CN110298857A - A kind of celadon glazed thickness method for automatic measurement based on SD-OCT image - Google Patents

Abstract

The invention discloses a method for automatically measuring the thickness of a celadon glaze layer based on an SD-OCT image. Carry out noise reduction and binarization processing on the SD-OCT image of the celadon glaze layer, detect the upper boundary of the glaze layer that contains the most pixels in the image, and optimize the upper boundary of the glaze layer by Lagrangian interpolation method; Separation of the background and target and image flattening; Locate the lower boundary of the glaze layer through morphological operations; Calibrate the pixel axial resolution of SD-OCT images of different types of celadon glaze layers by measuring the thickness of the glaze layer through the SD-OCT system; Calculate the pixel distance between the upper and lower boundaries of the glaze layer and combine the corresponding pixel axial resolution to calculate the thickness of the glaze layer. The present invention realizes the measurement of the thickness of various types of celadon glaze layers by calibrating the pixel axial resolution of SD‑OCT images of different types of celadon glaze layers, has strong robustness and applicability, and effectively improves the Efficiency of celadon glaze layer thickness measurement.

Description

An automatic measurement method of celadon glaze layer thickness based on SD-OCT image

technical field

The invention belongs to the field of automatic measurement of the thickness of celadon glaze layer, in particular to a method for measuring the thickness of celadon glaze layer based on SD-OCT images.

Background technique

The craftsmanship of celadon is the highest state in the history of porcelain making in my country, and it has huge market value. Since modern times, the development of celadon production technology has also encountered stagnation to varying degrees. There are few high-quality works and low yields hinder the development of Longquan celadon. In actual production, the thickness of the glaze layer affects the quality of the ceramics. As the thickness of the glaze layer increases, the saturation of the glaze color increases, and the thickness of the glaze layer affects the glaze stress between the glaze layer and the carcass. Controlling the thickness of the glaze layer can effectively improve and prevent cracking of the glaze layer and improve the quality of celadon. Therefore, in the production of celadon, the measurement of the thickness of the glaze layer has become an effective method to detect the quality of celadon.

Spectrum-domain optical coherence tomography (SD-OCT) technology is a new optical imaging technology based on confocal microscope and Michelson interference principle, which can reveal the near-surface structure of the glaze layer by detecting the backscattered light of celadon. It has the characteristics of high resolution and non-destructive testing. At present, SD-OCT imaging technology has been successfully applied to the structural research, classification and qualitative identification of ceramics. By combining SD-OCT imaging technology and image processing technology, the real-time, non-destructive and accurate measurement of the thickness of the celadon glaze layer can be realized, and the measurement position is not limited. Therefore, it has important research and reference value for the development of celadon production technology.

Contents of the invention

Aiming at the problems existing in the background technology, the present invention provides a method for measuring the thickness of the celadon glaze layer based on SD-OCT images to realize real-time and accurate measurement of the thickness of the celadon glaze layer. Quality, improve the yield of celadon has laid a technical foundation.

The present invention performs noise reduction and binarization processing on the SD-OCT image of the celadon glaze layer, detects the position of the target with the most pixels in the image, uses the uppermost pixel of each column of the target as the upper boundary pixel of the glaze layer, and passes Lagrangian interpolation optimizes the located upper boundary of the glaze layer. Separate image background and target and perform image flattening. By enhancing the grayscale contrast of the flattened image, combined with morphological operations, the lower boundary of the glaze layer is located. The pixel axial resolution of SD-OCT images of different types of celadon glaze layers was calibrated by measuring the thickness of the glaze layer with the OCT system. The product of the pixel distance between the upper and lower boundaries of the celadon glaze layer and the pixel axial resolution of the SD-OCT image of the corresponding type of celadon glaze layer is the thickness of the celadon glaze layer. The key lies in the calculated pixel axial resolution of SD-OCT images of different types of celadon glaze layers, software algorithms and the logic of the entire process.

The technical scheme adopted in the present invention comprises the following steps:

Step 1) collecting SD-OCT images of different types of celadon glaze layers;

Step 2) denoising the celadon glaze layer SD-OCT image collected in step 1);

Step 3) positioning the upper boundary of the glaze layer in the celadon glaze layer SD-OCT image collected;

Step 4) performing background separation and image flattening on the SD-OCT image of the celadon glaze layer;

Step 5) positioning the lower boundary of the glaze layer in the celadon glaze layer SD-OCT image collected;

Step 6) calibrating the pixel axial resolution of different types of celadon glaze layer SD-OCT images;

Step 7) Calculate the thickness of the glaze layer.

The step 2) specifically includes: using a template with a size of 3×5 to perform median filtering on the SD-OCT image of the celadon glaze layer collected in step 1).

