CN102353346A - Method and system for detection of verticality of CCD installation of laser cutting machine with automatically edge searching performance - Google Patents

Abstract

The invention relates to a method and a system for detection of a verticality of CCD installation of a laser cutting machine with an automatically edge searching performance. The method comprises the following steps that: a laser head of a laser cutting machine is directly utilized; a circle and a center of circle mark patterns that are formed by laser light spots are typed on calibration papers, which are paved fixedly at five different areas at a working plane of a cutting bed; marked images at the five different areas are shot through a CCD; an image processing technology is utilized and circumference point coordinates and center point coordinates of the marked images are respectively extracted; and an extreme values of distance of from the center point to all the circumference points are calculated and a verticality between a CCD shaft and the cutting bed is calculated by a calculation model. In addition, the invention also relates to a detection system for a verticality of CCD installation of a laser cutting machine with an automatical edge searching performance. And the detection system mainly comprises three software modules: a control module, an image processing module, and a data processing module. According to the invention, problems on slow detection speed, low efficiency and poor precision and the like during detection on a verticality of CCD installation of a laser cutting machine with an automatically edge searching performance are solved and an automatic capability of the system is also improved.

Description

Automatically seek detection method and system that limit laser cutting machine CCD installs verticality

Technical field

The present invention relates to the laser application technique field, particularly relate to a kind of intelligent detecting method and system that limit laser cutting machine CCD installs verticality of seeking automatically.

Background technology

Automatically seeking the limit technology based on the cut of computer vision is to be the basis with optics, and merging multi-door modern sciences such as electronics, control science, computer technology and image processing techniques is the system ensemble of one.It obtains the image of Laser Processing object through Vision Builder for Automated Inspection; In computing machine, utilize image processing algorithm to being handled then by the image of acquisition target; Automatically extract the processing object image edge information; And according to the corresponding relation and the proportionate relationship of image edge information and practical operation object; Be converted into the cephalomotor control information of cut; Pass to the control system of laser cutting machine, realize that target is sought the limit automatically accurately to be sheared.Improving the precision of seeking the limit automatically is the gordian technique that laser cutting machine is realized intelligent cutting, raising automatization level.Automatically seek in the system on limit at laser, the degree of accuracy of target object edge extracting directly influences aims of systems objects of cutting precision.Do not reach processing request if seek the limit precision, will cause line of cut and the real profile of being cut pattern to have the deviation that is difficult to tolerate, cause a large amount of cutting waste products.Influence machine vision edge extracting precision and have many-sided factor; Except effective image processing algorithm; Whether the optical axis direction of the installation of CCD is vertical with machined surface, directly restricting the quality that Vision Builder for Automated Inspection is obtained image, thereby influences the extraction precision of system to the processing object edge.In general, machine is before dispatching from the factory, and system installs and all will pass through strict detection, but because laser cutting machine is moved and vibration when working, tends to cause alignment error, need before machine run, the installment state to CCD detect.At present; CCD is installed whether satisfy its optical axis perpendicular cuts bed workplace requirement; Often adopt artificial visually examine's method or utilize spirit-leveling instrument to measure; Adopt three adjustment screws on the CCD erecting device that it is adjusted then; These methods are not only loaded down with trivial details, inefficiency; And precision also is difficult to meet the demands, and has a strong impact on the precision that realizing of Robot Vision is sought the limit automatically.

Summary of the invention

Technical problem to be solved by this invention provides a kind of intelligent detecting method and system that limit laser cutting machine CCD installs perpendicularity of seeking automatically; Under the condition of not adding miscellaneous equipment and device; Realize that by computer control regional laser gets spot pattern and carry out mark; Automated graphics is gathered and is handled; Accurately extract the coordinate of gauge point figure; And utilize given computation model to calculate the perpendicularity that CCD installs; Thereby can solve and seek limit laser cutting machine CCD automatically that the perpendicularity detection speed is installed is slow; Efficient low with problem such as low precision, improve the automatization level of system.

The technical solution adopted for the present invention to solve the technical problems is: a kind of detection method that limit laser cutting machine CCD installs verticality of seeking automatically is provided, may further comprise the steps:

(1) directly utilizes the laser head of laser cutting machine, on the fixing demarcation paper of laying of five zoness of different of cutting bed vise workplace, get circle and the center mark pattern of forming by laser facula;

(2) pass through the marking image that CCD absorbs five zoness of different;

(3) utilize image processing techniques, extract the circumference and the centre point coordinate of marking image respectively;

(4) calculate the extreme value of the center of circle, obtain the verticality of CCD optical axis and cutting bed vise by computation model to circumference each point distance.

Comprise following substep in the described step (1): at four angles of cutting bed vise and five zones in centre position; Tile regularly respectively five and demarcate paper; The laser head of control cutting machine is demarcated the marked circle and the center of circle of getting the constant radius of being made up of laser facula on the paper at each and is carried out mark; The marked region size is limited in the visual field of a CCD imaging; Location alignment point when the center mark point of each mark zone is adopted image for CCD; And in the process that verticality detects, five positions of demarcating paper remain unchanged.

Comprise following substep in the described step (2): each zone marker figure is gathered; Elder generation's driven CCD roughly moves to the zone of appointment; Field-of-view image is obtained in dynamic acquisition and processing then; Extract the coordinate of surveyed area center mark point; Be used to control the adjustment of CCD position; Make the center mark point adjust to the center of CCD images acquired, the marking image collection is carried out in the locking visual field.

Described step comprises following substep in (3): utilize image processing techniques; Size according to laser facula; Design size different disposal template is carried out the mixed noise Filtering Processing, and adopts the sub-pixel positioning algorithm, accurately extracts the coordinate of circumference gauge point and center mark point in image.

Comprise following substep in the described step (4): to each marker region image of gathering; After accurately extracting the marker graphic coordinate; Calculate center mark and put the extreme value of each circumference gauge point distance; Comprising two equal maximal values of different directions and two minimum value that do not wait of different directions and time minimum value; The substitution computation model calculates the verticality that CCD installs; The same processing done in five zones, tries to achieve the CCD verticality in five zones respectively, and the verticality of averaging and installing as CCD.

The technical solution adopted for the present invention to solve the technical problems is: a kind of detection system that limit laser cutting machine CCD installs verticality of seeking automatically also is provided, comprises cutting bed vise, described cutting bed vise is provided with can be along the mounting bracket of cutting bed vise both sides translation; Described mounting bracket is provided with can be along the laser cutting machine and the CCD of mounting bracket slip; Described laser cutting machine links to each other with computer system respectively with CCD, and described computer system comprises: control module, image processing module and data processing module; Described control module is used to control five zones of laser cutting machine on cutting bed vise, according to preassigned position and radius, accurately gets circular on the paper and mark is carried out in the center of circle demarcating, and controls described CCD calibration graph is gathered; Described image processing module is used for the image of gathering is carried out pre-service, utilizes the sub-pixel positioning algorithm, accurately extracts each point coordinate in the mark center of circle and circumference in every two field picture; Described data processing module is used for calculating required parameter according to the coordinate that extracts marker graphic, and the substitution computation model, obtains CCD optical axis and the verticality of cutting bed surface.

Beneficial effect

Owing to adopted above-mentioned technical scheme; The present invention compared with prior art; Have following advantage and good effect: the present invention utilizes the existing apparatus of cutting machine, adopts image process method, realizes seeking automatically the Intelligent Measurement that limit laser cutting machine CCD installs verticality.The present invention can replace artificial visually examine's method or adopt the method for spirit-leveling instrument to carry out the detection of CCD verticality; And it is slow to overcome manual detection speed; Efficient low with shortcoming such as low precision; What realization need not manual intervention has noncontact, highly sensitive, high precision, a high efficiency full-automatic detection, and what help to improve laser cutting machine seeks limit precision and automatization level.

Description of drawings

Fig. 1 seeks the detection system structural representation that limit laser cutting machine CCD installs verticality automatically;

Fig. 2 detects visual field sign figure distribution schematic diagram;

Fig. 3 seeks the process flow diagram that limit laser cutting machine CCD installs the detection method of verticality automatically;

Fig. 4 is a CCD mounting bracket adjustment synoptic diagram;

Fig. 5 is the projection model schematic diagram.

Embodiment

Below in conjunction with specific embodiment, further set forth the present invention.Should be understood that these embodiment only to be used to the present invention is described and be not used in the restriction scope of the present invention.Should be understood that in addition those skilled in the art can do various changes or modification to the present invention after the content of having read the present invention's instruction, these equivalent form of values fall within the application's appended claims institute restricted portion equally.

As shown in Figure 1, the present invention realizes that Intelligent Laser cutting machine vision CCD installs the automatic detection of verticality, and it need not add and install other extra equipment, only utilizes the intrinsic equipment of laser cutting machine, can accomplish detection.The detection system of seeking limit laser cutting machine CCD installation verticality automatically comprises cutting bed vise 1, and described cutting bed vise 1 is provided with can be along the mounting bracket 2 of cutting bed vise both sides translation; Described mounting bracket 2 is provided with can be along the laser cutting machine 3 and the CCD4 of mounting bracket slip; Described laser cutting machine 3 links to each other with computer system respectively with CCD4.Computer system comprises control module, image processing module and data processing module.Described control module is used to control five zones of laser cutting machine on cutting bed vise, according to preassigned position and radius, accurately gets circular on the paper and mark is carried out in the center of circle demarcating, and controls described CCD calibration graph is gathered.Described image processing module is used for the image of gathering is carried out pre-service, utilizes the sub-pixel positioning algorithm, accurately extracts each point coordinate in the mark center of circle and circumference in every two field picture.Described data processing module is used for calculating required parameter according to the coordinate that extracts marker graphic, and the substitution computation model, obtains CCD optical axis and the verticality of cutting bed surface.

This system adopts following equipment: (1) CCD model is HV1303UM, major parameter: resolution is the CMOS CCD of 1280*1024; Optical dimensions is 1/1.8; The highest 1,300,000 pixels; The highest horizontal resolution is 1280; The digital-to-analog conversion precision is 10bit; The high s/n ratio of 45dB (AGC OFF); Can control (AGC) by open/close automatic gain, the digital gain multiple is: * 2, * 1, * 0.5, * 0.25; For the light source of 550nm, its sensitivity is: 2.1V/Lux-s; Can be from the black-to-white level weighing apparatus correcting mode of motion tracking (ATW)/manually setting; Can accept two kinds of power supply supplies of 24V AC and 12V DC; (2) unit type of laser cutting machine: V-SPIN-1006-W60CA, controller model: CA-II.The X-axis ratio of gear: 1: 1, Y-axis ratio of gear: 2: 3; The X-axis drive model is 3ND583B; The Y-axis drive model is 3ND583B; X-axis motor model is 573S09-L; Y-axis motor model is 573S15-L; The focus lamp model is ZNSE2.0FL; Laser type: CO ₂Water-cooled amorphous laser (to nonmetal effective, voltage, the two controls of pulse); (3) led light source: YHFL-100-100-W electric parameter 24V/11.5W, eight grades of controlled change light intensity are arranged, when environment temperature is 25 ℃, surpass 30000 hours (damping capacity is) at 50% o'clock with the continuous reliably working of 50% white light source brightness; (4) computer system: Lenovo Qitian M7300:CPU, Intel Duo i3550, double-core/four threads, frequency 3.2GHz L2 cache 2 * 256KB, three grades of buffer memory 4MB; Board chip set Intel H57, the mainboard memory size is 2GB; Type of memory is DDR31333; Hard-disk capacity 500GB, rotating speed 7200 changes SATAII; The video card type is solely to show AMD Radeon HD4350, video memory capacity 512MB, and display sizes is 17 inches.

The detection system of seeking limit laser cutting machine CCD installation verticality automatically is behind system boot; The initialization of advanced line correlation; And adopt spirit-leveling instrument that the cutting bed vise of laser machine is carried out horizontal detection and horizontal adjustment; Carry out laser by the laser cutting machine of computer system control in five zones in four angles of cutting bed vise and centre position then and accurately get mark ready; The zone at four angles is designated as the a-quadrant respectively; The B zone; C zone and D zone; The zone in centre position is designated as the E zone; Rule is got the center of circle be made up of laser facula and the circular pattern of constant radius; Marker graphic distributes as shown in Figure 2, and its size is limited in the visual field of a CCD imaging.The alignment point of position when wherein, the center of circle of each mark zone is the CCD images acquired.

To the collection of marking image, drive the CCD that binds one with laser head by control module and accomplish, CCD is navigated to after the center mark point is positioned at the center of images acquired, the locking visual field and images acquired, as shown in Figure 1.After the image that obtains five marked regions carried out Filtering Processing respectively, gauge point (center of circle and circumference) is carried out sub-pixel positioning, accurately extract the sub-pix coordinate figure of each signature form point in image.Each is detected the visual field; Central coordinate of circle and each point coordinate of circumference according to the Flame Image Process extraction; Calculate the extreme value of the center of circle to circumference each point distance; Comprise two equal maximal values of different directions and two minimum value that do not wait of different directions and time minimum value, calculate the verticality that CCD installs according to computation model then.

Automatically seek the intelligent detecting method that limit laser cutting machine CCD installs verticality, as shown in Figure 3, implementation step is following:

(a) start-up system and carry out initialization is carried out the horizontal alignment of laser cutting machine cutting bed vise.

(b) open detection program, at first according to as shown in Figure 2 predefined five regional location a-quadrants, B zone, C is regional, D is regional and E is regional, fixing respectively tiling is gone up and is demarcated paper, and in testing process, its position strictness of demarcating paper remains unchanged; The laser head of controlling diced system is again got circle and the center of circle of constant radius as the test badge figure on each regional demarcation paper.

(c) control module is positioned ccd image sensor for 5 respectively by the zone of laser engraving mark; Handle through dynamic image; The position coordinates in the center of circle, the certification mark visual field; And be used to control the position adjustment of CCD; Be positioned at the center of image up to the center mark point; This moment, marker region image was gathered in the lock-in detection visual field.

(d) utilize filtering and noise reduction and binary conversion treatment, and adopt the sub-pixel positioning algorithm, extract the accurate coordinates of each two field picture center of circle and each gauge point of circumference respectively.

(e) to each marked region, calculate the extreme value of the mark center of circle in the image to each circumference gauge point distance, the substitution computation model is obtained the verticality that CCD installs.

(f) five zones are carried out the processing of the said method of (e) step respectively, calculate the installation verticality of each zone C CD, ask it average at last, the verticality information of installing as laser cutting machine CCD.

(g) show that the CCD that detects installs verticality information.

According to said method; Install in the adjustment process at the CCD of reality; Automatically detect CCD by system verticality information is installed; And according to showing detection information; The operator can judge whether the installment state of CCD meets the demands, and when CCD installation verticality does not meet the demands, then can adjust through 5 couples of CCD4 of adjusting mechanism shown in Figure 4; And duplicate detection-adjustment process repeatedly, satisfy its verticality precision requirement up to the installation of CCD4.

In testing process, only need adopt five to be fixed on five zones of cutting bed vise greatly in the demarcation paper in the CCD imaging visual field, as the sign of laser dotting, supply machine vision to gather marking image.The position of demarcating paper must remain unchanged in testing process in strictness.

In order to improve the accuracy of detection of marking image, said method adopts image filtering to handle the inhibition noise earlier, improves the signal to noise ratio (S/N ratio) of image; In the leaching process of laser labelling location and positional information, adopt the sub-pix algorithm, improve the gauge point image coordinate and extract precision.

The method of image denoising has a lot, and mean filter and medium filtering are two kinds of common methods.They are suitable for the Flame Image Process of different types of noise, and medium filtering has denoising ability preferably to salt-pepper noise, and mean filter has denoising ability preferably to Gaussian noise.But; Because it is comparatively complicated to seek the machine vision imaging circumstances of limit laser cutting machine automatically; In testing process; The image that the Vision Builder for Automated Inspection of laser cutting machine is obtained may receive two kinds of interference of noise simultaneously, and independent medium filtering or mean filter all are difficult to the denoising effect that reaches best.This method is according to the size of laser facula, and design 3*3 processing template is carried out the mixed noise wave filter, and with respect to traditional filtering algorithm, it is comparatively desirable to remove noise effects, helps to realize the stable accurate coordinates that extracts laser-marking.

In said method, consider the size of the pixel that laser facula is shared, at first selected Filtering Processing window is the 3*3 size, and adopts following method of discrimination that noise type is differentiated.At first obtain the variance Q of filter window interior pixel ², and set threshold values T.Q at that time ²During＞T, think the pollution that receives salt-pepper noise in this spectral window; Work as Q ²During＜T, then think the pollution that receives Gaussian noise in this spectral window.According to above-mentioned determination methods; If the filter window interior pixel is polluted by salt-pepper noise; Then obtain gray scale maximal value and minimum value in the spectral window; Remove interior those of spectral window then and equal the pixel of maximal value and minimum value; If residual pixel is non-vanishing in the spectral window; Then obtain the mean value M of residual pixel, and calculate the absolute value E of average grey scale pixel value and filter window mid point pixel grey scale value difference.This absolute value and preset threshold T relatively if absolute value E greater than threshold value T, then exports residual pixel average M, less than threshold value T, then export spectral window mid point grey scale pixel value as if absolute value E.Wherein the local property of the selection of threshold value T and image interrelates.Can suitably weaken denoising and preserve details in bright area; In dark areas, then suppress noise as far as possible.If residual pixel is zero in the filter window, then enlarges filter window and be of a size of 5 * 5, and repeat above algorithm.If polluted in the filter window, then calculate the gradient absolute value of filter window interior pixel earlier, if Grad greater than a certain given threshold value T, is then directly exported former pixel by salt-pepper noise.Otherwise, then export filter window interior pixel gray-scale value average.Repeat above-mentioned algorithm,, accomplish image and remove the processing of making an uproar until the Filtering Processing of accomplishing all pixels.

In order accurately to detect the verticality that CCD installs, the accurate location of regional laser labelling figure is crucial.In said method, adopt the sub-pixel positioning algorithm to extract the coordinate of identification point, guarantee the high precision of its location.Sub-pixel positioning algorithm based on second order spatial moment operator carries out hi-Fix to the laser facula center.At first image is being carried out on the pretreated basis; Gray level image to laser facula strengthens and binary conversion treatment; Adopt the LOG operator that image is carried out rim detection then; Make hot spot edge essence be determined to single pixel precision; Use second order spatial moment operator further to segment then; Make the location, edge reach sub-pixel precision, obtain the spot center coordinate figure through The Fitting Calculation again.Owing to consider the size of LASER SPECKLE; Be of a size of sample window with 5*5; Design six spatial moment computing templates; In comprising the marginal point neighborhood; Use these six templates and the hot spot circle of forming circle diagram shape to carry out convolution algorithm, judge, obtain the sub-pixel location at hot spot edge according to preset threshold; Through The Fitting Calculation, obtain the sub-pixel location coordinate of circular laser facula and center of circle laser facula again.Need to prove that in real system was measured, ellipse or irregular figure might appear in the laser facula of detection, at this moment, light spot image need be carried out match according to the circle model, so that accurate Calculation spot center sub-pixel location coordinate.

Obtaining the algorithm of CCD optical axis and cut surface verticality in the said method confirms by following method.Because the imageing sensor camera head of practical application laser cutting machine Vision Builder for Automated Inspection all is the perspective projection imaging system, the variation of optical axis can cause that scene point motion conditions on the picture plane of CCD is inconsistent.Transmission projection can produce transparent effect in image, forming images actual is the projection of three-dimensional scenic on the two-dimensional imaging plane.Because imaging plane changes with respect to the direction of scene, can cause optical axis direction to change.In the perspective projection system, the variation of optical axis can cause that projected image produces very large influence.Have only when the scene target leaves the imaging hole infinity, perspective projection just becomes parallel projection.Based on the pinhole imaging system principle of sensor and camera head, the essence of the perspective that the variation of optical axis produces is the variation of the relative viewing angle on the three dimensions, is embodied in the Flame Image Process of two dimension, is actually three-dimensional affine variation to two dimension.So, can utilize homogeneous coordinate transformation to set up coordinate transform relation according to the collinear theory of perspective projection from cut bed plane to the CCD imaging plane based on the laser cutting machine of machine vision.In said method, set up 3 coordinate systems earlier: (1) cut bed plane coordinate system OXYZ, this coordinate system is the plane at practical operation object place, promptly the OXY face remains unchanged in measuring process; (2) eye coordinates is O _sX _sY _sZ _s, this coordinate is true origin with the viewpoint, Z _sPoint to the initial point of coordinate system OXYZ; (3) CCD imaging plane coordinate system O ' X ' Y ', this is to be plane coordinate system, CCD imaging front is positioned at O ' X ' Y ' plane.The relation of three coordinate systems as shown in Figure 5.Projection transformational relation from cut bed plane coordinate system OXYZ to CCD imaging plane coordinate system O ' X ' Y ', available following coordinate transform is represented:

$x^{\′}=F\×\frac{-x\sin \mathrm{\α}+y\cos \mathrm{\α}}{-x\cos \mathrm{\α}\cos \mathrm{\β}-y\sin \mathrm{\α}\cos \mathrm{\β}+R}$ (1)

$y^{\′}=F\×\frac{-x\cos \mathrm{\α}\sin \mathrm{\β}-y\sin \mathrm{\α}\sin \mathrm{\β}}{-x\cos \mathrm{\α}\cos \mathrm{\β}-y\sin \mathrm{\α}\cos \mathrm{\β}+R}$

Wherein, (x, y) be on the OXY coordinate surface a bit, (x ', y ') is its subpoint on O ' X ' Y ' plane, F is the focal length of CCD, the A point is O _sPoint is at the subpoint on OXY plane, and R is O _sThe distance that point is ordered to O, α is the angle of OX axle and OA, and β is O _sThe angle of O and OA, it is the angle that CCD optical axis and laser are sheared the bed plane, the i.e. angle that will detect.Make (x y) puts initial point distance and is r, with the angle of Y-axis be θ, x=rcos θ is then arranged, y=rsin θ.Let (x, y) is a CCD image (x ', y'), (x ', y') to the origin point of the image plane origin O 'the distance and x = rcosθ, y = rsinθ into equation (1) , and let:

$\left\{\begin{array}{c}M=\cos \mathrm{\θ}\sin \mathrm{\α}\\ N=\cos \mathrm{\θ}\sin \mathrm{\θ}\sin \mathrm{\α}\cos \mathrm{\α}\\ G=\sin \mathrm{\θ}\cos \mathrm{\α}\\ H=\cos \mathrm{\θ}\cos \mathrm{\α}\sin \mathrm{\β}\\ W=2\cos \mathrm{\θ}\sin \mathrm{\θ}\sin \mathrm{\α}\cos \mathrm{\α}{\sin }^2\mathrm{\β}\\ J=\sin \mathrm{\θ}\sin \mathrm{\α}\sin \mathrm{\β}\end{array}\right.$

Then can obtain following relational expression:

$r^{\&prime;}=\left|F\right|\&times;\frac{\sqrt{M^2-2N+{\mathrm{<figure-callout\; id="2"\; label="G"\; filenames="HDA0000070466230000011.png,HDA0000070466230000031.png"\; state="\{\{state\}\}">G</figure-callout>}}^2+{\mathrm{<figure-callout\; id="2"\; label="H"\; filenames="HDA0000070466230000011.png,HDA0000070466230000031.png"\; state="\{\{state\}\}">H</figure-callout>}}^2+2W+{\mathrm{<figure-callout\; id="2"\; label="J"\; filenames="HDA0000070466230000011.png,HDA0000070466230000031.png"\; state="\{\{state\}\}">J</figure-callout>}}^2}}{|-x\cos \mathrm{\&alpha;}\cos \mathrm{\&beta;}-y\sin \mathrm{\&alpha;}\cos \mathrm{\&beta;}+R|}---\left(2\right)$

In order to obtain the value of β; Automatically seek in the edge system in cut; Use laser head to get radius and be r in the zone of appointment; The center of circle is the standard laser hot spot mark circular pattern of S, in testing process, sets S and coordinate origin O point and overlaps; And with the round center of circle of the center alignment tab of CCD; The image of picked-up marker graphic adopts the coordinate of image processing method certification mark circle diagram shape in imaging plane coordinate O ' X ' Y ', in the computed image central coordinate of circle to all circumferential point coordinates apart from r ' _iBecause the marker graphic that laser cutting machine is got is a standard circular, so α and θ angle do not influence the detection of vertical angle in the computing of reality, for convenience of calculation, getting it is 90 °, and then (2) formula abbreviation is:

$r_i^{\&prime;}=\left|F\right|\&times;\frac{r\sqrt{{\left(\cos \mathrm{\&theta;}\right)}^2+{\left(\sin \mathrm{\&theta;}\sin \mathrm{\&beta;}\right)}^2}}{|-y\cos \mathrm{\&beta;}+R|}---\left(3\right)$

Obvious condition R＞＞r is satisfied, and cos β ≈ 0 is then arranged, then | and-ycos β+R| ≈ R, but the following formula abbreviation is:

$r_i^{\&prime;}=K\&times;r\sqrt{1-{\sin }^2\mathrm{\&theta;}(1-{\sin }^2\mathrm{\&beta;})}---\left(4\right)$

The amplification coefficient of K=|F|/R imaging system wherein.The r ' that following formula records when being CCD optical axis and imaging plane out of plumb _i, can know variation, r ' along with the θ angle _iValue is arranged:

r′ _imax＝K×r (5)

r′ _imin＝K×rsinβ (6)

In standard marked circle image, along with the variation of angle θ, r ' _iTwo maximum value and two minimal values will be arranged, two maximum value maximal value for equating, unequal when two minimal values, and r ' is arranged _Imin1＜r ' _Imin2, can get by formula (4):

r′ _imax＝K×r (7)

So can obtain:

K＝r′ _imax/r (8)

Again because: r ' _Imin2=K * r * sin β can obtain calculating the formula of the verticality of CCD optical axis and cutting planes:

β＝arcsin(r′ _imin2/r′ _imax) (9)

Can find out that from formula (9) ask value and the laser cutting machine of the β radius of a circle r that makes marks irrelevant, the precision for guaranteeing to detect can make the mark radius of a circle more as far as possible according to the size in the CCD visual field, to improve accuracy of detection.Utilize formula (9) can obtain the vertical angle β of laser machine cutting bed vise plane and CCD optical axis.

Adopt above-mentioned relation, the coordinate of the mark coordinate of the reality through being positioned at cutting bed vise and image correspondence markings point that gauge point is gathered concerns, the above relational expression of substitution just can calculate the deflection angle of CCD, thereby calculate the verticality of the installation of CCD.In the calculating of above-mentioned corresponding relation; In order to take care of yourself the accuracy of detection that CCD installs deflection angle; The present invention proposes to ask average implementation method to improve accuracy of detection to the testing result of a plurality of subregions; Promptly in the centre of cut bed and the zone at close four angles a-quadrant as shown in Figure 2 is set respectively; B zone C zone, D zone and E zone, as shown in Figure 5; Detect the installation deviation angle of CCD respectively with identical method, ask its average installation deviation angle at last as final CCD.

Claims (6)

1. seek the detection method that limit laser cutting machine CCD installs verticality automatically for one kind, it is characterized in that, may further comprise the steps:

(1) directly utilizes the laser head of laser cutting machine, on the fixing demarcation paper of laying of five zoness of different of cutting bed vise workplace, get circle and the center mark pattern of forming by laser facula;

(2) pass through the marking image that CCD absorbs five zoness of different;

(3) utilize image processing techniques, extract the circumference and the centre point coordinate of marking image respectively;

(4) calculate the extreme value of the center of circle, obtain the verticality of CCD optical axis and cutting bed vise by computation model to circumference each point distance.

2. detection method of seeking limit laser cutting machine CCD installation perpendicularity automatically according to claim 1; It is characterized in that; Comprise following substep in the described step (1): at four angles of cutting bed vise and five zones in centre position; Tile regularly respectively five and demarcate paper; The laser head of control cutting machine is demarcated the marked circle and the center of circle of getting the constant radius of being made up of laser facula on the paper at each and is carried out mark; The marked region size is limited in the visual field of a CCD imaging; Location alignment point when the center mark point of each mark zone is adopted image for CCD; And in the process that perpendicularity detects, five positions of demarcating paper remain unchanged.

3. detection method of seeking limit laser cutting machine CCD installation verticality automatically according to claim 1; It is characterized in that; Comprise following substep in the described step (2): each zone marker figure is gathered; Elder generation's driven CCD roughly moves to the zone of appointment; Field-of-view image is obtained in dynamic acquisition and processing then; Extract the coordinate of surveyed area center mark point; Be used to control the adjustment of CCD position; Make the center mark point adjust to the center of CCD images acquired; The marking image collection is carried out in the locking visual field.

4. detection method of seeking limit laser cutting machine CCD installation verticality automatically according to claim 1; It is characterized in that; Described step comprises following substep in (3): utilize image processing techniques; Size according to laser facula; Design size different disposal template is carried out the mixed noise Filtering Processing; And adopt the sub-pixel positioning algorithm, accurately extract the coordinate of circumference gauge point and center mark point in image.

5. detection method of seeking limit laser cutting machine CCD installation verticality automatically according to claim 1; It is characterized in that; Comprise following substep in the described step (4): to each marker region image of gathering; After accurately extracting the marker graphic coordinate; Calculate center mark and put the extreme value of each circumference gauge point distance; Comprising two equal maximal values of different directions and two minimum value that do not wait of different directions and time minimum value; The substitution computation model calculates the verticality that CCD installs; The same processing done in five zones, tries to achieve the CCD verticality in five zones respectively, and the verticality of averaging and installing as CCD.

6. seek the detection system that limit laser cutting machine CCD installs verticality automatically for one kind, comprise cutting bed vise, described cutting bed vise is provided with can be along the mounting bracket of cutting bed vise both sides translation; Described mounting bracket is provided with can be along the laser cutting machine and the CCD of mounting bracket slip; Described laser cutting machine links to each other with computer system respectively with CCD, it is characterized in that, described computer system comprises: control module, image processing module and data processing module; Described control module is used to control five zones of laser cutting machine on cutting bed vise, according to preassigned position and radius, accurately gets circular on the paper and mark is carried out in the center of circle demarcating, and controls described CCD calibration graph is gathered; Described image processing module is used for the image of gathering is carried out pre-service, utilizes the sub-pixel positioning algorithm, accurately extracts each point coordinate in the mark center of circle and circumference in every two field picture; Described data processing module is used for calculating required parameter according to the coordinate that extracts marker graphic, and the substitution computation model, obtains CCD optical axis and the verticality of cutting bed surface.

Images