CN112893186B - Rapid visual detection method and system for electrifying LED lamp filament - Google Patents
Rapid visual detection method and system for electrifying LED lamp filament Download PDFInfo
- Publication number
- CN112893186B CN112893186B CN202110041253.3A CN202110041253A CN112893186B CN 112893186 B CN112893186 B CN 112893186B CN 202110041253 A CN202110041253 A CN 202110041253A CN 112893186 B CN112893186 B CN 112893186B
- Authority
- CN
- China
- Prior art keywords
- image
- light
- led filament
- led
- filament
- 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
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
- B07C5/342—Sorting according to other particular properties according to optical properties, e.g. colour
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/02—Measures preceding sorting, e.g. arranging articles in a stream orientating
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
-
- H—ELECTRICITY
- H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
- H05K—PRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
- H05K13/00—Apparatus or processes specially adapted for manufacturing or adjusting assemblages of electric components
- H05K13/08—Monitoring manufacture of assemblages
Landscapes
- Engineering & Computer Science (AREA)
- Operations Research (AREA)
- Manufacturing & Machinery (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention belongs to the technical field of LED detection, and particularly relates to a method and a system for quickly visually detecting the electrification of an LED filament, which comprises the following steps: s1, collecting a power-on image of an LED filament; s2, cutting the electrified image of the LED filament and performing threshold processing by a binary method; s3, performing gridding partitioning on the processed image according to the number of the LED filaments and the number of light-emitting points of each LED filament; and S4, setting a qualified threshold, calculating the light-emitting area of the light-emitting point in each grid, judging the light-emitting point with the light-emitting area within the qualified threshold interval as qualified, and judging other light-emitting points as unqualified. The method has simple and quick image processing, greatly improves the power-on detection efficiency of the LED lamp filament, and can be widely applied to the field of LED detection.
Description
Technical Field
The invention belongs to the technical field of LED detection, and particularly relates to a method and a system for quickly visually detecting power-on of an LED filament.
Background
In recent years, in the lighting industry, LED light emitting products have been on the market with advantages such as long power saving life. After years of development, the LED light-emitting products enter the red sea stage, and the key for the survival of various manufacturers is how to improve the productivity and reduce the cost in various links. In some high-end applications, all products need to be electrified and detected one by one, each luminous point is measured, and the products with excessive, insufficient or insufficient brightness need to be eliminated. A typical example is the production industry of LED filaments, the delivery price of finished products after dispensing and drying is less than 10% of the original delivery price, differential sales are often carried out for improving benefits, the best quality is selected and sold at a higher price to ensure the overall benefits, and the rest is subjected to secondary and tertiary screening and low-price treatment. To achieve this, all products need to be electrified for detection, the picture after electrification is shown in fig. 1, fig. 1 is a photograph of a part with 12 persons in a visual field, and in many cases, the number of the whole supports is 48 in one edition for improving the production efficiency; under the condition of higher requirements, all the luminous points on each lamp are required to have the brightness within the same narrow range, and filaments with excessive brightness, dark brightness or poor brightness appearing in individual points are eliminated. The work is carried out manually all the time, because the vision of people can be fatigued quickly and is easily influenced by external conditions, such as light, fatigue and other factors, and the subjective standards of all quality inspectors are greatly different, the production efficiency is low and the reject ratio of final products is high, so that the consistency of power-on detection of the LED lamp filaments is difficult to ensure by manual detection, and the efficiency is low.
There is also a conceivable technical solution to find a circle in the entire image by commercial image recognition software. However, in practical application, the effect is very poor, on one hand, the time is long, more than 100 seconds are needed for processing one picture frequently, on the other hand, the misjudgment rate is high, and because the light-emitting points are not completely circular, the problem that the light-emitting point position identification rate is low and the misjudgment rate is high exists when the scheme is applied. Therefore, it is very necessary to develop a rapid visual detection method for the power-on of the LED filament.
Disclosure of Invention
The invention overcomes the defects of the prior art, and solves the technical problems that: a method and a system for rapid visual inspection of LED filament power-on are provided.
In order to solve the technical problems, the invention adopts the technical scheme that: a quick visual detection method for power-on of an LED lamp filament comprises the following steps:
s1, collecting a power-on image of an LED filament plate;
s2, cutting the electrified image of the LED filament plate and performing threshold processing by a binary method, and removing areas irrelevant to the LED filament in the image and halos around each luminous point;
s3, carrying out gridding partitioning on the processed image according to the number of filaments in the LED filament plate and the number of light-emitting points of each LED filament;
and S4, setting a qualified threshold, calculating the light-emitting area of the light-emitting point in each grid, judging the light-emitting point with the light-emitting area within the qualified threshold interval as qualified, and judging other light-emitting points as unqualified.
The step S1 also comprises the following steps before:
s0, placing the material box into a material rack, automatically pushing out a filament plate through a machine, and electrifying;
the following steps are also included after the step S4:
and S5, recording the position of the unqualified filament, cutting off the unqualified filament, and conveying the remaining filament plate to a blanking box.
In step S3, a specific method for performing gridding partitioning on the image is as follows:
and (4) searching the image processed in the step (S2) line by line from top to bottom to find out the first bright pixel, and taking the line where the pixel is as the base line of the horizontal line of the partition.
In step S3, before performing the gridding partition on the image, a step of shifting and rotating the image is further included.
In step S3, the method for calculating the image shift amount and the rotation amount includes:
after the position angle of a standard product is adjusted in place in advance, recording the positions of four mark points forming a rectangle on the product in an image and the pixel equivalent weight of the image, and calculating the central position and the angle of the rectangle;
after the power-on image of the LED filament to be detected is acquired, the central position and the angle of the rectangle are calculated according to the positions of four mark points in the image, the central position and the angle are compared with data of a standard product, and the offset and the rotation amount are calculated.
In addition, the invention also provides a rapid visual inspection system for electrifying the LED lamp filament, which comprises:
a feeding unit: the automatic feeding device is used for automatically feeding materials;
an image acquisition unit: the LED filament plate is used for collecting images of the electrified LED filament plate;
a calculation unit: the device is used for calculating and processing the image of the LED filament plate and judging the qualified condition;
a PLC control unit: the servo mechanism is used for cutting off unqualified filaments according to the judgment result of the calculation unit;
a blanking unit: and the filament plate is used for conveying the treated filament plate to the blanking box and blanking.
The calculation unit calculates and processes the image of the LED filament plate, and the specific method for judging the qualified condition comprises the following steps:
cutting the electrified image of the LED filament and carrying out threshold processing by a binary method, and removing areas irrelevant to the LED filament and halos around each light-emitting point in the image;
performing gridding partitioning on the processed image according to the number of the LED filaments and the number of light-emitting points of each LED filament;
and setting a qualified threshold, calculating the area of the calculation light-emitting point in each grid, and judging the light-emitting point with the area in a qualified threshold interval as qualified, and judging other light-emitting points as unqualified.
The specific method for the calculation unit to perform gridding partition on the image comprises the following steps:
and searching the first bright pixel line by line from top to bottom of the processed image, and taking the line of the pixel as a horizontal line baseline of the partition.
Compared with the prior art, the invention has the following beneficial effects: the invention provides a method and a system for quickly visually detecting the electrification of an LED lamp filament, which can greatly improve the detection efficiency of an LED by electrifying and photographing the LED lamp filament and automatically identifying the LED lamp filament, have high accuracy and stable and reliable system and can be widely applied to the field of the electrification detection of the LED lamp filament.
Drawings
Fig. 1 is a control flow chart of a method for rapid visual inspection of power-on of an LED filament according to an embodiment of the present invention;
FIG. 2 is a flowchart of image processing according to one embodiment of the present invention;
fig. 3 is a schematic diagram illustrating a result of performing gridding partition processing on an image according to a first embodiment of the present invention;
FIG. 4 is a graph showing the results obtained in the example of the present invention;
FIG. 5 is a graph showing another result obtained in the example of the present invention;
fig. 6 is a block diagram of a structure of a rapid visual inspection system for power-on of an LED filament according to a second embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below, and it is obvious that the described embodiments are some embodiments of the present invention, but not all embodiments; all other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
Example one
As shown in fig. 1, an embodiment of the present invention provides a method for quickly visually detecting a power-on state of an LED filament, including the following steps:
s0, placing the material box into a material rack, automatically pushing out a filament plate through a machine, and electrifying.
S1, collecting power-on images of the LED lamp filaments.
And S2, cutting the electrified image of the LED filament and performing threshold processing by a binary method, and removing areas irrelevant to the LED filament and halos around each luminous point in the image.
And S3, carrying out gridding partitioning on the processed image according to the number of the LED lamp filaments and the number of the light-emitting points of each LED lamp filament.
The specific method for meshing and partitioning the image comprises the following steps:
and (3) searching the first bright pixel line by line from top to bottom of the image processed in the step (S2), and taking the line where the pixel is as the horizontal line base line of the partition.
When in photographing, the product smoothly passes through the flow channel and reaches the position right below the camera, a certain gap exists between the upper side and the lower side, and the image generated in the way has certain drift in the vertical direction. In order to improve the identification accuracy, the position where the light-emitting point starts is found to be used as a transverse line baseline, the consistency in the vertical direction can meet the requirement in the manufacturing of a product support, after the light-emitting point enters the visual field of a camera, the position of an LED filament plate is relatively fixed after the light-emitting point enters the visual field of the camera, the light-emitting point is mechanically limited and positioned and is detected by a proximity switch, each filament can be distributed in a grid after being uniformly divided in the direction every time on an image, the situation of crossing the grid can not occur, and the problem of the baseline of the filament is not considered. Since the LED filament plate is searched line by line from top to bottom in this embodiment, all filaments in the image, whichever one is, are used as the baseline as long as there is a point where the brightness meets the standard and the position is the highest, rather than searching from one filament only. In an extreme case, when the first light-emitting points of all the filaments are not turned on, the situation that the base line is shifted down by one light-emitting point occurs, and at the moment, all the points in the bottom row in the calculation result are considered not to be turned on, so that all the filaments are considered to be rejected and are completely cut off, and in this case, the machine is stopped and a human is informed to check.
Fig. 2 is a schematic diagram of an image after gridding and partitioning is performed on the image according to the embodiment of the present invention.
In addition, in this embodiment, before performing gridding partitioning on the image, a step of shifting and rotating the image is further included.
The method for calculating the image offset and the rotation amount comprises the following steps:
after the position angle of a standard product is adjusted in place in advance, recording the positions of four mark points forming a rectangle on the product in an image and the pixel equivalent weight of the image, and calculating the central position and the angle of the rectangle;
after the power-on image of the LED filament to be detected is collected, the central position and the angle of the rectangle are calculated according to the positions of four mark points in the image, the central position and the angle are compared with data of a standard product, and the offset and the rotation amount are calculated.
In specific implementation, a state matrix (for example, 100 × 50) may be established for all light-emitting points on the entire support, each filament corresponds to one row, and all points are initialized to zero by using a memory block operation instruction before replacing each support, that is, all light-emitting points are qualified by default, which is feasible because the actual situation is that most of the light-emitting points are qualified. Before detection, a completely qualified support is used as a template, a matrix area is uniformly divided on a screen, exactly one light-emitting point is arranged, and recording is carried out for later use. In order to improve the identification precision, after each photographing and binarization processing, the first bright pixel is found out line by line from top to bottom and is used as a base line of the zoning. And then, counting the area of the luminous pixel points in each region and marking on a screen.
Aiming at the conditions that the quality of some products is poor or the gap of a conveying flow channel is large, the condition that the light-emitting point is large after photographing can occur, and the collected photos need to be shifted and rotated.
Specifically, four circular MARK points forming a rectangular vertex are arranged on the frame of each support, the positions of the MARK points on different supports are very consistent, the positions of the points can be searched before each support is detected, the offset and the rotation angle of the support are calculated by using the difference between the coordinates of the MARK points in the image acquired in the measurement process and the coordinates of the MARK points in the image acquired when the support is at the standard position, and then the light-emitting point area is divided after the image is rotated and offset according to the two parameters.
In practical operation, a simple algorithm is provided, only three mark points are enough to be photographed, two points on the long side of the rectangle are used for calculating the rotation angle, and the middle point of the two points on the diagonal is used as the center, namely the position, of the rectangle. It has been practically proven that this accuracy can meet production requirements.
And S4, setting a qualified threshold, calculating the light-emitting area of the light-emitting point in each grid, judging the light-emitting point with the light-emitting area within the qualified threshold interval as qualified, and judging other light-emitting points as unqualified. Specifically, assume that the threshold interval is [ A, B ], where A < B. And if the light-emitting area is larger than A and smaller than B, judging the light-emitting area as qualified, and if the light-emitting area is smaller than A or larger than B, judging the light-emitting area as unqualified.
Each luminous point in the LED filament is light emitted after each luminous LED particle is electrified and penetrates through the color glue, so that the phenomenon of divergence inevitably occurs, the shape is different, the edge is fuzzy, and the LED filament is not round with the first impression. This also results in time and labor consuming and high error rates when identifying using circles as templates. In the embodiment, the acquired images are discretized, and after binarization processing, all points of the whole image are divided into a bright state and a non-bright state, so that the threshold value can be adjusted according to requirements, and the standard of qualified products can be controlled. The number of pixels in the bright area accurately reflects the area of the luminous point, so that the number of pixels with the standard brightness in each area is counted, the area of the luminous point is measured, a good and efficient method is provided, and the basic purpose of separating each luminous point by dividing the area is achieved.
As shown in fig. 4 and 5, which are schematic diagrams of displaying the inverted image and the final determination result on a display screen according to the embodiment of the present invention, during displaying, the light emitting areas of the light emitting points with the luminance not reaching the standard or the luminance exceeding the standard may be displayed in the corresponding areas, the unqualified area is enlarged on the lower side in fig. 4, the numbers thereof show the light emitting areas of the light emitting points with the luminance not reaching the standard, the unqualified area is enlarged on the right side in fig. 5, the numbers thereof show the light emitting areas of the light emitting points with the luminance exceeding the standard, and the area not displaying the numbers is a qualified product. Wherein, the luminous area of the luminous point is related to the number of pixels with the standard brightness in the area.
And S5, recording the position of the unqualified filament, cutting off the unqualified filament, and conveying the remaining filament plate to a blanking box.
This embodiment can also play the effect that reduces the defective rate through the detection analysis of going up electricity to the LED filament, has the guide effect to production process. Analyzing fig. 4, it can be seen that there is a large percentage of filament with insufficient brightness, the points with insufficient brightness are all concentrated at the lowest edge of the image, and the problem of the filament hardware itself can be basically eliminated, and the possibility of this happening is too small. Through the investigation and inquiry of the production field, in the production process, in order to ensure the attractive appearance of the product, the glue needle is not directly lifted out but has a back-wiping action in the final stage of glue collection, so that the glue quantity on the last luminous point is excessive, and the light is blocked from transmitting. Based on the viewpoint, the dispensing action is recommended to be operated for 2mm more in production, and then the wiping action is executed. By this improvement, the yield is greatly increased. Analyzing fig. 5, it is found that many light-emitting points of the rightmost filament exceed the brightness. There are basically two reasons for this, one is that the spot is off and light leaks out, which in this case is eliminated; another reason is that the yield of the filament at this position is much improved immediately after the replacement of the dispensing pin, which is suspected of being blocked by the insufficient amount of glue.
Example two
As shown in fig. 6, a second embodiment of the present invention provides a rapid visual inspection system for power-on of an LED filament, including:
an image acquisition unit: the LED filament plate is used for collecting images of the electrified LED filament plate;
a calculation unit: the device is used for calculating and processing the image of the LED filament plate and judging the qualified condition;
a PLC control unit: the servo mechanism is used for controlling the servo mechanism to cut off unqualified filaments according to the judgment result of the calculation unit;
specifically, in this embodiment, the calculating unit performs calculation processing on the image of the LED filament plate, and a specific method for determining the qualified condition is as follows:
1. cutting an electrified image of the LED filament and performing binary threshold processing, and removing areas irrelevant to the LED filament and halos around each light-emitting point in the image;
2. performing gridding partitioning on the processed image according to the number of the LED filaments and the number of light-emitting points of each LED filament;
3. and setting a qualified threshold, calculating the light-emitting area of the light-emitting point in each grid, judging the light-emitting point with the light-emitting area within the qualified threshold interval as qualified, and judging other light-emitting points as unqualified. Specifically, assuming that the threshold interval is [ a, B ], in the determination, when the light-emitting area is larger than a and smaller than B, the determination is made as pass, and when the light-emitting area is smaller than a or larger than B, the determination is made as fail.
Further, the rapid visual inspection system for power-on of the LED filament of this embodiment further includes:
a feeding unit: the automatic feeding device is used for automatically feeding materials;
a blanking unit: and the filament plate is used for conveying the processed filament plate to a blanking box and blanking.
Further, the quick visual detection system of electricity on LED filament of this embodiment, still include the input unit who is connected with PLC the control unit, output unit and motion control unit, wherein, the input unit includes various button switch, limit switch and origin switch, a each state for PLC the control unit monitored control system, motion control unit includes the material loading axle, the runner axle, the cutter advances and retreats, cutter lift and ejection of compact axle, PLC the control unit is according to the state of each sensor, control system carries out the operation such as material loading, the biography material, the cutter, unloading, furthermore, output unit includes relay, pneumatic valve etc., PLC the control unit is through the state of control relay and pneumatic valve, other operating procedure that can the control system.
In addition, in this embodiment, the image capturing unit may specifically be a commercial camera. The system may further comprise a touch display screen for displaying the captured images and for human-computer interaction.
Specifically, in this embodiment, the specific method for performing the gridding partition on the image by the computing unit is as follows:
and searching the first bright pixel line by line from top to bottom of the processed image, and taking the line of the pixel as a horizontal line baseline of the partition.
The LED filament electrifying rapid visual detection system provided by the embodiment of the invention has lower requirements on hardware such as a camera and the like and an operating system. The industrial camera is a USB3.0 interface, 500 ten thousand pixels and has the image resolution of 2448X2048. The computing unit can be written in C # language by utilizing Open CV image processing technology in a Visual Studio2019 environment under 64-bit Windows 10.
In summary, the invention provides a method and a system for rapid visual inspection of power-on of an LED filament, wherein the actual measurement time for identifying the LED filament of a whole support under a 6 generation i3 processor is within 0.2 second through the inspection research of power-on luminous points of the LED filament by computer vision. The 32K filaments are continuously selected on one machine when the detection system is applied to equipment inspection, and the filaments are all qualified after manual one-by-one re-inspection, so that the production efficiency and quality of the detection system can be improved, and the detection consistency is good. Through statistics, one machine and one staff carry out auxiliary work of loading and unloading the material boxes, which is equivalent to the recognition speed of 6 staff. Therefore, in actual production, one worker can keep track of three machines, the operation requirements on the worker are greatly reduced, and the unmanned control on the detection system is realized. In the processing algorithm, digital image processing tools such as a binary method and affine transformation are fully used, the influence of long detection time and high misjudgment rate is comprehensively overcome aiming at the target characteristics, the requirement of the detection system on accuracy is finally met, the capacity is improved, the cost is reduced, and a foundation is laid for future practicability and commercialization.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and these modifications or substitutions do not depart from the spirit of the corresponding technical solutions of the embodiments of the present invention.
Claims (5)
1. A quick visual detection method for power-on of an LED filament is characterized by comprising the following steps:
s1, collecting a power-on image of an LED filament plate;
s2, cutting the electrified image of the LED filament plate and performing threshold processing by a binary method, and removing areas irrelevant to the LED filament in the image and halos around each luminous point;
s3, carrying out gridding partitioning on the processed image according to the number of lamp filaments in the LED lamp filament plate and the number of light-emitting points of each LED lamp filament;
s4, setting a qualified threshold, calculating the light-emitting area of the light-emitting point in each grid, judging the light-emitting point with the light-emitting area within a qualified threshold interval as qualified, and judging other light-emitting points as unqualified;
in step S3, before performing gridding partition on the image, the method further includes a step of shifting and rotating the image, and the method for calculating the image shift amount and the rotation amount includes:
after the position angle of a standard product is adjusted in place in advance, recording the positions of four mark points forming a rectangle on the product in an image and the pixel equivalent weight of the image, and calculating the central position and the angle of the rectangle;
after the power-on image of the LED filament to be detected is acquired, the central position and the angle of the rectangle are calculated according to the positions of four mark points in the image, the central position and the angle are compared with data of a standard product, and the offset and the rotation amount are calculated.
2. The LED filament power-on rapid visual inspection method according to claim 1, further comprising the following steps before the step S1:
s0, placing the material box into a material rack, automatically pushing out a filament plate through a machine, and electrifying the filament plate;
the step S4 is followed by the following steps:
and S5, recording the position of the unqualified filament, shearing off the unqualified filament, and conveying the remaining filament plate to a discharge box.
3. The method for rapid visual inspection of the electrification of the LED filament according to claim 1, wherein in the step S3, the specific method for gridding and partitioning the image is as follows:
and (3) searching the first bright pixel line by line from top to bottom of the image processed in the step (S2), and taking the line where the pixel is as the horizontal line base line of the partition.
4. The utility model provides a quick visual inspection system of LED filament power-on which characterized in that includes:
a feeding unit: the automatic feeding device is used for automatically feeding materials;
an image acquisition unit: the LED filament plate is used for collecting images of the electrified LED filament plate;
a calculation unit: the device is used for calculating and processing the image of the LED filament plate and judging the qualified condition;
a PLC control unit: the servo mechanism is used for controlling the servo mechanism to cut off unqualified filaments according to the judgment result of the calculation unit;
a blanking unit: the filament plate is used for conveying the treated filament plate to a blanking box and blanking;
the calculation unit calculates and processes the image of the LED filament plate, and the specific method for judging the qualified condition comprises the following steps:
cutting the electrified image of the LED filament and carrying out threshold processing by a binary method, and removing areas irrelevant to the LED filament and halos around each light-emitting point in the image;
shifting and rotating the processed image, and carrying out gridding partitioning according to the number of the LED filaments and the number of light-emitting points of each LED filament; the method for calculating the image offset and the rotation amount comprises the following steps:
after the position angle of a standard product is adjusted in place in advance, recording the positions of four mark points forming a rectangle on the product in an image and the pixel equivalent weight of the image, and calculating the central position and the angle of the rectangle;
after a power-on image of the LED filament to be detected is acquired, calculating the central position and the angle of a rectangle according to the positions of four mark points in the image, comparing the central position and the angle with data of a standard product, and calculating the offset and the rotation amount;
and setting a qualified threshold, calculating the area of the calculation light-emitting point in each grid, judging the light-emitting point with the area within the qualified threshold interval as qualified, and judging other light-emitting points as unqualified.