The step 3) is specifically:

3.1) Use the OSTU method to calculate the binarization threshold t of the SD-OCT image of the celadon glaze layer, and use the binarization threshold t as the threshold of the canny operator to perform edge detection on the SD-OCT image after step 2) denoising, and obtain binary image Bw;

3.2) Define a circular structural element se with a radius of 5 pixels, and use the structural element se to perform morphological closing operation on the binary image Bw to obtain the image fc, which is calculated as follows:

In the formula, the symbol Indicates the morphological expansion operation, which completes the "growth" or "coarsening" of the object in the binary image Bw; the symbol Θ represents the morphological erosion operation, and completes the "shrinkage" or "thinning" of the object in the binary image Bw;

3.3) Search from top to bottom for the first pixel with a gray value of 1 appearing in each column of the image fc and record this point as the pixel to be fitted, and record the row position Top of the pixel to be fitted in each column (i), wherein i represents the column position where the pixel point to be fitted is located;

Use the pixel to be fitted in the connected domain with the largest number of pixels as the upper boundary pixel of the celadon glaze layer; if the pixel to be fitted does not belong to the upper boundary pixel of the celadon glaze layer, use the Lagrangian interpolation method to update the row The value of position Top(i), calculated as follows:

Top(i)=2×Top(x)-Top(y)

Among them, i represents the column position of the pixel to be fitted;

If there are upper boundary pixels of the celadon glaze layer in the columns on the left and right sides of the i column, then x indicates the left column with the upper boundary pixels closest to the i column, and y indicates the right column of the glaze layer upper boundary pixels closest to the i column ; If there is an upper boundary pixel in the column to the left of the ith column, and there is no upper boundary pixel in the right column, then x indicates the left column with the upper boundary pixel that is closest to the i column, and y indicates the upper boundary pixel that is closest to the i column The left column of the boundary pixels; if there is no boundary pixel on the glaze layer in the left column of the i-th column, and there is an upper boundary pixel on the glaze layer in the right column, then x represents the right column that has the boundary pixel on the glaze layer closest to the i-th column, y represents the right column of the border pixel on the glaze layer that is the closest to the i-th column.

Described step 4) specifically is:

4.1) For the image after median filtering, clear the gray value of the pixel whose position in each column and row in the image is smaller than Top(i), and separate the image background;

4.2) Take the minimum value in Top(i) as the upper boundary reference of the celadon glaze layer, and shift the position of each column of pixels upward so that the row positions of all upper boundary pixels are on the same horizontal line, thereby obtaining a flattened image fl , and the minimum value in Top(i) is used as the position U of the upper boundary of the glaze layer for the measurement of the thickness of the celadon glaze layer.

The step 5) is specifically described as:

5.1) Use the contrast-limited adaptive histogram equalization method to enhance the gray contrast of the flattened image fl to obtain the image fa after gray contrast enhancement, and then use the structural element se to perform morphological opening operations on the image fa to obtain the image fo , the calculation of the morphological opening operation is as follows:

In the formula, the symbol Indicates the morphological expansion operation, and the symbol Θ indicates the morphological erosion operation;

5.2) Use the flattened image fl as a mask, perform continuous expansion operations on the image fo, complete the morphological image reconstruction, and obtain the image fr;

5.3) Define a circular structural element with a radius of 3 pixels, and use this circular structural element to perform morphological expansion operation on image fr to obtain image fd;

5.4) Use the OSTU method to binarize the image fd, start traversing from the bottom of the glaze layer in the binarized image fd in row units, and traverse each row from bottom to top until the traversed row does not contain glaze Stop traversing the pixels in the region where the layer is located, and record this behavior Stop traversing the row, count the number of rows in each column where the pixel point with the smallest row position below the row where the traversal is stopped, and calculate the average value B of the number of rows in all columns, B is used as the row position of the lower boundary of the glaze layer in the SD-OCT image for the measurement of the thickness of the celadon glaze layer.

The step 6) is specifically:

6.1) Scan different types of celadon fragments respectively to obtain cross-sectional SD-OCT images of the celadon fragments;

6.2) use the caliper of the SD-OCT system to measure the thickness T _x of the glaze layer in the cross-sectional SD-OCT image of the celadon fragment;

6.3) scan the SD-OCT image of the celadon glaze layer in the same area as the cross-sectional SD-OCT image of the celadon fragments, and simultaneously locate the upper and lower boundaries of the celadon glaze layer according to the method of steps (2)-step (5) and obtain the distance between the upper and lower boundaries. The pixel distance D of is calculated as follows:

D=B-U

Among them, B is the lower boundary of the glaze layer, and U is the upper boundary of the glaze layer;

6.4) Calculate the pixel axial resolution Pr _x of SD-OCT images of different types of celadon glaze layers by using the following formulas:

The step 7) is specifically: calculate the pixel distance D between the upper and lower boundaries of the glaze layer in the celadon glaze layer SD-OCT image of the scanned celadon glaze layer SD-OCT image according to the following formula:

D=B-U

Among them, B is the lower boundary of the glaze layer, and U is the upper boundary of the glaze layer;

In step 6), select the pixel axial resolution Pr _x of the celadon glaze layer SD-OCT image of the same type as the celadon glaze layer scanned in step ₁ ), and use the following formula to calculate the thickness Tx of different types of celadon glaze layers:

Among them, Pr _x is the pixel axial resolution of the SD-OCT image of the celadon glaze layer scanned in step 1), and D is the pixel distance D between the upper and lower boundaries of the glaze layer.

The beneficial effects of the present invention are:

1) The present invention can accurately measure the thickness of the glaze layer at any position of different types of celadon in real time and non-destructively, and gets rid of the problem that the celadon glazing uniformity can only be judged by the number of times of glazing.

2) The present invention accurately locates the position of the upper and lower boundaries of the celadon glaze layer in the image, and by flattening the image, it overcomes the error problem caused by the curvature measurement, ensures the accuracy and robustness of the measurement, and simultaneously measures different types of celadon The pixel axial resolution of the SD-OCT image of the enamel layer is calibrated, and the thickness measurement of different types of celadon glaze layers is completed, and the universality of the method of the present invention for measuring the thickness of the celadon glaze layers is improved.

Description of drawings

Figure 1 is a flow chart of the method of the present invention.

Fig. 2 is the physical picture of the celadon samples whose thickness of the glaze layer is to be measured, and (a)-(f) respectively correspond to a kind of samples.

Figure 3 is the SD-OCT image of the celadon sample whose thickness of the glaze layer is to be measured, and (a)-(f) correspond to (a)-(f) in Figure 2 in turn.

Fig. 4 is the SD-OCT image of the embodiment of background separation and image flattening, and (a)-(f) correspond to (a)-(f) in Fig. 3 in turn.

Fig. 5 is the cross-sectional view of the celadon glaze layer and the corresponding SD-OCT image of the embodiment, (a) is the cross-sectional view of the celadon glaze layer, and (b) is the SD-OCT image of the celadon glaze layer cross-section.

Fig. 6 is an extraction effect diagram of the upper and lower boundaries of the celadon glaze layer of the embodiment, and (a)-(f) correspond to (a)-(f) in Fig. 3 in turn.

Detailed ways

The present invention will be described in further detail below in conjunction with the accompanying drawings and embodiments.

Specific examples:

A method for measuring the thickness of a celadon glaze layer based on an SD-OCT image, comprising the following steps:

1) collecting SD-OCT images of different types of celadon glaze layers;

The OQ LabScope system produced by Lumedica was used to collect SD-OCT images of different types of celadon glaze layers, the scanning range was set to 3.5mm, and the size of the collected images was 512pixel*512pixel. Fig. 2 is the physical picture of the celadon sample of the thickness of the glaze layer to be measured, (a)-(f) correspond to a sample respectively; Fig. 3 is the SD-OCT image of the celadon glaze layer of the thickness of the glaze layer to be measured, (a)-(f) ) corresponds to (a)-(f) in Figure 2 in turn.

2) Denoise the collected SD-OCT image of the celadon glaze layer: use a template with a size of 3×5 to perform median filtering on the SD-OCT image of the celadon glaze layer in step 1).

3) positioning the upper boundary of the glaze layer in the celadon glaze layer SD-OCT image collected;

3.1) Use the OSTU method to calculate the binarization threshold t of the SD-OCT image of the celadon glaze layer, and use the binarization threshold t as the threshold of the canny operator to perform edge detection on the SD-OCT image after step 2) denoising, and obtain binary image Bw;

3.2) Define a circular structural element se with a radius of 5 pixels, and use the structural element se to perform morphological closing operation on the binary image Bw to obtain the image fc, which is calculated as follows:

In the formula, the symbol Indicates the morphological expansion operation, which completes the "growth" or "coarsening" of the object in the binary image Bw; the symbol Θ represents the morphological erosion operation, and completes the "shrinkage" or "thinning" of the object in the binary image Bw;

3.3) Search from top to bottom for the first pixel with a gray value of 1 appearing in each column of the image fc and record this point as the pixel to be fitted, and record the row position Top of the pixel to be fitted in each column (i), wherein i represents the column position where the pixel point to be fitted is located;

3.4) Search for each adjacent pixel to form a connected domain in the form of eight connected regions, and obtain all connected domains in the image;

3.5) Use the pixel to be fitted in the connected domain with the largest number of pixels as the upper boundary pixel of the celadon glaze layer; if the pixel to be fitted does not belong to the upper boundary pixel of the celadon glaze layer, use the Lagrangian interpolation method Update the value of the row position Top(i) of the pixel to be fitted, calculated as follows:

