OCA bubble photographing detection method and device

Abstract

The invention relates to the technical field of OCA bubble detection, in particular to an OCA bubble photographing detection method and device.The method comprises the steps that the feeding result of a feeding carrying table is obtained, when a to-be-detected product exists, a code scanning gun records the product number, and when the to-be-detected product does not exist, feeding reminding is sent out, and the feeding result is obtained again; after the product number is recorded, the product assembling face is photographed, recognized and detected at the first detection position, and the lower surface of the product is photographed, recognized and detected at the second detection position; determining a final detection result according to the detection results of the first detection position and the second detection position; and the to-be-detected products on the transfer carrying platform are conveyed to the normal product discharging side or the abnormal product discharging side of the first linear conveying mechanism through the material conveying mechanism according to the final detection result. The full-process automation from feeding, detection to classified discharging is achieved, and the production efficiency and the product quality control level are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of OCA bubble detection, in particular to an OCA bubble photographing detection method and device. BACKGROUND

[0002] CG surface is the outermost protective glass of the mobile phone screen, and the lower surface of the CG needs to be coated with OCA glue. The OCA surface is the optical adhesive surface, which is located between the CG layer and the touch layer.

[0003] Before the introduction of an automatic system, the photographing process of OCA bubbles relies on manual operation of CCD, and no records are saved, which leads to the inability to trace back.

[0004] This manual operation method has several problems: first, the personnel operation efficiency is low because the manual operation speed is slow and prone to errors; second, there is a risk of secondary CG pollution, which may affect the cleanliness and quality of the product; in addition, because the manual operation is not as accurate as the machine, there is also a risk of product damage; finally, because there is no record saved, information omission or loss may occur, further increasing the management difficulty. SUMMARY

[0005] In view of this, the present application provides an OCA bubble photographing detection method and device to solve the problems in the prior art.

[0006] In one aspect, the present application provides an OCA bubble photographing detection method, comprising the following steps: Acquiring the feeding result detected by the sensor arranged on the feeding table, when the sensor determines that there is a product to be detected on the feeding table, the code scanning gun scans and records the product number, and when the sensor determines that there is no product to be detected on the feeding table, an feeding reminder is sent and the feeding result is re-acquired; After the code scanning gun scans and records the product number, the first linear conveying mechanism is used to convey the product to be detected to the first detection position to photograph and identify the product assembly surface for detection, and the detection result of the first detection position is acquired; When the product assembly surface detection is completed, the transfer mechanism arranged on the second linear conveying mechanism is used to suspend the product to be detected to the second detection position to photograph and identify the product lower surface for detection, and the detection result of the second detection position is acquired; The final detection result is determined according to the detection results of the first detection position and the second detection position; The product to be detected is transferred to the transfer table by the transfer mechanism, and the product to be detected on the transfer table is conveyed to the normal product discharge side or the abnormal product discharge side of the first linear conveying mechanism according to the final detection result by the material conveying mechanism.

[0007] In some embodiments of the present application, when the lower surface of the product to be detected is detected by the transfer mechanism suspended by the second linear conveying mechanism, the method comprises: The real-time adsorption force of the transfer mechanism is obtained to make a first judgment on the adsorption effect, and the specific calculation formula of the real-time adsorption force is as follows: ; ; Wherein, F is the real-time adsorption force; is the difference between the internal and external air pressures; P2 is the real-time air pressure in the sealed cavity formed when the transfer mechanism adsorbs the product to be detected; P1 is the real-time atmospheric pressure outside the sealed cavity; A is the effective adsorption area of the suction cup of the transfer mechanism; The effective adsorption force threshold Fa is preset; When F is greater than or equal to Fa, it is determined that the product to be detected is effectively adsorbed, and the next operation can be performed; When F is less than Fa, a second judgment on the adsorption effect is made.

[0008] In some embodiments of the present application, when F is less than Fa, a second judgment on the adsorption effect is made, which comprises: The adsorption force deviation absolute value of the effective adsorption force threshold Fa and the real-time adsorption force F is calculated , and the real-time pressure data detected by the pressure sensor in the transfer platform is obtained ; The adsorption force deviation threshold Fx is preset; When Fx is less than or equal to and the real-time pressure data is equal to 0, it is determined that the product to be detected is effectively adsorbed, and the next operation can be performed; When Fx is greater than or the real-time pressure data is greater than 0, it is determined that the product to be detected is not effectively adsorbed, and the adsorption operation is re-performed.

[0009] In some embodiments of the present application, when the final detection result is determined according to the detection results of the first detection site and the second detection site, the method comprises: The product defects on the product assembly surface or the product lower surface are predicted by constructing and training a convolutional neural network model, when the product assembly surface or the product lower surface has defects, it is determined that the final detection result of the product to be detected is an abnormal product, when the product assembly surface and the product lower surface do not have defects, it is determined that the final detection result of the product to be detected is a normal product.

[0010] In some embodiments of the present application, after determining the final detection result, when the final detection result is a normal product, the material conveying mechanism is used to convey the product on the transfer platform to the normal product unloading side of the first linear conveying mechanism; when the final detection result is an abnormal product, the material conveying mechanism is used to convey the product to the abnormal product unloading side of the first linear conveying mechanism, and the final detection result is combined with the corresponding product code and stored.

[0011] Compared with the prior art, the present application has the beneficial effects that through the multi-process cooperation of sensors, code scanning guns, and photographing recognition technologies, the whole process automation from feeding, detection to classification and unloading is realized, and the production efficiency and product quality control level are improved. First, the sensor installed on the feeding platform is used to detect whether there is a product to be detected, which ensures the accuracy and timeliness of subsequent operations. Second, the code scanning gun is used to record the unique identification of the product, i.e., the product number, for subsequent detection, processing, and tracing. Then, through the first linear conveying mechanism, the product to be detected is conveyed to the first detection position, and the camera is used to detect the assembly surface of the product by photographing and recognizing, and the corresponding detection result is generated. The transfer mechanism is used to transfer the product to be detected from the first detection position to the second detection position, and the lower surface of the product is detected by photographing and recognizing, which ensures the comprehensive detection of the product and avoids missing any possible problems. Finally, according to the detection results of the first detection position and the second detection position, the final detection result is determined, and according to the final detection result, the transfer mechanism transfers the product to the transfer platform, and then the material conveying mechanism conveys it to the normal product unloading side or the abnormal product unloading side of the first linear conveying mechanism.

[0012] On the other hand, the present application also provides an OCA bubble photographing detection device, which applies the OCA bubble photographing detection method, including a machine table, a first linear conveying mechanism for feeding and unloading is arranged on the machine table, and a material conveying mechanism for transfer is arranged on the machine table. The feeding side of the first linear conveying mechanism is provided with a movable feeding platform, and the feeding platform on which the product to be detected is placed is driven by the first linear conveying mechanism to move from the feeding port to the first detection position, and the first detection component arranged on the machine table is used to detect the assembly surface of the product by photographing. The material conveying mechanism and the end of the feeding side away from the feeding port are provided with a transfer platform. The first detection assembly is provided with a second linear conveying mechanism, and the second linear conveying mechanism is movably provided with a transfer mechanism. The transfer mechanism is used to suspend the product to be detected by the feeding table and place the product to be detected on a second detection position. After the second detection assembly provided on the machine table detects the lower surface of the product by taking a photo, the second linear conveying mechanism drives the transfer mechanism to place the product to be detected on a transfer table. The material conveying mechanism is used to transfer the product to be detected from the transfer table to a feeding table on the normal product discharge side or the abnormal product discharge side of the first linear conveying mechanism. The first linear conveying mechanism drives the feeding table on which the product to be detected is placed to move to the corresponding discharge port.

[0013] In some embodiments of the present application, the first detection assembly comprises a first support and a second support arranged on both sides of the feeding side, and two first cameras arranged on the first support and the second support respectively. The second linear conveying mechanism is arranged on the first support or the second support. The second detection assembly comprises a light source and a third camera arranged between the transfer table and the end close to the feeding side.

[0014] In some embodiments of the present application, one side of the machine table near the feeding port is provided with a code scanning gun.

[0015] In some embodiments of the present application, the material conveying mechanism comprises a material conveying guide rail arranged on the machine table, a second cylinder slidably arranged on the material conveying guide rail, and a second transfer plate connected with the driving end of the second cylinder.

[0016] In some embodiments of the present application, the transfer mechanism comprises a first cylinder slidably arranged on the second linear conveying mechanism, a motor connected with the driving end of the first cylinder, and a first transfer plate connected with the driving end of the motor. The first transfer plate and the second transfer plate adsorb the product to be detected by the suction cup.

[0017] Compared with the prior art, the OCA bubble photographing detection device has the beneficial effects that, first, the machine table is the basic platform of the entire detection device, supporting and integrating all other components. Specifically, the first linear conveying mechanism is responsible for the movement and positioning of the product, ensuring that the product can accurately reach the designated position for detection. The material conveying mechanism is responsible for the transfer work of the product, and the transfer table is used for temporarily storing the product to be detected, so that the product can be smoothly transferred between different stages. The first detection component is arranged to detect the assembly surface of the product, and the second detection component is arranged on the second linear conveying mechanism and is specially arranged for detecting the lower surface of the product. The image of the product surface is captured through the photographing technology, so as to identify whether there is a bubble defect. Finally, the transfer mechanism can move and adsorb the product to be detected in suspension, transfer the product from the feeding table to the second detection position, and place the product on the transfer table after the detection is completed, so as to facilitate the subsequent classification and unloading process. The OCA bubble photographing detection device of the present application realizes efficient detection and classification and unloading of OCA product bubbles through the cooperative work of the machine table, the linear conveying mechanism, the material conveying mechanism, the detection component and the transfer mechanism.

[0018] Further, the OCA bubble photographing detection device provided by the present application applies an OCA bubble photographing detection method and has the same beneficial effects as the OCA bubble photographing detection method, which will not be repeated here. BRIEF DESCRIPTION OF DRAWINGS

[0019] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not meant to limit the present application. Moreover, the same reference numerals in the attached drawings refer to the same or similar components throughout the several drawings. In the drawings: Figure 1 A flowchart of an OCA bubble photographing detection method provided by an embodiment of the present application; Figure 2 A schematic diagram of the overall structure of an OCA bubble photographing detection device provided by an embodiment of the present application; Figure 3 A schematic diagram of the internal structure of an OCA bubble photographing detection device provided by an embodiment of the present application.

[0020] In the figure: 1, machine table; 11, code scanning gun; 12, feeding port; 13, normal product discharge port; 14, abnormal product discharge port; 2, first linear conveying mechanism; 21, feeding side; 211, feeding platform; 22, normal product discharge side; 23, abnormal product discharge side; 221, discharge platform; 31, first support; 32, second support; 33, first camera; 4, second linear conveying mechanism; 5, transfer mechanism; 51, first air cylinder; 52, motor; 53, first transfer plate; 61, light source; 62, third camera; 7, transfer platform; 81, second air cylinder; 82, material conveying guide rail; 83, second transfer plate; 84, suction cup. DETAILED DESCRIPTION

[0021] Exemplary embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be accurately conveyed to those skilled in the art. It should be noted that the embodiments in the present disclosure and the features in the embodiments can be combined with each other without conflict. The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0022] Embodiment 1, see Figure 1 The present embodiment provides an OCA bubble photographing detection method, which comprises obtaining feeding results detected by a sensor arranged on the feeding platform 211. When the sensor determines that the feeding platform 211 has a product to be detected, the code scanning gun 11 scans and records the product number of the product to be detected. When the sensor determines that the feeding platform 211 does not have a product to be detected, a feeding reminder is issued, and the feeding results are re-obtained.

[0023] Specifically, the sensor of the feeding platform 211 can be a pressure sensor, which detects the pressure of the feeding platform 211. When the pressure of the feeding platform 211 is equal to the weight of the product to be detected, it is determined that the feeding platform 211 has a product to be detected. When the pressure of the feeding platform 211 is not equal to the weight of the product to be detected, it is determined that the feeding platform 211 does not have a product to be detected, and a feeding reminder is issued. After re-feeding, the feeding results are obtained again. The sensor of the feeding platform 211 can also be a photoelectric sensor, a depth sensor or other sensors. Any component that can achieve the purpose of detecting whether a product to be detected exists can be selected, and the present embodiment does not make specific limitations.

[0024] After the code scanning gun 11 scans and records the product number, the product to be detected is conveyed to the first detection position by the first linear conveying mechanism 2 to photograph and identify the product assembly surface for detection, and the detection results of the first detection position are obtained.

[0025] When the product assembly surface detection ends, the product to be detected is suspended by the transfer mechanism 5 arranged on the second linear conveying mechanism 4 to the second detection position to take a photo of the lower surface of the product for recognition and detection, and the detection result of the second detection position is obtained.

[0026] The final detection result is determined according to the detection results of the first detection position and the second detection position.

[0027] Preferably, when the product to be detected is suspended by the transfer mechanism 5 arranged on the second linear conveying mechanism 4 to the second detection position to detect the lower surface of the product, the following steps are included: The real-time adsorption force of the transfer mechanism 5 is obtained to make a first judgment on the adsorption effect, and the specific calculation formula of the real-time adsorption force is as follows: ; ; Wherein, F is the real-time adsorption force; is the difference between the internal and external air pressures; P2 is the real-time air pressure in the sealed cavity formed when the transfer mechanism 5 adsorbs the product to be detected; P1 is the real-time atmospheric pressure outside the sealed cavity; A is the effective adsorption area of the suction cup 84 of the transfer mechanism 5; An effective adsorption force threshold Fa is preset; When F is greater than or equal to Fa, it is determined that the product to be detected is effectively adsorbed, and the next operation can be performed; When F is less than Fa, a second judgment on the adsorption effect is made.

[0028] Preferably, when F is less than Fa, a second judgment on the adsorption effect is made, including: The adsorption force deviation absolute value of the effective adsorption force threshold Fa and the real-time adsorption force F is calculated , and the real-time pressure data detected by the pressure sensor in the transfer platform 7 is obtained ; An adsorption force deviation threshold Fx is preset; When Fx is less than or equal to and the real-time pressure data is equal to 0, it is determined that the product to be detected is effectively adsorbed, and the next operation can be performed; When Fx is greater than or the real-time pressure data is greater than 0, it is determined that the product to be detected is not effectively adsorbed, and the adsorption operation is re-performed.

[0029] It can be understood that the adsorption effect is judged twice in this embodiment. First, only the effective adsorption force threshold Fa and the real-time adsorption force F are compared in the first judgment, and only when it is greater than or equal to the effective adsorption force threshold Fa can it be determined as effective adsorption, so as to avoid the situation that the to-be-detected product falls off during the adsorption transfer process and ensure the safety and stability of the adsorption transfer process. Secondly, the second judgment calculates the adsorption force deviation absolute value Within the adsorption force deviation threshold Fx allowable range, according to the real-time pressure data Whether greater than 0 determines whether the to-be-detected product has been adsorbed and suspended, when being adsorbed and suspended and the adsorption force deviation absolute value Within the adsorption force deviation threshold Fx allowable range, it is determined that the to-be-detected product is effectively adsorbed, otherwise it is not effectively adsorbed, which avoids misjudgment and improves production detection efficiency.

[0030] Preferably, when determining the final detection result according to the detection results of the first detection position and the second detection position, it includes: By constructing and training a convolutional neural network model, the product defects on the product assembly surface or the product lower surface are predicted. When the product assembly surface or the product lower surface has defects, it is determined that the final detection result of the to-be-detected product is an abnormal product. When the product assembly surface and the product lower surface do not have defects, it is determined that the final detection result of the to-be-detected product is a normal product.

[0031] Specifically, when the product assembly surface or the product lower surface is photographed and recognized for detection, a convolutional neural network model can be constructed and trained to predict product defects on the product assembly surface or the product lower surface.

[0032] First, through image processing technology, the photographed product assembly surface image or product lower surface image is preprocessed, including denoising, contrast enhancement and other operations, so as to improve the accuracy of subsequent analysis. For example, a Gaussian filter is used to remove noise in the image, or histogram equalization is applied to enhance the contrast of the image.

[0033] Secondly, the convolutional neural network (CNN) is a deep learning model specially used for processing image data. CNN can automatically extract useful features from input images through convolutional layers, pooling layers and fully connected layers. For example, the product assembly surface image or the product lower surface image is input into the convolutional neural network model (such as ResNet), and the CNN can learn the features of the bubbles such as shape, size and color, so as to realize accurate recognition of the bubbles.

[0034] Thirdly, through multi-layer convolution operation, CNN can extract different levels of features of the image layer by layer. Low-level convolutional layers usually extract simple edge and texture information, while high-level convolutional layers can capture more complex patterns, such as the overall shape of the bubbles.

[0035] Fourth, a large number of labeled sample data is used to train the CNN model to accurately identify bubbles. During the training process, the backpropagation algorithm is used to adjust the network parameters to minimize the error between the predicted results and the actual labels. For example, the cross-entropy loss function can be used to measure the accuracy of the model prediction, and the gradient descent method is used to optimize the model parameters.

[0036] Finally, the fully trained CNN model can be used for actual prediction tasks. When a new product assembly surface image or product lower surface image is input into the model, the CNN will classify it according to the learned features to determine whether the image contains bubbles. For example, if a product assembly surface image or product lower surface image containing bubbles is input into the CNN model, the model will output a higher probability value indicating the presence of bubbles. The bubble presence probability predicted by the CNN model is compared with the preset minimum probability threshold. When the bubble presence probability predicted by the CNN model is greater than the minimum probability threshold, it is determined that the first detection site or the second detection site has defects in the product assembly surface or the product lower surface. Otherwise, there are no defects.

[0037] The product to be detected is transferred to the transfer platform 7 by the transfer mechanism 5, and the product to be detected on the transfer platform 7 is transported to the normal product discharge side 22 or the abnormal product discharge side 23 of the first linear conveying mechanism 2 according to the final detection result by the material conveying mechanism.

[0038] Preferably, after determining the final detection result, when the final detection result is a normal product, the product to be detected on the transfer platform 7 is transported to the normal product discharge side 22 of the first linear conveying mechanism 2 by the material conveying mechanism, and when the final detection result is an abnormal product, the product to be detected is transported to the abnormal product discharge side 23 of the first linear conveying mechanism 2 by the material conveying mechanism, and the final detection result is combined with the corresponding product code for storage.

[0039] It can be understood that in this embodiment, after determining the final detection result, appropriate processing will be performed. When the final detection result is a normal product, the material conveying mechanism transports the product to be detected on the transfer platform 7 to the normal product discharge side 22 of the first linear conveying mechanism 2. Conversely, if the detection result is an abnormal product, the material conveying mechanism transports it to the abnormal product discharge side 23. This process not only improves production efficiency, but also ensures product quality control. In addition, in order to achieve comprehensive quality traceability, the final detection result is combined with the corresponding product code for storage. For example, if a batch of products has a problem, it can be quickly traced to the specific production link through the product code, so that corrective measures can be taken in a timely manner.

[0040] Embodiment 2, see Figure 2 and Figure 3The application also provides an OCA bubble photographing detection device and an OCA bubble photographing detection method.

[0041] The feeding side 21 of the first linear conveying mechanism 2 is provided with a movable feeding platform 211. The first linear conveying mechanism 2 drives the feeding platform 211 on which the product to be detected is placed to move from the feeding port 12 to the first detection position, and the product assembly surface is detected by the first detection assembly arranged on the machine table 1.

[0042] Specifically, the feeding platform 211 is first located at the feeding port 12. After the product to be detected is placed on the feeding platform 211 from the feeding port 12, the first linear conveying mechanism 2 drives the feeding platform 211 to move along the guide rail of the feeding side 21 to the first detection position, and the photographing detection is performed by the first detection assembly.

[0043] The material conveying mechanism and the end of the feeding side 21 away from the feeding port 12 are provided with a transfer platform 7.

[0044] Specifically, the feeding platform 211 and the transfer platform 7 are provided with sensors for detecting whether the product to be detected exists, such as pressure sensors, photoelectric sensors, depth sensors, etc.

[0045] The first detection assembly is provided with a second linear conveying mechanism 4, and the second linear conveying mechanism 4 is movably provided with a transfer mechanism 5. The transfer mechanism 5 is used to suspend the product to be detected from the feeding platform 211 to the second detection position. After the lower surface of the product is detected by the second detection assembly arranged on the machine table 1, the second linear conveying mechanism 4 drives the transfer mechanism 5 to place the product to be detected on the transfer platform 7. The material conveying mechanism is used to transfer the product to be detected from the transfer platform 7 to the normal product discharge side 22 or the abnormal product discharge side 23 of the discharge platform 221 of the first linear conveying mechanism 2. The first linear conveying mechanism 2 drives the discharge platform 221 on which the product to be detected is placed to move to the corresponding discharge port.

[0046] Specifically, after the photographing detection of the product assembly surface of the product to be detected at the first detection position is completed, the transfer mechanism 5 suspends the product to be detected of the feeding platform 211 to the second detection position. The second detection position is above the first detection position, so as to realize the photographing detection of the lower surface of the product by the second detection assembly arranged on the machine table 1. After the photographing detection of the lower surface of the product to be detected at the second detection position is completed, the second linear conveying mechanism 4 drives the transfer mechanism 5 to move the product to be detected close to the transfer platform 7. The transfer mechanism 5 breaks the vacuum to place the product to be detected on the transfer platform 7.

[0047] Specifically, the machine table 1 is also provided with a controller (not shown), which is wirelessly connected or in communication connection with the first linear conveying mechanism 2, the second linear conveying mechanism 4, the transfer mechanism 5, the material conveying mechanism, the first camera 33 and the second camera. The controller obtains the photographed images of the first detection position and the second detection position, predicts the product defects on the product assembly surface or the product lower surface through the convolutional neural network model pre-stored in the controller, and outputs the final detection result. Based on the final detection result, the controller controls the material conveying mechanism to transfer the product to be detected on the transfer platform 7 to the unloading platform 221 on the normal product unloading side 22 or the abnormal product unloading side 23 of the first linear conveying mechanism 2, and drives the unloading platform 221 to the corresponding normal product unloading port 13 or abnormal product unloading port 14 by the first linear conveying mechanism 2.

[0048] Preferably, the first detection assembly includes a first support 31 and a second support 32 arranged on both sides of the feeding side 21, and two first cameras 33 arranged on the first support 31 and the second support 32 respectively; the second linear conveying mechanism 4 is arranged on the first support 31 or the second support 32; the second detection assembly includes a light source 61 and a third camera 62 arranged between the transfer platform 7 and the end close to the feeding side 21.

[0049] Specifically, the feeding side 21, the normal product unloading side 22 and the abnormal product unloading side 23 of the first linear conveying mechanism 2 are parallel to each other. The second linear conveying mechanism 4 is located above the first linear conveying mechanism 2 and parallel to the first linear conveying mechanism 2. The material conveying mechanism is arranged perpendicular to the extension direction of the first linear conveying mechanism 2.

[0050] It can be understood that in the embodiment, the first detection assembly is composed of a first support 31 and a second support 32 arranged on both sides of the feeding side 21, and two first cameras 33 are respectively installed on the two supports for capturing image information of the material. The first support 31 and the second support 32 provide a stable support structure to ensure that the camera can accurately align the product to be detected. The two first cameras 33 can provide more comprehensive detection data by capturing images at different angles, which helps to improve the detection accuracy. In addition, the second linear conveying mechanism 4 is arranged on the first support 31 or the second support 32 for accurate control of the movement of the material. The second detection assembly includes a light source 61 and a third camera 62 arranged between the transfer platform 7 and the end close to the feeding side 21. The function of the light source 61 is to provide uniform and sufficient illumination to ensure that the third camera 62 can clearly capture the details of the product. The first detection assembly and the second detection assembly are reasonably arranged to realize efficient and accurate detection of the product, which not only improves the reliability and accuracy of the detection, but also provides strong support for the optimization of the automatic production line.

[0051] Preferably, a code scanning gun 11 is arranged on one side of the machine table 1 at the feeding port 12 for scanning and recording the product code.

[0052] Preferably, the material conveying mechanism comprises a material conveying rail arranged on the machine table 1, a second cylinder 81 slidably arranged on the material conveying rail, and a second transfer plate 83 connected to the driving end of the second cylinder 81.

[0053] Preferably, the transfer mechanism 5 comprises a first cylinder 51 slidably arranged on the second linear conveying mechanism 4, a motor 52 connected to the driving end of the first cylinder 51, and a first transfer plate 53 connected to the driving end of the motor 52. The first transfer plate 53 and the second transfer plate 83 adsorb the product to be detected by the suction cup 84, and detect the internal and external pressure between the product to be detected and the suction cup 84 by, for example, a pressure sensor, to realize real-time detection of adsorption force.

[0054] It can be understood that the material conveying rail is used to guide and support the movement of the material on a specific path, and the second cylinder 81 is arranged on the material conveying rail and can slide along the rail. Specifically, when the second cylinder 81 receives a control signal, it pushes or pulls the second transfer plate 83 connected to the driving end thereof, thereby realizing the handling of the product. The second transfer plate 83 is connected to the driving end of the second cylinder 81 and is responsible for transferring the product from one location to another. The first cylinder 51 of the transfer mechanism 5 is slidably arranged on the second linear conveying mechanism 4. The driving end of the first cylinder 51 is connected to the motor 52, and the driving end of the motor 52 is connected to the first transfer plate 53, so that the product can be accurately positioned and transferred through multiple steps. The first transfer plate 53 and the second transfer plate 83 both adsorb the product to be detected by the suction cup 84, ensuring that the product will not fall or shift during transportation.

[0055] Those skilled in the art will appreciate that embodiments of the application can be provided as methods, systems or computer program products. Accordingly, the application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the application can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) embodying computer readable program code thereon.

[0056] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart or flowsheet block or blocks. Figure 1 one or more flowcharts and / or blocks Figure 1 means for functionally implementing the steps listed in the flowchart

[0057] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart or flowsheet block or blocks. Figure 1 one or more flowcharts and / or blocks Figure 1 means for functionally implementing the steps listed in the flowchart

[0058] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart or flowsheet block or blocks. Figure 1 one or more flowcharts and / or blocks Figure 1 means for functionally implementing the steps listed in the flowchart

[0059] Finally, it should be noted that the above-mentioned embodiments are merely intended for describing the technical solutions of the present application, but not for limiting thereof. Although the present application is described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalent replaced without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered in the protection scope of the claims of the present application.

Claims

1. A method for detecting OCA bubbles by photographing, characterized in that, Includes the following steps: The loading result detected by the sensor set on the feeding platform (211) is obtained. When the sensor determines that there is a product to be detected on the feeding platform (211), the barcode scanner (11) scans and records the product number of the product to be detected. When the sensor determines that there is no product to be detected on the feeding platform (211), a loading reminder is issued and the loading result is obtained again. After the barcode scanner (11) scans and records the product number, the product to be tested is transported to the first detection position through the first linear conveyor mechanism (2) to take pictures and identify the product assembly surface, and obtain the detection result of the first detection position; After the product assembly surface inspection is completed, the product to be inspected is adsorbed and suspended to the second inspection position by the transfer mechanism (5) set on the second linear conveying mechanism (4) to take pictures and identify the lower surface of the product, and obtain the inspection result of the second inspection position; The final detection result is determined based on the detection results of the first detection bit and the second detection bit; The product to be tested is transferred to the transfer platform (7) by the transfer mechanism (5), and the product to be tested on the transfer platform (7) is transported to the normal product unloading side (22) or abnormal product unloading side (23) of the first linear conveyor (2) according to the final test result by the material conveying mechanism.

2. The OCA bubble imaging detection method according to claim 1, characterized in that, When the product to be tested is adsorbed and suspended at the second detection position by the transfer mechanism (5) set on the second linear conveying mechanism (4) to detect the lower surface of the product, the process includes: The adsorption effect is judged once by obtaining the real-time adsorption force of the transfer mechanism (5). The specific calculation formula of the real-time adsorption force is as follows: ； Where F is the real-time adsorption force; P1 is the pressure difference between the inside and outside; P2 is the real-time air pressure in the sealed cavity formed when the transfer mechanism (5) adsorbs the product to be tested; P1 is the real-time atmospheric pressure outside the sealed cavity; A is the effective adsorption area of ​​the suction cup (84) of the transfer mechanism (5); Preset effective adsorption force threshold Fa; When F is greater than or equal to Fa, it is determined that the product to be tested has been effectively adsorbed, and the next step can be carried out. When F is less than Fa, the adsorption effect is judged a second time.

3. The OCA bubble imaging detection method according to claim 2, characterized in that, When F is less than Fa, the secondary judgment on the adsorption effect includes: Calculate the absolute value of the adsorption force deviation between the effective adsorption force threshold Fa and the real-time adsorption force F. And obtain the real-time pressure data detected by the pressure sensor in the transfer platform (7). ; Preset adsorption force deviation threshold Fx; When Fx is less than or equal to And real-time pressure data When the value equals 0, it is determined that the product to be tested has been effectively adsorbed, and the next step can be carried out. When Fx is greater than or real-time pressure data If the value is greater than 0, it is determined that the product to be tested has not been effectively adsorbed, and the adsorption operation should be repeated.

4. The OCA bubble imaging detection method according to claim 1, characterized in that, When determining the final detection result based on the detection results of the first detection bit and the second detection bit, the following is included: By constructing and training a convolutional neural network model, product defects on the product assembly surface or the product bottom surface are predicted. When there are defects on the product assembly surface or the product bottom surface, the final detection result of the product to be tested is determined to be an abnormal product. When there are no defects on the product assembly surface and the product bottom surface, the final detection result of the product to be tested is determined to be a normal product.

5. The OCA bubble imaging detection method according to claim 4, characterized in that, After determining the final test result, when the final test result is a normal product, the product to be tested on the transfer platform (7) is transported to the normal product unloading side (22) of the first linear conveyor (2) by the material conveying mechanism. When the final test result is an abnormal product, the product to be tested is transported to the abnormal product unloading side (23) of the first linear conveyor (2) by the material conveying mechanism, and the final test result is stored in combination with the corresponding product code.

6. An OCA bubble imaging and detection device, characterized in that, The OCA bubble imaging detection method according to any one of claims 1-5 includes a machine (1), wherein the machine (1) is provided with a first linear conveying mechanism (2) for loading and unloading materials and a material conveying mechanism for transfer. A movable feeding platform (211) is provided on the feeding side (21) of the first linear conveying mechanism (2). The first linear conveying mechanism (2) drives the feeding platform (211) on which the product to be tested is placed to move from the feeding port (12) to the first detection position, and takes pictures of the product assembly surface through the first detection component set on the machine tool (1). A transfer platform (7) is provided between the material conveying mechanism and the end of the feeding side (21) away from the feeding port (12). The first detection component is provided with a second linear conveying mechanism (4), and the second linear conveying mechanism (4) is movably provided with a transfer mechanism (5). The transfer mechanism (5) is used to suspend the product to be tested from the feeding platform (211) to the second detection position. After the second detection component on the machine (1) takes a picture of the lower surface of the product, the second linear conveying mechanism (4) drives the transfer mechanism (5) to place the product to be tested on the transfer platform (7). The material conveying mechanism is used to transfer the product to be tested from the transfer platform (7) to the unloading platform (221) of the normal product unloading side (22) or abnormal product unloading side (23) of the first linear conveying mechanism (2). The first linear conveying mechanism (2) drives the unloading platform (221) with the product to be tested to move to the corresponding unloading port.

7. The OCA bubble imaging and detection device according to claim 6, characterized in that, The first detection component includes a first bracket (31) and a second bracket (32) disposed on both sides of the feed side (21), and two first cameras (33) disposed on the first bracket (31) and the second bracket (32) respectively; the second linear conveying mechanism (4) is disposed on the first bracket (31) or the second bracket (32); the second detection component includes a light source (61) and a third camera (62) disposed between the transfer platform (7) and one end near the feed side (21).

8. The OCA bubble imaging and detection device according to claim 7, characterized in that, A barcode scanner (11) is provided on one side of the feed inlet (12) on the machine base (1).

9. An OCA bubble imaging and detection device according to claim 6, characterized in that, The material conveying mechanism includes a material conveying guide rail (82) disposed on the machine base (1), a second cylinder (81) slidably disposed on the material conveying guide rail (82), and a second transfer plate (83) connected to the drive end of the second cylinder (81).

10. An OCA bubble imaging and detection device according to claim 9, characterized in that, The transfer mechanism (5) includes a first cylinder (51) slidably disposed on the second linear conveying mechanism (4), a motor (52) connected to the drive end of the first cylinder (51), and a first transfer plate (53) connected to the drive end of the motor (52). The first transfer plate (53) and the second transfer plate (83) adsorb the product to be tested through a suction cup (84).

Images