5. The LED filament power-on rapid visual inspection system according to claim 4, wherein the specific method for the computing unit to perform gridding partitioning on the image is as follows:
and searching the first bright pixel of the processed image line by line from top to bottom, and taking the line where the pixel is as the horizontal line baseline of the partition.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110041253.3A CN112893186B (en) | 2021-01-13 | 2021-01-13 | Rapid visual detection method and system for electrifying LED lamp filament |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110041253.3A CN112893186B (en) | 2021-01-13 | 2021-01-13 | Rapid visual detection method and system for electrifying LED lamp filament |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112893186A CN112893186A (en) | 2021-06-04 |
CN112893186B true CN112893186B (en) | 2022-11-29 |
Family
ID=76112707
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110041253.3A Active CN112893186B (en) | 2021-01-13 | 2021-01-13 | Rapid visual detection method and system for electrifying LED lamp filament |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112893186B (en) |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4344539A (en) * | 1978-05-05 | 1982-08-17 | Lockett James F | Universal sorting apparatus |
GB0915587D0 (en) * | 2009-09-07 | 2009-10-07 | Sony Comp Entertainment Europe | Image processing apparatus, system and method |
CN102913795A (en) * | 2012-10-30 | 2013-02-06 | 江苏远大光电有限公司 | Light-emitting diode (LED) daylight lamp |
CN103804318A (en) * | 2014-02-14 | 2014-05-21 | 中山大学 | Benzothiazole derivative containing triphenylethylene or tetraphenylethylene structure and having aggregation-induced emission property and preparation method and application thereof |
CN107563380A (en) * | 2017-09-08 | 2018-01-09 | 上海理工大学 | A kind of vehicle license plate detection recognition method being combined based on MSER and SWT |
CN208401610U (en) * | 2018-03-19 | 2019-01-18 | 山西能源学院 | A kind of intelligent optimization device of renewable energy |
CN111938678A (en) * | 2020-08-10 | 2020-11-17 | 上海联影医疗科技股份有限公司 | Imaging system and method |
Family Cites Families (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3386579A (en) * | 1964-08-22 | 1968-06-04 | Schulze Ernst | Method of and apparatus for detecting light-deflecting flaws in hollow glass articles |
US9047657B2 (en) * | 2011-05-17 | 2015-06-02 | Gii Acquisition, Lcc | Method and system for optically inspecting outer peripheral surfaces of parts |
US8570504B2 (en) * | 2011-05-17 | 2013-10-29 | Gii Acquisition, Llc | Method and system for optically inspecting parts |
US9651658B2 (en) * | 2015-03-27 | 2017-05-16 | Google Inc. | Methods and systems for LIDAR optics alignment |
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 |
CN107064170B (en) * | 2017-04-11 | 2020-04-03 | 深圳市深视智能科技有限公司 | Method for detecting profile defect of mobile phone shell |
CN107492106A (en) * | 2017-08-24 | 2017-12-19 | 广州市鑫广飞信息科技有限公司 | A kind of water-surface areas dividing method |
CN108717692B (en) * | 2018-04-02 | 2022-04-08 | 浙江工业大学 | CCD image processing-based cut material deviation correcting method |
CN208729155U (en) * | 2018-07-21 | 2019-04-12 | 厦门海维尔自动化设备有限公司 | A kind of double-station feeding structure of filament automatic spot-welding device |
CN109323999B (en) * | 2018-10-24 | 2020-06-16 | 大连理工大学 | Spectrophotometric detection method based on image numerical analysis |
CN109561598B (en) * | 2018-12-19 | 2020-09-08 | 深圳市汉森软件有限公司 | Printing control method and device for PCB and storage medium |
CN109967389B (en) * | 2019-04-23 | 2020-04-21 | 南京斯杩克机器人技术有限公司 | Automatic detection system and detection method for defects of detonating tool |
CN111652022B (en) * | 2019-06-26 | 2023-09-05 | 广州虎牙科技有限公司 | Image data display method, image data live broadcast device, electronic equipment and storage medium |
CN111815757B (en) * | 2019-06-29 | 2023-04-07 | 浙江大学山东工业技术研究院 | Large member three-dimensional reconstruction method based on image sequence |
CN110455815B (en) * | 2019-09-05 | 2023-03-24 | 西安多维机器视觉检测技术有限公司 | Method and system for detecting appearance defects of electronic components |
CN110736701B (en) * | 2019-11-14 | 2021-12-07 | 林励 | Sample full-surface three-dimensional microscopic imaging system and method |
CN110939879A (en) * | 2019-12-31 | 2020-03-31 | 海宁市星达机械设备厂 | Automatic assembly device of LED filament |
CN112037318A (en) * | 2020-07-22 | 2020-12-04 | 山东大学 | Construction method and system of three-dimensional rock mass structure model and application of model |
CN111862220B (en) * | 2020-07-31 | 2022-10-11 | 广东利元亨智能装备股份有限公司 | Correction method and device for UVW platform calibration, deviation correction method and alignment system |
CN112058706A (en) * | 2020-08-13 | 2020-12-11 | 深圳市秀骑士科技有限公司 | Power-on self-inspection system for DMX lamp |
-
2021
- 2021-01-13 CN CN202110041253.3A patent/CN112893186B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4344539A (en) * | 1978-05-05 | 1982-08-17 | Lockett James F | Universal sorting apparatus |
GB0915587D0 (en) * | 2009-09-07 | 2009-10-07 | Sony Comp Entertainment Europe | Image processing apparatus, system and method |
CN102913795A (en) * | 2012-10-30 | 2013-02-06 | 江苏远大光电有限公司 | Light-emitting diode (LED) daylight lamp |
CN103804318A (en) * | 2014-02-14 | 2014-05-21 | 中山大学 | Benzothiazole derivative containing triphenylethylene or tetraphenylethylene structure and having aggregation-induced emission property and preparation method and application thereof |
CN107563380A (en) * | 2017-09-08 | 2018-01-09 | 上海理工大学 | A kind of vehicle license plate detection recognition method being combined based on MSER and SWT |
CN208401610U (en) * | 2018-03-19 | 2019-01-18 | 山西能源学院 | A kind of intelligent optimization device of renewable energy |
CN111938678A (en) * | 2020-08-10 | 2020-11-17 | 上海联影医疗科技股份有限公司 | Imaging system and method |
Also Published As
Publication number | Publication date |
---|---|
CN112893186A (en) | 2021-06-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111879781B (en) | Test paper quality detection system based on machine vision | |
CN104574389A (en) | Battery piece chromatism selection control method based on color machine vision | |
CN107895362B (en) | Machine vision method for detecting quality of miniature wiring terminal | |
CN109856156A (en) | A kind of display panel tiny flaw determination method and device based on AOI | |
CN109916910B (en) | Photovoltaic glass edge defect detection system and corresponding method | |
US11657599B2 (en) | Method for detecting appearance of six sides of chip multi-layer ceramic capacitor based on artificial intelligence | |
CN111047655A (en) | High-definition camera cloth defect detection method based on convolutional neural network | |
CN116188475B (en) | Intelligent control method, system and medium for automatic optical detection of appearance defects | |
CN110189670A (en) | A kind of LED display screen defect inspection method | |
CN112819844B (en) | Image edge detection method and device | |
US11947345B2 (en) | System and method for intelligently monitoring a production line | |
CN114119591A (en) | Display screen picture quality detection method | |
CN108802052A (en) | A kind of detecting system and its detection method about slide fastener defect | |
CN115908432A (en) | Material output quality detection system and prediction method | |
CN113592852A (en) | Carbon fiber carbon wire quality online detection system and method | |
CN110596118A (en) | Print pattern detection method and print pattern detection device | |
CN112893186B (en) | Rapid visual detection method and system for electrifying LED lamp filament | |
CN106501278B (en) | Surface of the light tube defect classification method and system based on invariable rotary textural characteristics | |
WO2021156873A1 (en) | System and method for imaging reflecting objects | |
CN105973910A (en) | Structure and texture characteristic-based lamp tube quality detection and flaw classification method and system thereof | |
CN116840240A (en) | Visual detection system of power supply distributor | |
CN110866917A (en) | Tablet type and arrangement mode identification method based on machine vision | |
CN114125435B (en) | Intelligent image defect detection method for network camera | |
CN211263229U (en) | Bottled liquid detection optical system | |
CN210953857U (en) | Product operation flow monitoring system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |