Roll-to-roll high uniformity coating method for flexible circuit board reinforcements

Abstract

The application discloses a roll-to-roll high-uniformity coating method for a flexible circuit board reinforcing piece, relates to the technical field of reinforcing piece coating, and comprises the following steps: after the coating of the strip is completed, a corresponding gray-scale line scanning image of the strip coating is collected to obtain line scanning image data; adaptive image preprocessing is carried out based on the line scanning image data, and a normal gray-scale range of the strip coating is adaptively obtained to obtain normal gray-scale range information; binarization processing is carried out on the line scanning image data according to the normal gray-scale range information, and defect information in the image is obtained to obtain coating defect information; uniformity judgment is carried out on the strip coating, corresponding early warning and re-coating are carried out, and the application is used to solve the problem that the existing reinforcing piece coating technology cannot adaptively set a reliable threshold range through the line scanning image of the strip coating to reliably detect coating defects when the uniformity of the coating of the reinforcing piece on the strip is detected.

Description

Technical Field

[0001] This invention relates to the field of reinforcing component coating technology, specifically to a roll-to-roll high uniformity coating method for reinforcing components of flexible circuit boards. Background Technology

[0002] Reinforcing component coating technology is the core precision coating process in the manufacturing of flexible printed circuit board (FPC) reinforcing components. It refers to the process of continuously and precisely coating a coating system with adhesive coating as the core on a designated surface of the FPC reinforcing component substrate using specialized coating equipment and processes. This system can be overlaid with coatings that provide functions such as thermal conductivity, shielding, bending resistance, and temperature resistance as needed. After testing, gradient drying, and post-treatment, the coating and substrate are firmly bonded together, ultimately giving the reinforcing component reliable adhesion to the FPC board and other core functions.

[0003] Existing reinforcement coating technologies often rely on manual observation to detect coating defects when inspecting the uniformity of reinforcement coatings on strips; or they rely on acquiring grayscale images of the coating and detecting defects based on a fixed grayscale threshold. However, most core defects in reinforcement coatings are micro-defects, such as pinholes and microbubbles, below the minimum resolution limit of the human eye and completely undetectable by manual observation. Only macroscopic defects such as streaks and large-area pinholes can be detected, resulting in insufficient detection accuracy and real-time performance. Furthermore, defect judgment relies entirely on the experience and subjective judgment of quality inspectors, leading to inconsistent judgment standards and poor detection consistency. Detecting coating defects based on a fixed grayscale threshold requires ensuring the accuracy of the acquired grayscale images. While the uniformity of the external environment is important, the grayscale image background of the reinforcing component coating can experience dynamic grayscale fluctuations due to factors such as light source brightness attenuation, flicker, coating speed fluctuations, slight changes in coating thickness, and slight camera drift. Fixed thresholds, being statically set, lack adaptive adjustment capabilities. If the threshold is set too strictly, normal pixels with background fluctuations may be misjudged as defects; if the threshold is set too loosely, defect pixels with small grayscale differences may be missed, leading to frequent misjudgments and missed detections. Therefore, existing reinforcing component coating technologies cannot reliably detect coating defects by adaptively setting a reliable threshold range based on the row scan image of the strip coating when performing coating uniformity detection on reinforcing components on the strip. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in the prior art. After the strip is coated, a grayscale line scan image of the corresponding strip coating is acquired to obtain line scan image data. Adaptive image preprocessing is then performed to adaptively obtain the normal grayscale range of the strip coating, yielding normal grayscale range information. The line scan image data is then binarized to obtain defect information in the image, resulting in coating defect information. Furthermore, the uniformity of the strip coating is assessed, and corresponding warnings and recoating are initiated. This addresses the problem in existing reinforcing component coating technologies where, when detecting coating uniformity on reinforcing components on strips, a reliable threshold range cannot be adaptively set using the line scan image of the strip coating to reliably detect coating defects.

[0005] To achieve the above objectives, this application provides a roll-to-roll high uniformity coating method for flexible circuit board reinforcement, comprising the following steps:

[0006] After the strip is coated, a grayscale line scan image of the corresponding strip coating is acquired to obtain line scan image data.

[0007] Adaptive image preprocessing is performed based on line scan image data, and the normal grayscale range of the strip coating is adaptively obtained to obtain normal grayscale range information;

[0008] The line scan image data is binarized based on the normal grayscale range information, and the defect information in the image is obtained to obtain the coating defect information.

