CN104034516B - Machine vision based LED detection device and detection method thereof - Google Patents
Machine vision based LED detection device and detection method thereof Download PDFInfo
- Publication number
- CN104034516B CN104034516B CN201410283584.8A CN201410283584A CN104034516B CN 104034516 B CN104034516 B CN 104034516B CN 201410283584 A CN201410283584 A CN 201410283584A CN 104034516 B CN104034516 B CN 104034516B
- Authority
- CN
- China
- Prior art keywords
- led
- positioning
- camera
- machine vision
- industrial computer
- 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
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention relates to the technical field of machine vision, in particular to a machine vision based LED detection device and a detection method thereof. The LED detection device comprises a carrier plate clamp, a camera obscura and an industrial personal computer, an industrial camera connected with the industrial personal computer is arranged in the camera obscura, a prime lens is arranged on the industrial camera, a condensation plate disposed at the front end of the prime lens is further arranged in the camera obscura, at least two positioning lamps are arranged on a plane of the condensation plate, through holes are formed in a coordinate system plane formed by the positioning plates, and the carrier plate clamp is connected with the through holes through a conduction fiber. The machine vision based LED detection device can detect LEDs quickly, efficiently and stably with low cost, and is applicable to any application environment needing to detect a large quantity of printed circuit board LEDs; the detection method can quickly and efficiently complete detection of dozens of LEDs, and is simple, convenient and rapid to operate.
Description
Technical field
The present invention relates to technical field of machine vision, more specifically, relate to a kind of LED pick-up unit based on machine vision and detection method thereof.
Background technology
Machine vision, as the term suggests be exactly make machine have visual performance as people, thus realizes the functions such as various detection, judgement, identification, measurement.Vision Builder for Automated Inspection is generally made up of video camera, image pick-up card, computing machine, light source etc., and its principle of work is: under certain illumination condition, with video camera, the target image that is ingested of three-dimensional scenic is collected computer-internal formation original image; Then, use image processing techniques to carry out pre-service to improve picture quality to original image, partition graph, extract characteristic element, form the description to image; Finally, mode identification technology is adopted to carry out tagsort, and according to pre-conditioned Output rusults.
Machine vision have contactless with the object be observed, objective to subject not damaged, observation process, differentiate the feature that result reliability is high; Simultaneously machine vision has widened human visible range, cannot the occasion of perception in many human visions, and as the perception etc. of high-risk scene under industrial environment, machine vision has more advantage; And machine vision can quick obtaining bulk information, being easy to automatic process, being also easy to realize information integerated, is the basic technology realizing computer integrated manufacturing system.
The existing LED checkout equipment based on machine vision generally has two kinds, and one uses color sensor to detect, as spectroscope; Another kind detects by measuring LED conduction voltage drop.But the cost utilizing sensor to carry out detecting is high, does not support that many LED detect simultaneously, have the shortcomings such as detection efficiency is low, this approach utilizing sensor to carry out detecting generally is applicable to laboratory test, is not suitable for the high volume applications of production line; And be a kind of indirect inspection approach by the mode that LED conduction voltage drop carries out detecting, really can not measure the characteristic feature of LED, as color, brightness etc.
Summary of the invention
The present invention is intended to solve the problems of the technologies described above at least to a certain extent.
Primary and foremost purpose of the present invention is to provide a kind of LED pick-up unit based on machine vision, and this device can complete with lower cost and rapidly and efficiently stably detect LED, and is applicable to the applied environment of any needs a large amount of detection circuitry plate LED.
A further object of the present invention is to provide a kind of detection method adopting the above-mentioned LED pick-up unit based on machine vision, uses the method can complete the detection of dozens of LED quickly and efficiently simultaneously, fast simple, convenient.
For solving the problems of the technologies described above, the technical solution used in the present invention is:
A kind of LED pick-up unit based on machine vision is provided, comprise the support plate fixture for placing LED to be measured, camera bellows and industrial computer, the industrial camera be connected with described industrial computer is provided with in described camera bellows, described industrial camera is provided with tight shot, the solar panel being placed in described tight shot front end is also provided with in described camera bellows, described solar panel plane is provided with at least two positioning lamps, arrange through hole in the coordinate plane consisted of described positioning lamp, described support plate fixture is connected with described through hole by conduction optical fiber.
Further, described solar panel plane is provided with the different positioning lamp of three colors, and these three positioning lamps are arranged on any three angles in described solar panel plane four angles respectively.The accuracy direct relation of positioning lamp the location recognition of LED to be measured.
Preferably, described three positioning lamps choose red, green, blue three kinds of colors respectively, identify the position coordinates of described three colored positioning lamps when using so that follow-up by the principle of three primary colours.
Preferably, described through hole is provided with several, and through hole described in several forms matrix form through hole.Through hole is except adopting the layout of matrix form, linear placement can also be adopted, but in order to reach preferably effect, adopt the design of matrix form through hole, effectively can utilize the sensor area of industrial camera, also facilitate the design of location algorithm, and reduce the accuracy requirement to positioning lamp, thus reach better stability.
Further, described support plate fixture is provided with fixture connection interface, and described fixture connection interface is provided with fiber optic conduction module, and one end of described conduction optical fiber is connected with described through hole with described fiber optic conduction model calling, the other end.
The present invention also provides a kind of detection method adopted as the above-mentioned LED pick-up unit based on machine vision, comprises the following steps:
S1. LED to be measured lights luminescence, LED light to be conducted on the solar panel in camera bellows by conduction optical fiber;
S2. industrial camera gathers the original image on solar panel and reaches industrial computer by PCI-Express;
S3. described industrial computer carries out pre-service to the original image collected, and first identifies positioning lamp;
S4. described industrial computer calculates area-of-interest after obtaining positioning lamp, and by this area maps on original image, identifies the position coordinates of positioning lamp further;
S5. by the position coordinates adjustment image direction of step S4, redraw coordinate system, and be according to identifying other LED to be measured further with this coordinate;
S6. according to the M*N matrix template of design in advance, calculated the coordinate position of each LED in matrix by the coordinate system of step S5, and determine its area-of-interest according to the coordinate position of each LED;
S7. each area-of-interest that step S6 obtains is mapped on original image, the LED eigenwert identified in single region by the composition analyzed one by one in single region;
S8. the LED eigenwert obtained according to the criterion preset and step S7 contrasts, and as unanimously being then considered as passing through, otherwise is considered as bad;
If be S9. judged to bad by step S8, described industrial computer then preserves bad original image for failure analysis foundation.
Wherein, the positioning lamp in described step S3 is the colored positioning lamp that three colors are respectively red, green, blue; In described step S4, identified the position coordinates of described three colored positioning lamps by the principle of three primary colours.
Compared with prior art, the invention has the beneficial effects as follows:
(1) cost is low.The LED pick-up unit that the present invention is based on machine vision carrys out concentrated photographic subjects by conduction optical fiber and solar panel, effective minimizing visual field area, thus do not need the equipment such as high high pixel camera, wide-angle lens and adjusting triangle support just can realize the accurate detection of extensive number LED; Also improve the applicability of device, not by the impact of the layout of LED to be measured own simultaneously.
(2) high-level efficiency is detected.The LED pick-up unit that the present invention is based on machine vision can complete the detection of the performances such as the color of tens of LEDs and brightness simultaneously, and the time used only needs several seconds.
(3) stability is high.The LED pick-up unit that the present invention is based on machine vision accurately identifies LED by three look positioning lamps, reduces widely because camera lens rocks, the reason such as to loosen and cause camera to lose the fault of focus, the stability of this device is promoted greatly.
(4) applied widely.The LED pick-up unit that the present invention is based on machine vision only needs suitably to change support plate fixture, and can complete the test of the LED product of more than ten different model on same device, other all parts are shared to save cost.
