CN110196021A - Coating layer thickness and its application are measured based on Optical coherence tomography technology - Google Patents

Abstract

The invention discloses a coating thickness measurement based on optical coherence tomography imaging technology and its application. The method includes the following steps: building an optical coherence tomography (OCT) test system; using the OCT system to measure the depth direction of the sample material Scan the two-dimensional cross-section of the surface, and finally obtain the three-dimensional scanning image of the measured object; after the obtained three-dimensional scanning image is subjected to noise reduction processing, it is converted into a binary image, and then the upper and lower boundaries of the coating are segmented by edge tracking method , to calculate the coating thickness. The method can not only realize the non-destructive automatic detection of the coating, but also can realize the fast, accurate and large-scale measurement of the thickness of the coating. It can be applied to measure the thickness of thin film coatings, including conformal coatings on the surface of printed circuit boards, pigmented polymer coatings, protective coatings on porcelain, plasma sprayed ceramics, biodegradable composite coatings, varnish coatings, transparent or translucent tablet films coating.

Description

Coating Thickness Measurement and Its Application Based on Optical Coherence Tomography

technical field

The invention belongs to the detection field based on image analysis, and in particular relates to an optical coherence tomography-based imaging technique for measuring coating thickness and its application.

Background technique

Optical coherence tomography (OCT) is a low-coherence optical interference imaging technique. This technology measures the light signals scattered/reflected at different depths in the material by means of interference, thereby realizing three-dimensional imaging of the interior of the material. Based on its imaging mechanism, this technique is well suited for imaging and thickness measurement of multilayer structures. This imaging technique has the advantages of being non-destructive, fast and highly sensitive. OCT was originally invented for medical applications, and has successfully developed into a routine examination and diagnosis method for ophthalmic diseases. At the same time, OCT has shown great potential in other clinical fields, such as dermatology, cardiology, and gastroenterology. In recent years, the application of OCT in industrial non-destructive testing has also been greatly developed, especially the thickness measurement of various industrial materials and products, such as automotive coatings, egg shells and multi-layer foils. In addition, OCT is also used to detect jade, silicon integrated circuits, industrial ceramics, etc. Our recent research has found that OCT can perform three-dimensional imaging through the conformal paint coating on the surface of the printed circuit board, thereby realizing the measurement of the coating thickness.

For example, printed circuit boards (Printed Circuit Board, PCB) are widely used in automotive, consumer electronics, industrial and communication fields, which requires that printed circuit boards can work normally in various environments, including high temperature, high humidity and Corrosive and other harsh environments. In actual use, a layer of conformal paint is generally applied to the surface of the printed circuit board to ensure that the printed circuit board can work normally under the above harsh conditions or to prolong its service life. The three anti-paint coating is generally a thin layer with a thickness of tens to hundreds of microns, usually made of synthetic resin or plastic. Conformal paint coating can protect PCB from harsh environments such as electrical, mechanical and chemical, such as moisture, dust, mechanical stress, thermal stress, corrosion, solvents, chemical vapors, etc. Therefore, conformal coatings have been widely used to improve the reliability and lifespan of PCBs, especially in automotive, avionics, and military electronics. In order to ensure effective protection, the quality of the three anti-paint coatings needs to be strictly controlled, and the thickness of the three anti-paint coatings is an important indicator of the quality of the three anti-paint coatings. The unevenness of the coating thickness will lead to the decline of its protective ability. At the same time, the coating thickness needs to be within the standard range, but the thickness of the coating cannot be determined after spraying, and it needs to be tested after curing. Therefore, it is very necessary to measure the thickness of the three anti-paint. The current three anti-paint thickness measurement methods include non-destructive and destructive. A simple non-destructive method is direct measurement with a micrometer, but requires measurements on an uncoated PCB as a reference. Near-infrared spectrometer, Raman spectrometer, and β backscattering method are also non-destructive measurement methods, but the measurement method is not direct, and the uniformity of coating distribution cannot be measured. Ultrasonic technology and acoustic microscope are also non-destructive testing methods, but they need to immerse the object under test in liquid and have low resolution. As for destructive measurement, a common method is to cut the PCB and then observe its cross-section under a microscope to measure the thickness. This method can only be performed on a spot check basis due to its destructive nature; it is also time-consuming and subject to human error. In addition to the above-mentioned disadvantages, the above-mentioned non-destructive measurement and destructive measurement methods are only suitable for point measurements rather than regional measurements. However, in industrial practice, area scanning is required to assess conformal coating coverage and uniformity, two other key factors that determine protection effectiveness.

