Collection electron visual inspection system
By combining a single camera with a three-mirror design and image processing algorithms, the integrated electronic vision inspection system solves the problem of separate inspection of multiple devices in the electroplating target production line, achieving efficient, stable, and accurate dual-sided synchronous inspection, adapting to various production line layouts and providing full-chain product traceability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SYNPOWER
- Filing Date
- 2026-01-05
- Publication Date
- 2026-05-12
AI Technical Summary
Existing integrated circuit testing technologies suffer from low efficiency, poor accuracy, cumbersome maintenance, and difficulty in adaptation. In particular, they cannot achieve simultaneous and high-precision testing of multiple separate devices in electroplating target production lines. Furthermore, traditional equipment occupies a large space and requires production line modifications.
It adopts a single camera combined with a three-mirror design, and achieves dual-sided synchronous image acquisition through optical path design. Combined with image processing algorithms, it judges the surface defects of the object to be inspected, and is equipped with sensors and barcode readers for automatic identification and recording.
It achieves efficient, stable, and accurate dual-sided synchronous inspection, improving inspection efficiency, reducing maintenance costs, adapting to various production line layouts, requiring no modification to existing equipment, and providing end-to-end product traceability capabilities.
Smart Images

Figure CN122016799A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of visual inspection technology for electronic components, specifically referring to a static double-sided synchronous inspection system for "pure copper body collectors". It is specifically used to detect two types of surface defects, namely "side wear" and "bottom electrolytic corrosion", of collectors with a size of 75×30×8mm (length×width×height). It is especially suitable for online quality inspection scenarios in electroplating target production lines, which need to be carried out in an environment with a brushed stainless steel background (about 100mm away from the product), and solves the problems of low efficiency and insufficient accuracy of traditional multi-device separate inspection. Background Technology
[0002] In the production of current collectors for electroplated targets, side wear and bottom electrolytic corrosion are key defects that directly affect product lifespan. The former requires determining whether the wear reaches "half a circle" (i.e., wear exceeding the radius is considered NG), while the latter requires detecting the presence of electrolytic corrosion pits on the bottom. However, existing detection technologies have the following core shortcomings, failing to meet actual production needs.
[0003] Multi-device split inspection: Because the inspection angles of the side and bottom are different, the traditional solution requires two independent vision cameras to capture images separately. This not only occupies factory space (requiring separate planning of side / bottom inspection stations), but also the data synchronization delay between the two devices leads to low inspection efficiency, with a single piece inspection time exceeding 300ms, which cannot match the production line cycle of "more than 8 products per second".
[0004] Poor stability of the mechanical mechanism: Some improvement solutions attempt to use an electric turntable to drive the collector electrons to switch angles in order to achieve single-camera detection. However, wear and tear on the electric mechanism will lead to positioning errors (exceeding 0.5mm after long-term use), and the motor / belt needs to be replaced regularly, increasing maintenance costs. At the same time, the turntable is prone to causing secondary scratches on the surface of the pure copper body during operation (brushed lines appear on the bottom of the collector electrons), affecting product quality.
[0005] Reflective interference and insufficient accuracy: The collector is made of pure copper, which has strong surface reflection. Traditional light source designs (such as direct top light) are prone to forming light spots at the brushed texture, resulting in the failure to detect micro-defects such as "bottom electrical erosion < 0.3 × 0.3 mm". In addition, side wear detection mostly relies on visual comparison, and the repeatability accuracy can only reach 1 mm, which cannot meet the production requirement of "repeatability within 0.2 mm".
[0006] Poor on-site adaptability: In actual production line scenarios, collectors are mounted on fixtures in groups of two with a spacing of about 500mm, with a total of 4 slots. The materials to be tested are randomly supplied, and the spacing between the front and rear slots is 60mm. Traditional testing equipment is large in size and has a rigid fixing method, which cannot be compatible with this slot layout. Large-scale modifications to the production line are required, which is costly.
[0007] The aforementioned problems result in four major pain points in the existing technology for electronic component inspection: low efficiency, poor accuracy, complicated maintenance, and difficult adaptation. There is an urgent need for a static vision system that can match production line scenarios, eliminate the need for mechanical movement, and perform simultaneous double-sided inspection. Summary of the Invention
[0008] The main objective of this invention is to provide an electronic vision inspection system for visually inspecting whether the surface defects of at least one object to be inspected meet a predetermined standard in an environment with a light source.
[0009] To achieve the above objectives, the present invention provides an electronic vision inspection system, including a machine body, a controller, a first imaging device, a first reflector, a second reflector, and a third reflector. The first imaging device is disposed on the machine body and connected to the controller, and performs imaging on the object to be inspected according to instructions from the controller. The first reflector is disposed on the machine body and located in front of the first imaging device, with a first distance between them. The second reflector is disposed on the machine body and located in front of the first imaging device, with a second distance between them, wherein the second distance is greater than the first distance, and the object to be inspected is located to the right of the second reflector. The third reflector is disposed on the machine body and located to the right of the first reflector, with a third distance between them. The first photographic imaging device captures an image of the bottom surface of the object to be tested via a first optical path and through the first and third reflectors, and captures an image of the side surface of the object to be tested via a second optical path and through the second reflector.
[0010] In one embodiment of the present invention, the integrated electronic vision inspection system further includes a sensor. The sensor is disposed on the main body of the machine and connected to the controller, and the sensor is used to sense whether the object to be inspected is present in a detection area.
[0011] In one embodiment of the present invention, the integrated electronic vision inspection system further includes a barcode reader. The barcode reader is disposed on the main body of the machine and connected to the controller. If the sensor detects the presence of the object to be inspected in the detection area, the controller instructs the barcode reader to read the barcode of the object to obtain its identification number or unique information.
[0012] In one embodiment of the present invention, the first photographic imaging device transmits a bottom image of the bottom surface of the object to be inspected and a side image of the side surface of the object to be inspected to the controller, and the controller determines whether the bottom image and the side image meet the predetermined standard according to an image algorithm.
[0013] In one embodiment of the present invention, if the controller determines that both the bottom image and the side image meet the predetermined standard, the object to be tested is determined to be a qualified product.
[0014] In one embodiment of the present invention, if the controller determines that at least one of the bottom image and the side image does not meet the predetermined standard, the object to be inspected is determined to be a defective product and is subsequently automatically recorded or notified to relevant personnel.
[0015] In one embodiment of the present invention, the first optical path is incident from the first photographic imaging device to the first reflector, and then the first optical path is reflected by the first reflector to the third reflector, and then the first optical path is reflected by the third reflector to a bottom surface of the object to be detected.
[0016] In one embodiment of the present invention, the second optical path is incident from the first photographic imaging device to the second reflector, and then reflected by the second reflector to one side of the object to be detected.
[0017] In summary, the electronic vision inspection system provided by this invention can bring the following benefits:
[0018] 1. Detection efficiency has increased dramatically;
[0019] 2. Stability and protection optimizations;
[0020] 3. Improved detection accuracy and reliability;
[0021] 4. Strong adaptability to on-site conditions; and
[0022] 5. End-to-end data traceability.
[0023] The following detailed description through specific embodiments will make it easier to understand the purpose, technical content, features and effects achieved by the present invention. Attached Figure Description
[0024] Figure 1 This is a schematic diagram of the overall architecture of the integrated electronic vision inspection system of the present invention.
[0025] Figure 2 This is a side view of the integrated electronic vision inspection system of the present invention.
[0026] Figure 3 This is a schematic diagram of the optical path reflection of the integrated electronic vision inspection system of the present invention.
[0027] Explanation of reference numerals in the attached drawings: 100 - Integrated electronic vision inspection system; 110 - Main body of the machine; 120 - Controller; 130 - First photographic imaging device; 140 - First reflecting mirror; 150 - Second reflecting mirror; 160 - Third reflecting mirror; 170 - Sensor; 180 - Code reader; TA - Object to be inspected; TA10 - Bottom surface; TA20 - Side surface; L1 - First optical path; L2 - Second optical path; d1 - First distance; d2 - Second distance; d3 - Third distance. Detailed Implementation
[0028] This invention aims to address four major pain points of existing integrated circuit (IC) inspection technologies: 1. Eliminating the inefficiency of multi-device split inspection and achieving simultaneous dual-sided imaging with a single camera; 2. Eliminating the electric mechanism, improving the long-term stability of the system and product protection; 3. Optimizing the optical path and algorithm to solve reflection interference, achieving a repeatability accuracy of 0.2mm and a detection limit of 0.3×0.3mm for electrical erosion; 4. Compatible with a random material arrival layout of 4 slots on the production line (60mm spacing), requiring no on-site modifications.
[0029] Please see Figure 1 , Figure 1 This is a schematic diagram of the overall architecture of the integrated electronic vision inspection system of the present invention. Figure 2 This is a side view of the integrated electronic vision inspection system of the present invention. Figure 3This is a schematic diagram of the optical path reflection of the collector-emitter vision inspection system of the present invention. As shown, the collector-emitter vision inspection system is used to visually inspect whether the surface defects of at least one object TA meets a predetermined standard in an environment with a light source. The collector-emitter vision inspection system includes a machine body 110, a controller 120, a first photographic imaging device 130, a first reflector 140, a second reflector 150, and a third reflector 160. The first photographic imaging device 130 is disposed on the machine body 110 and connected to the controller 120. The first photographic imaging device 130 takes photographs of the object TA according to the instructions of the controller 120. The first reflector 140 is disposed on the machine body 110 and located in front of the first photographic imaging device 130, and the straight-line distance between the two is a first distance d1. The second reflector 150 is disposed on the main body 110 of the machine and located in front of the first photographic imaging device 130. The straight-line distance between the two is a second distance d2, wherein the second distance d2 is greater than the first distance d1, and the object to be tested TA is located to the right of the second reflector 150. The third reflector 160 is disposed on the main body 110 of the machine and located to the right of the first reflector 140, and the straight-line distance between the two is a third distance d3. The first photographic imaging device 130 takes a photograph of the bottom surface TA10 of the object to be tested TA through a first optical path L1 and through the first reflector 140 and the third reflector 160, and takes a photograph of the side surface TA20 of the object to be tested TA through a second optical path L2 and through the second reflector 160. Furthermore, the first optical path L1 originates from the first imaging device 130, enters the first reflecting mirror 140, is reflected by the first reflecting mirror 140 to the third reflecting mirror 160, and is then reflected by the third reflecting mirror 160 to a bottom surface TA10 of the object to be tested TA. The second optical path L2 originates from the first imaging device 130, enters the second reflecting mirror 150, and is reflected by the second reflecting mirror 150 to a side surface TA20 of the object to be tested TA.
[0030] The integrated electronic vision inspection system further includes a sensor 170 and a barcode reader 180. The sensor 170 is mounted on the main body 110 of the machine and connected to the controller 120. The sensor 170 is used to sense whether the object to be inspected (TA) is present in a detection area. The barcode reader 180 is mounted on the main body 110 of the machine and connected to the controller 120. If the sensor 170 detects the presence of the object to be inspected (TA) in the detection area, the controller 120 instructs the barcode reader 180 to read the code of the object to be inspected (TA) to obtain its identification number or unique information.
[0031] The first imaging device 130 transmits a bottom image of the bottom surface TA10 and a side image of the side surface TA20 of the object to be inspected TA to the controller 120. The controller 120 then uses an image algorithm to determine whether the bottom image and the side image meet a predetermined standard. If the controller 120 determines that both the bottom image and the side image meet the predetermined standard, the object to be inspected TA is deemed a qualified product. If the controller 120 determines that at least one of the bottom image or the side image does not meet the predetermined standard, the object to be inspected TA is deemed a defective product, and this is automatically recorded or reported to relevant personnel.
[0032] Furthermore, the following description will be based on a specific embodiment, but this embodiment is not intended to be limiting.
[0033] Regarding the installation and parameter settings of the main body 110: Place the main body 110 of the machine at the production line inspection station, and adjust the bottom feet to make the equipment level (error < 0.1mm / m); install the first imaging device 130 on the top crossbeam of the machine, ensuring it is tilted at 15° to the horizontal direction with the lens facing the inspection area. Calibrate the distances using a laser rangefinder: First distance d1 = 100mm (between the first imaging device 130 and the first reflector 140); second distance d2 = 600mm (between the first imaging device 130 and the second reflector 150); third distance d3 = 50mm (between the first reflector 140 and the third reflector 160).
[0034] Regarding the installation of auxiliary components for equipment installation and parameter settings: Sensor 170 is installed on the left side of the entrance to the detection area, with the detection distance adjusted to 80mm to ensure stable triggering when the object to be detected (TA) enters the detection area. The barcode reader 180 is installed 50mm behind sensor 170, with the scanning angle adjusted to 45° to ensure clear reading of the QR code on the fixture. Regarding the light source configuration, a "bottom light source" (ring-shaped diffused light source, wavelength 650nm) is installed to the side of the first reflector 140, and a "side light source" (strip light source, adjustable brightness) is installed below the second reflector 150 to prevent direct light from hitting the surface of the object to be detected (TA).
[0035] Regarding the parameter settings for controller 120: Image parameters can be set to a resolution of 2448×2048 for the first photographic recording device 130, an exposure time of 8ms, and a gain of 1.0. The detection standards are set within controller 120 to "acceptable side distance range 70-80mm" and "acceptable bottom pit pixel range <10 pixels," with the "adjustable range" function enabled. The trigger logic can be configured as follows: "sensor 170 senses material → delay 0.5s to trigger barcode reader 180 → successful barcode reading 0.3s to trigger first photographic imaging device 130 to capture image," avoiding signal interference.
[0036] The complete detection process of the integrated electronic vision inspection system of the present invention is as follows.
[0037] Step 1: The object to be tested is in place.
[0038] The objects to be tested, TA (two in a set, 500mm apart), are transported to the testing area of the main body 110 of the machine with the fixture. After the fixture stops, it remains still for 5 minutes (each pause is 5 minutes to take a still photo for detection) to ensure that there is no vibration interference.
[0039] Step 2: Material Sensing and Identification. Sensor 170 detects the presence of the item to be detected, TA, in the detection area and sends a "material arrival signal" to controller 120. After receiving the signal, controller 120 drives barcode reader 180 to scan the QR code on the fixture, read the identification number of the item to be detected, TA (e.g., "Batch-20250827-Slot3", representing the batch dated August 27, 2025, slot number 3), and uploads the identification information to the data storage unit.
[0040] Step 3: Synchronous Image Acquisition. After the controller 120 confirms the identity number upload, it sends an "image acquisition command" to the first imaging device 130 and the light source. The bottom light source and the side light source are simultaneously turned on, and the first imaging device 130 emits the first light path L1 and the second light path L2. The first light path L1 illuminates the bottom surface TA10 of the object to be detected TA through the first reflector 140 (45° reflection) and the third reflector 160 (45° reflection), forming a "bottom image". The second light path L2 illuminates the side surface TA20 of the object to be detected TA through the second reflector 150 (30° reflection), forming a "side image". The first imaging device 130 transmits the bottom image and the side image back to the image processing module of the controller 120 in real time.
[0041] Step 4: Image analysis and result determination. The image processing module of the controller 120 executes the following algorithm process. Image preprocessing: Perform "Gaussian blur (kernel size 3×3) → grayscale conversion → threshold segmentation (dynamic threshold of 20 - 50 grayscale values)" on the bottom image and the side image to remove reflective noise and background interference (such as brushed stainless steel background). Side TA20 analysis: Use the Hough transform algorithm to extract the circular arc features of the side contour and calculate the "distance from the center of the circle to the contour line". If the distance is within the range of 70 - 80 mm, it is determined that the side wear is qualified; if the distance < 70 mm or > 80 mm (such as worn to half of the circle), it is determined that the side wear is NG.
[0042] Bottom TA10 analysis: Use the "background difference method" (with the image of the qualified sample as the background template) to extract the bottom pit area and calculate the pixel area of the pit. If the pixel area < 10 Pixel, it is determined that the bottom electrolytic corrosion is qualified. If the pixel area ≥ 10 Pixel (i.e., within the range of 10 - 240 Pixel), it is determined that the bottom electrolytic corrosion is NG.
[0043] Comprehensive determination: If both the side and the bottom are qualified, the object under test TA is determined to be a qualified product. The controller 120 records the "qualified result + identification number + image" and sends a "release signal" to the production line, and the fixture enters the next process. If at least one of the side and the bottom is unqualified NG, the object under test TA is determined to be a defective product, and the controller 120 performs the following actions: 1. Store the defective image (mark the NG area); 2. Trigger the red warning light on the machine body 110 + NG condition; 3. Trigger the red warning light on the machine body 110; 4. Push the NG information to the on-site management terminal through the RS485 communication interface to notify the inspection personnel for review.
[0044] Step 5: Loop detection. Regardless of whether the determination result is qualified or unqualified, the controller 120 clears the current detection data cache, waits for the next trigger signal from the sensor 170, and enters the detection loop of the next object under test TA to achieve continuous automatic detection.
[0045] Regarding on-site adaptation and maintenance. Slot compatibility: There are 4 windows corresponding to the slots in the detection area of the machine body 110. When the object under test TA comes in randomly, the sensor 170 and the barcode reader 180 can automatically identify the slot number. The controller 120 calls the barcode reader 180 of the corresponding slot to automatically identify the slot number, and the controller 120 calls the detection value parameters of the corresponding slot (such as the distance compensation slot) without switching. Daily maintenance: Clean the mirrors of the first mirrors 140 / 150 / 160 every week (wipe with a dust-free cloth dipped in alcohol), and calibrate the focal length and mirror angle of the first imaging device 130 every month (test the accuracy with a standard sample and adjust when the deviation exceeds 0.05 mm). The maintenance cycle and cost are both lower than those of traditional equipment.
[0046] In summary, the electronic vision inspection system provided by this invention can bring the following benefits:
[0047] 1. Significantly improved detection efficiency: Through the “single CCD + three reflection mirrors” design, images of the bottom surface TA10 and the side surface TA20 can be acquired simultaneously in one photo. The algorithm detection time is only about 110ms, and more than 8 products can be detected per second. This is 3 times more efficient than the traditional two-camera solution, which is perfectly matched to the production line cycle time.
[0048] 2. Stability and protection optimization: The solution does not require an electric mechanism or autofocus lens, avoiding the decrease in accuracy caused by mechanical wear; at the same time, the object under test (TA) remains stationary throughout the process, without rotation / movement, effectively protecting the brushed surface of the pure copper body, reducing the secondary damage rate to 0%.
[0049] 3. Improved Detection Accuracy and Reliability: Side TA20: After algorithm simulation, the repeatability detection accuracy reaches 4 pixels (corresponding to 0.058mm / pixel×4=0.232mm), with a detection accuracy of over 99.5%; no missed detections within the 10-240 pixel range, with a positioning accuracy of over 99.5%. Reflection Suppression: The polarization design of the 160° third reflector and the specific angle optical path reduce reflective noise in the bottom brushed texture area by 80%, improving the image signal-to-noise ratio to over 30dB.
[0050] 4. Strong on-site adaptability: The main body of the machine (110) can directly connect to four slots on the production line (four slots in total, with materials arriving randomly each time), without requiring modification to existing fixtures or production line layout. Equipment installation only requires fixed foot cups, with installation time less than 2 hours, making it suitable for production line setup.
[0051] 5. End-to-end data traceability: Through the process of sensor 170 triggering - barcode reader 180 scanning - controller 120 storing, a complete data chain is established from "TA identification number of the object to be inspected → bottom / side image → judgment result → inspection time", which can trace the inspection process of any product and meet the requirements of the quality management system.
[0052] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of the invention. Therefore, all equivalent variations or modifications made in accordance with the features and spirit described in the claims of the present invention should be included within the protection scope of the present invention.
Claims
1. An electronic vision inspection system for visually inspecting whether surface defects of at least one object conform to a predetermined standard in an environment with a light source, characterized in that, This electronic vision inspection system includes: Main body of the machine; One controller; A first photographic imaging device is mounted on the main body of the machine and connected to the controller. The first photographic imaging device takes photographic images of the object to be inspected according to the instructions of the controller. A first reflector is disposed on the main body of the machine and located in front of the first photographic imaging device, and the straight-line distance between the two is a first distance; A second reflector is disposed on the main body of the machine and located in front of the first photographic imaging device, and the straight-line distance between the two is a second distance, wherein the second distance is greater than the first distance, and the object to be detected is located to the right of the second reflector; and A third reflector is mounted on the main body of the machine and located to the right of the first reflector, with a third distance between them. The first photographic imaging device captures an image of the bottom surface of the object to be tested via a first optical path and through the first and third reflectors, and the first photographic imaging device captures an image of the side surface of the object to be tested via a second optical path and through the second reflector.
2. The electronic vision inspection system as described in claim 1, characterized in that, Also includes: A sensor is mounted on the main body of the machine and connected to the controller. The sensor is used to sense whether the object to be detected is in a detection area.
3. The electronic vision inspection system as described in claim 2, characterized in that, Also includes: A barcode reader is mounted on the main body of the machine and connected to the controller. If the sensor detects the presence of the object to be detected in the detection area, the controller instructs the barcode reader to read the barcode of the object to obtain its identification number or unique information.
4. The electronic vision inspection system as described in claim 1, characterized in that, The first imaging device transmits a bottom image of the bottom surface of the object to be inspected and a side image of the side surface of the object to be inspected to the controller, and the controller determines whether the bottom image and the side image meet the predetermined standard based on an image algorithm.
5. The electronic vision inspection system as described in claim 4, characterized in that, If the controller determines that both the bottom image and the side image meet the predetermined standard, the object to be inspected is judged to be a qualified product.
6. The electronic vision inspection system as described in claim 4, characterized in that, If the controller determines that at least one of the bottom image and the side image does not meet the predetermined standard, the object to be inspected is judged as a defective product and is subsequently automatically recorded or reported to relevant personnel.
7. The electronic vision inspection system as described in claim 1, characterized in that, The first light path enters from the first imaging device and enters the first reflector, then is reflected by the first reflector to the third reflector, and then is reflected by the third reflector to a bottom surface of the object to be tested.
8. The electronic vision inspection system as described in claim 1, characterized in that, The second optical path enters from the first photographic imaging device, enters the second reflector, and is then reflected by the second reflector to one side of the object to be detected.