An intelligent production control system for PCB circuit boards

Abstract

This invention discloses an intelligent production control system for PCB circuit boards. It relates to the field of PCB manufacturing technology and solves the problem of insufficient precision control for PCBs with slight bends. The invention identifies the raw materials of the PCB board through image determination and analysis. Based on the analysis process, it determines whether the raw material image meets the standards. For compliant parts, drilling is performed directly. For non-compliant parts, pixel analysis is used to identify whether there is bending on the surface of the non-compliant parts and determine the specific area associated with the bending. The determined specific area is then reconfirmed to lock the corresponding associated planar vector and the specific location of the corresponding hole. This allows for more precise drilling even on bent boards, effectively ensuring the control precision during the drilling process.

Description

Technical Field

[0001] This invention relates to the field of PCB manufacturing technology, specifically to an intelligent production control system for PCB circuit boards. Background Technology

[0002] A PCB, or Printed Circuit Board, is an important component of electronic devices. It is formed by printing, etching, and other processes on an insulating substrate material to create conductive lines and component mounting areas, which are used to connect various electronic components to achieve circuit functions.

[0003] Application CN112492760A discloses a PCB manufacturing method that requires the use of a drilling device. This drilling device includes a processing frame, a telescopic cylinder, a sliding control frame, a drilling execution device, and a control and adjustment device. This invention can solve the following problems existing in the current PCB manufacturing process: A) During the processing of copper plates, manual drilling is required. With numerous holes on the copper plate, manual drilling of each hole is time-consuming and results in poor drilling accuracy, affecting the quality of the finished PCB. B) When manually drilling copper plates, only one copper plate can be processed at a time, making it impossible to drill multiple copper plates simultaneously, resulting in low processing efficiency and affecting PCB production efficiency.

[0004] During the manufacturing process of PCB boards, it is generally necessary to perform preliminary drilling on the PCB board raw materials based on a preset template. However, in the actual drilling process, if the corresponding raw materials are slightly bent, it will cause a decrease in drilling accuracy, which will make it difficult to install electronic components on the corresponding PCB board later, thus causing the circuit board to be scrapped. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent production control system for PCB circuit boards, which solves the problem of not being able to perform precision control for circuit boards with slight bending.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent production control system for PCB circuit boards, comprising:

[0007] At the machine vision end, images of the initial board material involved in the hole processing of the circuit board are acquired and high-definition images of the board material are obtained. The acquired high-definition images of the board material are then transmitted to the circuit board calibration end.

[0008] The template library contains preset standard templates, which are preset templates.

[0009] The circuit board calibration end compares the acquired high-resolution image of the board material with the preset standard templates in the template library. Based on the comparison results, it identifies whether the initial board material is a qualified part. If it is a qualified part, drilling is performed directly through the execution end. If it is not a qualified part, it is calibrated as a part to be identified, and pixel analysis is performed based on the gradient analysis processing end. The specific method is as follows:

[0010] Based on the acquired high-resolution image of the sheet metal, the overall edge contour of the high-resolution image is determined, and the center point of the high-resolution image is determined based on the overall edge contour of the sheet metal. The specific method is as follows:

[0011] The overall edge contour is decomposed into several contour points, and then these contour points are simultaneously placed in a set of two-dimensional coordinate systems. Based on the corresponding two-dimensional coordinate systems, the two-dimensional coordinates of each set of contour points can be determined. Then, based on the two-dimensional coordinates associated with several sets of contour points, the two-dimensional coordinates are averaged to determine the mean coordinates. The determined mean coordinates are then used to determine the associated coordinate points and simultaneously marked within the high-definition image of the sheet material to determine the center point of this high-definition image of the sheet material.

[0012] Based on the preset center point in the standard template, the center point of the high-definition image is combined with the preset center point of the standard template, and it is identified whether the high-definition image completely overlaps with the standard template. If they completely overlap, the initial board material is marked as a qualified part; if they do not completely overlap, the initial board material is marked as a part to be identified.

[0013] For the calibrated qualified parts, the corresponding execution end confirms the pre-set associated hole position features in the standard template according to the point relationship, and then drills holes on the initial board material based on the confirmed associated hole position features. The associated hole position features are vector features from the center point to the center point of the corresponding hole. Based on this vector feature, the specific position of the corresponding hole is locked on the initial board material and drilled.

[0014] For the identified object, pixel analysis is performed using subsequent gradient analysis processing.

[0015] The gradient analysis processing end then checks the surface flatness of the part to be identified. Based on the different pixel features between different pixels on the surface of the part to be identified, it identifies whether the surface of the part to be identified is curved. If curvature exists, the part to be identified is marked as a curved part; if no curvature exists, the part to be identified is marked as a dimensionally abnormal part and directly displayed. The specific method is as follows:

[0016] The surface of the object to be identified is sequentially calibrated, and the pixel value associated with each pixel is determined. The different pixel values ​​associated with different pixels are labeled as X. i , where i represents different pixels;

[0017] Perform difference analysis on a set of pixels, mark the neighboring pixels of this pixel as the pixels to be checked, and identify the pixel difference between this pixel and the pixels to be checked, where the pixel difference is ≥ 0. Then, average the several sets of pixel differences identified for this pixel to lock the associated feature value J. i ;

[0018] Identify the associated feature values ​​J of several pixels on the surface of the object to be identified. i Do they all satisfy: J i ≤Y1, where Y1 is a preset value. If all conditions are met, the object to be identified is marked as an abnormal size object and displayed directly. If not all conditions are met, the area associated with the unmet pixels is marked, and the object to be identified is marked as a bent object.

[0019] The associated adjustment end, for the marked curved part, confirms the associated area marked on the surface of the curved part, and simultaneously adjusts the curvature of the standard template so that the pixel values ​​of the points inside the standard template are consistent with those on the surface of the curved part. Then, based on the positional features of the corresponding associated holes in the standard template after curvature adjustment, the vector features are confirmed, and the drilling process of the associated holes is performed through the execution end. The specific method is as follows:

[0020] Based on the associated region marked on the surface of the bent part, the curvature of the surface of the standard template is adjusted until it is adjusted to a region that is completely consistent with the pixel feature changes inside this associated region. The region identified in the standard template is marked as the adjustment region. Keeping the direction of the adjustment region unchanged, the adjustment region is moved within the standard template. During the movement, the actual size of the standard template will change accordingly. It is identified whether there is a set of standard templates that completely overlap with the bent part during several movement processes. If so, the standard templates in the overlapping state are marked as feature templates.

[0021] Based on the feature template and the edge contour of the associated curved part, determine its center point, combine the two center points, and based on the preset associated hole position in the feature template and the center point inside the feature template, confirm the plane vector of the corresponding associated hole. The initial point of the plane vector is the center point of the feature template, and the end point is the center point of the associated hole.

[0022] The confirmed planar vector and the size of the associated hole are transmitted to the execution end. The execution end performs the drilling process of the associated hole on the surface of the bent part based on the confirmed center point of the bent part. The specific location of the corresponding associated hole is confirmed based on the planar vector, and the drilling process of the hole is completed on the surface of the bent part according to the size of the associated hole.

[0023] If, during several of the aforementioned movements, there is no set of standard templates that completely overlap with the bent part, then the bent part is directly marked as a dimensionally abnormal part and directly displayed.

[0024] This invention provides an intelligent production control system for PCB circuit boards. Compared with the prior art, it has the following advantages:

[0025] This invention identifies the raw materials of a PCB board by image determination, analyzes the identified raw material images, and identifies whether the raw material images meet the standards based on the analysis process. For compliant parts, drilling is performed directly.

[0026] For other parts that do not meet the standards, pixel point analysis is used to identify whether there is bending on the surface of the non-compliant parts, and to determine the specific area associated with the bending. The determined specific area is then reconfirmed to lock the corresponding associated planar vector and the specific location of the corresponding hole. During the drilling process of the corresponding hole, the bent sheet material can achieve a more accurate drilling effect, and the control precision of the drilling production process can be effectively guaranteed. Attached Figure Description

[0027] Figure 1 This is a schematic diagram of the principle framework of the present invention;

[0028] Figure 2 This is a schematic diagram illustrating the determination of the planar vector within the compliant component of the present invention. Detailed Implementation

[0029] 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.

[0030] First Embodiment

[0031] Please see Figure 1 This application provides an intelligent production control system for PCB circuit boards, including a machine vision end, a circuit board calibration end, a template library, a gradient analysis and processing end, an association adjustment end, and an execution end. The machine vision end, circuit board calibration end, gradient analysis and processing end, association adjustment end, and execution end are electrically connected from the output node to the input node in sequence. The template library is electrically connected to the input node of the circuit board calibration end or the association adjustment end, and the circuit board calibration end is electrically connected to the input node of the execution end.

[0032] In the machine vision end, images of the initial board material involved in the hole processing of the circuit board are acquired and high-definition images of the board material are obtained. The acquired high-definition images of the board material are then transmitted to the circuit board calibration end. The vision probe associated with the image acquisition is an ultra-high-definition processing probe, which acquires high-definition images of the surface of the initial board material. The initial board material is cut from the original mold material, which is a composite material mainly composed of glass fiber cloth and epoxy resin.

[0033] The template library contains preset standard templates. These standard templates are pre-prepared by relevant operators and are used to compare with high-definition images of the sheet metal to identify whether the acquired high-definition images of the sheet metal are consistent with these standard templates.

[0034] In the circuit board calibration section, the acquired high-definition image of the board material is compared with a preset standard template in the template library. Based on the comparison result, it is determined whether the initial board material is a qualified part. If it is a qualified part, drilling is performed directly through the execution end. If it is not a qualified part, it is calibrated as a part to be identified, and pixel point analysis is performed based on the gradient analysis processing end. The specific comparison method is as follows:

[0035] Based on the acquired high-resolution image of the sheet metal, the overall edge contour of the high-resolution image of the sheet metal is confirmed (since the background corresponding to the image is white when it is acquired, the edge contour of the high-resolution image can be directly confirmed). Based on the overall edge contour of the high-resolution image of the sheet metal, the center point of the high-resolution image of the sheet metal is confirmed. Specifically, the overall edge contour is decomposed into several contour points, and then several contour points are simultaneously placed in a set of two-dimensional coordinate systems. Based on the corresponding two-dimensional coordinate systems, the two-dimensional coordinates of each set of contour points can be confirmed. Then, based on the two-dimensional coordinates associated with several sets of contour points, the two-dimensional coordinates are averaged to determine the mean coordinates. Then, the determined mean coordinates are used to determine the associated coordinate points and are simultaneously marked in the high-resolution image of the sheet metal to determine the center point of the high-resolution image of the sheet metal.

[0036] Based on the preset center point in the standard template, the center point of the high-definition image is combined with the preset center point of the standard template, and it is identified whether the high-definition image completely overlaps with the standard template (there is no non-overlapping area, which is a state of complete overlap). If they completely overlap, the initial board material is marked as a qualified part; if they do not completely overlap, the initial board material is marked as a part to be identified.

[0037] For the calibrated compliant parts, the corresponding execution end confirms the pre-set associated hole position features within the standard template based on the point-to-point relationship. Then, based on the confirmed associated hole position features, it drills holes on the initial sheet material. The associated hole position features are vector features from the center point to the corresponding hole center point. Based on this vector feature, the specific location of the corresponding hole is locked on the initial sheet material and drilled (e.g., there is a corresponding center point and several hole features within the standard template, and a center point within the hole features. Based on the center point inside the template and the center point of the hole, the corresponding planar vector can be confirmed. Based on this planar vector and the center point confirmed above the initial sheet material, the specific location of the corresponding hole is confirmed. Then, the corresponding specific location point is used as the center point to drill the associated hole, completing the hole drilling process for the corresponding sheet material). Figure 2 Based on the center point marked within the compliant part and the specific location of the corresponding associated hole, determine the center point of the associated hole, and then start from the center point of the compliant part and end at the center point of the associated hole to lock the corresponding planar vector.

[0038] For the calibrated object to be identified, gradient analysis is performed using a subsequent gradient analysis processing terminal.

[0039] In the gradient analysis processing section, the surface flatness of the object to be identified is confirmed. Based on the different pixel features between different pixels on the surface of the object, it is determined whether the surface of the object is curved. If curvature exists, the object is marked as a curved object; if no curvature exists, the object is marked as a dimensionally abnormal object and displayed directly. The specific identification method is as follows:

[0040] The surface of the object to be identified is sequentially calibrated, and the pixel value associated with each pixel is determined. The different pixel values ​​associated with different pixels are labeled as X. i , where i represents different pixels;

[0041] For a given group of pixels, perform difference analysis, identify the neighboring pixels of this pixel as the pixels to be checked, and determine the pixel difference between this pixel and the pixels to be checked, where the pixel difference is ≥ 0 (that is, when the pixel value of the corresponding pixel is differing from the pixel value of the pixel to be checked, the minimum value is subtracted from the maximum value to determine the corresponding pixel difference). Then, average the several groups of pixel differences identified for this pixel to lock the associated feature value J. i ;

[0042] Identify the associated feature values ​​J of several pixels on the surface of the object to be identified. i Do they all satisfy: J i≤Y1, where Y1 is a preset value. Its specific value is determined by the operator based on experience. If all conditions are met, it means that there are no pixels with abrupt changes on the surface of the object to be identified, that is, a flat state. Therefore, the object to be identified is marked as a size abnormality object and is directly displayed. If not all conditions are met, the area associated with the unmet pixels is marked, and the object to be identified is marked as a bent object.

[0043] Specifically, if the corresponding processed part is not consistent with the standard template, and the surface of the processed part does not show any bending, then the processed part has a corresponding dimensional non-compliance. Before processing, external personnel will inspect the processed part to identify whether it has defects. Processed parts with defects will be directly rejected and will not participate in the etching and drilling process. When there is partial bending or slight bending, it cannot be observed by the human eye in time. When the size is slightly abnormal, it cannot be observed by the human eye in time either. Therefore, in order to ensure the accuracy of hole drilling and etching, it is necessary to analyze such processed parts, evaluate whether the corresponding identification part is a bent part, and then carry out drilling processing.

[0044] Second Embodiment

[0045] In the specific implementation process, compared with the above embodiments, this embodiment mainly focuses on the drilling process of the corresponding hole, mainly for the bent part, and is executed by the corresponding associated adjustment end;

[0046] The associated adjustment end, for the marked curved part, confirms the associated area marked on the surface of the curved part, and simultaneously adjusts the curvature of the standard template so that the pixel values ​​of the points inside the standard template are consistent with those on the surface of the curved part. Then, based on the positional features of the corresponding associated holes in the standard template after curvature adjustment, the vector features are confirmed, and the drilling process of the associated holes is performed through the execution end. The specific method for confirming the vector features is as follows:

[0047] Based on the associated area marked on the surface of the bent part, the curvature of the surface of the standard template is adjusted until it is adjusted to a region that is completely consistent with the pixel feature changes inside this associated region. The region identified in the standard template is marked as the adjustment region. Keeping the direction of the adjustment region unchanged, the adjustment region is moved within the standard template. During the movement, the actual size of the standard template will change accordingly. It is identified whether there is a set of standard templates that completely overlap with the bent part during several movements. If they do, the standard templates in the overlapping state are marked as feature templates. If they do not exist, it means that this bent part not only has a bent area, but also a missing area, which leads to the situation where no overlap can be found. In this case, the bent part is directly marked as a dimensional abnormal part and directly displayed.

[0048] Based on the feature template and the edge contour of the associated curved part, determine its center point (place the edge contour in a two-dimensional coordinate system and determine the center point by determining the mean coordinate). Combine the center points of the two, and based on the preset associated hole position in the feature template and the center point inside the feature template, confirm the plane vector of the corresponding associated hole. The initial point of the plane vector is the center point of the feature template, and the end point is the center point of the associated hole.

[0049] The confirmed planar vector and the dimensions of the associated holes are transmitted to the execution end. Based on the confirmed center point of the bent part, the execution end performs the drilling process of the associated holes on the surface of the bent part. Based on the planar vector, the specific location of the corresponding associated holes is confirmed, and then the drilling process of the holes is completed on the surface of the bent part according to the dimensions of the associated holes, thereby completing the etching and drilling work on the surface of the bent part.

[0050] Specifically, when the corresponding part to be identified is a curved part, then its surface is determined to have a corresponding curved area. Within the standard template, the associated area with the same pixel features is preferentially identified. This associated area is completely consistent with the curved area. After the corresponding associated area is identified within the standard template, the overall size and outline of the standard template will change. Then, by moving the position of the associated area, when the curved part only has a curved state and there is no missing state, then the corresponding completely overlapping situation can be locked.

[0051] Based on the corresponding overlapping conditions, the planar vector is determined to lock the corresponding drilling process. Based on the corresponding planar vector, the location of the corresponding drilling hole is determined and drilling is carried out to complete the overall drilling process of the corresponding sheet material.

[0052] Third Embodiment

[0053] In its specific implementation, this embodiment includes the entire implementation process of the two sets of embodiments described above.

[0054] Some of the data in the above formulas are numerical calculations with dimensions removed, and the contents not described in detail in this specification are all prior art known to those skilled in the art.

[0055] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. An intelligent production control system for PCB circuit boards, characterized in that, include: At the machine vision end, images of the initial board material involved in the hole processing of the circuit board are acquired and high-definition images of the board material are obtained. The acquired high-definition images of the board material are then transmitted to the circuit board calibration end. The template library contains preset standard templates, which are preset templates. The circuit board calibration end compares the high-definition image of the board material with the preset standard template in the template library. Based on the comparison result, it identifies whether the initial board material is a qualified part. If it is a qualified part, it directly performs drilling processing through the execution end. If it is not a qualified part, it is calibrated as a part to be identified, and pixel point analysis is performed based on the gradient analysis processing end. The gradient analysis processing end then confirms the surface flatness of the object to be identified. Based on the different pixel features between different pixels on the surface of the object to be identified, it identifies whether the surface of the object to be identified is curved. If there is a curve, the object to be identified is marked as a curved object. If there is no curve, the object to be identified is marked as a dimensional abnormality object and is directly displayed. The associated adjustment end identifies the associated area marked on the surface of the calibrated curved part and simultaneously adjusts the curvature of the standard template so that the pixel values ​​of the points inside the standard template are consistent with those of the curved part. Then, based on the positional features of the associated holes in the standard template after curvature adjustment, the vector features are identified, and the drilling process of the associated holes is performed through the execution end.

2. The intelligent production control system for PCB circuit boards according to claim 1, characterized in that, The specific method by which the circuit board calibration terminal compares the high-definition image of the board material with the preset standard template in the template library is as follows: Based on the acquired high-resolution image of the sheet metal, the overall edge contour of the high-resolution image of the sheet metal is confirmed, and based on the overall edge contour of the high-resolution image of the sheet metal, the center point of the high-resolution image of the sheet metal is confirmed. Based on the preset center point in the standard template, the center point of the high-definition image is combined with the preset center point of the standard template, and it is identified whether the high-definition image completely overlaps with the standard template. If they completely overlap, the initial board material is marked as a qualified part; if they do not completely overlap, the initial board material is marked as a part to be identified. For the calibrated qualified parts, the corresponding execution end confirms the pre-set associated hole position features in the standard template according to the point relationship, and then drills holes on the initial board material based on the confirmed associated hole position features. The associated hole position features are vector features from the center point to the center point of the corresponding hole. Based on this vector feature, the specific position of the corresponding hole is locked on the initial board material and drilled. For the calibrated object to be identified, pixel analysis is performed using subsequent gradient analysis processing.

3. The intelligent production control system for PCB circuit boards according to claim 2, characterized in that, The specific method for confirming the center point of the high-definition image of the circuit board at the circuit board calibration terminal is as follows: The overall edge contour is decomposed into several contour points, and then these contour points are simultaneously placed in a set of two-dimensional coordinate systems. Based on the corresponding two-dimensional coordinate systems, the two-dimensional coordinates of each set of contour points can be determined. Then, based on the two-dimensional coordinates associated with several sets of contour points, the two-dimensional coordinates are averaged to determine the mean coordinates. The determined mean coordinates are then used to determine the associated coordinate points and simultaneously marked on the high-definition image of the sheet material to determine the center point of this high-definition image of the sheet material.

4. The intelligent production control system for PCB circuit boards according to claim 1, characterized in that, The gradient analysis processing terminal identifies whether the surface of the object to be identified is curved in the following specific way: The surface of the object to be identified is sequentially calibrated, and the pixel value associated with each pixel is determined. The different pixel values ​​associated with different pixels are labeled as X. i , where i represents different pixels; Perform difference analysis on a set of pixels, mark the neighboring pixels of this pixel as the pixels to be checked, and identify the pixel difference between this pixel and the pixels to be checked, where the pixel difference is ≥ 0. Then, average the several sets of pixel differences identified for this pixel to lock the associated feature value J. i ; Identify the associated feature values ​​J of several pixels on the surface of the object to be identified. i Do they all satisfy: J i ≤Y1, where Y1 is a preset value. If all conditions are met, the part to be identified will be marked as a part with abnormal size and displayed directly.

5. The intelligent production control system for PCB circuit boards according to claim 4, characterized in that, The associated feature value J of several pixels on the surface of the object to be identified i If not all J is satisfied i If ≤Y1, the region associated with the unsatisfied pixel is calibrated, and the object to be identified is calibrated as a curved object.

6. The intelligent production control system for PCB circuit boards according to claim 1, characterized in that, The specific method for confirming vector features at the correlation adjustment end is as follows: Based on the associated region marked on the surface of the bent part, the curvature of the surface of the standard template is adjusted until it is adjusted to a region that is completely consistent with the pixel feature changes inside this associated region. The region identified in the standard template is marked as the adjustment region. Keeping the direction of the adjustment region unchanged, the adjustment region is moved within the standard template. During the movement, the actual size of the standard template will change accordingly. It is identified whether there is a set of standard templates that completely overlap with the bent part during several movement processes. If so, the standard templates in the overlapping state are marked as feature templates. Based on the feature template and the edge contour of the associated curved part, determine its center point, combine the two center points, and based on the preset associated hole position in the feature template and the center point inside the feature template, confirm the plane vector of the corresponding associated hole. The initial point of the plane vector is the center point of the feature template, and the end point is the center point of the associated hole. The confirmed planar vector and the dimensions of the associated holes are transmitted to the execution end. The execution end performs the drilling process of the associated holes on the surface of the bent part based on the confirmed center point of the bent part. The specific location of the corresponding associated holes is confirmed based on the planar vector, and the drilling process of the holes is completed on the surface of the bent part according to the dimensions of the associated holes.

7. The intelligent production control system for PCB circuit boards according to claim 6, characterized in that, If, during several of the aforementioned movements, there is no set of standard templates that completely overlap with the bent part, then the bent part is directly marked as a dimensionally abnormal part and directly displayed.

Images