[0009] The uniformity of the strip coating is judged based on the coating defect information, and corresponding warnings and recoating are issued.

[0010] Further, after the strip has completed the coating process, a grayscale line scan image of the corresponding strip coating is acquired to obtain the line scan image data, which includes the following sub-steps:

[0011] The strip of the flexible circuit board reinforcement is referred to as the reinforcement strip, the coating applied to the reinforcement strip is referred to as the strip coating, and the device for applying the strip coating is referred to as the coating device.

[0012] At the outlet side of the coating device, a line scan camera is installed, referred to as the line scanning device, and the line scanning device acquires images of the reinforcing strip with the strip coating; any one line of grayscale image acquired by the line scanning device is recorded as a line scan image.

[0013] Furthermore, after the strip has completed the coating process, acquiring the corresponding grayscale line scan image of the strip coating to obtain the line scan image data also includes the following sub-steps:

[0014] The line scanning device is set to line frequency F0. Line scan images of the reinforcing strip after the strip coating is completed are continuously acquired based on F0 and recorded as line scan image data.

[0015] The currently acquired line scan image is denoted as the current line scan image. The line scan images acquired k1 times before the current line scan image are obtained and merged with the current line scan image, and denoted as the current line scan image data, where k1 is the set number.

[0016] Furthermore, adaptive image preprocessing is performed based on the line scan image data, and the normal grayscale range of the strip coating is adaptively obtained. The normal grayscale range information includes the following sub-steps:

[0017] Based on the current line scan image data, all line scan images are stitched together in the acquisition order to form the corresponding image, which is recorded as the initial scan image; the pixels of each row of the initial scan image are recorded as basic scan row 1 to basic scan row k2, where k2=k1+1.

[0018] Set the window size of the one-dimensional median filter to 3×1, perform one-dimensional median filtering on the basic scan line 1, and repeat the one-dimensional median filtering on all scan lines in the initial scan image in turn. After completion, the first scan image is obtained.

[0019] Remove the k3 columns of pixels on the left and right sides of the first scan image to obtain the second scan image. Then, label the basic scan rows 1 to k2 in the second scan image as the filtered scan rows 1 to k2, where k3 is the set column number.

[0020] Furthermore, adaptive image preprocessing based on line scan image data and adaptive acquisition of the normal grayscale range of the strip coating, obtaining normal grayscale range information, also includes the following sub-steps:

[0021] For filter scan line 1, take the k4 nearest filter scan lines and filter scan line 1 to form the neighbor processing window of filter scan line 1, where k4 is the set number;

[0022] Obtain the median of the gray values ​​of all pixels in the filtered scan row 1, denoted as the gray median, and calculate the corresponding absolute deviation of the median, denoted as the median deviation MA.

[0023] Pixels with an absolute deviation of no more than 2×MA in filter scan line 1 are recorded as background pixels; background pixels in all filter scan lines in all adjacent processing windows are repeatedly obtained, and the median gray value MG0 of all background pixels is obtained; the median gray value MG1 of background pixels in filter scan line 1 is obtained.

[0024] Add (MG0-MG1) to the gray values ​​of all pixels in filter scan line 1; repeat the process for all pixels in all filter scan lines of the second scan image to obtain the third scan image.

[0025] Furthermore, adaptive image preprocessing based on line scan image data and adaptive acquisition of the normal grayscale range of the strip coating, obtaining normal grayscale range information, also includes the following sub-steps:

[0026] Obtain all background pixels in all filtered scan rows of the second scan image, and obtain the median gray value of all background pixels in the third scan image, denoted as the background median AE;

[0027] Let any pixel in the third scanned image be the first pixel, and let the gray value of the first pixel be denoted as AF; get the pixels in the 8-neighborhood of the first pixel, and let them be denoted as the neighboring pixels; get the median BE of the gray values ​​of the neighboring pixels, and calculate the corresponding median absolute deviation, denoted as MF;

[0028] For any neighboring pixel, if the absolute deviation of the corresponding gray value from BE is not less than 2×MF, then it is marked as the corresponding neighboring background pixel; if there are no less than k5 neighboring background pixels in the 8-neighborhood of the first pixel, then calculate AF-(BE-AE) and replace AF, where k5 is the set number.

[0029] Repeat the process on all pixels in the second scan image to obtain the baseline scan image.

[0030] Furthermore, adaptive image preprocessing based on line scan image data and adaptive acquisition of the normal grayscale range of the strip coating, obtaining normal grayscale range information, also includes the following sub-steps:

[0031] Obtain the median AE of the gray values ​​of all background pixels in the reference scan image, denoted as the reference median WE; calculate the corresponding absolute deviation of the median, denoted as the reference median deviation WR;

[0032] For any pixel in the reference scan image, if the absolute difference between the corresponding gray value and WE is no greater than 2×WR, it is marked as a candidate background pixel.

[0033] For any candidate background pixel, obtain the number of corresponding neighboring background pixels in the 8-neighborhood, denoted as AN; if AN / 9 is not less than k6, then mark it as a coating pixel, where k6 is the set proportional threshold.

[0034] Further, perform adaptive image preprocessing based on the line-scanned image data, and adaptively obtain the normal gray level range of the strip coating. Obtaining the normal gray level range information further includes the following sub-steps:

[0035] Obtain the median VM of the gray values of all coating pixels and the corresponding median absolute deviation VA; set the threshold coefficient as e0; if VA ≤ r1, then let e0 = 2.5; if r1 < VA ≤ r2, then let e0 = 3.0; if r2 < VA ≤ 2×r2, then let e0 = 3.2; if 2×r2 < VA, give an alarm; where r1 and r2 are set thresholds;

[0036] If VA ≤ 2×r2, then record [VM - e0×VA, VM + e0×VA] as the coating gray level range, which is recorded as the normal gray level range information.

[0037] Further, perform binarization processing on the line-scanned image data according to the normal gray level range information, and obtain the defect information in the image. Obtaining the coating defect information includes the following sub-steps:

[0038] Record the pixel points in the reference scanned image whose gray values are not within the coating gray level range as defect pixel points, and record the pixel points whose gray values are within the coating gray level range as coating pixel points;

[0039] Perform binarization processing on the reference scanned image according to the defect pixel points and coating pixel points to obtain a coating binary image, and obtain the connected domain composed of defect pixel points in the coating binary image, which is recorded as the defect region;

[0040] Record any one defect region as the first region, obtain the number of defect pixel points in the first region, which is recorded as the pixel number AX; and repeat to obtain the pixel numbers of all defect regions, which is recorded as the coating defect information.

[0041] Further, perform uniformity judgment on the strip coating according to the coating defect information, and perform corresponding alarm and recoating, including the following sub-steps:

[0042] For the first region, if AX is less than k7, then judge the first region as a false defect region, otherwise judge the first region as a true defect region, where k7 is the set number threshold; repeat the judgment for all defect regions;

[0043] Record the reinforcing strip area corresponding to the reference scanned image as the current detection area; if there is a true defect region in the reference scanned image, then the uniformity of the strip coating in the current detection area is unqualified, give an alarm, and recoat the strip coating on the current detection area again.

[0044] The beneficial effects of this invention are as follows: After the strip is coated, the invention acquires grayscale line scan images of the corresponding strip coating to obtain line scan image data; adaptive image preprocessing is performed based on the line scan image data, and the normal grayscale range of the strip coating is adaptively obtained to obtain normal grayscale range information; binarization processing is performed on the line scan image data according to the normal grayscale range information, and defect information in the image is obtained to obtain coating defect information; uniformity judgment of the strip coating is performed based on the coating defect information, and corresponding early warning and recoating are performed; when detecting the coating uniformity of the reinforcing parts on the strip, a reliable threshold range can be adaptively set through the line scan image of the strip coating to reliably detect coating defects and improve the uniformity of the coating of the produced reinforcing parts;

[0045] This invention effectively eliminates inter-row brightness drift caused by lens position, edge shadows, or strip vibration by stitching together row scan images, removing edge columns from the left and right, and correcting the grayscale value of the current row using the background median of neighboring rows. This ensures that subsequent judgments are based on the true grayscale of the coating rather than equipment bias. The invention calculates the median and absolute deviation of the median based on the selected coating pixels, and then adaptively selects a threshold coefficient according to the magnitude of the absolute deviation. This adapts to the grayscale characteristics under different batches, materials, and lighting conditions, avoiding manual parameter tuning batch by batch and improving the reliability of defect identification. Pixels that are close to the benchmark median are recorded as candidate background pixels, and the proportion of background pixels in their 8-neighborhood is used to determine whether they are identified as coating pixels. This combines coating pixel determination with neighborhood consistency, reducing misjudgments caused by isolated noise. Attached Figure Description

[0046] Figure 1 This is a flowchart illustrating the steps of the method of the present invention;

[0047] Figure 2 This is a scatter plot of the amplitude of the present invention;

[0048] Figure 3 This is a scatter plot of the pulse width of the present invention;

[0049] Figure 4 This is a schematic diagram of the electronic device of the present invention. Detailed Implementation

[0050] 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.

[0051] Example 1, please refer to Figure 1As shown, this application provides a roll-to-roll high uniformity coating method for flexible circuit board reinforcement, comprising the following steps:

[0052] Step S1: After the strip coating is completed, acquire the corresponding grayscale line scan image of the strip coating to obtain line scan image data; Step S1 includes the following sub-steps:

[0053] Step S101: The strip of the flexible circuit board reinforcement is referred to as the reinforcement strip, the coating applied to the reinforcement strip is referred to as the strip coating, and the device for applying the strip coating is referred to as the coating device.

[0054] Step S102: Install a line scan camera on the outlet side of the coating device, which is called the line scan device, and make the line scan device acquire an image of the reinforcing strip with the strip coating; record any line of grayscale image acquired by the line scan device as a line scan image. The outlet side is the first sampleable position after the coating process is completed, and the coating defects can be detected in the instant before the subsequent processor.

[0055] Line scan images refer to the acquisition of a single row of pixel data on a strip banner at a specific instant by a line scan camera. Line scan images are not taken at once to capture a certain length of the image. Instead, a row of pixels is captured, the strip moves a little, another row of pixels is captured, and then multiple rows of pixels are stitched together to form the required image. The horizontal row of pixels in the image corresponds to the strip banner, and the vertical column of pixels corresponds to the length of the strip movement.

[0056] Step S103: Set the line frequency of the line scanning device to F0. Based on F0, continuously collect line scan images of the reinforcing strip after the strip coating is completed, and record them as line scan image data. The line frequency of the line scanning device is calculated to determine the acquisition frequency of the line scanning device. It can be set according to the transmission speed of the strip. In this embodiment, F0=50kHz.

[0057] Step S104: Record the currently acquired line scan image as the current line scan image, obtain the line scan images acquired k1 times before the current line scan image, and merge them with the current line scan image to record the current line scan image data, where k1 is the set number;

[0058] In this embodiment, k1=99, which can be flexibly set, but is generally not less than 20; that is, the single-line pixel data on the current instantaneous strip banner is collected by the line scan camera, and then the continuous longitudinal movement of the strip is used to record multiple single-line pixels collected in time sequence as line scan image data.

[0059] In practice, line scan cameras have extremely short single-line long exposure times, and can clearly capture single-line pixels of the horizontal cross section even when the strip is moving at high speed. Area scan cameras, on the other hand, are prone to motion blur due to their long overall exposure time. Therefore, line scan cameras can capture more accurate image data.

[0060] Step S2 involves adaptive image preprocessing based on the line scan image data, and adaptively obtaining the normal grayscale range of the strip coating to obtain normal grayscale range information. Step S2 includes the following sub-steps:

[0061] Step S201: Based on the current line scan image data, stitch all the line scan images into the corresponding image according to the acquisition order, and record it as the initial scan image; record the pixels of each row of the initial scan image as basic scan row 1 to basic scan row k2, where k2=k1+1; that is, stitch together multiple rows of pixels to form the required image.

[0062] Step S202: Set the window size of the one-dimensional median filter to 3×1, perform one-dimensional median filtering on the basic scan line 1, and repeat the one-dimensional median filtering on all scan lines in the initial scan image in turn to obtain the first scan image; one-dimensional median filtering only filters along the line direction and does not process across lines to avoid blurring defects; it can effectively remove salt and pepper noise without changing the grayscale distribution of the normal background.

[0063] Step S203: Remove the k3 columns of pixels on the left and right sides of the first scan image to obtain the second scan image. Record the basic scan rows 1 to k2 in the second scan image as the filtered scan rows 1 to k2, where k3 is the set column number. In this embodiment, k3=5, which can be flexibly set and is generally [3, 8]. The edge area of ​​the reinforcing strip has no effective coating and the gray level has no practical meaning. Removing its detection attribute can avoid misjudging edge gray level jumps as defects and avoid edge pseudo-defects.

[0064] Step S204: For filter scan line 1, take the k4 nearest filter scan lines and filter scan line 1 to form the neighbor processing window of filter scan line 1, where k4 is the set number; in this embodiment, k4=20, which can be set flexibly, generally [15, 25]; if the neighbor processing window is too small, the statistical benchmark will be unstable, and if the neighbor processing window is too large, the amount of calculation will increase, generally around 20 lines;

[0065] Step S205: Obtain the median of the gray values ​​of all pixels in the filtered scan row 1, denoted as the gray median, and calculate the corresponding median absolute deviation, denoted as the median deviation MA.

[0066] The median is far more resistant to outliers than the mean, avoiding deviations in row baseline calculations due to defects within the window; the median absolute deviation (the median of the absolute difference between the grayscale and the row median) is more resistant to outliers than the standard deviation, making it more suitable for real-world scenarios where there are a few defective pixels within the adjacent processing window.

[0067] Step S206: Record the pixels in the filtered scan line 1 with an absolute deviation of no more than 2×MA as background pixels; repeatedly obtain the background pixels of all filtered scan lines in all adjacent processing windows, and obtain the median MG0 of the gray values ​​of all background pixels; obtain the median MG1 of the gray values ​​of the background pixels in the filtered scan line 1.

[0068] For the threshold 2×MA, it can be flexibly adjusted according to the actual application scenario. Obtaining the background pixels provides anomaly-free background samples for the calculation of MG0, that is, the gray value of the normal coating, to ensure the accuracy of the adaptive benchmark, i.e., MG0; MG0 represents the normal gray value of the coating in the adjacent processing window, which is dynamically calculated by the image itself and can adapt to the fluctuation of image features, such as fluctuations caused by light source brightness and transmission speed; MG1 represents the normal gray value of the coating in the current row.

[0069] Step S207: Add (MG0-MG1) to the gray values ​​of all pixels in the filtered scan row 1; level the gray values ​​of the current row and the adjacent processing windows; repeat the processing of all pixels in all filtered scan rows in the second scan image to obtain the third scan image; due to slight fluctuations in the production line, the gray value of the normal coating in the current row may deviate from the gray value of the normal coating in the previous adjacent processing window. Correcting this deviation can avoid affecting the subsequent defect judgment.

[0070] For example, filter scan line 1 is {148, 150, 149, 151, 120, 147, 152, 150, 149, 151}, and the adjacent processing windows are {152, 150, 153, 149, 151, 148, 150, 170, 147, 152} and {147, 149, 150, 152, 148, 151, 149, 150}. {152, 148}, {151, 148, 149, 150, 152, 147, 153, 150, 149, 151}, {149, 152, 148, 151, 150, 152, 147, 150, 172, 149}; then the median gray value of filtered scan line 1 is 149.5, the median deviation is 1.5, and the background pixels of filtered scan line 1 are {1 The background pixels of the other rows are calculated repeatedly: {152, 150, 153, 149, 151, 148, 150, 152}, {147, 149, 150, 152, 148, 151, 149, 150, 152}, {147, 149, 150, 152, 148, 151, 149, 150, 152, 148}, {151, 148}. If the median MG0 is 150 and MG1 is 149, then the filtered scan line 1 after processing is {149, 151, 150, 152, 148, 151, 150, 152, 147, 150, 149}.

[0071] Step S208: Obtain all background pixels in all filtered scan rows of the second scan image, and obtain the median gray value of all background pixels in the third scan image, denoted as the background median AE; the background median represents the grayscale reference of the normal coating in the processed third scan image; this reference is used to map the pixels from the local median scale to the global scale in the future to ensure that all local corrections have a unified reference.

[0072] Step S209: Record any pixel in the third scanned image as the first pixel and the gray value of the first pixel as AF; obtain the pixels in the 8-neighborhood of the first pixel and record them as neighboring pixels; obtain the median BE of the gray values ​​of the neighboring pixels and calculate the corresponding median absolute deviation, recorded as MF; BE reflects the typical gray level of the local area; MF reflects the local gray level distribution.

[0073] Step S210: For any neighboring pixel, if the absolute deviation of the corresponding gray value from BE is not greater than 2×MF, it is marked as the corresponding neighboring background pixel; if there are no less than k5 neighboring background pixels in the 8-neighborhood of the first pixel, it means that the first pixel is a pixel with normal coating, and the gray value of the pixel is corrected to eliminate local background micro-uniformity and gray value jump of row splicing; then calculate AF-(BE-AE) and replace AF, where k5 is the set number; if there are less than k5, it means that the first pixel is a pixel suspected of being a defect, and no processing is performed, and the defect features are retained; in this embodiment, k5=5, which can be flexibly set, and is generally [3.6];

[0074] For example, if the background median AE = 149, the grayscale value of the first pixel is 152, and the grayscale values ​​of the 8-neighborhood are {149, 149, 150, 150, 150, 150, 151, 151}, then BE = 150, MF = 0.5, and the number of background pixels in the neighborhood is 8. Therefore, AF - (BE - AE) = 152 - (150 - 149) = 151.

[0075] Step S211: Repeat the processing of all pixels in the second scan image to obtain the reference scan image.

[0076] Step S212: Obtain the median AE of the gray values ​​of all background pixels in the reference scan image, denoted as the reference median WE; calculate the corresponding absolute deviation of the median, denoted as the reference median deviation WR.

[0077] Step S213: For any pixel in the reference scan image, if the absolute difference between the corresponding gray value and WE is not greater than 2×WR, it is marked as a candidate background pixel; use the global scale to first coarsely screen the pixels, screen out the pixels that are close to the global baseline as candidate background, thereby excluding pixels that are large in difference from the baseline and may be defective pixels.

[0078] Step S214: For any candidate background pixel, obtain the number of corresponding neighboring background pixels in the 8-neighborhood. The size of the neighborhood can be adjusted appropriately according to the requirements, in the same way as obtaining the neighboring background pixels of the first pixel; denoted as AN; if AN / 9 is not less than k6, it is marked as a coating pixel, where k6 is the set proportional threshold; in this embodiment, k6=0.6, which can be flexibly set, at least 0.5; a single pixel may be accidental or noise if its grayscale is close to WE. Adding neighborhood consistency can ensure that the points identified as coating pixels are also continuous in space, thereby reducing isolated misjudgments and ensuring the accuracy of the selected coating pixels.

[0079] Step S215, obtain the median VM of the gray values of all coating pixels and the corresponding median absolute deviation VA; set the threshold coefficient as e0; if VA ≤ r1, then let e0 = 2.5; if r1 < VA ≤ r2, then let e0 = 3.0; if r2 < VA ≤ 2×r2, then let e0 = 3.2; if 2×r2 < VA, give an alarm, where r1 and r2 are set thresholds. In this embodiment, r1 = 0.5 and r2 = 1.5, which can be flexibly set; for the range division of VA and the value of e0, it can also be set according to the actual application scenario;

[0080] Step S216, if VA ≤ 2×r2, then record [VM - e0×VA, VM + e0×VA] as the coating gray range, which is recorded as the normal gray range information;

[0081] In the specific implementation process, if 2×r2 < VA, it means that the coating dispersion is extremely large. This situation is not the normal small fluctuation of the coating background, but is caused by significant abnormalities in the production line or the acquisition system, such as large-scale strobing of the light source, oil stains on the substrate surface, sudden change in the thickness of the coating adhesive layer, vibration of the line-scanning camera, dirty lens, etc., and immediate handling is required without subsequent judgment.

[0082] Step S3, perform binarization processing on the line-scanned image data according to the normal gray range information, and obtain the defect information in the image to obtain the coating defect information; Step S3 includes the following sub-steps:

[0083] Step S301, record the pixel points whose gray values in the reference scanned image are not within the coating gray range as defect pixel points, and record the pixel points whose gray values are within the coating gray range as coating pixel points;

[0084] Step S302, perform binarization processing on the reference scanned image according to the defect pixel points and coating pixel points to obtain a coating binary image, and obtain the connected domain composed of defect pixel points in the coating binary image, which is recorded as the defect area;

[0085] Step S303, record any one defect area as the first area, obtain the number of defect pixel points in the first area, which is recorded as the pixel number AX; and repeat to obtain the pixel numbers of all defect areas, which is recorded as the coating defect information;

[0086] In the specific implementation process, AX is the basic measurement of each defect area, used for subsequent true and false defect judgment, and more indicators can also be obtained according to the actual application scenario for judgment, such as aspect ratio, circularity, compactness, etc.

[0087] Step S4, judge the uniformity of the strip coating according to the coating defect information, and give corresponding alarms and re-coat; Step S4 includes the following sub-steps:

[0088] Step S401: For the first region, if AX is less than k7, the first region is judged as a false defect region; otherwise, the first region is judged as a true defect region. Here, k7 is the set number threshold. Repeat the judgment for all defect regions. In this embodiment, k7=3, which can be set flexibly. Generally, it is [3, 5]. If the number of pixels in the defect region is too small, it may be caused by noise, small particles, or occasional dirt. Judging it as a false defect region can reduce false alarms.

[0089] Step S402: Record the reinforcing strip area corresponding to the reference scan image as the current detection area; if there is a true defect area in the reference scan image, the uniformity of the strip coating in the current detection area is unqualified, and an early warning is issued, and the strip coating is reapplied to the current detection area.

[0090] In the actual implementation process, a certain threshold can be set for the true defect area according to actual needs to control whether to issue an early warning and recoat. For example, a small area of ​​true defect does not affect product quality and can be ignored.

[0091] Example 2, please refer to Figure 4 As shown, Figure 4 A schematic diagram of an electronic device is provided, which may include a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other via the communication bus. The memory stores computer-readable instructions, and the processor can call these instructions. When the processor executes a computer-readable instruction, it performs steps such as those in a roll-to-roll high-uniformity coating method for flexible circuit board reinforcement, to achieve the following functions: after coating is applied to the strip, a grayscale line scan image of the corresponding strip coating is acquired to obtain line scan image data; adaptive image preprocessing is performed based on the line scan image data, and the normal grayscale range of the strip coating is adaptively obtained to obtain normal grayscale range information; binarization processing is performed on the line scan image data based on the normal grayscale range information, and defect information in the image is obtained to obtain coating defect information; uniformity judgment of the strip coating is performed based on the coating defect information, and corresponding warnings and recoating are initiated.

[0092] Furthermore, when the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0093] Example 3: This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it performs steps such as those in the roll-to-roll high uniformity coating method for flexible circuit board reinforcements to achieve the following functions: after the strip is coated, a grayscale line scan image of the corresponding strip coating is acquired to obtain line scan image data; adaptive image preprocessing is performed based on the line scan image data, and the normal grayscale range of the strip coating is adaptively obtained to obtain normal grayscale range information; binarization processing is performed on the line scan image data according to the normal grayscale range information, and defect information in the image is obtained to obtain coating defect information; uniformity judgment of the strip coating is performed according to the coating defect information, and corresponding warnings and recoating are performed.

[0094] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the above technical solutions, in essence or in terms of their contribution to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments.

[0095] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A roll-to-roll high-uniformity coating method for flexible circuit board reinforcing components, characterized in that, Includes the following steps: After the strip is coated, a grayscale line scan image of the corresponding strip coating is acquired to obtain line scan image data. Adaptive image preprocessing is performed based on line scan image data, and the normal grayscale range of the strip coating is adaptively obtained to obtain normal grayscale range information; The line scan image data is binarized based on the normal grayscale range information, and the defect information in the image is obtained to obtain the coating defect information. The uniformity of the strip coating is judged based on the coating defect information, and corresponding warnings and recoating are issued. After the strip is coated, a grayscale line scan image of the corresponding strip coating is acquired. The line scan image data includes the following sub-steps: The strip of the flexible circuit board reinforcement is referred to as the reinforcement strip, the coating applied to the reinforcement strip is referred to as the strip coating, and the device for applying the strip coating is referred to as the coating device. At the outlet side of the coating device, a line scan camera is installed, which is called the line scanning device. The line scanning device acquires images of the reinforcing strip with the strip coating. Any line of grayscale images acquired by the line scanning device at one time is recorded as a line scan image. The line scanning device is set to line frequency F0. Line scan images of the reinforcing strip after the strip coating is completed are continuously acquired based on F0 and recorded as line scan image data. The currently acquired line scan image is recorded as the current line scan image. The line scan images acquired k1 times before the current line scan image are obtained and merged with the current line scan image, and recorded as the current line scan image data, where k1 is the set number. Adaptive image preprocessing based on line scan image data, and adaptive acquisition of the normal grayscale range of the strip coating, to obtain normal grayscale range information includes the following sub-steps: Based on the current line scan image data, all line scan images are stitched together in the acquisition order to form the corresponding image, which is recorded as the initial scan image; the pixels of each row of the initial scan image are recorded as basic scan row 1 to basic scan row k2, where k2=k1+1. Set the window size of the one-dimensional median filter to 3×1, perform one-dimensional median filtering on the basic scan line 1, and repeat the one-dimensional median filtering on all scan lines in the initial scan image in turn. After completion, the first scan image is obtained. Remove the k3 columns of pixels on the left and the k3 columns of pixels on the right of the first scan image to obtain the second scan image. Record the basic scan rows 1 to k2 in the second scan image as the filtered scan rows 1 to k2, where k3 is the set column number. For filter scan line 1, take the k4 nearest filter scan lines and filter scan line 1 to form the neighbor processing window of filter scan line 1, where k4 is the set number; Obtain the median of the gray values ​​of all pixels in the filtered scan row 1, denoted as the gray median, and calculate the corresponding absolute deviation of the median, denoted as the median deviation MA. Pixels with an absolute deviation of no more than 2×MA in filter scan line 1 are recorded as background pixels; background pixels in all filter scan lines in all adjacent processing windows are repeatedly obtained, and the median gray value MG0 of all background pixels is obtained; the median gray value MG1 of background pixels in filter scan line 1 is obtained. Add the gray value of all pixel points in the filtered scan line 1 by (MG0 - MG1); repeat the process for all pixel points in all filtered scan lines of the second scan image, and after completion, obtain the third scan image; Obtain all background pixel points in all filtered scan lines of the third scan image, and obtain the median of the gray values of all background pixel points in the third scan image, denoted as the background median AE; Denote any pixel point in the third scan image as the first pixel point, and denote the gray value of the first pixel point as AF; obtain the pixel points within the 8-neighborhood of the first pixel point, denoted as neighborhood pixel points, obtain the median BE of the gray values of the neighborhood pixel points, and calculate the corresponding median absolute deviation, denoted as MF; For any neighborhood pixel point, if the absolute deviation of the corresponding gray value from BE is not less than 2×MF, then mark it as the corresponding neighborhood background pixel point; if the number of neighborhood background pixel points within the 8-neighborhood of the first pixel point is not less than k5, then calculate AF - (BE - AE) and replace AF, where k5 is the set number; Repeat the process for all pixel points in the third scan image, and after completion, obtain the reference scan image; Obtain the median of the gray values of all background pixel points in the reference scan image, denoted as the reference median WE; calculate the corresponding median absolute deviation, denoted as the reference median deviation WR; For any pixel point in the reference scan image, if the absolute difference between the corresponding gray value and WE is not greater than 2×WR, then mark it as a candidate background pixel; For any candidate background pixel, obtain the number of corresponding neighborhood background pixel points within the 8-neighborhood, denoted as AN; if AN / 9 is not less than k6, then mark it as a coating pixel, where k6 is the set ratio threshold; Obtain the median VM and the corresponding median absolute deviation VA of the gray values of all coating pixels; set the threshold coefficient as e0; if VA ≤ r1, then let e0 = 2.5; if r1 < VA ≤ r2, then let e0 = 3.0; if r2 < VA ≤ 2×r2, then let e0 = 3.2; if 2×r2 < VA, then give an early warning; where r1 and r2 are the set thresholds; If VA ≤ 2×r2, then denote [VM - e0×VA, VM + e0×VA] as the coating gray range, denoted as the normal gray range information.

2. The roll-to-roll high uniformity coating method for flexible circuit board reinforcement components according to claim 1, characterized in that, Perform binary processing on the line scan image data according to the normal gray range information, and obtain the defect information in the image. The coating defect information includes the following sub-steps: Denote the pixel points in the reference scan image whose gray values are not within the coating gray range as defect pixel points, and denote the pixel points whose gray values are within the coating gray range as coating pixel points; Perform binary processing on the reference scan image according to the defect pixel points and coating pixel points to obtain a coating binary image, and obtain the connected domain composed of defect pixel points in the coating binary image, denoted as the defect region; Denote any defect region as the first region, obtain the number of defect pixel points within the first region, denoted as the pixel number AX; and repeat to obtain the pixel numbers of all defect regions, denoted as the coating defect information.

3. The roll-to-roll high uniformity coating method for flexible circuit board reinforcement components according to claim 2, characterized in that, The uniformity of the strip coating is assessed based on coating defect information, and corresponding warnings and recoating are initiated, including the following sub-steps: For the first region, if AX is less than k7, the first region is determined to be a false defect region; otherwise, the first region is determined to be a true defect region, where k7 is the set threshold for the number of defects. The judgment is repeated for all defect regions. The area of ​​the reinforcing strip corresponding to the reference scan image is recorded as the current detection area; If a true defect area is found in the reference scan image, the uniformity of the strip coating in the current detection area is not up to standard, an early warning is issued, and the strip coating in the current detection area is reapplied.

Images