(5) volume is little, modularization.The LED pick-up unit that the present invention is based on machine vision both can independent utility, also can be integrated in other automatization test systems.
(6) the present invention adopts the detection method of the above-mentioned LED pick-up unit based on machine vision, simple to operate, convenient and swift, can complete the detection of dozens of LED quickly and efficiently simultaneously.
Accompanying drawing explanation
Fig. 1 is the structural representation of embodiment one based on the LED pick-up unit of machine vision.
Fig. 2 is the solar panel planar structure schematic diagram of embodiment one based on the LED pick-up unit of machine vision.
Fig. 3 is the X-Y plane coordinate system that embodiment one or three positioning lamps are formed.
Embodiment
Below in conjunction with embodiment, the present invention is further illustrated.Wherein, accompanying drawing only for exemplary illustration, expression be only schematic diagram, but not pictorial diagram, can not be interpreted as the restriction to this patent; In order to better embodiments of the invention are described, some parts of accompanying drawing have omission, zoom in or out, and do not represent the size of actual product; For a person skilled in the art, in accompanying drawing, some known features and explanation thereof may be omitted is understandable.
The corresponding same or analogous parts of same or analogous label in the accompanying drawing of the embodiment of the present invention; In describing the invention, it will be appreciated that, if have term " on ", the orientation of the instruction such as D score, "left", "right" or position relationship be for based on orientation shown in the drawings or position relationship, only the present invention for convenience of description and simplified characterization, instead of indicate or imply that the device of indication or element must have specific orientation, with specific azimuth configuration and operation, therefore the term describing position relationship in accompanying drawing, only for exemplary illustration, can not be interpreted as the restriction to this patent.
Further, above-mentioned part term represents except orientation or position relationship except may be used for, and also may be used for representing other implications, such as term " on " also may be used in some cases representing certain relations of dependence or annexation.For those of ordinary skills, these terms concrete meaning in the present invention can be understood as the case may be.
Embodiment one
As shown in Figure 1 to Figure 3 for the present invention is based on the embodiment of the LED pick-up unit of machine vision.As Fig. 1, shown in Fig. 2, this LED pick-up unit comprises the support plate fixture 1 for placing LED to be measured, camera bellows 2 and industrial computer 3, the industrial camera 21 be connected with industrial computer 3 is provided with in camera bellows 2, industrial camera 21 is by camera support 22 fixed support be arranged in camera bellows 2, industrial camera 21 is provided with tight shot 23, the solar panel 24 being placed in tight shot 23 front end is also provided with in camera bellows 2, solar panel 24 plane is provided with the positioning lamp 25 that three colors are different, through hole 26 is set in the coordinate plane formed by positioning lamp 25, support plate fixture 1 is connected with through hole 26 by conduction optical fiber 4.Wherein, support plate fixture 1 is also provided with fixture connection interface 11, this fixture connection interface 11 is provided with fiber optic conduction module (not shown), and one end of conduction optical fiber 4 is connected with through hole with fiber optic conduction model calling, the other end.
In the present embodiment, as shown in Figure 2, the positioning lamp 25 that three colors in solar panel 24 plane are different, these three positioning lamps 25 are arranged on any three angles in solar panel 24 plane four angles respectively.The accuracy direct relation of positioning lamp the location recognition of LED to be measured.As shown in Figure 3, three positioning lamps form an X-Y plane coordinate system, are divided into some matrixes according to this in plane, the position of the corresponding LED of each matrix unit, the i.e. position of figure interior joint.
For the ease of identifying the position coordinates of three colored positioning lamps during follow-up use by the principle of three primary colours, three positioning lamps 25 of the present embodiment choose red, green, blue three kinds of colors respectively.
In addition, the through hole arranged in the coordinate plane consisted of three positioning lamps 25 has several, and this several through hole forms matrix form through hole.Be understandable that, these through holes are except adopting the layout of matrix form, linear placement can also be adopted, but in order to reach preferably effect, the present embodiment adopts the design of matrix form through hole, effectively can utilize the sensor area of industrial camera, also facilitate the design of location algorithm, and the accuracy requirement reduced positioning lamp, thus reach better stability.
The size multiselect of the support plate fixture 1 of the present embodiment, replaceable.When detecting the LED of different model, only need suitably to change support plate fixture, can complete the test of the LED product of more than ten different model on same device, other all parts are shared to save cost.
Embodiment two
The present invention is the embodiment of the detection method adopting the above-mentioned LED pick-up unit based on machine vision, comprises the following steps:
S1. LED to be measured lights luminescence, LED light to be conducted on the solar panel in camera bellows by conduction optical fiber;
S2. industrial camera gathers the original image on solar panel and reaches industrial computer by PCI-Express;
S3. described industrial computer carries out pre-service to the original image collected, and first identifies the colored positioning lamp that three colors are respectively red, green, blue;
S4. described industrial computer calculates area-of-interest after obtaining positioning lamp, and by this area maps on original image, identifies the position coordinates of three colored positioning lamps further by the principle of three primary colours; After positioning lamp is identified, its absolute position is known, so-called absolute position, the position namely obtained relative to original photo.
S5. by the position coordinates adjustment image direction of step S4, redraw coordinate system, this coordinate system is the X-Y coordinate system built according to the absolute position of positioning lamp, and is according to identifying other LED to be measured further with this coordinate; In addition, the present embodiment adopts the mode of Fiber isolation conduction, avoids influencing each other of the test between each LED.
S6. according to the M*N matrix template of design in advance, calculated the coordinate position of each LED in matrix by the coordinate system of step S5, and determine its area-of-interest according to the coordinate position of each LED; Be understandable that, M*N matrix template mainly designs according to the quantity of LED to be measured, and it comprises size and each LED relative position in a matrix of planning matrix in advance.
S7. each area-of-interest that step S6 obtains is mapped on original image, the LED eigenwert identified in single region by the composition analyzed one by one in single region; Be specially and the rgb value of LED is compared.
S8. the LED eigenwert obtained according to the criterion preset and step S7 contrasts, and as unanimously being then considered as passing through, otherwise is considered as bad;
If be S9. judged to bad by step S8, described industrial computer then preserves bad original image for failure analysis foundation.
In step s 6, generally carry out design matrix template according to the quantity of LED to be measured, as the present embodiment can be designed to the matrix size of 12*12, so just can support identifying of maximum 144 LED; If the quantity of LED is less than 100, just 10*10 can be designed to.So just effectively can make full use of the sensor area of industrial camera.
In compute matrix during the coordinate position of LED, basic principle is as follows, for 12*12 matrix, 1,300,000 pixel industrial cameras.The valid pixel of 1300000 pixels is 1280*960, and in this device, available maximum pixel is 960*960, wherein removes positioning lamp again and stays white region, and 580*580 elemental area can be used for the layout of LED to be measured.Like this, each LED can take 48*48 elemental area.If be (0,0) with the initial point of X-Y coordinate system, then the coordinate of LED is then followed successively by (N*24+24, M*24+24).
Adopt the detection efficiency of the present embodiment detection method high, through test, use said apparatus and adopt this detection method to detect, completing the color of 60 plurality of LEDs and the detection of brightness characteristics simultaneously, only need the time of 2 ~ 3 seconds.
Obviously, the above embodiment of the present invention is only for example of the present invention is clearly described, and is not the restriction to embodiments of the present invention.For those of ordinary skill in the field, can also make other changes in different forms on the basis of the above description.Here exhaustive without the need to also giving all embodiments.All any amendments done within the spirit and principles in the present invention, equivalent to replace and improvement etc., within the protection domain that all should be included in the claims in the present invention.
Claims (3)
1. the detection method based on the LED pick-up unit of machine vision, this pick-up unit comprises the support plate fixture for placing LED to be measured, camera bellows and industrial computer, the industrial camera be connected with described industrial computer is provided with in described camera bellows, described industrial camera is provided with tight shot, the solar panel being placed in described tight shot front end is also provided with in described camera bellows, described solar panel plane is provided with at least three positioning lamps, these three positioning lamps are arranged on any three angles in described solar panel plane four angles respectively, in the coordinate plane formed by described positioning lamp, through hole is set, described support plate fixture is connected with described through hole by conduction optical fiber, described industrial computer is by first identifying the position coordinates of these three positioning lamps, and redraw coordinate system according to the position coordinates obtained, the coordinate position of each LED in coordinate system is calculated by the coordinate system redrawn, coordinate position according to each LED determines its area-of-interest, the LED eigenwert identified in single region by the composition analyzed one by one in single region, and carry out contrasting to realize the detection to LED according to the LED eigenwert in the criterion preset and each single region, described detection method comprises the steps:
S1. LED to be measured lights luminescence, LED light to be conducted on the solar panel in described camera bellows by described conduction optical fiber;
S2. described industrial camera gathers the original image on described solar panel and reaches described industrial computer;
S3. described industrial computer carries out pre-service to the original image collected, and first identifies three positioning lamps;
S4. described industrial computer calculates area-of-interest after obtaining positioning lamp, and by this area maps on original image, identifies the position coordinates of three positioning lamps further;
S5. by the position coordinates adjustment image direction of step S4, redraw coordinate system, and be according to identifying other LED to be measured further with this coordinate;
S6. according to the M*N matrix template of design in advance, calculated the coordinate position of each LED in matrix by the coordinate system of step S5, and determine its area-of-interest according to the coordinate position of each LED;
S7. each area-of-interest that step S6 obtains is mapped on original image, the LED eigenwert identified in single region by the composition analyzed one by one in single region;
S8. the LED eigenwert obtained according to the criterion preset and step S7 contrasts, and as unanimously being then considered as passing through, otherwise is considered as bad;
If be S9. judged to bad by step S8, described industrial computer then preserves bad original image for failure analysis foundation.
2. detection method according to claim 1, is characterized in that, the positioning lamp in described step S3 is the colored positioning lamp that three colors are respectively red, green, blue.
3. detection method according to claim 2, is characterized in that, in described step S4, is identified the position coordinates of three colored positioning lamps by the principle of three primary colours.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410283584.8A CN104034516B (en) | 2014-06-23 | 2014-06-23 | Machine vision based LED detection device and detection method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410283584.8A CN104034516B (en) | 2014-06-23 | 2014-06-23 | Machine vision based LED detection device and detection method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104034516A CN104034516A (en) | 2014-09-10 |
| CN104034516B true CN104034516B (en) | 2015-06-03 |
Family
ID=51465373
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410283584.8A Active CN104034516B (en) | 2014-06-23 | 2014-06-23 | Machine vision based LED detection device and detection method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104034516B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105241638A (en) * | 2015-09-09 | 2016-01-13 | 重庆平伟光电科技有限公司 | Vision-based quick LED module brightness uniformity detection method |
| CN105703827B (en) * | 2016-04-22 | 2018-04-24 | 成都英鑫光电科技有限公司 | A kind of optical connector fault detection method and device and controller |
| CN106053020A (en) * | 2016-06-07 | 2016-10-26 | 北京安达维尔民用航空技术有限公司 | Optical detection device |
| CN108205101A (en) * | 2016-12-19 | 2018-06-26 | 常州星宇车灯股份有限公司 | The detecting system and its method of rear combination lamp PCBA board based on machine vision |
| CN109000890A (en) * | 2018-05-02 | 2018-12-14 | 四川斐讯信息技术有限公司 | LED light detection device and method |
| CN108414915B (en) * | 2018-06-13 | 2020-12-08 | 苏州晶品新材料股份有限公司 | LED light source nondestructive real-time chip distinguishable measuring device |
| CN108981943B (en) * | 2018-06-13 | 2021-08-13 | 苏州晶品新材料股份有限公司 | Nondestructive real-time measurement method for junction temperature of LED light source |
| CN110062502B (en) * | 2019-04-22 | 2020-05-19 | 北京航空航天大学 | Machine vision-based online residual life prediction and reliability evaluation method for LED lighting lamp |
| TWI729836B (en) * | 2020-06-04 | 2021-06-01 | 和碩聯合科技股份有限公司 | Light-emitting element inspection device |
| CN111721507B (en) * | 2020-06-30 | 2022-08-19 | 东莞市聚明电子科技有限公司 | Intelligent detection method and device for keyboard backlight module based on polar coordinate identification |
| CN112461503A (en) * | 2020-11-15 | 2021-03-09 | 珠海速乐科技有限公司 | LED lamp panel visual detection device and detection method |
| CN114751195A (en) * | 2021-01-11 | 2022-07-15 | 北京小米移动软件有限公司 | Material taking and placing device, carrier plate, material testing system and material testing method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002296115A (en) * | 2001-04-02 | 2002-10-09 | Unitec:Kk | Method of measuring tone of light emitting body, measuring device for the same, and measuring device for brightness of light emitting body |
| CN1982857B (en) * | 2005-12-14 | 2011-06-29 | 深圳迈瑞生物医疗电子股份有限公司 | Method for automatically measuring luminescent device |
| CN101187684B (en) * | 2006-11-15 | 2011-06-22 | 鸿富锦精密工业（深圳）有限公司 | Main board light-emitting diode detection device and method |
| CN202382852U (en) * | 2012-02-29 | 2012-08-15 | 华高科技（苏州）有限公司 | Detecting device for LED (Light-Emitting Diode) lamp |
| CN203178063U (en) * | 2013-04-08 | 2013-09-04 | 苏州索拉科技有限公司 | LED color coordinate test system |
-
2014
- 2014-06-23 CN CN201410283584.8A patent/CN104034516B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN104034516A (en) | 2014-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104034516B (en) | Machine vision based LED detection device and detection method thereof | |
| CN103048333B (en) | Appearance detection equipment and method | |
| US8797388B2 (en) | Mounting accuracy inspection method and inspection apparatus using the inspection method | |
| US9977981B2 (en) | Method and apparatus for calibrating a three-dimensional (3D) image in a tiled display | |
| US9454024B1 (en) | Composite repairing apparatus for liquid crystal panel | |
| CN201707304U (en) | Device for automated optical inspection of surface of product | |
| CN204649634U (en) | A kind of infrared imaging pick-up unit for surface-mounted devices | |
| CN209803004U (en) | Device for simultaneously carrying out optical detection on front surface and back surface of object | |
| CN109765234A (en) | Two surfaces positive and negative to object carry out the device and method of optical detection simultaneously | |
| CN103033127B (en) | Base plate pre-alignment pose measuring method | |
| CN109862345B (en) | Method and system for testing field angle | |
| CN102879404B (en) | System for automatically detecting medical capsule defects in industrial structure scene | |
| CN109596054A (en) | The size detection recognition methods of strip workpiece | |
| CN104330749A (en) | Front-side-illuminating LED light strip test device and method | |
| CN104376323B (en) | A kind of method and device for determining target range | |
| CN108469591A (en) | A kind of detection method and system of RGB LED displays | |
| CN101929843B (en) | Device and method for automatically measuring spike length of rice | |
| CN114820439A (en) | PCB bare board defect detection system and method based on AOI | |
| CN202952978U (en) | Train part information collector | |
| CN108072664B (en) | Appearance detection system and method | |
| KR20140075042A (en) | Apparatus for inspecting of display panel and method thereof | |
| CN104344944A (en) | Photoelectrical detecting system of LED (light emitting diode) digital tube and photoelectrical detecting method of photoelectrical detecting system | |
| CN105184739A (en) | Printed circuit board AOI detection image stitching method | |
| CN108353159A (en) | Measure the rotation position of column lens | |
| CN102865849A (en) | Camera device for ranging and ranging method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |