Circuit board solder paste detection machine and detection method thereof

Abstract

The invention discloses a circuit board solder paste detector and a detection method thereof, and belongs to the field of circuit board processing quality detection, and the circuit board solder paste detector comprises a defect identification workbench, a solder paste area detection module, a three-dimensional laser scanner and a processor. The method comprises the following steps: placing a to-be-tested circuit board on a movable support plate; acquiring a plane image of the circuit board to be detected, identifying the solder paste area of the solder paste area, and comparing the solder paste area with the solder paste standard area to obtain a difference coefficient # imgabs0 #; the three-dimensional laser scanner obtains the thickness data of the solder paste area on the circuit board to be detected, and the difference coefficient # imgabs1 # of the solder paste thickness between the solder paste standard area and the solder paste area is calculated; and evaluating the processing defect of the solder paste area on the circuit board to be detected. The circuit board solder paste detection machine can accurately identify the processing defects of the solder paste area, the solder paste processing quality is detected from the perspective of multi-dimensional data, and the detection of the solder paste processing quality has enough accuracy and reliability.

Description

Technical Field

[0001] The present invention relates to the field of quality inspection in circuit board production, and particularly to a solder paste inspection machine for circuit boards and its inspection method. Background Art

[0002] In the process of assembling electronic components, solder paste is a key material, and its quality directly affects the welding quality and reliability of printed circuit boards (PCBs). Therefore, in the process of circuit board production, the quality inspection of solder paste is very important and is also a key link in circuit board quality assessment. Traditional solder paste inspection methods, such as manual visual inspection or sensors, are susceptible to changes in environmental conditions and equipment aging, have insufficient detection accuracy, and low detection efficiency. Therefore, there is an urgent need to develop a high-precision and high-efficiency solder paste inspection technology. Summary of the Invention

[0003] Aiming at the above deficiencies in the prior art, the present invention provides a solder paste inspection machine for circuit boards and its inspection method, which combines image recognition technology and three-dimensional laser technology to realize the detection and evaluation of solder paste processing quality defects in the solder paste area on the circuit board.

[0004] In order to achieve the above invention purpose, the technical solution adopted by the present invention is as follows: Provide a solder paste inspection machine for circuit boards, which includes: A defect recognition workbench, above which a solder paste area detection module and a three-dimensional laser scanner are arranged. A circuit board conveying mechanism is arranged on the defect recognition workbench for sequentially conveying the circuit board to be tested under the solder paste area detection module and the three-dimensional laser scanner; The solder paste area detection module includes an image acquisition probe and an image processing module. The image acquisition probe is used to acquire a planar image of the surface of the circuit board to be tested and send it to the image processing module; the image processing module is used to process the planar image to identify the solder paste area and the area of the solder paste area in the planar image; A three-dimensional laser scanner, which uses the laser scanning component on the three-dimensional laser scanner to measure the solder paste thickness data of the corresponding solder paste area on the circuit board; A processor, the image processing module is built in the processor, and both the image acquisition probe and the three-dimensional laser scanner are electrically connected to the processor.

[0005] Further, the circuit board conveying mechanism includes two single-slide linear movement modules arranged on both sides of the defect recognition workbench. A moving support plate is arranged between the sliders of the two single-slide linear movement modules through a support rod; On both sides of the movable support plate, there are double-slider linear motion modules. Between the two sliders of the double-slider linear motion modules on both sides, there are two parallel fixed bars. The two fixed bars move towards or away from each other, and on the opposite side surfaces of the two fixed bars, there are strip-shaped notches for placing the circuit board to be tested; the drive motors of the single-slider linear motion module and the double-slider linear motion module are both electrically connected to the controller.

[0006] Further, a first travel switch and a second travel switch are respectively arranged on the solder paste detection area and the three-dimensional laser scanning area of the defect recognition workbench. The movable support plate triggers the first travel switch and the second travel switch during the movement process. An optical switch for detecting whether there is a circuit board to be tested is arranged on the movable support plate. The first travel switch, the second travel switch, and the optical switch are all electrically connected to the controller.

[0007] Provide a method for detecting solder paste on a circuit board, which is applied to the above-mentioned circuit board solder paste detector, and it includes: Step S1: Place the circuit board to be tested on the movable support plate. After the optical switch is triggered to detect the circuit board to be tested, the double-slider linear motion modules on both sides drive the two fixed bars to move towards each other and clamp the circuit board to be tested; Step S2: Then, the two single-slider linear motion modules drive the movable support plate to move to the solder paste detection area of the defect recognition workbench and trigger the first travel switch. The single-slider linear motion module stops working, and the image acquisition probe works to obtain the planar image of the circuit board to be tested and send the planar image to the image processing module; Step S3: The image processing module identifies the solder paste area on the planar image, obtains the solder paste area of the solder paste area, and compares it with the solder paste standard area to obtain the difference coefficient of the solder paste area between the solder paste area and the solder paste standard area ; Step S4: After identifying the solder paste area of the circuit board to be tested, the two single-slider linear motion modules continue to drive the movable support plate to move to the three-dimensional laser scanning area, trigger the second travel switch, and the three-dimensional laser scanner obtains the thickness data of the solder paste area on the circuit board to be tested, and calculates the difference coefficient of the solder paste thickness between the solder paste standard area and the solder paste area ; Step S5: Use the difference coefficient and the difference coefficient to evaluate the processing defects of the solder paste area on the circuit board to be tested.

[0008] Further, step S3 includes: Step S31: Establish a three-dimensional model of the circuit board according to the size parameters of the circuit board to be tested, perform gray-scale processing on the planar image to obtain the gray-scale image of the planar image, and screen the solder paste area according to the gray-scale value of each pixel in the gray-scale image screen the solder paste area,n is the pixel number; Step S32: Calculate the difference coefficient of the solder paste area with respect to the solder paste standard area based on the pixel coordinates within the solder paste area .

[0009] Furthermore, step S31 includes: Step S311: Set the standard gray value of the pixels in the solder paste area , calculate the difference between the gray value and the standard gray value , and compare the difference with the preset gray error tolerance ; If , then determine that the pixel n is a non-solder paste pixel; if , then determine that the pixel n is a solder paste pixel; Step S312: After obtaining all the solder paste pixels in the gray image, based on the pixel coordinates of each solder paste pixel, select a reference pixel from any solder paste pixel, and filter the pixel set of the continuous solder paste area around the reference pixel ; Step S313: Obtain the data set N of the pixel sets of [[ID= forty]] solder paste areas is the pixel set of the [[ID= forty-three]] N th solder paste area selected, and traverse the pixel coordinates of the solder paste pixels in each pixel set to calculate the repetition rate of the solder paste pixels in any two pixel sets Step S314: Set the threshold of the repetition rate ; If , then determine that the solder paste area corresponding to the pixel set overlaps with the solder paste area corresponding to the pixel set , merge the solder paste pixels within the pixel set and the pixel set , and delete the solder paste pixels with repeated pixel coordinates to form a new pixel set; the solder paste pixels with repeated pixel coordinates satisfy the constraint condition: ; If , then determine that the solder paste area corresponding to the pixel set does not overlap with the solder paste area corresponding to the pixel set , the pixel set With the pixel set Do not merge; Step S315: After the merging of the pixel sets in the data set inside N through Step S24, a new data set is formed , B is the number of pixel sets reduced after merging; The solder paste pixels of each pixel set in the data set form independent solder paste regions.

[0010] Furthermore, Step S32 includes: Step S321: Calculate the center coordinates of the solder paste region corresponding to the pixel set according to the pixel coordinates inside the solder paste region ; ; Step S322: Mesh the surface of the circuit board 3D model, and the size of the mesh is the same as the size of the pixels. Calculate the center coordinates of the solder paste standard region according to the coordinates of the meshes inside the solder paste standard region designed on the circuit board 3D model ; Step S323: Align the center coordinates of the solder paste standard region with the center coordinates of the solder paste region, and screen the non-corresponding pixel regions between the aligned solder paste region and the solder paste standard region , and generate the pixel set of the non-corresponding pixel region ; Step S324: Calculate the difference coefficient between the solder paste standard region and the solder paste region with respect to the solder paste area according to the number q of the solder paste pixels inside the pixel set ; .

[0011] Furthermore, the method for calculating the difference coefficient between the solder paste standard region and the solder paste region with respect to the solder paste thickness in Step S4 includes: Step S41: Use the thickness data of the solder paste region obtained by scanning the surface of the circuit board to be measured with a 3D laser scanner to obtain the 3D point cloud data of the surface of the circuit board to be measured; Extract the 3D point cloud data inside the solder paste region , and calculate the solder paste thickness of the solder paste region ; Step S42: Use the solder paste thickness of the solder paste region The solder paste thickness in the standard area of the solder paste Calculate the difference coefficient of the solder paste thickness between the standard area and the solder paste area of the solder paste 。

[0012] Further, step S5 includes: Calculate the solder paste area The detected error coefficient Compare the error coefficient with the threshold of the error coefficient , and , are the influence weight coefficients of the solder paste area and the solder paste thickness on the solder paste defect respectively; If , it is determined that the solder paste quality in the solder paste area on the circuit board to be tested does not meet the requirements and there are processing defects; if , it is determined that the solder paste quality in the solder paste area on the circuit board to be tested meets the requirements and there are no processing defects.

[0013] The beneficial effects of the present invention are as follows: The present invention uses image recognition technology to obtain the solder paste area and the area data of the solder paste area on the circuit board to be tested, and conducts a standardized comparison to obtain the difference in the solder paste area on the solder paste area; at the same time, the thickness data of the solder paste area on the circuit board to be tested is used to obtain the difference in the solder paste thickness on the solder paste area; then, the solder paste quality on each solder paste area is comprehensively evaluated by combining the differences between the two, so as to know the processing defects of the solder paste on the circuit board, providing data support for the subsequent solder paste processing technology. The present invention can accurately identify the processing defects in the solder paste area by using the circuit board solder paste detector, detect the solder paste processing quality from the perspective of multi-dimensional data, and the detection of the solder paste processing quality has sufficient accuracy and reliability. Description of the Drawings

[0014] Figure 1 It is a top view of the defect recognition workbench.

[0015] Figure 2 It is a front view of the defect recognition workbench.

[0016] Figure 3 It is a structural diagram of the fixing bar.

[0017] Figure 4 It is a flowchart of the circuit board solder paste detection method.

[0018] ​Among them, 1. Defect recognition workbench, 2. Single-sliding-table linear movement module, 3. First travel switch, 4. Second travel switch, 5. Double-sliding-table linear movement module, 6. Moving support plate, 7. Photoelectric switch, 8. Fixed strip, 9. Support rod, 10. Strip-shaped notch, 11. Image acquisition probe, 12. 3D laser scanner. Specific embodiments

[0019] The following describes the specific embodiments of the present invention to facilitate those skilled in the art of this technology to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art of this technology, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are within the scope of protection.

[0020] As Figures 1 - 3 shown, a solder paste inspection machine for circuit boards includes: A defect recognition workbench 1, above which a solder paste area detection module and a 3D laser scanner 12 are provided. A circuit board conveying mechanism is provided on the defect recognition workbench 1 for sequentially conveying the circuit board to be tested below the solder paste area detection module and the 3D laser scanner 12; The solder paste area detection module includes an image acquisition probe 11 and an image processing module. The image acquisition probe 11 is used to acquire the planar image of the surface of the circuit board to be tested and send it to the image processing module; the image processing module is used to process the planar image to identify the solder paste area in the planar image and the area of the solder paste area; A 3D laser scanner 12, which uses the laser scanning component on the 3D laser scanner 12 to measure the solder paste thickness data of the corresponding solder paste area on the circuit board; A processor, the image processing module is built in the processor, and both the image acquisition probe 11 and the 3D laser scanner 12 are electrically connected to the processor.

[0021] In this embodiment, the circuit board conveying mechanism includes two single-sliding-table linear movement modules 2 arranged on both sides of the defect recognition workbench 1. A moving support plate 6 is provided between the sliders of the two single-sliding-table linear movement modules 2 through a support rod 9.

[0022] Double-sliding-table linear movement modules 5 are provided on both sides of the moving support plate 6. Two parallel fixed strips 8 are provided between the two sliders of the double-sliding-table linear movement modules 5 on both sides. The two fixed strips 8 move towards or away from each other, and strip-shaped notches 10 for placing the circuit board to be tested are provided on the opposite sides of the two fixed strips 8; the drive motors of the single-sliding-table linear movement modules 2 and the double-sliding-table linear movement modules 5 are electrically connected to the controller.

[0023] The first travel switch 3 and the second travel switch 4 are respectively provided in the solder paste detection area and the three-dimensional laser scanning area of the defect identification workbench 1. The movable support plate 6 triggers the first travel switch 3 and the second travel switch 4 during the movement. The movable support plate 6 is provided with a photoelectric switch 7 for detecting whether there is a circuit board to be tested. The first travel switch 3, the second travel switch 4 and the photoelectric switch 7 are all electrically connected to the controller.

[0024] like Figure 4 As shown, a circuit board solder paste detection method is applied to the above-mentioned circuit board solder paste detection machine, comprising: Step S1: Place the circuit board to be tested on the movable support plate 6. After the photoelectric switch 7 is triggered to detect the circuit board to be tested, the double-slide linear moving modules 5 on both sides drive the two fixing bars 8 to move toward each other to clamp the circuit board to be tested.

[0025] Step S2: Afterwards, the two single-slide linear moving modules 2 drive the movable support plate 6 to move to the solder paste detection area of the defect identification workbench 1, and trigger the first travel switch 3. The single-slide linear moving module 2 stops working, and the image acquisition probe 11 works to obtain a planar image of the circuit board to be tested, and sends the planar image to the image processing module.

[0026] Step S3: The image processing module identifies the solder paste area on the plane image, obtains the solder paste area of the solder paste area, and compares it with the solder paste standard area to obtain the difference coefficient between the solder paste area and the solder paste standard area. Step S3 specifically includes: Step S31: Establish a three-dimensional model of the circuit board according to the size parameters of the circuit board to be tested, grayscale the plane image to obtain a grayscale image of the plane image, and calculate the grayscale value of each pixel in the grayscale image. Screening solder paste area, n is the pixel number. Step S31 specifically includes: Step S311: Setting the standard grayscale value of the solder paste area pixels , calculate the gray value With standard gray value The difference between , and the difference Grayscale error tolerance Make comparisons; like , then determine the pixel n For non-solder paste pixels; if , then pixel n is determined to be a solder paste pixel; Step S312: After obtaining all solder paste pixels in the grayscale image, based on the pixel coordinates of each solder paste pixel, take any solder paste pixel as a reference pixel and filter the pixel set of the continuous solder paste area around the reference pixel. , the solder paste pixels within the pixel set satisfy the constraint conditions: a ; ; Among them, are the pixel coordinates of two adjacent solder paste pixels in the solder paste area, are the gray values of two adjacent solder paste pixels in the solder paste area, d is the distance between two adjacent solder paste pixels, is the error threshold between solder paste pixels; If each solder paste pixel is used as a reference pixel, each solder paste pixel can form a pixel set. However, since there are multiple solder paste pixels in a solder paste area, there will be a large number of repetitions of the solder paste pixels in the pixel set. Therefore, according to the repetition situation of the pixels in the pixel set, the pixel sets of the same solder paste area can be screened out and merged to ensure that there is only one pixel set for a solder paste area.

[0027] Step S313: Obtain the data set of the pixel sets of N solder paste areas , is the pixel set of the N th solder paste area screened out, and traverse the pixel coordinates of the solder paste pixels in each pixel set to calculate the repetition rate of the solder paste pixels in any two pixel sets ; ; Among them, is the solder paste pixel number within the pixel set , is the solder paste pixel number within the pixel set , are respectively the pixel coordinates of the solder paste pixels in the pixel set , the pixel set ; u is the pixel set , the pixel set The number of repeated solder paste pixels within; indicates taking the maximum value, is the pixel set The number of solder paste pixels within; is the pixel set The number of solder paste pixels within; Step S314: Set the threshold of the repetition rate ; The repetition rate represents the probability that two pixel sets correspond to a solder paste area, and the threshold of the repetition rate Indicates the standard for judging whether two pixel sets correspond to a solder paste area, repetition rate The larger the value, the greater the probability that two pixel sets correspond to a solder paste area, and vice versa. By setting the repetition rate threshold It can avoid merging two adjacent but non-overlapping solder paste areas because the repetition rate of two adjacent but non-overlapping solder paste areas is very low; like , then determine the pixel set Corresponding solder paste area With pixel collection Corresponding solder paste area If overlap occurs, the pixel collection and pixel collection The solder paste pixels in the image are merged and the solder paste pixels with repeated pixel coordinates are deleted to form a new pixel set; the solder paste pixels with repeated pixel coordinates meet the constraints: ; like , then determine the pixel set Corresponding solder paste area With pixel collection Corresponding solder paste area No overlap occurs, pixel collection With pixel collection No merger will be conducted; Step S315: Dataset Inside N After the pixel sets are merged in step S24, a new data set is formed. , B is the number of pixel sets reduced after merging; Dataset The solder paste pixels within each pixel set form an independent solder paste area.

