AOI imaging system and use method

Abstract

The invention belongs to the technical field of optical imaging, and particularly relates to an AOI imaging system and a use method.The AOI imaging system comprises a dispensing unit, the dispensing unit comprises a dispensing machine for reinforcing a key component, i.e., a Tape-C terminal, and the dispensing machine uses UV red glue to accurately coat the positions of pins; the imaging unit comprises a top lamp and a side lamp, the top lamp is used for testing the area of the red glue, the side lamp is used for detecting the recess of the red glue, and a color digital camera is arranged right above the top lamp and the side lamp. By arranging the imaging unit, when the AOI imaging equipment detects a product subjected to dispensing, the top lamp can be coaxially and vertically irradiated on an element and a welding point, the area of red glue is tested through the black and white imaging effect, meanwhile, the side lamp can detect the sunken position of the red glue through the true color imaging effect, and the detection accuracy is improved. And whether the welding point is good or bad is fully judged, so that the dispensing position of the element is detected more comprehensively.

Description

Technical Field

[0001] The present invention belongs to the technical field of optical imaging, and in particular to an AOI imaging system and a use method thereof. Background Art

[0002] In the production process of components such as Tapy-C terminals, a dispensing process is usually required. Its main function is to reinforce and protect. It can prevent mechanical wear, environmental corrosion (such as moisture and chemicals) and improve durability. In addition, dispensing can also provide electrical insulation to prevent short circuits or electrical interference.

[0003] In the dispensing process, glue needs to be applied to the pin position, and specific glue dots or coatings are formed by precisely controlling the injection or coating of the glue. In this process, it is necessary to first set parameters such as glue flow, dispensing speed and dispensing position on the dispensing machine. The dispensing machine or spray gun will accurately apply the glue to the specified position according to the set parameters. Depending on the type of glue, the glue needs to be cured by ultraviolet (UV) lamps, heating, etc. Finally, check whether the dispensing effect meets the requirements, including coating uniformity and position accuracy.

[0004] However, in the existing dispensing process, when checking the dispensing effect, it is often done through two methods: visual inspection by staff and conventional AOI inspection. Among them, the visual inspection method of staff is inefficient, prone to visual fatigue for a long time, and the risk of omission increases. At the same time, there are two fixed people at the day and night shift dispensing stations, and the long-term labor cost is high. In the conventional AOI inspection method, since the UV-curing red glue is liquid and adheres to the uneven surface of the pin, the shape and volume vary during the dispensing process. For example, the degree of reflection of the concave and sloped surfaces is different, which will lead to poor detection effect, and the interception effect of defective products is also relatively poor, which reduces the quality of the product.

[0005] Therefore, there is an urgent need for an AOI imaging system and a method of use that can perform more comprehensive detection of the dispensing position. Summary of the invention

[0006] The purpose of the present invention is to address the problems raised in the above-mentioned background technology and to provide an AOI imaging system and a method for using the system, which can coaxially and vertically project a top light onto a component and a welding point, test the area of ​​the red glue through the black and white imaging effect, and detect the recessed position of the red glue through the true color imaging effect of the side light, so as to fully judge the quality of the welding point.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions: An AOI imaging system, comprising: A glue dispensing unit, the glue dispensing unit includes a glue dispensing machine for reinforcing the Tapy-C terminals of key components, and the glue dispensing machine uses UV red glue to accurately apply to the pin position; An imaging unit, the imaging unit comprising a top light for testing the red glue area and a side light for detecting the red glue depression, a color digital camera is provided directly above the top light and the side light, the color digital camera is used to collect images of the glue dispensing position under the top light and the side light, and a telecentric lens is provided between the color digital camera and the top light and the side light to ensure imaging consistency and accuracy; An interception unit, after the imaging unit works, detects and intercepts multiple types of defects such as glue leakage, insufficient glue, excessive glue, dents, residual glue, and missing pins of the red glue; The algorithm unit is centered around the location of the pin dispensing, and the editing detection box establishes a "tree" hierarchical relationship, and the sub-group box is linked with the parent attribute box.

[0008] Preferably, the dispensing output of a single tube of red glue of the dispensing machine is 250-400 pcs, and the glue amount at a single point is 100-150 mg.

[0009] Preferably, the workflow of the interception unit is as follows: when the AOI imaging device reports, it waits for personnel to process, and the PCB stops waiting at the exit, accompanied by a buzzer prompt, until the personnel completes the processing, and the PCB flows into the next workstation. At the same time, the products that pass will also alarm, increasing the personnel processing time.

[0010] Preferably, the imaging effect of the top light is black and white, and the imaging effect of the side light is true color.

[0011] Preferably, the algorithm unit includes multiple defect types to optimize the detection conditions. The defect types may be lack of glue, insufficient glue, exposed pins, excessive glue, depression, residual glue, and glue sticking to the body. Different lighting compensations are set to amplify the difference between defective products and qualified products.

[0012] Preferably, the algorithm unit includes a positioning algorithm for position compensation during high-speed movement of the color digital camera, an auxiliary algorithm for arranging detection frames at multiple positions, and a detection algorithm capable of detecting the position of the detection frame.

[0013] Preferably, when the color digital camera moves at high speed, there will be a deviation between the detection frame and the detected object in the picture. The position is compensated by the positioning algorithm. The conventional detection algorithm hierarchy is first positioned, and then a new detection frame is called by the algorithm for layout. Mark positioning and slice positioning are used as positioning algorithms to ensure that the position of the detection frame is fixed at the detected position. Window generation is an auxiliary algorithm to facilitate the layout of detection frames in multiple positions. The detection algorithms are brightness extraction, color extraction, brightness gradient bilateral distance, etc., to detect the position of the detection frame.

[0014] Preferably, color extraction is specifically as follows: within the area specified by the detection window, the number of pixels that meet the qualified range of HSV colors is obtained as a percentage of the window pixels. After the desired color is extracted and filtered out, the qualified range is set to distinguish between good and bad colors, and the percentage is directly calculated to obtain the percentage ratio of the extracted part to the entire detection window.

[0015] Preferably, if the parameter in the algorithm unit is within the setting range, the device determines that the product is PASS and flows into the back station. If the detected product is not within the setting range, the device determines that it is NG and outputs the judgment result, and reports the measured position and the picture of the defect.

[0016] The present invention also discloses a method for using the AOI imaging system, which specifically includes the following steps: S1: The dispensing unit uses a dispensing machine to precisely coat the UV red glue on the pin position. The dispensing machine can dispense 350pcs of red glue per tube, and the glue volume per point is 130mg. S2: The imaging unit illuminates the dispensing position with top and side lights, and the color digital camera collects image data through a telecentric lens. The top light is coaxially and vertically projected onto the component and the soldering point, and the area of ​​the red glue is tested through the black and white imaging effect. The side light can detect the sunken position of the red glue through the true color imaging effect, and fully judge the quality of the soldering point. S3: The interception unit can detect and intercept components with dispensing defects according to the algorithm unit. The algorithm unit includes: MARK positioning, slice positioning, window generation, color extraction, brightness extraction, brightness gradient bilateral distance, and color extraction; S4: The setting value of the parameter in the algorithm unit is the standard for the detected position. If it is within the range, it is judged as OK, otherwise it is NG.

[0017] Compared with existing technologies, the advantages of this AOI imaging system and method of use are: The present invention sets an imaging unit. When the AOI imaging device detects the product after the dispensing, the top light can be coaxially and vertically projected on the component and the welding point, and the area of ​​the red glue can be tested through the black and white imaging effect. At the same time, the side light can project the three-color light on the component and the welding point from different angles according to the principle of three-primary color, refraction and diffuse reflection imaging, presenting an angle (0°-30° is a flat area, 30°-60° is a steep area, and 60°-90° is a steep area). Then, the distribution of RGB three colors of the welding point image reflected by the three-color light in the color digital camera is analyzed, and the concave position of the red glue is detected through the true color imaging effect, and the quality of the welding point is fully judged, so as to more comprehensively detect the dispensing position of the component, improve the detection and interception effect of the AOI imaging device on the component, improve the product quality, and solve the problem of high labor cost of the staff during visual inspection. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a structural block diagram of an AOI imaging system provided by the present invention; Figure 2 It is the qualified standard of UV curing red glue in the present invention; Figure 3 It is a bad picture when the glue leaks in the present invention; Figure 4 It is a bad picture when there is more glue in the present invention; Figure 5 It is a bad picture when there is less glue in the present invention; Figure 6 It is a bad picture when it is concave in the present invention; Figure 7 It is a bad picture when the pin is leaking in the present invention; Figure 8 It is a bad picture when the residual glue in the present invention. DETAILED DESCRIPTION

[0019] The following examples are for illustrative purposes only and are not intended to limit the scope of the present invention.

[0020] Example: Refer to Figures 1 to 8 , an AOI imaging system, comprising: A glue dispensing unit, the glue dispensing unit includes a glue dispensing machine for reinforcing the Tapy-C terminals of key components, and the glue dispensing machine uses UV red glue to accurately apply to the pin position; Among them, the dispensing machine can dispense about 350 pcs of red glue in a single tube, and the amount of glue at a single point is about 130 mg. Therefore, when reinforcing the Tapy-C terminals of key components, the dispensing machine can accurately apply the UV red glue to the pin position to ensure 100% coverage on the Tapy-c pin surface, thereby protecting the pin, resisting tensile forces and falling, and extending the service life of the product, thereby improving product quality.

[0021] An imaging unit, the imaging unit comprising a top light for testing the red glue area and a side light for detecting the red glue depression, a color digital camera is provided directly above the top light and the side light, the color digital camera is used to collect images of the glue dispensing position under the top light and the side light, and a telecentric lens is provided between the color digital camera and the top light and the side light to ensure imaging consistency and accuracy; Among them, the imaging effect of the top light is black and white, and the imaging effect of the side light is true color.

[0022] The imaging unit here is an AOI imaging device.

[0023] In view of the problem that visual inspection by staff in the prior art is prone to visual fatigue, the risk of omission increases, and the labor cost is high, and the conventional AOI inspection has poor inspection effect due to the variable shape and volume during the dispensing process, such as different degrees of reflection of depressions and slopes, and the interception effect of defective products is also relatively poor, the present invention sets an imaging unit, and when the AOI imaging device inspects the dispensing products, the top light can be coaxially and vertically projected on the components and welding points, and the area of ​​the red glue can be tested through the black and white imaging effect, and the side light can be based on the three primary colors, refraction, and diffuse reflection imaging. The principle is to project three-color light onto the components and welding points from different angles, presenting angles (0°-30° is a flat area, 30°-60° is a steep area, and 60°-90° is a steep area). Then, the distribution of RGB colors of the welding point image reflected by the three-color light into the color digital camera is analyzed, and the sunken position of the red glue is detected through the true color imaging effect to fully judge the quality of the welding point, thereby more comprehensively detecting the component dispensing position, improving the detection and interception effect of AOI imaging equipment on components, improving product quality, and solving the problem of high labor costs for staff during visual inspection.

[0024] An interception unit, after the imaging unit works, detects and intercepts multiple types of defects such as glue leakage, insufficient glue, excessive glue, dents, residual glue, and missing pins of the red glue; The working process of the interception unit is as follows: when the AOI imaging device reports, it waits for personnel to process, and the PCB stops waiting at the exit, accompanied by a buzzer prompt, until the personnel completes the processing and the PCB flows into the next workstation. Unlike conventional inspected products, since the AOI imaging device detects the appearance of the surface of the component fixed object, the red glue material is a paste-like object. With the weight of the glue dispensed, the shape changes, and the size and height will be affected at the same time. Therefore, the detection of non-fixed objects will cause the equipment to misjudge. Therefore, the interception unit 1 can also alarm the passed products, increasing the processing time of personnel.

[0025] The algorithm unit is centered around the location of the pin dispensing, and the editing detection box establishes a "tree" hierarchical relationship, and the sub-group box is linked with the parent attribute box.

[0026] Among them, the algorithm unit applies a variety of algorithms, with the three key points of "accurate positioning", "stable algorithm" and "good light source" coordinating with each other, to intercept as many diverse defects as possible and improve the accuracy of component interception detection.

[0027] Among them, the algorithm unit covers enough defect types to optimize the detection conditions, such as lack of glue, insufficient glue, exposed pins, excessive glue, depressions, residual glue, and body glue. Different lighting compensations are set to amplify the differences between defective products and qualified products, which can more comprehensively intercept defective workpieces and improve the comprehensiveness of detection.

[0028] The algorithm unit includes a positioning algorithm for position compensation during high-speed movement of a color digital camera, an auxiliary algorithm for arranging detection frames at multiple positions, and a detection algorithm capable of detecting the position of the detection frame.

[0029] It is worth mentioning that during the high-speed movement of the color digital camera, there will be deviations between the detection frame and the detected object in the picture. The position is compensated through the positioning algorithm. The conventional detection algorithm first locates the hierarchical relationship, and then a new detection frame is created and the algorithm is called for layout. Mark positioning and slice positioning are used as positioning algorithms to ensure that the position of the detection frame is fixed at the detected position. Window generation is an auxiliary algorithm to facilitate the layout of detection frames in multiple positions. The detection algorithms include brightness extraction, color extraction, brightness gradient bilateral distance, etc., to detect the position of the detection frame.

[0030] Among them, color extraction is specifically as follows: in the area specified by the detection window, obtain the percentage data of the number of pixels in the qualified range that conforms to the HSV color (white is the valid color) in the window pixels. After the desired color is extracted and filtered out, set the qualified range to distinguish between good and bad. The percentage (in the red box) directly calculates the percentage ratio of the extracted part to the entire detection window. If the red proportion is higher when testing for lack of glue, it proves that there is red glue at the position. Set the qualified range to the left, and the smaller the proportion, the better. If the red percentage is 0% when testing for lack of glue, that is, red is 0 detection window, that is, the alarm is for lack of glue, and so on.

[0031] Specifically, the light sources used for MARK positioning are top*2 and top.r. The center of the Mark point can be automatically searched and located within the area specified in the detection window through brightness filtering and the shape and size setting of the Mark point, so as to achieve a more accurate position matching between the component window position and the actual image of each PCB to be tested. The parameter setting range of the top*2 light source and the top.r light source are both 75%-100%.

[0032] Specifically, the light source used for chip positioning is 1-top.lum, which can automatically track and locate the position of the component in the area specified in the detection window according to the specified component length and width dimensions and electrode dimensions, and can output exact offset data. The 1-top.lum light source parameter setting range is 0.3-1.

[0033] Specifically, the light source used for window generation is top, which can quickly generate a series of detection windows in a matrix / array form in a real-time response manner within the area specified by the detection window.

[0034] Specifically, the light source used for color extraction (HSV) is side.(rg)*2. The percentage data of the number of pixels that meet the HSV color qualified range to the total number of window pixels can be obtained in the area specified by the detection window. The setting range of the side.(rg)*2 light source parameter is 75%-100%.

[0035] Specifically, the light source used for brightness extraction is top.(rg)*8. In the area specified by the detection window, the percentage data of the number of pixels that meet the specified brightness qualified range to the total number of pixels in the window can be obtained. The value range of the top.(rg)*8 light source parameter is 14000-23200.

[0036] Specifically, the light source used for the brightness gradient bilateral distance is side.(rg)*5. Based on the two boundaries obtained by "brightness gradient bilateral positioning", the distance between the two boundaries can be calculated. The value range of the side.(rg)*5 light source parameter is 1.2-2.0.

[0037] Specifically, the light source used for color extraction (RGB quantity) is top.(rg)*5. In the area specified by the detection window, data on the number of pixels that meet the HSV color qualified range and the total number of pixels in the window can be obtained. The value range of the top.(rg)*5 light source parameter is 800-2700.

[0038] The parameter setting in the algorithm is the standard for the detected position. It is judged OK within the range, otherwise it is NG. For example, if the mark positioning is 75%-100%, the similarity of the Mark point in the detection area exceeds 75%, it will pass. After the detection is qualified, the coordinates of the detection frame are compensated to the correct detection position through the fixed position of the Mark point. If it is lower than 75%, the device will judge NG and alarm and wait for personnel to handle it. The engineering personnel will debug the lighting or position and then test normally.

[0039] If the parameter is within the setting range, the device will judge it as PASS and flow into the back station. If the detected product is not within the setting range, the device will judge it as NG and output the judgment result, report the measured position and the picture of the defect, and wait for personnel to repair.

[0040] Reference Figure 2 , the qualified standards for UV curing red glue are: 1. The surface of the red glue must be filled to completely cover the pins, with a glossy surface, no bubbles, depressions or cracks, etc., and the upper and lower ends must be filled to the positioning pins; 2. The glue amount cannot be offset to the right side of the gold line, and must be in a straight line within the range; 3. The surface of the connector body on the left side of the red glue should not be stained with glue, the area around area ① should be kept clean, and the thickness of the red glue should not exceed 2mm of the connector height.

[0041] The present invention also discloses a method for using the AOI imaging system, which specifically includes the following steps: S1: The dispensing unit uses a dispensing machine to precisely coat the UV red glue on the pin position. The dispensing machine can dispense about 350pcs of red glue per tube, and the amount of glue at a single point is about 130mg. S2: The imaging unit (i.e., AOI imaging equipment) illuminates the dispensing position with top and side lights. The color digital camera collects image data through a telecentric lens. The top light is coaxially and vertically projected onto the component and the soldering point. The area of ​​the red glue is tested through the black and white imaging effect. The side light can detect the sunken position of the red glue through the true color imaging effect, fully judging the quality of the soldering point. S3: The interception unit can detect and intercept components with dispensing defects according to the algorithm unit. The algorithm unit includes: MARK positioning, slice positioning, window generation, color extraction (HSV), brightness extraction, brightness gradient bilateral distance, color extraction (RGB quantity); S4: The parameter setting in the algorithm unit is the standard for the detected position. It is judged OK within the range, otherwise it is NG. For example, if the mark positioning is 75%-100%, and the similarity of the Mark point in the detection area exceeds 75%, it will pass. After the detection is qualified, the detection frame coordinates are compensated to the correct detection position through the fixed position of the Mark point. If it is lower than 75%, the device will determine NG and alarm and wait for personnel to handle it. The engineering personnel will debug the light or position and then test normally.

[0042] The detection and interception of the dispensing position by the above-mentioned AOI imaging system can improve the pass rate and yield of red glue, as shown in the following table: 1. The pass rate of red glue AOI inspection equipment is greater than 98%, and the yield rate is greater than 99.9%, eliminating the visual inspection of personnel (the weekly yield rate is as follows).

[0043] 2. When a defect occurs that requires replacement of glue, the equipment will stop and alarm and wait for personnel to make a decision and handle it.

[0044] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An AOI imaging system, characterized in that: include: A glue dispensing unit, the glue dispensing unit includes a glue dispensing machine for reinforcing the Tapy-C terminals of key components, and the glue dispensing machine uses UV red glue to accurately apply to the pin position; An imaging unit, the imaging unit comprising a top light for testing the red glue area and a side light for detecting the red glue depression, a color digital camera is provided directly above the top light and the side light, the color digital camera is used to collect images of the glue dispensing position under the top light and the side light, and a telecentric lens is provided between the color digital camera and the top light and the side light to ensure imaging consistency and accuracy; An interception unit, which detects and intercepts multiple types of defects such as glue leakage, insufficient glue, excessive glue, depression, residual glue, and missing pins of the red glue after the imaging unit works; The algorithm unit is centered around the location of the pin dispensing, and the editing detection box establishes a "tree" hierarchical relationship, and the sub-group box is linked with the parent attribute box.

2. The AOI imaging system according to claim 1, characterized in that: The dispensing machine can dispense 250-400 pcs of red glue per tube, and the amount of glue at a single point is 100-150 mg.

3. The AOI imaging system according to claim 2, characterized in that: The workflow of the interception unit is as follows: when the AOI imaging device reports an error, it waits for personnel to process it. The PCB stops and waits at the exit, accompanied by a buzzer prompt, until the personnel completes the processing and the PCB flows into the next workstation. At the same time, the products that pass will also alarm, increasing the personnel processing time.

4. The AOI imaging system according to claim 3, characterized in that: The imaging effect of the top light is black and white, and the imaging effect of the side light is true color.

5. The AOI imaging system according to claim 4, characterized in that: The algorithm unit includes multiple defect types to optimize the detection conditions. The defect types may be lack of glue, insufficient glue, exposed pins, excessive glue, depression, residual glue, and glue sticking to the body. Different lighting compensations are set to amplify the difference between defective products and qualified products.

6. The AOI imaging system according to claim 5, characterized in that: The algorithm unit includes a positioning algorithm for position compensation during high-speed movement of a color digital camera, an auxiliary algorithm for arranging detection frames at multiple positions, and a detection algorithm capable of detecting the position of the detection frame.

7. The AOI imaging system according to claim 6, characterized in that: During the high-speed movement of a color digital camera, there will be a deviation between the detection frame and the detected object in the image. The position is compensated through the positioning algorithm. The conventional detection algorithm first locates the hierarchical relationship, and then a new detection frame is called for layout. Mark positioning and slice positioning are used as positioning algorithms to ensure that the position of the detection frame is fixed at the detected position. Window generation is an auxiliary algorithm to facilitate the layout of detection frames in multiple positions. The detection algorithms include brightness extraction, color extraction, brightness gradient bilateral distance, etc., to detect the position of the detection frame.

8. The AOI imaging system according to claim 7, characterized in that: Color extraction is as follows: within the area specified by the detection window, obtain the percentage data of the number of pixels that meet the qualified range of HSV colors to the window pixels. After the desired color is extracted and filtered out, set the qualified range to distinguish between good and bad, and the percentage directly calculates the percentage ratio of the extracted part to the entire detection window.

9. The AOI imaging system according to claim 8, characterized in that: If the parameter in the algorithm unit is within the range of the setting value, the device determines that the product is PASS and flows into the back station. If the detected product is not within the range of the setting value, the device determines that it is NG and outputs the judgment result, and reports the measured position and the picture of the defect.

10. A method for using the AOI imaging system according to claim 9, characterized in that: The following steps are involved: S1: The dispensing unit uses a dispensing machine to precisely coat the UV red glue on the pin position. The dispensing machine can dispense 350pcs of red glue per tube, and the glue volume per point is 130mg. S2: The imaging unit illuminates the dispensing position with top and side lights, and the color digital camera collects image data through a telecentric lens. The top light is coaxially and vertically projected onto the component and the soldering point, and the area of ​​the red glue is tested through the black and white imaging effect. The side light can detect the sunken position of the red glue through the true color imaging effect, and fully judge the quality of the soldering point. S3: The interception unit can detect and intercept components with dispensing defects according to the algorithm unit. The algorithm unit includes: MARK positioning, slice positioning, window generation, color extraction, brightness extraction, brightness gradient bilateral distance, and color extraction; S4: The setting value of the parameter in the algorithm unit is the standard for the detected position. If it is within the range, it is judged as OK, otherwise it is NG.

Images