Utility model content
The purpose of this utility model is in order to solve the defect of above-mentioned prior art, provides a kind of can realization based on machine vision technique to complete welding work pieces track at the welding track correction Special testing device based on machine vision of x axle and the detection of y direction of principal axis side-play amount.
The purpose of this utility model can be by taking following technical scheme to reach:
Welding track based on machine vision is proofreaied and correct Special testing device, it is characterized in that: comprise workbench, precision positioning mechanism, machine vision collecting device and control appliance, described precision positioning mechanism is connected with control appliance respectively with machine vision collecting device, described precision positioning mechanism, machine vision collecting device and control appliance are arranged on workbench, and described machine vision collecting device is positioned at the top of precision positioning mechanism; Described precision positioning mechanism comprises x direction of principal axis driver element, y direction of principal axis driver element, x direction of principal axis position detection unit and y direction of principal axis position detection unit; When test, described y direction of principal axis driver element drives welding work pieces to move along y direction of principal axis, and described x direction of principal axis driver element drives y direction of principal axis driver element that welding work pieces is moved along x direction of principal axis.
Preferably, described x direction of principal axis driver element comprises the first AC servo motor, the first shaft coupling and the first ball screw connecting successively, is provided with x direction of principal axis travelling carriage on described the first ball screw; Described y direction of principal axis driver element comprises the second AC servo motor, the second shaft coupling and the second ball screw connecting successively, is provided with y direction of principal axis travelling carriage on described the second ball screw; Described y direction of principal axis driver element is placed on x direction of principal axis travelling carriage, and described y direction of principal axis travelling carriage is used for placing welding work pieces when test; Described x direction of principal axis position detection unit and y direction of principal axis position detection unit adopt respectively the first linear grid ruler and the second linear grid ruler, described the first linear grid ruler is arranged on a side of x direction of principal axis driver element, and described the second linear grid ruler is arranged on a side of y direction of principal axis driver element.
Preferably, described machine vision collecting device comprises industrial camera, camera lens and light source, and described camera lens is connected with industrial camera and aims at precision positioning mechanism, and described distribution of light sources at camera lens around.
Preferably, described control appliance comprises industrial control host, motion controller, the first AC servo driver and the second AC servo driver, described the first AC servo driver is connected with the first AC servo motor, described the second AC servo driver is connected with the second AC servo motor, described the first AC servo driver and the second AC servo driver are set up communication by motion controller and industrial control host respectively, described the first linear grid ruler is connected with motion controller by communication interface respectively with the second linear grid ruler, described industrial control host is connected with industrial camera by bus communication card, described industrial control host is connected with for realizing the display of man-machine interaction.
Preferably, described workbench adopts frame structure, one side of this frame structure is provided with vertical supports, in described vertical supports, be vertically fixed with horizontal support, described precision positioning mechanism is arranged in the top planes of frame structure, described control appliance is arranged on the inside of frame structure, and described machine vision collecting device is fixed on horizontal support.
Preferably, the bottom of described frame structure is provided with four wheels that can make movable workbench.
Preferably, the frame structure that the frame structure that described workbench adopts is 1000*800*700mm.
Another object of the present utility model can be by taking following technical scheme to reach:
The utility model has following beneficial effect with respect to prior art:
Welding track based on machine vision of the present utility model is proofreaied and correct Special testing device, by precision positioning mechanism and machine vision collecting device are set, can realize based on machine vision technique and complete welding work pieces track in x axle and the detection of y direction of principal axis side-play amount, and utilize industrial control host testing result is assessed and verified, when meeting required precision, can be integrated in industrial robot control system and apply, the reason that analytical error produces when not meeting required precision, continue to optimize the image processing method in machine vision technique, thereby realize the object that welding track side-play amount accurately detects.
The specific embodiment
Embodiment 1:
As shown in Figures 2 and 3, the welding track based on machine vision of the present embodiment is proofreaied and correct Special testing device, comprise workbench 1, precision positioning mechanism, machine vision collecting device and control appliance, described workbench 1 adopts frame structure 1, the frame structure that this frame structure 1 is 1000*800*700mm, adopt 40*40 standard aluminum section bar to form, its bottom is provided with four wheels 2 that can make workbench 1 move, one side is provided with vertical supports 3, in described vertical supports 3, be vertically fixed with horizontal support 4, described precision positioning mechanism is arranged in the top planes of frame structure 1, described control appliance is arranged on the inside of frame structure 1, described machine vision collecting device is fixed on horizontal support 4, described machine vision collecting device is positioned at the top of precision positioning mechanism, described precision positioning mechanism comprises x direction of principal axis driver element 5, y direction of principal axis driver element 6, x direction of principal axis position detection unit 7 and y direction of principal axis position detection unit 8,
Described x direction of principal axis driver element 5 comprises the first AC servo motor 5-1, the first shaft coupling 5-2 and the first ball screw 5-3 connecting successively, is provided with x direction of principal axis travelling carriage 5-4 on described the first ball screw 5-3; Described y direction of principal axis driver element 6 comprises the second AC servo motor 6-1, the second shaft coupling 6-2 and the second ball screw 6-3 connecting successively, is provided with y direction of principal axis travelling carriage 6-4 on described the second ball screw 6-3; It is upper that described y direction of principal axis driver element 6 is placed in x direction of principal axis travelling carriage 5-4, and described y direction of principal axis travelling carriage 6-4 is used for placing welding work pieces when test; Described x direction of principal axis position detection unit 7 and y direction of principal axis position detection unit 8 adopt respectively the first linear grid ruler 7 and the second linear grid ruler 8, described the first linear grid ruler 7 is arranged on a side of x direction of principal axis driver element 5, and described the second linear grid ruler 8 is arranged on a side of y direction of principal axis driver element 6;
Described machine vision collecting device comprises industrial camera 9, camera lens 10 and light source 11, and described camera lens 10 is connected with industrial camera 9 and aims at precision positioning mechanism, and described light source 11 is distributed in camera lens 10 around;
Described control appliance comprises industrial control host 12, motion controller 13, the first AC servo driver 14 and the second AC servo driver 15.
In the present embodiment, the HC-KFS-23A model motor of described the first AC servo motor 5-1Wei Japanese mitsubishi electric Co., Ltd, the HC-KFS-43A model motor of described the second AC servo motor 6-1Wei Japanese mitsubishi electric Co., Ltd; Described the first ball screw 5-3 and the second ball screw 6-3 all adopt the KK6005P-600A1-FE-CS2 model screw mandrel of Taiwan Shang Yin Co., Ltd, and its helical pitch is 5mm, and effectively rail length is 600mm, and precision is ± 0.01mm; Described the first linear grid ruler 7 and the second linear grid ruler 8 all adopt the Spain MKT-52 of Fa Ge FAGOR company model, and effective travel is 520mm, and resolution ratio is 5um, and precision is ± 10um; Described industrial camera 9 adopts 800,000 pixel 1394 cameras of German Ying Meijing Co., Ltd (IMAGINGSOURCE), model is DMK31AF03, resolution ratio is 1024x768, Pixel Dimensions horizontal direction is 4.65um, vertical direction is 4.65um, the M2514-MP2 model mega pixel camera lens of the Japanese Computar of camera lens 10 employing company, focal length is 25mm; GTS-400-PV (G)-PCI movement sequence controller that described motion controller adopts solid High Seience Technology Co., Ltd. to produce; Described the first AC servo driver 14 adopts the MR-J2S-20A model servo-driver of Japanese mitsubishi electric Co., Ltd, and described the second AC servo driver 15 adopts the MR-J2S-40A model servo-driver of Japanese mitsubishi electric Co., Ltd.
In conjunction with the welding track based on machine vision shown in Fig. 4, proofread and correct Special testing device control circuit figure, described the first AC servo driver 14 is connected with the first AC servo motor 5-1, described the second AC servo driver 15 is connected with the second AC servo motor 6-1, described the first AC servo driver 14 and the second AC servo driver 15 are set up communication by motion controller 13 and industrial control host 12 respectively, described the first linear grid ruler 7 is connected with motion controller 13 with CN13 communication interface by CN12 respectively with the second linear grid ruler 8, described industrial control host 12 is connected with industrial camera 9 by 1394 bus communication cards 16, described industrial control host 12 is connected with display 17, can realize man-machine interaction, by the man-machine interface shown in Fig. 5, set the kinematic parameter of x axle and y axle, motion controller 13 is to the first AC servo driver 14 and the second AC servo driver 15 output pulse and direction signals, drive the firstth AC servo motor 5-1 to drive x direction of principal axis travelling carriage 5-4, and drive the second AC servo motor 6-1 to drive y direction of principal axis travelling carriage 6-4 to arrive anchor point, the first linear grid ruler 7 and the second linear grid ruler 8 detect the x shaft position signal of x direction of principal axis travelling carriage 5-4 and the y shaft position signal of y direction of principal axis travelling carriage 6-4, and be input to motion controller 13 through 2 tunnel quadruple increment type auxiliaring coding devices, in man-machine interface, can obtain testing result.
The welding track based on machine vision of the present embodiment is proofreaied and correct the method for testing of Special testing device, it is characterized in that comprising the following steps:
1) adopt scaling board to demarcate, obtain industrial camera intrinsic parameter and outer parameter;
2) welding work pieces is positioned on the y direction of principal axis travelling carriage of precision positioning mechanism, according to the current location of welding work pieces, by industrial camera, captures welding work pieces image and set up welding work pieces template;
3) by industrial control host, set welding work pieces x axle and the y axle offset amount on workbench, controlling the first AC servo motor and the first ball screw drives welding work pieces to move to the x axle offset amount position of setting, control the y axle offset amount position that the second AC servo motor and the second ball screw driving welding work pieces move to setting, by the first linear grid ruler and the second linear grid ruler, obtain respectively x axle and the y axle real offset (x of current welding work pieces
r, y
r);
4) after welding work pieces skew, by industrial camera, again capture the welding work pieces image of current location, and adopt image noise reduction algorithm to process image, specific as follows:
After welding work pieces skew, by industrial camera, again capture the welding work pieces image of current location, this image is 24bit coloured image, the gray level image that transfers 256 grades by image binaryzation to, adopt again mean filter to complete noise reduction process, wherein mean filter mask size is m * n, and in image, any point (x, y) response is:
Wherein, g (x, y) is mask pixel average,
5) adopt the template matching algorithm based on gray value, obtain welding work pieces with respect to x axle and the axial side-play amount (x of y of template workpiece
m, y
m), specific as follows:
A) adopt the template matching algorithm based on gray value, by calculating normalizated correlation coefficient NCC, determine the similarity of current welding work pieces image and template image coupling, determine thus welding work pieces with respect to the side-play amount of template workpiece x axle and y axle in image coordinate system (x '
m, y '
m):
Wherein, t (i, j) is welding work pieces template gray value, and f (x+i, y+j) is current welding work pieces gradation of image value, m
ttemplate average gray value,
the variance of all gray values of template, m
f(x, y) and m
f(x, y) is average gray value and the variance of each point in the current welding work pieces template of translation; When normalizated correlation coefficient NCC=± 1, between welding work pieces template and current welding work pieces image, mate completely, and normalizated correlation coefficient NCC absolute value more approaches 1, represent that welding work pieces template is more approaching with the welding work pieces image detecting;
B) industrial camera obtains intrinsic parameter and outer parameter after demarcating, x axle above-mentioned steps a) being obtained through seven submatrixs conversion and affine transformation and the side-play amount of y axle (x '
m, y '
m) be converted to x axle and y axle offset amount (x in tool coordinates system
m, y
m);
6) compare (x
r, y
r) and (x
m, y
m) value, according to welding robot, reproduce process allowable error, judge whether the precision of Machine Vision Detection meets the demands, judge whether to meet
Δ wherein
xand Δ
yerror for welding system permission; If meet the demands, can be integrated in industrial robot control system and apply; If do not meet the demands, the reason that analytical error produces, is optimized from industrial camera demarcation, image noise reduction Processing Algorithm and template matching algorithm, improves certainty of measurement, thereby meets instructions for use.
In sum, welding track correction Special testing device based on machine vision of the present utility model can be realized based on machine vision technique and complete welding work pieces track in x axle and the detection of y direction of principal axis side-play amount, and testing result is assessed and verified, the reason that analytical error produces, continue to optimize the image processing method in machine vision technique, thereby realize the object that welding track side-play amount accurately detects.
The above; it is only the utility model patent preferred embodiment; but the protection domain of the utility model patent is not limited to this; anyly be familiar with those skilled in the art in the disclosed scope of the utility model patent; according to the technical scheme of the utility model patent and utility model design thereof, be equal to replacement or changed, all being belonged to the protection domain of the utility model patent.