Contents of the invention

In view of the above problems, the present invention proposes a measurement of paint coating thickness based on optical coherence tomography imaging technology and its application.

Realize above-mentioned technical purpose, reach above-mentioned technical effect, the present invention realizes through the following technical solutions:

A method for measuring coating thickness based on optical coherence tomography imaging technology, comprising the following steps:

An optical coherence tomography (OCT) test system is built. The laser light emitted from the light source end of the OCT test system is divided into two beams, which are irradiated to the reference arm and the sample arm respectively, and the beams reflected from the reference arm and the sample arm interfere. Obtain the interference spectrum signal;

Use the spectrometer in the OCT system to measure the interference spectrum signal, including performing inverse Fourier transformation on the interference spectrum signal to obtain the reflectivity profile of the measured position along the depth direction, and using the beam to scan the surface of the measured object and obtain two Combined with the obtained reflectivity profile along the depth direction to obtain a three-dimensional scanning image of the measured object;

After denoising the obtained 3D scanning image, it is converted into a binary image, and then the upper and lower boundaries of the coating are segmented by the edge tracking method to calculate the thickness of the coating.

As a further improvement of the present invention, the spectrometer collects interference spectra in pixel or wavelength space, and converts the output interference spectra into interference spectra in wavenumber space through calibration.

As a further improvement of the present invention, the calibration method includes: extracting the remapping vector from the interference spectrum of the space through Hilbert transform, and then obtaining the interference spectrum of the wavenumber space through linear difference; or, performing dispersion by using linear regression Compensate to obtain an interference spectrum in wavenumber space.

As a further improvement of the present invention, the noise reduction method includes a mean filtering method and a Gaussian filtering method.

As a further improvement of the present invention, the process of using the edge tracking method to segment the upper and lower boundaries of the coating on the binary image includes:

Take the first pixel as the starting point to mark the intensity 1, compare the intensity of the pixels in the neighborhood one by one in the vertical or horizontal direction, and select the pixel points with the intensity of 0 in the neighborhood as tracking points, and finally obtain all Connect the trace points of , and confirm the upper and lower boundaries.

As a further improvement of the present invention, the spectral range emitted by the laser source in the constructed OCT system is 770-950nm.

Apply the above-mentioned method of measuring coating thickness based on optical coherence tomography imaging technology, which is used to measure the thickness of thin film coatings, including conformal coatings on the surface of printed circuit boards, colored polymer coatings, porcelain protective coatings, plasma sprayed ceramics , biodegradable composite coatings, varnish coatings, transparent or translucent tablet film coatings, and the measured film coating thickness is from micron to millimeter.

Beneficial effects of the present invention: the present invention not only realizes the non-destructive automatic detection of the coating through the ultra-high resolution optical coherence tomography imaging technology, but also realizes the rapid, accurate and large-scale measurement of the thickness of the coating.

Description of drawings

Fig. 1 is the structural diagram of the OCT system built among the present invention;

Fig. 2 (a) and Fig. 2 (b) are in an embodiment of the present invention, conformal paint is coated on the structural representation of two kinds of printed circuit boards with different structures;

Fig. 3 (a) and Fig. 3 (b) are the OCT cross-sectional images obtained corresponding to Fig. 2 (a) and Fig. 2 (b);

Figure 4 is a diagram of the image processing process: (a) the original OCT cross-sectional image, (b) the noise-reduced image, (c) the binary image, (d) the boundary of the conformal coating produced by edge tracking Enclosed envelope, (e) the envelope of the lower boundary of the conformal paint and the lower region generated by edge tracking, (f) the detected top conformal coating and the bottom boundary;

Fig. 5 (a) is the PCB photo adopted in the embodiment of this patent, and Fig. 5 (b) is the front plan view in the rectangular frame area shown in Fig. 5 (a);

Figure 6 shows that two different methods are used to test the thickness of the coating, Figure 6(a) uses OCT technology, and Figure 6(b) uses microscope technology;

Figure 7 is a coating thickness distribution diagram and a coating thickness difference diagram obtained by two different methods;

Fig. 8 is a distribution histogram and a fitted Gaussian curve diagram of the thickness difference in Fig. 7(b);

Fig. 9 is a graph of coating thickness generated by the OCT technique of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention more clear, the present invention will be further described in detail below in conjunction with the examples. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

The application principle of the present invention will be described in detail below in conjunction with the accompanying drawings.

In the present invention, the thickness measurement of the conformal paint coating on the printed circuit board (Printed Circuit Board, PCB) is taken as an example for specific description. As shown in Figure 5, the printed circuit board selected in this embodiment is sprayed with the three anti-paint layer produced by Peters Company, and the rectangular frame area in Figure 5 (a) is selected for scanning to obtain the corresponding Figure 5 (b) , and imaged for Figure 5(b).

The test method used mainly includes three steps: ultra-high resolution OCT system design and development, data acquisition and processing, image edge extraction and thickness calculation. The specific description is as follows:

(1) Build an ultra-high resolution OCT test system to obtain interference spectrum signals

As shown in Figure 1, the ultra-high resolution OCT test system designed by the present invention is mainly composed of a supercontinuum laser light source, a sample arm, a reference arm, a data acquisition card, and a spectrometer (Spectrometer). The central wavelength of the laser light source is 810nm, the half-peak bandwidth is 180nm, and the output power is 15.6mW. The light from the light source is divided into two parts by the beam splitter, one part enters the reference arm and the other enters the sample arm. In the reference arm, the lens assembly composed of L2, L3 and L4 arranged in sequence is irradiated onto the reference mirror (RM), and the sample arm is irradiated to the sample through the lens assembly composed of L5, L6 and L7 arranged in sequence ( Sample). Light returning from the reference and sample arms is recombined into the same single-mode fiber (SMF), interferes with each other, and is finally coupled to the spectrometer. The reference and sample arms use the same optical lens set (L2 and L5, L3 and L6, L4 and L7) to minimize chromatic aberration. An X-Y scanning galvanometer (GS) is also set on the sample arm to realize beam scanning, and the track of beam scanning is controlled by an analog voltage signal (AO) generated by a data acquisition card.

The interference spectrum signal formed by OCT is detected by a self-designed spectrometer. The spectrometer is composed of a collimator lens (L9), a diffraction grating (Grating), an imaging lens (CL), and a line scan camera (LSC). The signal detected by the line scan camera is converted into a digital signal by a 12-bit image acquisition card (IMAQ) and stored in a computer.

(2) OCT signal processing and image reconstruction

In spectrometer-based OCT, the interference spectra are sampled in detector pixel space (n-space (n-space), where n is the index of the detector pixel), rather than equally spaced in wavenumber space (k-space) Evenly distributed. Therefore, the measured interference spectrum needs to be resampled before performing the inverse Fourier transform, so that it is evenly spaced in the wavenumber space. In this patent, the calibration method we adopt for pixel space to wavenumber space conversion is: when using a single mirror as a sample, OCT should generate a perfect sinusoidal interferogram in k-space, so the mapping vector can be found to remap n- A previously imperfect interferogram of space. Specifically, the calibration method consists of two main steps: first, two interferograms are recorded at different optical path lengths, subsequently the remapping vectors are extracted by Hilbert transform, and then the k-space interferometric spectra are obtained by linear interpolation; secondly , using linear regression to implement the dispersion compensation process to obtain the desired k-space interference spectrum.

After the spectral signal is converted from wavenumber space to k-space, the inverse Fourier transform of the interference spectral signal can be performed to obtain a cross-sectional image (B-scan). Combining multiple cross-sectional images together produces a three-dimensional (3D) image.

(3) Image edge extraction and thickness calculation

As shown in Figure 2(a), PCB is usually a multi-layer structure. There is usually a layer of solder mask under the conformal paint, and there is a layer of copper foil circuit under the solder mask in some areas (as shown in Figure 2(b)) . Due to the shallow depth of OCT imaging, the above-mentioned two-layer or three-layer structure is usually imaged. The discontinuity of the refractive index between the conformal coating and the adjacent coating will cause the conformal coating to usually show two different boundaries when OCT imaging, as shown in Figure 3. We use these two boundaries and use the following algorithm to calculate and measure the thickness of the three-proof paint. The specific process is as follows.

First, the obtained original OCT cross-sectional image (as shown in Figure 4(a)) is denoised using the mean filter method and Gaussian filter method to obtain a denoised image (as shown in Figure 4(b)); then , each image is converted into a binary image (as shown in Figure 4(c)); finally, the edge tracking method is used to segment the upper and lower boundaries of the coating.

Taking Figure 4(c) as an example, the process of dividing the upper and lower boundaries of the coating with the edge tracking method is as follows: traverse each pixel in the first column on the left from top to bottom, and use the first pixel with the selection intensity of 1 as As a starting point, the following strategy is used for edge tracking. Specifically, (1) use the first pixel with intensity of 1 in the first column on the left as the starting point, and find pixels with intensity of 1 in the neighborhood in turn; (2) select pixels with intensity of 1 above Select the next tracking point among the points, and the selection condition is that the neighborhood of the pixel point contains a pixel point with an intensity of 0; (3) Repeat the above process. This tracking process finally produces a closed envelope as shown in Fig. 4(d). Set the middle of the envelope as the upper boundary of the conformal paint coating (Fig. 4 (e)). After detecting the upper boundary of the coating, the algorithm selects a new tracking starting point to continue edge tracking. Search for the first pixel with an intensity of 1 in the first column of the binary image at the bottom as a new tracking starting point, and edge tracking produces another closed envelope. As shown in Fig. 4(f), finally, the coating thickness can be obtained by measuring the vertical distance between the detected upper and lower boundaries.

Experimental results:

From the results shown in Figure 6(a), it can be seen that there are two obvious upper and lower boundaries in the OCT cross-sectional image, and the coating thickness is measured based on these two boundaries. In addition, the results shown in this figure include two-layer and three-layer structures, the two-layer area is shorter, and the three-layer area is longer.

The image of the coating obtained by microscopy technique shown in Fig. 6(b) is to further evaluate the accuracy of the thickness measurement method proposed in the present invention. After the PCB is imaged with ultra-high resolution OCT, the same sample is sent to a third party for metallographic sectioning to measure the thickness. The method is to cut the PCB, then polish it smooth, and image it under a microscope to measure its thickness. This method will permanently damage the PCB. Therefore, only one cross-section of the PCB was photographed under the microscope, as shown in Fig. 6(b). At the same time, it can be seen that the microscope images and the cross-sectional images generated by super-resolution OCT are in good agreement. Parts of the copper layer in both images show strong reflections. Thickness measurements on microscopic images were achieved by manually segmenting the coating.

Figure 7(a) depicts the comparison of the coating thicknesses measured by the two different methods described above. Overall, the measurements from the two methods show good agreement over a range of about 5 mm. As shown in Figure 7(b), inconsistency is assessed by computing the difference between two measurements. The differences are all distributed within the range of plus or minus five microns, mainly within the range of three microns, with an average difference of 0.37 μm and a standard deviation of 1.38 μm. Fig. 8 is the distribution histogram of the thickness difference in Fig. 7(b), and the following curve is a Gaussian fit on the thickness difference histogram, with a full width at half maximum of 3.01um. This difference is partly attributed to the axial resolution (1.72 μm) of ultra-high resolution OCT, which determines the measurement accuracy. Again, another part is attributable to measurement errors in traditional methods. The above results demonstrate that ultra-high resolution OCT can provide accurate thickness measurements for conformal paint coatings on PCBs

In addition, the three-dimensional imaging capability of OCT can be used to generate a thickness map, as shown in Figure 9, through the color coding of the thickness, it is easy to grasp the thickness change. This paves the way for rapid detection of PCB coating thickness.

Finally, we also evaluate our proposed algorithm in terms of computational efficiency. The algorithm is implemented on the Matlab platform and tested on a computer with Intel(R) Xeon(R) CPU E5-1650v2@3.5GHz and 24GB RAM, where only a single core is used. 350 consecutive images were processed and the running time of the algorithm was about 36 seconds. For comparison, the metallographic sectioning method takes about two hours, including cutting the PCB, polishing and measuring. From these data, our method can be used for online detection in pipelines.

The basic principles and main features of the present invention and the advantages of the present invention have been shown and described above. Those skilled in the industry should understand that the present invention is not limited by the above-mentioned embodiments. What are described in the above-mentioned embodiments and the description only illustrate the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention will also have Variations and improvements are possible, which fall within the scope of the claimed invention. The protection scope of the present invention is defined by the appended claims and their equivalents.

Claims (7)

1. A method for measuring coating thickness based on optical coherence tomography imaging technology, is characterized in that, comprises the following steps: To build an optical coherence tomography (OCT) test system, the laser light emitted from the light source end of the OCT test system is divided into two beams, which are irradiated to the reference arm and the sample arm respectively, and the beams reflected from the reference arm and the sample arm interfere. Obtain the interference spectrum signal; Use the spectrometer in the OCT system to measure the interference spectrum signal, including performing inverse Fourier transformation on the interference spectrum signal to obtain the reflectivity profile of the measured position along the depth direction, and using the beam to scan the surface of the measured object and obtain two Combined with the obtained reflectivity profile along the depth direction to obtain a three-dimensional scanning image of the measured object; After denoising the obtained 3D scanning image, it is converted into a binary image, and then the upper and lower boundaries of the coating are segmented by the edge tracking method to calculate the thickness of the coating. 2. The method for measuring coating thickness based on optical coherence tomography imaging technology according to claim 1, characterized in that: what the spectrometer collects is the interference spectrum of the pixel or wavelength space, and the output interference spectrum is converted by calibration into the interference spectrum in wavenumber space. 3. the method for measuring coating thickness based on optical coherence tomography imaging technology according to claim 2, is characterized in that, described calibration method comprises: Extract the remapping vector from the interference spectrum of the space through the Hilbert transform, and then obtain the interference spectrum of the wavenumber space through linear difference; Alternatively, obtain the interference spectrum in wavenumber space by performing dispersion compensation using linear regression. 4. The method for measuring coating thickness based on optical coherence tomography imaging technology according to claim 1, characterized in that: said noise reduction method includes mean filtering method and Gaussian filtering method. 5. the method for measuring coating thickness based on optical coherence tomography imaging technology according to claim 1, is characterized in that, the process of adopting edge tracking method to the upper and lower boundary of binary image segmentation coating comprises: Take the first pixel as the starting point to mark the intensity 1, compare the intensity of the pixels in the neighborhood one by one in the vertical or horizontal direction, and select the pixel points with the intensity of 0 in the neighborhood as tracking points, and finally obtain all Connect the trace points of , and confirm the upper and lower boundaries. 6. The method for measuring coating thickness based on optical coherence tomography imaging technology according to claim 1, characterized in that: the spectral range emitted by the laser source in the constructed OCT system is 770-950nm. 7. application according to the method for measuring coating thickness based on optical coherence tomography imaging technology according to claim 1-6, is characterized in that: adopt described method to be applied to the thickness of measuring film coating and comprise three parts of printed circuit board surface Anti-paints, pigmented polymer coatings, protective coatings on porcelain, plasma sprayed ceramics, biodegradable composite coatings, varnish coatings, transparent or translucent tablet film coatings.