CN209680599U - A kind of automatic detection device of automobile parts - Google Patents
A kind of automatic detection device of automobile parts Download PDFInfo
- Publication number
- CN209680599U CN209680599U CN201821382519.0U CN201821382519U CN209680599U CN 209680599 U CN209680599 U CN 209680599U CN 201821382519 U CN201821382519 U CN 201821382519U CN 209680599 U CN209680599 U CN 209680599U
- Authority
- CN
- China
- Prior art keywords
- product
- guide rail
- axis displacement
- displacement guide
- automobile parts
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
A kind of automatic detection device of automobile parts of the utility model is related to a kind of automatic machinery vision-based detection, the especially defects detection of Automobile Plastic Parts.It include: to grasp sucker 1, upper light source 2, industrial camera 3, lower light source 4, the fixed load plate 5 of product, X-axis displacement guide rail 6, Y-axis displacement guide rail 7, mechanical arm 10, conveyer belt 11, processor 12, body 13 including new product.Make processor 12 by Image Acquisition using aforesaid way, to judge required the defects of identifying object, judging the disconnected glue of terminal hole and plug-hole or flash, avoid worker's great work intensity, it works long hours and easily causes visual fatigue, judge incorrectly, influence the quality testing of product.
Description
Technical field
The utility model relates to a kind of automatic machinery vision-based detections, the especially defects detection of Automobile Plastic Parts.
Background technique
Currently, traditional industrial production and manufacturing, the method that artificial detection is still mainly used due to the limitation of science and technology
The defect of testing product is gone, this method is due to the backwardness of artificial limitation and technology, and not only the speed of testing product is slow, efficiency
Lowly, and during detection it is easy error, so as to cause the inaccurate of testing result.Today's society, vision-based detection with
Going deep into for computer technology, the emergence and development of the science and technology such as artificial intelligence, and research, occur based on machine vision
The surface defects detection technology of technology.The appearance of this technology substantially increases the efficiency of production operation, avoids because of operation item
Part, subjective judgement etc. influence the accuracy of testing result, and realization more preferably can more accurately carry out surface defects detection, more quickly
Identification product defects.
Summary of the invention
It is low, at high cost in order to solve manual detection efficiency, it is difficult to which that the problems such as realizing full inspection, a kind of automobile of the utility model are matched
The automatic detection device of part, it is characterised in that: including new product grasp sucker 1, upper light source 2, industrial camera 3, lower light source 4,
Product fixes load plate 5, X-axis displacement guide rail 6, Y-axis displacement guide rail 7, mechanical arm 10, conveyer belt 11, processor 12, body 13, institute
It states X-axis displacement guide rail 6 to connect with the 7 lower section sliding block of Y-axis displacement guide rail, can be moved horizontally in the X-direction, the Y-axis
Displacement guide rail 7 is connect with the fixed 5 lower section sliding block of load plate of the product, can be moved horizontally in the Y direction, the product is fixed
5 chassis upper surface of load plate is equipped with the lower light source 4, and there is an industrial camera 3 in the top of the fixed load plate 5 of the product, described
Industrial camera 3 is fixed by fixed bracket, and the upper light source 2 is housed immediately below the industrial camera 3,10 top of mechanical arm with
Product grasps sucker 1 and is flexibly connected, and bottom end is connected with the body 13, and the conveyer belt 11 is in the case where product grasps sucker 1 just
Side, the processor 12 are connected with mechanical arm 10, industrial camera 3 and X-axis displacement guide rail 6, Y-axis displacement guide rail 7 respectively.
A kind of automatic detection device of the automobile parts, it is characterised in that: the industrial camera 3 by camera 8 with
Telecentric lens 9 form, wherein the camera 8 is millions camera, the telecentric lens 9 are the telecentricity that enlargement ratio is 0.09 times
Camera lens.
The automatic detection device of a kind of automobile parts, it is characterised in that: the upper light source 2 is ring light.
The automatic detection device of a kind of automobile parts, it is characterised in that: the mechanical arm 10 can 6DOF fortune
It is dynamic.
The automatic detection device of a kind of automobile parts, it is characterised in that: 10 top of mechanical arm and product
Grasp 1 activity axis connection of sucker.
The automated detection method of the automatic detection device of a kind of automobile parts, it is characterised in that: it is detected
Step is as follows:
A: the processor 12 sends crawl product signal to the mechanical arm 10, and the mechanical arm 10 leads to after grabbing product
It crosses the mechanical arm point that sets up an office in advance and rotates counterclockwise to fixed 5 top of load plate of the product, and product is put into the product and is consolidated
Determine in load plate 5；
B: when product is put into the fixed load plate 5 of the product, the processor 12 is triggered to institute by weight sensor
It states industrial camera 3 and sends shooting high-definition image 1 and open instruction, then the processor 12 is to the X-axis displacement guide rail 6 and the Y
Axial displacement guide rail 7 issues move；
C: when the X-axis displacement guide rail 6 is mobile with the Y-axis displacement guide rail 7, the upper light source 2, the processing are opened
Device 12 sends the random instruction capturing high-definition image 12 and opening to the industrial camera 3, and establishes corresponding Image Acquisition library；
D: testing product nut whether there is neglected loading phenomenon: the processor 12 is by the image in Image Acquisition library come really
Determine nut appearance profile, by shape matching algorithm, obtains nut appearance profile number of matches in image, and the nut to obtain
Quantity and the product nut number of original setting compare, to determine the nut in product with the presence or absence of neglected loading phenomenon；
E: test side sub-aperture whether there is disconnected glue: open lower light source 4, the processor 12 sends to camera 3 shoot again
12 high-definition image instructions, establish Image Acquisition library by obtained high-definition image, by the image in Image Acquisition, through shape
Matching algorithm determines terminal hole, determine terminal hole and the terminal also by obtained terminal hole between white dot, In
Area is preset to this white area in processor 12, this area is indicated with pixel, according to the gray value of black white image into
Row segmentation, segmentation threshold are b (range 0-255), the terminal hole and terminal also between white dot and processor 12
In area is preset to this white dot fixed value be compared, determine that terminal hole whether there is disconnected glue phenomenon with this；
F: it test side sub-aperture plug-hole or flash defect: after E step, is examined according to terminal hole obtained in E step
It surveys, whether there is black splotch inside the sub-aperture of test side, determine the terminal hole with the presence or absence of plug-hole phenomenon with this black splotch；
G: judge whether product is qualified according to step D, E, F.
The automated detection method of the automatic detection device of a kind of automobile parts, it is characterised in that: described
Shape matching algorithm is that the elementary contour for needing shape matching object is pre-designed in processor 12, passes through the image in step C
The each image in library is acquired, the outer profile of the matching object of shape needed for each image is arranged searches for number x, and figure puts deviation angle
Degree is y1-y2(any range between -360 ° to 360 °), according to the gray value of black white image, selection Grads threshold is that (range is z
0-255, this experiment in be set as 40), to workpiece to be checked carry out shape match beyond the region of objective existence outline,
The automated detection method of the automatic detection device of a kind of automobile parts, it is characterised in that: in D step
Selection Grads threshold be z be 40.
The automated detection method of the automatic detection device of a kind of automobile parts, it is characterised in that: in E step
Selection Grads threshold be z be 60, white dot area is 5 pixels, segmentation threshold 127.
The automated detection method of the automatic detection device of a kind of automobile parts, it is characterised in that: in F-step
Selection Grads threshold be z be 45.
Make processor 12 by Image Acquisition using aforesaid way, to identify object needed for judging, judge terminal hole break glue with
The defects of plug-hole or flash, avoids worker's great work intensity, works long hours and easily cause visual fatigue, judges incorrectly, shadow
Ring the quality testing of product.
Detailed description of the invention
Fig. 1: the automated detection method and schematic device of a kind of automobile parts of the utility model；
Fig. 2: the automated detection method and device of a kind of automobile parts of the utility model detect the big high-precision light in the visual field
Learn detection device schematic diagram；
Fig. 3: the automated detection method and device overhaul flow chart of a kind of automobile parts of the utility model；
Fig. 4: whether the automated detection method and device nut of a kind of automobile parts of the utility model do not fill defects detection
Schematic diagram；
Fig. 5: the whether disconnected glue defect inspection of the automated detection method and device end sub-aperture of a kind of automobile parts of the utility model
It measures and is intended to.
Specific embodiment
With reference to the accompanying drawings and examples the automated detection method to a kind of automobile parts of the utility model and device and
Its application scenarios is illustrated.
As shown in Fig.1 and Fig.2, a kind of automated detection method of automobile parts of the utility model and device include new product
Grasping sucker 1, upper light source 2, industrial camera 3, lower light source 4, the fixed load plate 5 of product, X-axis displacement guide rail 6, Y-axis displacement guide rail 7,
Mechanical arm 10, conveyer belt 11, processor 12, body 13, X-axis displacement guide rail 6 are connect with 7 lower section sliding block of Y-axis displacement guide rail, can be
X-direction moves horizontally, and Y-axis displacement guide rail 7 is connect with the fixed 5 lower section sliding block of load plate of product, can move horizontally in Y direction,
The fixed 5 chassis upper surface of load plate of product is equipped with lower light source 4, and there are an industrial camera 3, industrial camera 3 in the top of the fixed load plate 5 of product
It is fixed by fixed bracket, upper light source 2 is housed, 10 top of mechanical arm and 1 activity of product grasping sucker are even immediately below industrial camera 3
Connect, bottom end is connected with body 13, conveyer belt 11 immediately below product grasping sucker 1, processor 12 respectively with mechanical arm 10, work
Industry camera 3 and X-axis displacement guide rail 6, Y-axis displacement guide rail 7 are connected.
In order to which high-definition image industrial camera 3 is made of camera 8 and telecentric lens 9, camera 8 is millions camera, telecentric mirror
First 9 be the telecentric lens that enlargement ratio is 0.09 times, and upper light source 2 is ring light.Mechanical arm 10 can 360 degree rotation；Mechanical arm 10
Top and product grasp 1 activity axis connection of sucker.
As shown in figure 3, figure 4 and figure 5: the automated detection method and device of a kind of automobile parts of this utility model are to vapour
The detection of vehicle accessory is as follows, by taking tesla's plastic cement automobile parts as an example, to its accessory detection testing product nut with the presence or absence of leakage
For phenomenon, test side sub-aperture are filled with the presence or absence of disconnected glue and test side sub-aperture plug-hole or flash defect, retrieving is as follows:
Before present apparatus work, the fixed load plate 5 of product is on the position respectively set with mechanical arm 10, host computer handle
The automobile parts of required detection are transported to conveyer belt 11,11 setting in motion of conveyer belt, when conveyer belt 11 transports automobile parts
Triggering processor 12 grabs the instruction of automobile parts to mechanical arm 10 when to 10 positive laxative remedy of mechanical arm, and mechanical arm 10 drives new product
Grasping inhales sucker 1 and removes crawl automobile parts, and then mechanical arm 10 passes through on preset counterclockwise movement to the fixed load plate 5 of product
Side, and automobile parts are put into the fixed load plate 5 of the product；When automobile parts are put into the fixed load plate 5 of the product, lead to
Cross weight sensor trigger processor 12 to industrial camera 3 send shooting high-definition image 1 open instruction, then the processor 12 to
The X-axis displacement guide rail 6 issues move with the Y-axis displacement guide rail 7；When X-axis displacement guide rail 6 and Y-axis displacement guide rail 7 move
When dynamic, light source 2 in opening, processor 12 sends the random instruction capturing high-definition image 12 and opening to industrial camera 3, and establishes corresponding
Image Acquisition library；
Testing product nut whether there is neglected loading phenomenon first: set nut shape to be detected in processor 12 in advance
Then profile information and number are searched for by the image in Image Acquisition library with shape matching algorithm to obtain in image outside nut
Shape outline quantity, and compared with the product nut number of obtained nut quantity and original setting, to determine in product
Nut whether there is neglected loading phenomenon, wherein shape matching algorithm are as follows: a nut foreign steamer is arranged in the image information for inputting acquisition
Wide learning template, for each image nut outer profile being arranged and searches for number x, figure puts deviation angle to nut profiles position
Degree is y1-y2(any range between -360 ° to 360 °), according to the gray value of black white image, selection Grads threshold is that (range is z
0-255, this experiment in be set as 40), to automobile parts to be checked carry out nut profiles matching, according to the nut foreign steamer being matched to
The number set in wide quantity and processor 12 compares, and if matching value, less than explanation if desirable value, there are the feelings that nut fills less
Condition, control mechanical arm 10 substandard products are recycled, if each nut profiles information can match there is no nut feed into less it is next
Walk process.Fig. 4 shows the position nut normal mounting, does not occur nut and fills less.
Then test side sub-aperture whether there is disconnected glue: open lower light source 4, the processor 12 sends to camera 3 clap again
12 high-definition image instructions are taken the photograph, Image Acquisition library is established by obtained high-definition image, by the image in Image Acquisition, are passed through
Shape matching algorithm determines terminal hole, determine terminal hole and the terminal also by obtained terminal hole between leukasmus
Point presets area to this white area in processor 12, this area is indicated with pixel, this time white dot in detection
Area be 5 pixels, be split according to the gray value of black white image, segmentation threshold be b (range 0-255), this time
Segmentation threshold is that b is 127, between the terminal hole and terminal hole white dot and processor (12) in this white dot
5 pixels of fixed value for presetting area are compared, if less than 5 pixels of detected white dot, table
Bright terminal hole then enters next step process without disconnected glue phenomenon；If more than or equal to 5 pixels, then it is disconnected to show that terminal hole exists
Glue phenomenon, mechanical arm 10 recycle substandard products.
Last test side sub-aperture plug-hole or flash defect: in test side sub-aperture break glue phenomenon on the basis of, processor 12
This is detected according to the terminal hole detected, according to the gray value of black white image, selection Grads threshold is z (range 0-
255, it is set as in this experiment 45) inside the sub-aperture of test side with the presence or absence of black splotch, terminal hole is determined with this black splotch
With the presence or absence of plug-hole phenomenon, if there are black splotch among terminal hole, for plug-hole, if there are black spots in terminal bore edges
Point is then flash, if plug-hole recycles substandard products with flash phenomenon mechanical arm 10；If no plug-hole will with flash phenomenon mechanical arm 10
Automobile parts are referred to qualified product region, rotate to the top of conveyer belt 11 again, go to grab next wanted automobile to be checked and match
Part.
In entire detection process, conveyer belt is constantly in transport status, conveying speed are as follows: according to detection speed between
Conveyer belt travelling speed.
The preferred concrete mode of the utility model or embodiment are explained in detail above in conjunction with attached drawing, but this is practical
It is novel to be not limited to the above embodiment and embodiment, it within the knowledge of those skilled in the art, can also be
Do not depart from the utility model design under the premise of various changes can be made.
Claims (5)
1. a kind of automatic detection device of automobile parts, it is characterised in that: grasp sucker (1) including new product, upper light source
(2), industrial camera (3), lower light source (4), product is fixed load plate (5), X-axis displacement guide rail (6), Y-axis displacement guide rail (7), mechanical
Arm (10), conveyer belt (11), processor (12), body (13), the X-axis displacement guide rail (6) and the Y-axis displacement guide rail (7)
The connection of lower section sliding block, can move horizontally in X-direction, below the Y-axis displacement guide rail (7) and the product fixed load plate (5)
Sliding block connection, can move horizontally in Y direction, and fixed load plate (5) the chassis upper surface of the product is equipped with the lower lower light source
(4), there is an industrial camera (3) above the product fixed load plate (5), the industrial camera (3) is consolidated by fixed bracket
It is fixed, the upper upper light source (2) is housed, mechanical arm (10) top and product grasp sucker immediately below the industrial camera (3)
(1) it is flexibly connected, bottom end is connected with the body (13), and the conveyer belt (11) is under product grasping sucker (1) just
Side, the processor (12) respectively with the mechanical arm (10), the industrial camera (3) and the X-axis displacement guide rail (6), institute
Y-axis displacement guide rail (7) is stated to be connected.
2. a kind of automatic detection device of automobile parts according to claim 1, it is characterised in that: the industrial camera
(3) it is made of camera (8) and telecentric lens (9), wherein the camera (8) is millions camera, the telecentric lens (9) are to put
The telecentric lens that big multiplying power is 0.09 times.
3. a kind of automatic detection device of automobile parts according to claim 1, it is characterised in that: the upper light source
It (2) is ring light.
4. according to a kind of automatic detection device of automobile parts described in claim 1, it is characterised in that: the mechanical arm (10)
It can the movement of 360 freedom degrees.
5. a kind of automatic detection device of automobile parts according to claim 1, it is characterised in that: the mechanical arm
(10) top and product grasp sucker (1) activity axis connection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201821382519.0U CN209680599U (en) | 2018-08-24 | 2018-08-24 | A kind of automatic detection device of automobile parts |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201821382519.0U CN209680599U (en) | 2018-08-24 | 2018-08-24 | A kind of automatic detection device of automobile parts |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN209680599U true CN209680599U (en) | 2019-11-26 |
Family
ID=68593605
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201821382519.0U Active CN209680599U (en) | 2018-08-24 | 2018-08-24 | A kind of automatic detection device of automobile parts |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN209680599U (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110856849A (en) * | 2018-08-24 | 2020-03-03 | 东莞中科蓝海智能视觉科技有限公司 | Automatic detection method and device for automobile parts |
| CN111189844A (en) * | 2020-02-25 | 2020-05-22 | 威海远航科技发展股份有限公司 | Automatic detection system for packaging quality of venous blood taking needle |
-
2018
- 2018-08-24 CN CN201821382519.0U patent/CN209680599U/en active Active
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110856849A (en) * | 2018-08-24 | 2020-03-03 | 东莞中科蓝海智能视觉科技有限公司 | Automatic detection method and device for automobile parts |
| CN111189844A (en) * | 2020-02-25 | 2020-05-22 | 威海远航科技发展股份有限公司 | Automatic detection system for packaging quality of venous blood taking needle |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102539443B (en) | Bottle body defect automatic detection method based on machine vision | |
| CN109772724B (en) | Flexible detection and analysis system for major surface and internal defects of castings | |
| CN106802301A (en) | The machine vision on-line detecting system and method for a kind of engine valve | |
| CN110216080A (en) | A kind of Quality Monitoring Control System of the PCB processing producing line based on image comparison | |
| CN209680599U (en) | A kind of automatic detection device of automobile parts | |
| Liu et al. | Automatic detection technology of surface defects on plastic products based on machine vision | |
| CN109127464A (en) | A kind of vision detects sorting equipment and its control method automatically | |
| CN106501272B (en) | Machine vision soldering tin positioning detection system | |
| CN111539927A (en) | Detection process and algorithm of automobile plastic assembly fastening buckle lack-assembly detection device | |
| CN112718552A (en) | Automatic quality defect inspection device for LED circuit board and working method thereof | |
| CN106645185A (en) | Method and device for intelligently detecting surface quality of industrial parts | |
| CN202471610U (en) | Automatic bottle body defect detecting device based on machine vision | |
| CN104764377B (en) | The high-precision formula automatic detection comparison device of cleaner oil-out screw thread | |
| CN206146851U (en) | Intellectual detection system industrial part surface quality's device | |
| Gayubo et al. | On-line machine vision system for detect split defects in sheet-metal forming processes | |
| CN205507114U (en) | Engine cylinder lid screw neglected loading detection device | |
| CN208928631U (en) | A kind of vision detects sorting equipment automatically | |
| CN112088387A (en) | System and method for detecting defects in imaged articles | |
| Giesko et al. | Advanced mechatronic system for in-line automated optical inspection of metal parts | |
| JP2010217069A (en) | Device and method for inspection of fluid leakage | |
| CN114518079A (en) | Hole internal feature detection system and detection method | |
| CN214310215U (en) | Workpiece defect detection system for automobile clutch pressure plate production line | |
| CN208125628U (en) | A kind of engine commutator quality detecting system based on machine vision | |
| Lei et al. | Automated vision inspection system for the size measurement of workpieces | |
| CN109632815A (en) | Surface defect on-line detecting system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |