Transformation method of visual system of automatic parallel seam welder

Abstract

The invention relates to a method for transforming a visual system of an automatic parallel seam welder, and relates to the field of integrated circuit airtight packaging. According to the technical scheme, by constructing the targeted problem analysis model and the modification model, the root of the visual alignment deviation is accurately positioned, an efficient solution is quickly output, and the pain point that the alignment effect of traditional equipment is poor is fundamentally solved. After transformation, the alignment precision of the seam welding upper cover is remarkably improved, the defect of cover deviation is greatly reduced, the reliability and air tightness of integrated circuit packaging are guaranteed, and the problem that the yield is low due to deviation accumulation in seam welding of the large-size tube shell is effectively solved. Meanwhile, equipment does not need to be reconstructed in the transformation scheme, upgrading is achieved through paths such as light source optimization with the optimal cost, low investment and high return are both considered, the economical efficiency and the production practicability of the scheme are fully reflected, and the method is easy to apply and popularize in the industry.

Description

Technical Field

[0001] This invention relates to the field of integrated circuit hermetic packaging technology, and in particular to a method for modifying an automatic parallel seam welding machine vision system. Background Technology

[0002] One of the functions of the automatic parallel seam welding machine is the automatic cover-up function based on vision alignment. Its vision system relies on a dual-camera acquisition module to achieve this function: after the two cameras capture images of the cover plate and the shell respectively, the center point coordinates of the two are calculated through image algorithms, and then the cover deviation ∆x, ∆y, ∆θ are accurately obtained. Finally, the robotic arm and the transplanting platform are guided to complete the real-time compensation of the cover position during the automatic cover-up process, ensuring the basic alignment accuracy.

[0003] However, the vision system of traditional automatic parallel seam welding machines has the following technical limitations: the imaging quality of the casing is easily affected by fluctuations in the flatness of the casing itself, making it difficult to guarantee the alignment accuracy between the cover plate and the casing, resulting in frequent misalignment defects, which directly affect the reliability and hermeticity of integrated circuit packaging. Especially in the seam welding scenario of large-size integrated circuits, existing technologies mostly use the four-corner positioning method to obtain the casing position. Misalignment defects will cause the positioning deviation to accumulate and amplify, further aggravating the alignment error, ultimately resulting in an extremely low yield rate for large-size seam welded integrated circuits. Summary of the Invention

[0004] The purpose of this invention is to provide a method for modifying the vision system of an automatic parallel seam welding machine to solve the problems existing in the prior art.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a method for modifying an automatic parallel seam welding machine vision system, the method comprising: Based on the functional modules and optical principles of the vision system of the automatic parallel seam welding machine, at least two analytical models are established, which are used to locate the technical limitations of the vision system. The limitations of the vision system in terms of pixel accuracy and light source illumination method are obtained through the analysis model. A vision system modification model was established based on optical principles, and the capability enhancement and modification of the vision system of the automatic parallel seam welding machine was completed by combining the aforementioned limiting parameters.

[0006] In some embodiments, the at least two analysis models include a pixel accuracy analysis model and a light source imaging analysis model. The analysis models are established based on the imaging optical path principle, photoelectric conversion mechanism, and light source illumination characteristics of the automatic parallel seam welding machine vision system.

[0007] In some implementations, the process of establishing the pixel precision analysis model includes: Based on Fermat's principle, the imaging optical path of the vision system is simplified, and a similar triangle model of the imaging optical path is constructed. By combining the visual image acquisition method and optical principles of the automatic parallel seam welding machine, and incorporating parameters such as camera field of view size, imaging target size, lens working distance, and lens focal length, a complete pixel accuracy analysis model is formed.

[0008] In some implementations, the process of establishing the light source imaging analysis model includes: The original red direct ring light illumination method of the automatic parallel seam welding machine was simplified, and the relative positional relationship between the light source, lens, and pipe shell was clarified. Based on the propagation characteristics of reflected light from the tube and incident light from the lens, a light source imaging analysis model is established by incorporating parameters of the camera lens module, tube, and transplanting platform components.

[0009] In some implementations, obtaining the limiting parameters of the vision system in terms of pixel accuracy and light source illumination through the analysis model includes: The positive correlation between pixel accuracy and top cover deviation was determined by a pixel accuracy analysis model, and the deviation parameters corresponding to insufficient pixel accuracy were obtained. The illumination limitation parameters of the tube shell that cause abnormal grayscale imaging when the tube shell is tilted are obtained by using a light source imaging analysis model, which is based on the single incident angle and small exit angle of the direct ring light.

[0010] In some implementations, the capability enhancement and modification of the vision system for the automatic parallel seam welding machine in conjunction with the aforementioned limiting parameters includes: To address the insufficient pixel accuracy, the proposed improvements are to reduce pixel size, shrink camera field of view, shorten lens working distance, increase lens focal length, or increase camera resolution. To address the limitations of the light source, the proposed modification direction was to increase the angle of light incidence, and cost comparison confirmed this as the most cost-effective modification method.

[0011] In some implementations, the step of establishing a vision system modification model based on optical principles includes: Replace the small-sized direct ring light of the vision system of the automatic parallel seam welding machine with a large-sized diffuse light source to broaden the incident angle of the light. The modified vision system ensures that when the tube shell is tilted, the change in grayscale value of the tube shell image is controlled within the allowable range of the image algorithm's threshold segmentation.

[0012] In some implementations, the pixel accuracy analysis model calculates the top cover deviation based on pixel accuracy according to the following formula: N = B * D / E = B * C / A; Where N is the top cover deviation; B is the camera pixel size; D is the lens working distance; E is the lens focal length; C is the camera field of view; and A is the camera resolution.

[0013] In some embodiments, the defects of the original light source in the light source imaging analysis model are manifested as follows: When the tilt angle of the tube exceeds a preset threshold, the reflected light exceeds the receiving range of the camera lens, causing the gray value of the metal area of ​​the tube to be lower than the normal threshold.

[0014] In some implementations, the modified vision system needs to undergo mass production verification. The verification indicators include the improvement rate of vision alignment stability and the yield rate of large-size tube shell seam welds. The verification indicators need to meet preset industry standards. In addition, it also includes dynamically adjusting the parameters of the analysis model and the modification model based on the results of mass production verification. The parameters include the lens working distance compensation value and the diffuse light source brightness adjustment coefficient.

[0015] The beneficial effects of the technical solution provided by this invention include at least the following: This technical solution, by constructing targeted problem analysis and modification models, accurately pinpoints the root cause of visual alignment deviations and quickly outputs efficient solutions, fundamentally overcoming the pain point of poor alignment performance in traditional equipment. After the modification, the alignment accuracy of the welded cap is significantly improved, and the defect of misaligned caps is greatly reduced. This not only ensures the reliability and hermeticity of integrated circuit packaging but also effectively solves the problem of low yield caused by the accumulation of deviations in the welded caps of large-size tubes. Furthermore, the modification solution does not require equipment reconstruction; upgrades are achieved through cost-effective methods such as light source optimization, balancing low investment with high returns. This fully demonstrates the economic efficiency and practicality of the solution, making it easy to promote and apply within the industry. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0017] Figure 1 The diagram shows a flowchart illustrating a method for modifying an automatic parallel seam welding machine vision system according to an exemplary embodiment of the present invention.

[0018] Figure 2 This diagram illustrates a pixel accuracy analysis model of a method for modifying an automatic parallel seam welding machine vision system according to an exemplary embodiment of the present invention.

[0019] Figure 3 This illustration shows a schematic diagram of a prior art red direct ring light provided by an exemplary embodiment of the present invention.

[0020] Figure 4 This diagram illustrates a light source imaging analysis model for a method of modifying an automatic parallel seam welding machine vision system, provided by an exemplary embodiment of the present invention.

[0021] Figure 5 This illustration shows a schematic diagram of prior art imaging defect cause analysis provided by an exemplary embodiment of the present invention.

[0022] Figure 6 The diagram shows a vision system modification model of a vision system modification method for an automatic parallel seam welding machine provided by an exemplary embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0025] Figure 1 The diagram illustrates a flowchart of a method for modifying an automatic parallel seam welding machine vision system according to an exemplary embodiment of the present invention. The method includes: Step 101: Based on the functional modules and optical principles of the automatic parallel seam welding machine vision system, establish at least two analytical models. These analytical models are used to identify the technical limitations of the vision system.

[0026] In some embodiments, at least two analysis models include a pixel accuracy analysis model and a light source imaging analysis model. The analysis models are established based on the imaging optical path principle, photoelectric conversion mechanism, and light source illumination characteristics of the automatic parallel seam welding machine vision system.

[0027] In some embodiments, the process of establishing a pixel precision analysis model includes: Based on Fermat's principle, the imaging optical path of the vision system is simplified, and a similar triangle model of the imaging optical path is constructed. By combining the visual image acquisition method and optical principles of the automatic parallel seam welding machine, and incorporating parameters such as camera field of view size, imaging target size, lens working distance, and lens focal length, a complete pixel accuracy analysis model is formed.

[0028] Furthermore, in the pixel accuracy analysis model, the calculation of the top cover deviation corresponding to pixel accuracy satisfies the following formula: N = B * D / E = B * C / A; Where N is the top cover deviation (unit: micrometers); B is the camera pixel size; D is the lens working distance; E is the lens focal length; C is the camera field of view; and A is the camera resolution.

[0029] In this embodiment, the imaging principle of the tube shell detection image of the automatic parallel seam welding machine includes: reflected light from the tube shell surface within the camera's field of view is focused into a true image by the lens and projected onto the target surface of the camera's photoelectric sensor; the target surface converts the light signal into a digital image; the imaging optical path model of the vision system is simplified based on Fermat's principle, a pixel accuracy analysis model of similar triangles in the imaging optical path is established, and a complete pixel accuracy analysis model is constructed based on the equipment's visual image acquisition method and optical principles. (See also...) Figure 2 The pixel accuracy analysis model includes the camera field of view (FOV) size 101, imaging optical path 102, device telecentric lens 103, camera imaging target size 104, lens working distance 105, and lens focal length 106.

[0030] As a supplementary explanation, the pixel accuracy analysis model simplifies complex imaging problems, helping equipment pinpoint the root cause of cover deviation. First, it simplifies the imaging optical path of the vision system using Fermat's Last Theorem, visually representing the laws of light propagation with a similar triangle model, avoiding tedious optical calculations. Next, considering the actual working method of the welding machine, it incorporates hardware parameters such as camera field of view, target size, lens distance, and focal length into the model, ensuring a perfect match between the model and real-world conditions. The formula N=B*D / E=B*C / A, by substituting visible parameters like camera pixel size and resolution, allows for the calculation of the specific value of the cover deviation.

[0031] In some embodiments, the process of establishing a light source imaging analysis model includes: The original red direct ring light illumination method of the automatic parallel seam welding machine was simplified, and the relative positional relationship between the light source, lens, and pipe shell was clarified. Based on the propagation characteristics of reflected light from the tube and incident light from the lens, a light source imaging analysis model is established by incorporating parameters of the camera lens module, tube, and transplanting platform components.

[0032] In the embodiments of this application, see Figure 3 Existing technology uses a red direct-firing ring light, which has a small beam angle and a relatively large distance between the light source, lens, and tube housing. (See also...) Figure 4 The image shows the camera lens module 201, the reflected light from the tube shell and the incident light from the lens 202 are both perpendicular to the tube shell, the tube shell 203, and the transfer platform 204.

[0033] Furthermore, in the light source imaging analysis model, the defects of the original light source are manifested as follows: When the tilt angle of the tube exceeds a preset threshold, the reflected light exceeds the receiving range of the camera lens, causing the gray value of the metal area of ​​the tube to be lower than the normal threshold.

[0034] In the embodiments of this application, see Figure 5 The existing technology has a relatively single incident angle of illumination light, and the incident angle is small. When the tube shell is tilted at a large angle, a large amount of light is reflected outside the lens range, causing the gray value of the originally bright metal area to darken. This results in some areas of the tube shell being darkened after imaging, causing the visual algorithm to make a deviation in judging the position.

[0035] As a supplementary explanation, the light source imaging analysis model mainly identifies the matching problem between illumination and imaging, providing a basis for the visual system to supplement lighting and correct errors. During modeling, the original working method of the direct red ring light was first simplified, clarifying the positional relationship between the light source, lens, and tube shell. Then, parameters of actual components such as the camera lens module and the transfer platform were incorporated to ensure the model perfectly matches the welding machine's operating conditions. In this case, it was found that the original light source had a small and singular light angle and was far from the tube shell. If the tube shell warped slightly, the reflected light would escape the lens's receiving range, causing the originally bright metal area to appear dark. This explains why the visual algorithm experienced positioning errors.

[0036] Step 102: Obtain the limiting parameters of the vision system in terms of pixel accuracy and light source illumination method through analysis model.

[0037] In some embodiments, the limitations of the vision system in terms of pixel accuracy and light source illumination are obtained through analysis models, including: The positive correlation between pixel accuracy and top cover deviation was determined by a pixel accuracy analysis model, and the deviation parameters corresponding to insufficient pixel accuracy were obtained. The illumination limitation parameters of the tube shell that cause abnormal grayscale imaging when the tube shell is tilted are obtained by using a light source imaging analysis model, which is based on the single incident angle and small exit angle of the direct ring light.

[0038] In this embodiment, the pixel accuracy analysis model no longer remains at the theoretical level, but transforms insufficient accuracy into explicit deviation parameters by quantifying the correlation between pixel accuracy and top cover deviation. The light source imaging analysis model captures parameters of abnormal image grayscale when the tube shell warps due to angle issues caused by direct ring light. These two sets of parameters, starting from positioning calculation and image quality respectively, fully reflect the shortcomings of the vision system, making the upgrade solution more targeted.

[0039] Step 103: Establish a vision system modification model based on optical principles, and combine the limiting parameters to complete the capability improvement and modification of the vision system of the automatic parallel seam welding machine.

[0040] In some embodiments, the capability enhancement and modification of the vision system for an automated parallel seam welding machine is achieved by incorporating limiting parameters, including: To address the insufficient pixel accuracy, the proposed improvements are to reduce pixel size, shrink camera field of view, shorten lens working distance, increase lens focal length, or increase camera resolution. To address the limitations of the light source, the proposed modification direction was to increase the angle of light incidence, and cost comparison confirmed this as the most cost-effective modification method.

[0041] In some embodiments, see Figure 6 A vision system modification model was established based on optical principles, including: Replace the small-sized direct ring light of the vision system of the automatic parallel seam welding machine with a large-sized diffuse light source to broaden the incident angle of the light. The modified vision system ensures that when the tube shell is tilted, the change in grayscale value of the tube shell image is controlled within the allowable range of the image algorithm's threshold segmentation.

[0042] In this embodiment, addressing the shortcoming of pixel accuracy, the modified model directly addresses the correlation between hardware parameters and accuracy. By reducing pixels and shortening lens distance, the accuracy of coordinate calculation is improved from the source. The light source modification focuses on balancing cost and effectiveness. Based on the principle of light propagation, the direct ring light is replaced with a large-size diffused light source, thereby widening the incident angle to solve the imaging problem when the tube shell is warped. After the modification, the grayscale variation of the tube shell is controlled within the algorithm's allowable range, solving the problem of insufficient illumination from the original light source while ensuring the stability of the visual algorithm's judgment.

[0043] In some embodiments, the modified vision system needs to undergo mass production verification. The verification indicators include the improvement rate of vision alignment stability and the yield rate of large-size tube shell seam welding. The verification indicators need to meet the preset industry standards. In addition, it also includes dynamically adjusting the parameters of the analysis model and the modification model based on the mass production verification results. The parameters include the lens working distance compensation value and the diffuse light source brightness adjustment coefficient.

[0044] In summary, this technical solution, by constructing targeted problem analysis and modification models, accurately identifies the root cause of visual alignment deviations and quickly outputs efficient solutions, fundamentally addressing the pain point of poor alignment performance in traditional equipment. After the modification, the alignment accuracy of the welded cap is significantly improved, and the defect of misaligned caps is greatly reduced. This not only ensures the reliability and hermeticity of integrated circuit packaging but also effectively solves the problem of low yield caused by the accumulation of deviations in the welded caps of large-size tubes. Furthermore, the modification solution does not require equipment reconstruction; upgrades are achieved through cost-effective methods such as light source optimization, balancing low investment with high returns. This fully demonstrates the economic efficiency and practicality of this solution, making it easy to promote and apply within the industry.

[0045] It is understood that the specific examples in this document are only intended to help those skilled in the art better understand this disclosure, and are not intended to limit the scope of the invention.

[0046] It is understood that in the various embodiments of this specification, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of this disclosure.

[0047] It is understood that the various implementation methods described in this specification can be implemented individually or in combination, and this disclosure does not limit them.

[0048] Unless otherwise stated, all technical and scientific terms used in this disclosure have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of this specification. The term "and / or" as used in this specification includes any and all combinations of one or more of the associated listed items. The singular forms "a," "the," and "the" as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0049] The above description is merely a specific embodiment of this specification, but the scope of protection of this invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this specification should be included within the scope of protection of this specification. Therefore, the scope of protection of this invention should be determined by the scope of the claims.

Claims

1. A method for modifying the vision system of an automatic parallel seam welding machine, characterized in that, The method includes: Based on the functional modules and optical principles of the vision system of the automatic parallel seam welding machine, at least two analytical models are established, which are used to locate the technical limitations of the vision system. The limitations of the vision system in terms of pixel accuracy and light source illumination method are obtained through the analysis model. A vision system modification model was established based on optical principles, and the capability enhancement and modification of the vision system of the automatic parallel seam welding machine was completed by combining the aforementioned limiting parameters.

2. The method for modifying the vision system of the automatic parallel seam welding machine according to claim 1, characterized in that, The at least two analysis models include a pixel accuracy analysis model and a light source imaging analysis model. The analysis models are established based on the imaging optical path principle, photoelectric conversion mechanism, and light source illumination characteristics of the automatic parallel seam welding machine vision system.

3. The method for modifying the vision system of the automatic parallel seam welding machine according to claim 2, characterized in that, The process of establishing the pixel accuracy analysis model includes: Based on Fermat's principle, the imaging optical path of the vision system is simplified, and a similar triangle model of the imaging optical path is constructed. By combining the visual image acquisition method and optical principles of the automatic parallel seam welding machine, and incorporating parameters such as camera field of view size, imaging target size, lens working distance, and lens focal length, a complete pixel accuracy analysis model is formed.

4. The method for modifying the vision system of the automatic parallel seam welding machine according to claim 2, characterized in that, The process of establishing the light source imaging analysis model includes: The original red direct ring light illumination method of the automatic parallel seam welding machine was simplified, and the relative positional relationship between the light source, lens, and pipe shell was clarified. Based on the propagation characteristics of reflected light from the tube and incident light from the lens, a light source imaging analysis model is established by incorporating parameters of the camera lens module, tube, and transplanting platform components.

5. The method for modifying the vision system of the automatic parallel seam welding machine according to claim 1, characterized in that, The step of obtaining the limitation parameters of the vision system in terms of pixel accuracy and light source illumination method through the analysis model includes: The positive correlation between pixel accuracy and top cover deviation was determined by a pixel accuracy analysis model, and the deviation parameters corresponding to insufficient pixel accuracy were obtained. The illumination limitation parameters of the tube shell that cause abnormal grayscale imaging when the tube shell is tilted are obtained by using a light source imaging analysis model, which is based on the single incident angle and small exit angle of the direct ring light.

6. The method for modifying the vision system of the automatic parallel seam welding machine according to claim 5, characterized in that, The improvement and modification of the vision system for the automatic parallel seam welding machine, based on the aforementioned limiting parameters, includes: To address the insufficient pixel accuracy, the proposed improvements are to reduce pixel size, shrink camera field of view, shorten lens working distance, increase lens focal length, or increase camera resolution. To address the limitations of the light source, the proposed modification direction was to increase the angle of light incidence, and cost comparison confirmed this as the most cost-effective modification method.

7. The method for modifying the vision system of the automatic parallel seam welding machine according to claim 6, characterized in that, The vision system modification model established based on optical principles includes: Replace the small-sized direct ring light of the vision system of the automatic parallel seam welding machine with a large-sized diffuse light source to broaden the incident angle of the light. The modified vision system ensures that when the tube shell is tilted, the change in grayscale value of the tube shell image is controlled within the allowable range of the image algorithm's threshold segmentation.

8. The method for modifying the vision system of the automatic parallel seam welding machine according to claim 3, characterized in that, In the pixel accuracy analysis model, the calculation of the upper cover deviation corresponding to pixel accuracy satisfies the following formula: N = B * D / E = B * C / A; Where N is the top cover deviation; B is the camera pixel size; D is the lens working distance; E is the lens focal length; C is the camera field of view; and A is the camera resolution.

9. The method for modifying the vision system of the automatic parallel seam welding machine according to claim 4, characterized in that, In the aforementioned light source imaging analysis model, the defects of the original light source are manifested as follows: When the tilt angle of the tube exceeds a preset threshold, the reflected light exceeds the receiving range of the camera lens, causing the gray value of the metal area of ​​the tube to be lower than the normal threshold.

10. The method for modifying the vision system of the automatic parallel seam welding machine according to claim 7, characterized in that, The modified vision system needs to be verified through mass production. The verification indicators include the improvement rate of vision alignment stability and the yield rate of large-size tube shell seam welding. The verification indicators need to meet the preset industry standards. In addition, it also includes dynamically adjusting the parameters of the analysis model and the modification model based on the results of mass production verification. The parameters include the lens working distance compensation value and the diffuse light source brightness adjustment coefficient.

Images