean square deviations of value minimum

as adjacent two width image NMvtemp _nand NMvtemp _n+1between thick registration position, and note

Step (6-4): repeating step (6-2)～step (6-3), to image sequence

in every two width adjacent images carry out thick registration, obtain thick registration position sequence

Step 7: bar code image essence registration:

Step (7-1): the adjacent two width image NMvtemp of calculating that adopt classical similarity measure method _nand NMvtemp _n+1at matched position g ⁿthe matching degree R at place ⁿ(g ⁿ):

$R^n\left(g^n\right)=\frac{\underset{i=0}{\overset{g^n-1}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{y_s-1}{\mathrm{\&Sigma;}}}({\mathrm{pix}}_{w-i-g^n,j}^n\&times;{\mathrm{pix}}_{i,j}^{n+1})}{\sqrt{\underset{i=0}{\overset{g^n-1}{\mathrm{\&Sigma;}}}\underset{j=0}{\overset{y_s-1}{\mathrm{\&Sigma;}}}{\left({\mathrm{pix}}_{w-i-g^n,j}^n\right)}^2}\sqrt{\underset{i=0}{\overset{g^n-1}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{y_s-1}{\mathrm{\&Sigma;}}}{\left({\mathrm{pix}}_{i,j}^{n+1}\right)}^2}},g^n\&Element;U(g_1^n,\mathrm{\&delta;})\&cap;U(g_2^n,\mathrm{\&delta;})\&cap;U(g_3^n,\mathrm{\&delta;})$

The columns δ of fine setting gets δ _g+ 1;

Step (7-2): get matching degree R ⁿ(g ⁿ) maximal value be adjacent two width image NMvtemp _nand NMvtemp _n+1between optimum matching degree, be designated as R ⁿ=max{R ⁿ(g ⁿ), and by max{R ⁿ(g ⁿ) corresponding position g ⁿas adjacent two width image NMvtemp _nand NMvtemp _n+1between smart registration position, and remember smart registration position C ⁿ=g ⁿ;

Step (7-3): repeating step (7-1)～step (7-2), sequence of computed images

in optimum matching degree and the smart registration position of every two width adjacent images, obtain optimum matching number of degrees group

with smart registration position array

Step 8: Image Mosaics merges and bar-code identification:

Step (8-1): traversal optimum matching number of degrees group

with two location point n of matching degree numerical value minimum ₁and n ₂as waypoint, by image sequence

be divided into three parts;

Step (8-2): the image of each part is fade-in to the method for weighted mean gradually going out according to the smart registration position employing between image and splices fusion, obtain the composograph Part of three parts ₀, Par _t1, Part ₂, three width picture traverses are respectively w ₀, w ₁, w ₂, be highly y _s, the smart registration position between three width images is respectively

${\mathrm{PC}}_1=C^{n_2};$

Step (8-3): calculate

middle image NMvtemp ₀transverse gradients:

${x\mathrm{grad}}_i^{\mathrm{NMvtem}{p}_0}=\underset{j=0}{\overset{y_s-1}{\mathrm{\&Sigma;}}}({\mathrm{pix}}_{i+1,j}^{\mathrm{NMvtem}{p}_0}-{\mathrm{pix}}_{i,j}^{\mathrm{NMvtem}{p}_0}),i\&Element;[0,w)$

representative image NMvtemp ₀the transverse gradients of middle i row; Obtain array

and calculate

in maximal value

position x _lbe image NMvtemp ₀middle bar code region and white space boundary position;

Step (8-4): calculate middle image NMvtemp _m-1transverse gradients:

${x\mathrm{grad}}_i^{\mathrm{NMvtem}{p}_{M-1}}=\underset{j=0}{\overset{y_s-1}{\mathrm{\&Sigma;}}}({\mathrm{pix}}_{i+1,j}^{\mathrm{NMvtem}{p}_{M-1}}-{\mathrm{pix}}_{i,j}^{\mathrm{NMvtem}{p}_{M-1}}),i\&Element;[0,w)$

representative image NMvtemp _m-1the transverse gradients of middle i row; Obtain array and calculate

in maximal value

position x _rbe image NMvtemp _m-1middle bar code region and white space boundary position;

Step (8-5): by Part ₀, Part ₁, Part ₂width after three width Image Mosaics merge is w _m=(x _l+ y _s+ w-x _r), be highly y _s; Set up image memory buffer zone, size is w _m× y _s; By image Part ₀put into left side, buffer zone, by image Part ₂put into right side, buffer zone; Judgement | (w _m-w ₀-w ₂)-(w ₁-PC ₀-PC ₁) |≤10, if meet Rule of judgment by image Part1 according to Part ₀, Part ₂smart registration position PC ₀, PC ₁put into buffer zone, overlapping region employing is fade-in the method for weighted mean gradually going out and completes Image Mosaics fusion, enters step (8-7), enters step (8-6) if do not meet Rule of judgment;

Step (8-6): in image memory buffer zone by Part ₁be placed between two width images starting position and Part ₀coincidence w row, then Part ₁stepping moves right, until Part ₁with Part ₂coincidence w only classifies as; By Part in step motion process ₁with Part ₀, Part ₂associating overlapping region is merged in the region overlapping respectively, calculates associating overlapping region matching degree, the same step of matching degree computing method (7-1), and the position of getting the maximum matching degree in associating overlapping region in step motion process is as registration position, by image Part ₁put into buffer zone according to registration position, overlapping region employing is fade-in the method for weighted mean gradually going out and completes Image Mosaics fusion;

Step (8-7): splicing in image memory buffer zone is merged to the two-dimensional barcode image obtaining, carry out cutting according to width, get two-dimensional barcode image x _l～x _l+ y _spart, obtains the y that a width is new _s× y _sdata Matrix two-dimensional barcode image, barcode block size is M _s× M _s; Use decode system to read the bar code information in new Data Matrix two-dimensional barcode image, decode system is decoded and error correction to it according to decoding principle and Reed-Solomon error correction algorithm.

Images