Method for detecting solder resist deviation of printed circuit board

Abstract

The invention discloses a method for detecting resistance welding deviation of a printed circuit board, relates to the technical field of printed circuit boards, and solves the problem that resistance welding thickness detection generally adopts a single-wavelength laser scanning or contact type thickness measuring method, and single-wavelength laser is easy to interfere with surface flatness of a resistance welding layer. A dual-wavelength laser layered detection scheme is adopted, the characteristics that visible light cannot penetrate through a solder resist oil film and near-infrared light can penetrate through the oil film are utilized, an angle verification mechanism of a flat point position is combined, solder resist reflection points and base material reflection points are accurately locked, and direct measurement of the solder resist thickness is achieved; the method does not need to depend on an additional reference area, the defect that traditional single-wavelength laser detection is affected by surface reflection and ink color is overcome, it is ensured that only thickness calculation is conducted on the flat point position through angle matching, the thickness detection accuracy is further improved, and reliable data support is provided for accurate judgment of a thin area and a thick area.

Description

Technical Field

[0001] This invention relates to the field of printed circuit board technology, specifically to a method for detecting solder resist deviation in printed circuit boards. Background Technology

[0002] As the core interconnect carrier of electronic devices, the precision of the solder mask coating on the surface of printed circuit boards (PCBs) directly determines the insulation performance, soldering reliability and service life of the product.

[0003] The uniformity of solder mask thickness and the accuracy of its position are the core quality control indicators of the solder mask process. If the solder mask is too thin, it is easy to cause exposed circuit oxidation and increase the risk of short circuits. If the solder mask is too thick, it may cover the pads and cause poor soldering and desoldering of components. Furthermore, misdivision of the solder mask area will directly lead to deviation of the detection target area, resulting in the outflow of batches of defective products.

[0004] In traditional PCB solder mask inspection processes, area division often relies on manual visual inspection or single visual threshold segmentation. This method is greatly affected by ink color, surface reflection, and lighting conditions, making it prone to misjudging the substrate area and the solder mask area, and it cannot effectively screen atypical areas. Solder mask thickness inspection generally uses single-wavelength laser scanning or contact thickness measurement methods. Single-wavelength lasers are easily affected by the flatness of the solder mask surface and cannot penetrate the ink to directly obtain the substrate surface position. They require additional reference area calibration, resulting in low detection efficiency and insufficient accuracy. Contact thickness measurement methods can cause physical damage to the solder mask layer and are not suitable for finished product inspection.

[0005] Meanwhile, traditional testing methods only assess the quality of individual PCBs in isolation, lacking the ability to analyze the correlation of common anomalies among multiple PCBs. This makes it impossible to promptly identify systemic process problems caused by defects in production equipment (such as surface cracks on the compaction board or misalignment of the exposure machine). When equipment malfunctions, the root cause can often only be traced after a batch of defective products has accumulated, leading to a significant increase in production losses and making it difficult to meet the demands of large-scale production of high-density, high-precision PCBs.

[0006] Therefore, there is an urgent need to develop a solder mask deviation detection method that integrates precise area division, high-precision thickness detection, and systematic anomaly tracing to solve the pain points of traditional technology, such as high misjudgment rate of zoning, low thickness detection accuracy, and weak anomaly tracing capability, so as to achieve closed-loop control of the entire process of PCB solder mask process quality. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides a method for detecting solder resist deviation in printed circuit boards. This method solves the problem that solder resist thickness detection commonly uses single-wavelength laser scanning or contact thickness measurement methods, and that single-wavelength lasers are easily affected by the surface flatness of the solder resist layer.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a method for detecting solder resist deviation on a printed circuit board, comprising the following steps: Step 1: Perform machine vision scanning on the printed circuit board after the solder mask coating treatment to confirm the RGB values ​​associated with different points. Then, perform a summation and verification process on the RGB values ​​to confirm the RGB features associated with different points. Based on the RGB features, perform point partitioning planning to divide the surface area of ​​the printed circuit board into substrate areas or solder mask areas. The specific method is as follows: The printed circuit board (PCB) is visually scanned to identify the RGB values ​​associated with different points within the PCB. The RGB values ​​associated with each point are then summed to determine the total RGB value associated with that point, which is denoted as Z. i , where i represents different points; The associated total value Z i The value is compared and verified with a preset numerical range, which includes either a substrate range or a solder resist range. The endpoint values ​​of both ranges are preset values. If Z... i If it belongs to the substrate area, then mark the corresponding point as the substrate point. If Z i If it belongs to the solder mask area, then mark the corresponding point as a solder mask point. If Z i If it does not belong to the substrate area or the solder mask area, then no marking is required; The area covered by several substrate points is designated as the substrate area, and the area covered by several solder resist points is designated as the solder resist area. Step 2: Perform laser displacement scanning on the marked solder resist area. Use lasers of different wavelengths to perform the scanning process and record the bounce position of each reflection line. Lock the reflection lines of different wavelengths at the same reflection position, and based on the angular characteristics, lock the solder resist reflection points and the substrate reflection points. Then confirm the solder resist thickness at the corresponding points. The specific method is as follows: The solder resist area is laser displacement scanning using visible light with a wavelength of 550nm. For different points within the solder resist area, the emission angle A1 of the emitted laser is gradually adjusted so that the laser reflection point falls precisely on the corresponding point. The reflection line associated with the emitted laser is then confirmed, and the landing point of the reflection line is confirmed on the receiving panel. The angle A2 between the reflection line and the receiving panel is also confirmed. Points where A1 and A2 are exactly the same are recorded as flat points. Otherwise, no calibration is performed. For several sets of flat points marked in the solder resist area, a laser displacement scanning process is performed on the substrate area below the flat points using near-infrared light with a wavelength of 1350nm. The reflection point of this near-infrared light is prioritized to fall on the flat points, and then the emission angle of this near-infrared light is gradually expanded downward. The reflection point of the near-infrared light is recorded in real time. The process stops when the recorded reflection point and the reflection point of visible light on the flat points are at the same point. The current state is recorded as the standard state. Identify the emission angle F1 of near-infrared light and the emission angle F2 of visible light in the standard state, and then confirm the corresponding reflection line and the receiving angles J1 and J2 generated by the receiving panel, where J1 is the reflection line of near-infrared light and J2 is the reflection line of visible light. Based on the emission angle F1 and the receiving angle J1 of near-infrared light, generate the substrate reflection point 1 of near-infrared light, and then simultaneously confirm the solder resist reflection point 2 of visible light. Identify the straight-line distance between the substrate reflection point 1 and the solder resist reflection point 2, and use this straight-line distance as the solder resist thickness of the corresponding point. Confirm the solder resist thickness involved in each flat point in turn. Step 3: Based on the different solder resist thicknesses associated with different points within the solder resist area, identify and mark the abnormal features associated with different solder resist areas, and then display the marked processed images. The specific method is as follows: Identify the total number of flat points in different solder resist areas. If the total number is 1, the solder resist thickness associated with the corresponding flat point is recorded as the regional feature of the current solder resist area. If the total number exceeds 1, the average value of the multiple sets of solder resist thicknesses associated with multiple sets of flat points in the corresponding area is processed, and the average value obtained is used as the regional feature associated with the corresponding area. If the area feature is <5um, the corresponding solder mask area is marked as a thin area; if the area feature is >8um, the corresponding solder mask area is marked as a thick area. Solder mask areas with area features between 5-8um are not marked. If there are no flat points within a certain solder resist area, then this solder resist area is marked as an area to be determined.

[0009] Preferred options also include: Step 4: Perform feature verification on printed circuit boards with abnormal features, identify the abnormal vectors related to the abnormal features within the printed circuit board, and synchronously compare the identified abnormal vectors of different printed circuit boards to identify whether the abnormal features associated with several printed circuit boards have similar features. If they exist, the signal is directly displayed; if they do not exist, no processing is required. Identify the center point of the printed circuit board and record it as the main point. Then, based on the abnormal features marked on the surface of the corresponding printed circuit board, identify the associated solder mask area and identify the center point of this solder mask area and record it as the secondary point. Using the main point as the starting point and the secondary point as the ending point, generate an abnormal vector associated with the corresponding solder mask area, and then identify the multiple sets of abnormal vectors existing in the printed circuit board in sequence. Simultaneously, the abnormal vectors of other printed circuit boards with abnormal characteristics are confirmed to identify whether there are similar vectors between the abnormal vectors located on different printed circuit boards: a set of circles with a radius of 1um is constructed with the end point of a certain set of abnormal vectors as the center. If the end point of other abnormal vectors is located inside the circle constructed by this abnormal vector, it is recorded as a similar vector. If it exists, the two sets of printed circuit boards are marked as undetermined circuit boards; otherwise, no marking is made. The abnormal vectors existing in the circuit board to be determined are checked and compared in turn. The similar vectors existing in the circuit board are marked in turn, and the total number of similar vectors QG is recorded. Then, the total number of abnormal vectors in a single circuit board to be determined is confirmed. The proportion of similar vectors associated with a single circuit board to be determined is confirmed by QG÷ZG=ZB. The proportions generated by the two circuit boards to be determined are averaged. If the average value obtained by the processing exceeds 85%, a similar signal is generated and displayed. Otherwise, no marking is performed.

[0010] This invention provides a method for detecting solder mask deviation on printed circuit boards. Compared with existing technologies, it has the following advantages: By adopting the RGB total value partitioning method, the substrate area and solder mask area are quickly defined by preset value ranges, realizing the automatic and accurate division of the two types of areas. This effectively avoids the subjective errors and missed judgments of traditional manual partitioning or single visual recognition. At the same time, non-target area points are not marked to reduce invalid data interference and ensure that subsequent laser scanning focuses only on the solder mask area, which greatly improves detection efficiency and targeting, especially suitable for the rapid partitioning needs in mass production scenarios. A dual-wavelength laser layer detection scheme is adopted, which utilizes the characteristic that visible light cannot penetrate the solder resist film while near-infrared light can. Combined with an angle verification mechanism for flat points, the solder resist reflection points and substrate reflection points are accurately located, enabling direct measurement of solder resist thickness. This method does not rely on an additional reference area, avoiding the drawbacks of traditional single-wavelength laser detection being affected by surface reflection and ink color. Furthermore, angle matching ensures that thickness calculation is performed only on flat points, further improving the accuracy of thickness detection and providing reliable data support for the accurate determination of thin and thick areas. By processing the average thickness of flat points within a region to generate regional features, the system can automatically mark thin and thick solder mask areas. At the same time, areas without flat points are marked as pending areas, balancing the automation needs of batch inspection with manual verification of special areas to avoid missing anomalies. In addition, by constructing anomaly vectors and verifying similarity, the system can accurately identify common anomalies among multiple circuit boards, quickly pinpointing systemic problems caused by production equipment rather than random errors of a single circuit board, providing a clear direction for quality traceability. Attached Figure Description

[0011] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0012] 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.

[0013] First Embodiment Please see Figure 1 This application provides a method for detecting solder mask deviation on printed circuit boards, comprising the following steps: Step 1: Perform machine vision scanning on the printed circuit board after the solder mask coating process to identify the RGB values ​​associated with different points. Then, perform a summation verification on the RGB values ​​to identify the RGB features associated with different points. Based on the RGB features, perform point partitioning planning to divide the surface area of ​​the printed circuit board into substrate areas or solder mask areas. Specifically, in the corresponding printed circuit board image acquisition process, different points have different RGB values. Based on the specific numerical processing process, the substrate area and coating area within the corresponding circuit board can be effectively identified. The coating area is the corresponding solder mask area. In the coating process, it is necessary to identify the corresponding total features based on the RGB features associated with the points and then perform effective partitioning planning to facilitate subsequent laser scanning processing. The specific methods for dividing a printed circuit board into a substrate area or a solder mask area are as follows: The printed circuit board (PCB) is visually scanned to identify the RGB values ​​associated with different points within the PCB. The RGB values ​​associated with each point are then summed to determine the total RGB value associated with that point, which is denoted as Z. i , where i represents different points; The associated total value Z iThe values ​​are compared and verified against preset ranges. For the solder mask area, the associated total value range is generally between 100 and 300; for the substrate area, the associated total value range is generally between 600 and 765. The value range includes either the substrate area or the solder mask area, and the endpoint values ​​of both ranges are preset values, determined in advance by relevant personnel based on experience. If Z... i If it belongs to the substrate area, then mark the corresponding point as the substrate point. If Z i If it belongs to the solder mask area, then mark the corresponding point as a solder mask point. If Z i If it does not belong to the substrate area or solder mask area, no marking will be made (that is, during the scanning process, the points in other areas may also be scanned, which will make it impossible to quickly and effectively confirm the range to which the corresponding value belongs when the corresponding range is determined). The area covered by several substrate points is marked as the substrate area, and the area covered by several solder mask points is marked as the solder mask area (that is, the relevant area connected by several adjacent points is the corresponding substrate area or solder mask area, which facilitates subsequent laser scanning processing to identify the corresponding specific partition). Step 2: Perform laser displacement scanning on the marked solder resist area. Use lasers of different wavelengths to perform the scanning process and record the bounce position of each reflection line. Lock the reflection lines of different wavelengths at the same reflection position. Based on the angle characteristics, lock the solder resist reflection points and the substrate reflection points. Then confirm the solder resist thickness at the corresponding points. Specifically, for the solder resist area, there is a set of solder resist oil film areas. The lower end of the solder resist oil film area is the substrate area. First, use visible light of 400-760nm for laser scanning. This visible light cannot penetrate the oil film, so the resulting reflection points are on the oil film. Then use near-infrared light of 800-1700nm for laser scanning. This near-infrared light can penetrate the oil film, so the resulting reflection points are on the substrate. Based on the different reflection points generated, identify the specific associated reflection points to comprehensively confirm the location of each point. The specific method for confirming the solder mask reflection points and the substrate reflection points is as follows: Visible light with a wavelength of 550nm is used to perform laser displacement scanning on the solder mask area. For different points within the solder mask area, the emission angle A1 of the emitted laser is gradually adjusted to ensure that the laser reflection point falls precisely on the corresponding point. Then, the reflection line associated with the emitted laser is confirmed, and the landing point of the reflection line is confirmed on the receiving panel. The angle A2 formed by the reflection line and the receiving panel (where the receiving panel and the laser emission point are parallel to each other) is also confirmed. Points where A1 and A2 are exactly the same are recorded as flat points. Otherwise, no calibration is performed. Only when the corresponding plane is flat will there be two states where the angles are completely consistent. When the emitted laser is emitted, the emission plane is parallel to the plane on which the circuit board is placed. Therefore, when the emitted laser is emitted at the corresponding angle, there is a corresponding emission angle. After being reflected by the flat surface, there is a reflection line, which will generate a receiving point on the corresponding receiving plane. By following the corresponding reflection and receiving process, it is possible to effectively confirm whether the corresponding point is flat, which is convenient for subsequent confirmation of the point height. For the several sets of flat points marked in the solder mask area (there are usually many sets of flat points because after the oil film coating is completed, it is necessary to perform compaction to ensure that the thickness of all oil films is relatively consistent), a laser displacement scan is performed on the substrate area below the flat points using near-infrared light with a wavelength of 1350nm. The reflection point of this near-infrared light is prioritized to fall on the flat point, and then the emission angle of this near-infrared light is gradually increased downward (that is, the corresponding emission point is rotated downward, and the corresponding reflection point is gradually moved downward to the flat point, so that the reflection point of the near-infrared light and the flat point are on the same vertical horizontal line). The reflection point of the near-infrared light is recorded in real time. The process stops when the recorded reflection point and the reflection point of visible light on the flat point are at the same point, and the current state is recorded as the standard state. Identify the emission angle F1 of near-infrared light and the emission angle F2 of visible light (flat point) in the standard state, and then confirm the corresponding reflection line and the receiving angles J1 and J2 generated by the receiving panel, where J1 is the reflection line of near-infrared light and J2 is the reflection line of visible light. Based on the emission angle F1 and the receiving angle J1 of near-infrared light, generate the substrate reflection point 1 of near-infrared light (based on the two angles, the two sets of corresponding light rays can be identified, and thus the corresponding angles can be identified). Then, simultaneously confirm the solder resist reflection point 2 of visible light, identify the straight-line distance between the substrate reflection point 1 and the solder resist reflection point 2, and use this straight-line distance as the solder resist thickness of the corresponding point. Confirm the solder resist thickness involved in each flat point in turn. Step 3: Based on the different solder resist thicknesses associated with different points within the solder resist area, identify and mark the abnormal features associated with different solder resist areas, and then display the marked processed images. The specific method for identifying and marking the abnormal features associated with different solder mask areas is as follows: Identify the total number of flat points in different solder resist areas. If the total number is 1, the solder resist thickness associated with the corresponding flat point is recorded as the regional feature of the current solder resist area. If the total number exceeds 1, the average value of the multiple sets of solder resist thicknesses associated with multiple sets of flat points in the corresponding area is processed, and the average value obtained is used as the regional feature associated with the corresponding area. If the area feature is <5um, the corresponding solder mask area is marked as a thin area; if the area feature is >8um, the corresponding solder mask area is marked as a thick area. Solder mask areas with area features between 5-8um are not marked. If there are no flat points in a certain solder resist area, this solder resist area is marked as a pending area. When relevant personnel find a pending area on the corresponding image, they will use relevant instruments to detect it, indicating that there are no flat points in this area. Specifically, each different solder mask area has different thickness characteristics. Based on the corresponding thickness characteristics, the relevant solder mask areas are marked, and the marked areas are displayed to facilitate subsequent personnel to promptly identify specific problems in the solder mask process of the corresponding circuit board and take timely countermeasures.

[0014] Second Embodiment In this embodiment, the main focus is on situations involving multiple consecutive abnormalities. The main implementation steps are as follows: Step 4: Perform feature verification on printed circuit boards with abnormal features, identify the abnormal vectors related to the abnormal features within the printed circuit board, and synchronously compare the identified abnormal vectors of different printed circuit boards to identify whether the abnormal features associated with several printed circuit boards have similar features. If they exist, the signal is directly displayed; if they do not exist, no processing is required. The specific method for confirming anomaly vectors is as follows: Identify the center point of the printed circuit board and record it as the principal point (based on the outline of the corresponding printed circuit board and the two-dimensional coordinate system, the center point of the corresponding printed circuit board can be quickly and effectively identified). Then, based on the abnormal features marked on the surface of the corresponding printed circuit board, identify the associated solder mask area and identify the center point of this solder mask area and record it as the secondary point (combining the two-dimensional coordinate system and the coordinates of different contour points on the outline, the mean coordinates can be quickly and effectively identified, and the point where the mean coordinates are located is the location of the center point). Starting from the principal point and ending from the secondary point, generate an anomaly vector associated with the corresponding solder mask area, and confirm the multiple sets of anomaly vectors existing in the printed circuit board in sequence. Simultaneously, the abnormal vectors of other printed circuit boards with abnormal characteristics are confirmed to identify whether there are similar vectors between the abnormal vectors located on different printed circuit boards: a set of circles with a radius of 1um is constructed with the end point of a certain set of abnormal vectors as the center. If the end point of other abnormal vectors is located inside the circle constructed by this abnormal vector, it is recorded as a similar vector. If it exists, the two sets of printed circuit boards are marked as undetermined circuit boards; otherwise, no marking is made. The abnormal vectors existing in the circuit board to be determined are checked and compared in turn. The similar vectors existing in the circuit board are marked in turn, and the total number of similar vectors QG is recorded. Then, the total number of abnormal vectors in a single circuit board to be determined is confirmed. The proportion of similar vectors associated with a single circuit board to be determined is confirmed by QG÷ZG=ZB. The proportions generated by the two circuit boards to be determined are averaged. If the average value obtained by the processing exceeds 85%, a similar signal is generated and displayed. Otherwise, no marking is performed.

[0015] Specifically, based on the similar signals generated, relevant personnel need to inspect the production equipment. This is because if there are cracks or holes in a certain area of ​​the surface of the compacted plate during the production process, the oil film generated in that area cannot be effectively compacted, resulting in an oil film that is too thick or too thin. Therefore, it is necessary to promptly inspect and handle the equipment based on the corresponding similar signals and abnormal characteristics, perform timely maintenance, identify the cause of the abnormality, and avoid affecting the subsequent production process of other printed circuit boards.

[0016] 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.

[0017] 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. A method for detecting solder resist deviation on a printed circuit board, characterized in that, Includes the following steps: Step 1: Perform machine vision scanning on the printed circuit board after the solder mask coating is completed to confirm the RGB values ​​associated with different points. Then, perform summation verification on the RGB values ​​to confirm the RGB features associated with different points. Finally, based on the RGB features, perform point partitioning planning to divide the surface area of ​​the printed circuit board into a substrate area or a solder mask area. Step 2: Perform laser displacement scanning on the marked solder resist area. Use lasers of different wavelengths to perform the scanning process and record the bounce position of each reflection line. Lock the reflection lines of different wavelengths at the same reflection position. Based on the angle characteristics, lock the solder resist reflection point and the substrate reflection point, and then confirm the solder resist thickness at the corresponding point. Step 3: Based on the different solder resist thicknesses associated with different points within the solder resist area, identify and mark the abnormal features associated with different solder resist areas, and then display the marked processed images.

2. The method for detecting solder resist deviation on a printed circuit board according to claim 1, characterized in that, In step one, the specific method for dividing the surface area of ​​the printed circuit board into a substrate area or a solder mask area is as follows: The printed circuit board (PCB) is visually scanned to identify the RGB values ​​associated with different points within the PCB. The RGB values ​​associated with each point are then summed to determine the total RGB value associated with that point, which is denoted as Z. i , where i represents different points; The associated total value Z i The value is compared and verified with a preset numerical range, which includes either a substrate range or a solder resist range. The endpoint values ​​of both ranges are preset values. If Z... i If it belongs to the substrate area, then mark the corresponding point as the substrate point. If Z i If it belongs to the solder mask area, then mark the corresponding point as a solder mask point. If Z i If it does not belong to the substrate area or the solder mask area, then no marking is made; The area covered by several substrate points is designated as the substrate area, and the area covered by several solder resist points is designated as the solder resist area.

3. The method for detecting solder resist deviation on a printed circuit board according to claim 1, characterized in that, In step two, the specific method for using lasers of different wavelengths to perform the scanning process is as follows: The solder resist area is laser displacement scanning using visible light with a wavelength of 550nm. For different points within the solder resist area, the emission angle A1 of the emitted laser is gradually adjusted so that the laser reflection point falls precisely on the corresponding point. The reflection line associated with the emitted laser is then confirmed, and the landing point of the reflection line is confirmed on the receiving panel. The angle A2 between the reflection line and the receiving panel is also confirmed. Points where A1 and A2 are exactly the same are recorded as flat points. Otherwise, no calibration is performed. For the several sets of flat points marked in the solder mask area, a laser displacement scanning process is performed on the substrate area below the flat points using near-infrared light with a wavelength of 1350nm. The reflection point of this near-infrared light is prioritized to fall on the flat points, and then the emission angle of this near-infrared light is gradually expanded downward. The reflection point of the near-infrared light is recorded in real time. The process stops when the recorded reflection point and the reflection point of visible light on the flat points are at the same point. The current state is recorded as the standard state.

4. The method for detecting solder resist deviation on a printed circuit board according to claim 3, characterized in that, In step two, the specific method for locking the solder mask reflection point and the substrate reflection point is as follows: Identify the emission angle F1 of near-infrared light and the emission angle F2 of visible light in the standard state, and then confirm the corresponding reflection line and the receiving angles J1 and J2 generated by the receiving panel, where J1 is the reflection line of near-infrared light and J2 is the reflection line of visible light. Based on the emission angle F1 and the receiving angle J1 of near-infrared light, generate the substrate reflection point 1 of near-infrared light, and then simultaneously confirm the solder resist reflection point 2 of visible light. Identify the straight-line distance between the substrate reflection point 1 and the solder resist reflection point 2, and use this straight-line distance as the solder resist thickness of the corresponding point. Confirm the solder resist thickness involved in each flat point in turn.

5. The method for detecting solder resist deviation on a printed circuit board according to claim 1, characterized in that, In step three, the specific method for confirming and marking abnormal features within the solder mask area is as follows: Identify the total number of flat points in different solder resist areas. If the total number is 1, the solder resist thickness associated with the corresponding flat point is recorded as the regional feature of the current solder resist area. If the total number exceeds 1, the average value of the multiple sets of solder resist thicknesses associated with multiple sets of flat points in the corresponding area is processed, and the average value obtained is used as the regional feature associated with the corresponding area. If the area feature is <5um, the corresponding solder mask area is marked as a thin area; if the area feature is >8um, the corresponding solder mask area is marked as a thick area. Solder mask areas with area features between 5-8um are not marked. If there are no flat points within a certain solder resist area, then this solder resist area is marked as an area to be determined.

6. The method for detecting solder resist deviation on a printed circuit board according to claim 1, characterized in that, Also includes: Step 4: Perform feature verification on printed circuit boards with abnormal features, identify the abnormal vectors related to the abnormal features within the printed circuit board, and synchronously compare the identified abnormal vectors of different printed circuit boards to identify whether the abnormal features associated with several printed circuit boards have similar characteristics. If they exist, the signal is directly displayed; if they do not exist, no processing is required.

7. The method for detecting solder resist deviation on a printed circuit board according to claim 6, characterized in that, In step four, the specific method for confirming the abnormal vector is as follows: Identify the center point of the printed circuit board and record it as the main point. Then, based on the abnormal features marked on the surface of the corresponding printed circuit board, identify the associated solder mask area and identify the center point of this solder mask area and record it as the secondary point. Using the main point as the starting point and the secondary point as the ending point, generate an abnormal vector associated with the corresponding solder mask area, and then identify the multiple sets of abnormal vectors existing in the printed circuit board in sequence. Simultaneously, the abnormal vectors of other printed circuit boards with abnormal characteristics are confirmed to identify whether there are similar vectors between abnormal vectors located on different printed circuit boards: a set of circles with a radius of 1um is constructed with the end point of a certain set of abnormal vectors as the center. If the end point of other abnormal vectors is located inside the circle constructed by this abnormal vector, it is recorded as a similar vector. If it exists, the two sets of printed circuit boards are marked as undetermined circuit boards; otherwise, no marking is made.

8. The method for detecting solder resist deviation on a printed circuit board according to claim 7, characterized in that, In step four, the specific method for identifying abnormal features is as follows: The abnormal vectors existing in the circuit board to be determined are checked and compared in turn. The similar vectors existing in the circuit board are marked in turn, and the total number of similar vectors QG is recorded. Then, the total number of abnormal vectors in a single circuit board to be determined is confirmed. The proportion of similar vectors associated with a single circuit board to be determined is confirmed by QG÷ZG=ZB. The proportions generated by the two circuit boards to be determined are averaged. If the average value obtained by the processing exceeds 85%, a similar signal is generated and displayed. Otherwise, no marking is performed.

Images