[0028] Step S32: Calculate the difference coefficient between the solder paste area and the solder paste standard area based on the pixel coordinates in the solder paste area. Step S32 specifically includes: Step S321: According to the pixel set Corresponding solder paste area Pixel coordinates within, calculate solder paste area The center coordinates of ; ; in, i Solder paste area The number of the inner solder paste pixels, I Solder paste area The number of inner solder paste pixels, is the solder paste area inside the i coordinates of the th solder paste pixel; ; ; Among them, U is the number of grids in the solder paste standard area, u is the grid number, is the u coordinates of the grid; In this embodiment, the grid coordinates and pixel coordinates adopt the same coordinate system, and when aligning the solder paste standard area and the solder paste area, it is also carried out in the same coordinate system.

[0029] The three-dimensional model of the circuit board is obtained according to the design size parameters of the circuit board to be measured. The surface of the three-dimensional model of the circuit board reflects the standard solder paste structure on each solder paste area, and the standard solder paste structure includes the solder paste area and the solder paste thickness characteristics. The planar image has the actual solder paste area on the circuit board to be measured, which can be used to calculate the area of the solder paste area, and the point cloud data scanned by the three-dimensional laser scanner 12 contains the thickness data of the actual solder paste area, which can be used to calculate the thickness of the solder paste in the solder paste area.

[0030] Step S323: Align the center coordinate of the solder paste standard area with the center coordinate of the solder paste area , and screen the non-corresponding pixel areas between the aligned solder paste area and the solder paste standard area, and generate a pixel set of the non-corresponding pixel areas , and the solder paste pixels in the pixel set satisfy the constraint condition: i ; Step S324: Calculate the difference coefficient of the solder paste area between the solder paste standard area and the solder paste area q according to the number of the solder paste pixels in the pixel set ; ; Among them, is the area of the grid, is the area of the solder paste standard area, is the area of the circuit board.​

[0031] Step S4: After identifying the solder paste area of the circuit board to be tested, the two single-slide linear movement modules 2 continue to drive the moving support plate 6 to move to the three-dimensional laser scanning area, trigger the second travel switch 4, and the three-dimensional laser scanner 12 obtains the thickness data of the solder paste area on the circuit board to be tested, and calculates the difference coefficient of the solder paste thickness between the solder paste standard area and the solder paste area. In step S4, the method for calculating the difference coefficient of the solder paste thickness between the solder paste standard area and the solder paste area includes: Step S41: Using the thickness data of the solder paste area obtained by scanning the surface of the circuit board to be tested with the three-dimensional laser scanner 12, the three-dimensional point cloud data of the surface of the circuit board to be tested; extracting the three-dimensional point cloud data within the solder paste area and calculating the solder paste thickness of the solder paste area ; ; ; Among them, e is the number of the three-dimensional point cloud data within the solder paste area , is the solder paste thickness corresponding to the e th three-dimensional point cloud data, E is the number of the three-dimensional point cloud data within the solder paste area ; Step S42: Using the solder paste thickness of the solder paste area and the solder paste thickness of the solder paste standard area to calculate the difference coefficient of the solder paste thickness between the solder paste standard area and the solder paste area ; .

[0032] Step S5: Using the difference coefficient and the difference coefficient to evaluate the processing defects of the solder paste area on the circuit board to be tested. Step S5 specifically includes: Calculating the error coefficient detected in the solder paste area , comparing the error coefficient with the threshold of the error coefficient , and , are the influence weight coefficients of the solder paste area and the solder paste thickness on the solder paste defects respectively; If , it is determined that the solder paste quality of the solder paste area on the circuit board to be tested does not meet the requirements and there are processing defects; if , it is determined that the solder paste area on the circuit board to be tested has solder paste quality meeting the requirements and no processing defects.

[0033] The present invention uses image recognition technology to obtain the solder paste area of the circuit board to be tested and the area data of the solder paste area, and conducts a standardized comparison to obtain the difference in solder paste area on the solder paste area; meanwhile, the difference in solder paste thickness on the solder paste area is obtained by using the thickness data of the solder paste area on the circuit board to be tested; afterwards, the solder paste quality on each solder paste area is comprehensively evaluated by combining the differences between the two, and thus the processing defects of the solder paste on the circuit board can be known, providing data support for the subsequent solder paste processing technology. The present invention can accurately identify the processing defects of the solder paste area by using a circuit board solder paste detector, detect the solder paste processing quality from the perspective of multi-dimensional data, and the detection of the solder paste processing quality has sufficient accuracy and reliability.

Claims

1. A circuit board solder paste inspection machine, characterized in that, Including: A defect recognition workbench, above which a solder paste area detection module and a three-dimensional laser scanner are arranged. On the defect recognition workbench, a circuit board conveying mechanism is arranged for sequentially conveying the circuit board to be tested under the solder paste area detection module and the three-dimensional laser scanner; The solder paste area detection module includes an image acquisition probe and an image processing module. The image acquisition probe is used to acquire a planar image of the surface of the circuit board to be tested and send it to the image processing module; the image processing module is used to process the planar image to identify the solder paste area and the area of the solder paste area in the planar image; A three-dimensional laser scanner, using a laser scanning component on the three-dimensional laser scanner to measure the solder paste thickness data corresponding to the solder paste area on the circuit board; A processor, the image processing module is built in the processor, and the image acquisition probe and the three-dimensional laser scanner are both electrically connected to the processor.

2. The solder paste inspection machine for circuit boards according to claim 1, wherein The circuit board conveying mechanism includes two single-slide linear motion modules arranged on both sides of the defect recognition workbench. A moving support plate is arranged between the sliders of the two single-slide linear motion modules through a support rod; Double-slide linear motion modules are arranged on both sides of the moving support plate. Two parallel fixed bars are arranged between the two slides of the double-slide linear motion modules on both sides. The two fixed bars move towards or away from each other, and strip-shaped notches for placing the circuit board to be tested are arranged on the opposite side surfaces of the two fixed bars; the drive motors of the single-slide linear motion modules and the double-slide linear motion modules are both electrically connected to the controller.

3. The solder paste inspection machine for circuit boards according to claim 2, characterized in that, A first travel switch and a second travel switch are respectively arranged on the solder paste detection area and the three-dimensional laser scanning area of the defect recognition workbench. The moving support plate triggers the first travel switch and the second travel switch during the moving process. An optical switch for detecting whether there is a circuit board to be tested is arranged on the moving support plate. The first travel switch, the second travel switch and the optical switch are all electrically connected to the controller.

4. A method for detecting solder paste on a circuit board, which is applied to the circuit board solder paste detector according to any one of claims 1-3, and is characterized in that, Including: Step S1: Place the circuit board to be tested on the moving support plate. After the optical switch is triggered to detect the circuit board to be tested, the double-slide linear motion modules on both sides drive the two fixed bars to move towards each other to clamp the circuit board to be tested; Step S2: Then, the two single-slide linear motion modules drive the moving support plate to move to the solder paste detection area of the defect recognition workbench and trigger the first travel switch. The single-slide linear motion module stops working, and the image acquisition probe works to obtain a planar image of the circuit board to be tested and send the planar image to the image processing module; Step S3: The image processing module identifies the solder paste area on the planar image, obtains the solder paste area of the solder paste area, and compares it with the solder paste standard area to obtain the difference coefficient of the solder paste area between the solder paste area and the solder paste standard area ; Step S4: After identifying the solder paste area of the circuit board to be tested, the two single-slide linear motion modules continue to drive the moving support plate to move to the three-dimensional laser scanning area, trigger the second travel switch, and the three-dimensional laser scanner obtains the thickness data of the solder paste area on the circuit board to be tested, and calculates the difference coefficient of the solder paste thickness between the solder paste standard area and the solder paste area ; Step S5: Use the coefficient of variation and the coefficient of variation to evaluate the processing defects in the solder paste area on the circuit board to be tested.

5. The method for detecting solder paste on a circuit board according to claim 4, wherein The step S3 includes: Step S31: Establish a three-dimensional model of the circuit board according to the size parameters of the circuit board to be measured, perform grayscale processing on the planar image to obtain a grayscale image of the planar image, and according to the grayscale value of each pixel in the grayscale image screen the solder paste area, n number the pixels; Step S32: Calculate the difference coefficient of the solder paste area between the solder paste area and the standard solder paste area based on the pixel coordinates within the solder paste area .

6. The solder paste inspection method for a circuit board according to claim 5, characterized in that, The step S31 includes: Step S311: Set the standard gray value of the solder paste area pixels , calculate the gray value and the standard gray value to obtain the difference , and compare the difference with the preset allowable gray error value ; If , then the pixel n is determined to be a non-solder paste pixel; if , then the pixel n is determined to be a solder paste pixel; Step S312: After obtaining all the solder paste pixels in the grayscale image, according to the pixel coordinates of each solder paste pixel, using any solder paste pixel as a reference pixel, filter the pixel set of the continuous solder paste area around the reference pixel ; Step S313: Obtain N a data set of pixel sets of solder paste regions , which is the pixel set of the N th solder paste region selected, and traverse the pixel coordinates of the solder paste pixels in each pixel set to calculate the repetition rate of the solder paste pixels in any two pixel sets Step S314: Set the threshold of the repetition rate ; If , it is determined that the solder paste regions corresponding to the pixel sets overlap with the solder paste regions corresponding to the pixel sets . The solder paste pixels within the pixel sets and the pixel sets are merged, and the solder paste pixels with duplicate pixel coordinates are deleted to form a new pixel set; the solder paste pixels with duplicate pixel coordinates satisfy the constraint condition: ; ;​ If , it is determined that the solder paste area corresponding to the pixel set does not overlap with the solder paste area corresponding to the pixel set , and the pixel set and the pixel set are not merged; and the pixel set are not merged; Step S315: Dataset within N After the merging of the pixel sets in step S24, a new dataset is formed , B which is the number of pixel sets reduced after merging; Dataset The solder paste pixels of each pixel set within form independent solder paste regions.

7. The method for detecting solder paste on a circuit board according to claim 6, characterized in that, The step S32 includes: Step S321: According to the pixel set corresponding solder paste area inside the pixel coordinates, calculate the center coordinates of the solder paste area ; ; Step S322: Mesh the surface of the circuit board three-dimensional model, and the size of the mesh is the same as the size of the pixel. Calculate the center coordinates of the solder paste standard area according to the coordinates of the meshes within the solder paste standard area designed on the circuit board three-dimensional model ; Step S323: Align the center coordinates of the solder paste standard area with the center coordinates of the solder paste area , and filter the non-corresponding pixel areas between the aligned solder paste area and the solder paste standard area to generate a pixel set for the non-corresponding pixel area , generating a pixel set for the non-corresponding pixel area ; Step S324: According to the pixel set the number of solder paste pixels q , calculate the difference coefficient of the solder paste area between the solder paste standard area and the solder paste area . .

8. The method for detecting solder paste on a circuit board according to claim 7, wherein In the step S4, the method for calculating the coefficient of difference regarding the solder paste thickness between the standard area of the solder paste and the solder paste area includes: Step S41: Using a three-dimensional laser scanner to scan the surface of the circuit board to be measured to obtain thickness data of the solder paste area, and three-dimensional point cloud data of the surface of the circuit board to be measured; extracting the three-dimensional point cloud data within the solder paste area and calculating the solder paste thickness of the solder paste area ; within the solder paste area ; Step S42: Use the solder paste area of the solder paste thickness and the solder paste thickness of the solder paste standard area to calculate the difference coefficient of the solder paste thickness between the solder paste standard area and the solder paste area .​ 9. The method for detecting solder paste on a circuit board according to claim 8, wherein The step S5 includes: Calculate the solder paste area Detected error coefficient , the error coefficient is compared with the threshold of the error coefficient , and , are the influence weight coefficients of the solder paste area and the solder paste thickness on the solder paste defects respectively; If , it is determined that the solder paste quality in the solder paste area of the circuit board to be tested does not meet the requirements and there are processing defects; if , it is determined that the solder paste quality in the solder paste area of the circuit board to be tested meets the requirements and there are no processing defects.

Images