Top(i)=2×Top(x)-Top(y)

Among them, i represents the column position of the pixel to be fitted;

If there are upper boundary pixels of the celadon glaze layer in the columns on the left and right sides of the i column, then x indicates the left column with the upper boundary pixels closest to the i column, and y indicates the right column of the glaze layer upper boundary pixels closest to the i column ; If there is an upper boundary pixel in the column to the left of the ith column, and there is no upper boundary pixel in the right column, then x indicates the left column with the upper boundary pixel that is closest to the i column, and y indicates the upper boundary pixel that is closest to the i column The left column of the boundary pixels; if there is no boundary pixel on the glaze layer in the left column of the i-th column, and there is an upper boundary pixel on the glaze layer in the right column, then x represents the right column that has the boundary pixel on the glaze layer closest to the i-th column, y represents the right column of the border pixel on the glaze layer that is the closest to the i-th column.

4) Perform background separation and image flattening on the SD-OCT image of the celadon glaze layer;

4.1) Corresponding the value of Top(i) to the image after median filtering, clearing the gray value of each column and row position in the image smaller than the pixel of Top(i), and separating the image background;

4.2) Take the minimum value in Top(i) as the upper boundary reference of the celadon glaze layer, and shift the position of each column of pixels upward so that the row positions of all upper boundary pixels are on the same horizontal line, thereby obtaining a flattened image fl , and the minimum value in Top(i) is used as the position U of the upper boundary of the glaze layer for the measurement of the thickness of the celadon glaze layer. Figure 4 is the SD-OCT image after background separation and image flattening, and (a)-(f) correspond to (a)-(f) in Figure 3 in turn.

5) Locate the lower boundary of the glaze layer of the celadon glaze layer SD-OCT image;

5.1) Use the contrast-limited adaptive histogram equalization method to enhance the gray contrast of the flattened image fl to obtain the image fa after gray contrast enhancement, and then use the structural element se to perform morphological opening operations on the image fa to obtain the image fo , the calculation of the morphological opening operation is as follows:

In the formula, the symbol Indicates the morphological expansion operation, and the symbol Θ indicates the morphological erosion operation;

5.2) Use the flattened image fl as a mask, perform continuous expansion operations on the image fo, complete the morphological image reconstruction, and obtain the image fr;

5.3) Define a circular structural element with a radius of 3 pixels, and use this circular structural element to perform morphological expansion operation on image fr to obtain image fd;

5.4) Use the OSTU method to binarize the image fd, start traversing from the bottom of the glaze layer in the binarized image fd in row units, and traverse each row from bottom to top until the traversed row does not contain glaze Stop traversing the pixels in the region where the layer is located, and record this behavior Stop traversing the row, count the number of rows in each column where the pixel point with the smallest row position below the row where the traversal is stopped, and calculate the average value B of the number of rows in all columns, B is used as the row position of the lower boundary of the glaze layer in the SD-OCT image for the measurement of the thickness of the celadon glaze layer.

6) Calibrate the pixel axial resolution of SD-OCT images of different types of celadon glaze layers;

6.1) Set the scanning parameters of the OQ LabScope system produced by Lumedica: the horizontal scanning range is 3.5mm, the output image size is 512pixel*512pixel, and different types of celadon fragments are scanned respectively to obtain cross-sectional SD-OCT images of the fragments. Fig. 5 is embodiment celadon glaze layer sectional view and corresponding SD-OCT image, (a) is celadon glaze layer sectional view, (b) is the SD-OCT image of celadon glaze layer cross section;

6.2) Use the built-in caliper of the SD-OCT system to measure the thickness T _x of different types of celadon glaze layers;

6.3) For the same area of the scanned celadon fragment section SD-OCT image, scan the SD-OCT image of the celadon glaze layer according to step (1), and at the same time locate the upper and lower boundaries of the glaze layer according to the steps described in (2)-(5) and Find the pixel distance D between the upper and lower boundaries, calculated as follows:

D=B-U

Among them, B is the lower boundary of the glaze layer, and U is the upper boundary of the glaze layer;

6.4) Calculate the pixel axial resolution Pr _x of SD-OCT images of different types of celadon glaze layers by using the following formulas:

7) Calculate the thickness of the glaze layer.

Calculation step 1) the pixel distance D between the upper and lower boundaries of the celadon glaze layer in the celadon glaze layer SD-OCT image collected, in conjunction with the pixel axial resolution of the celadon glaze layer SD-OCT image of the scanned type in step 6), Use the formula described in step 6.4) to calculate the thickness T _x of different types of celadon glaze layers. Figure 6 is an extraction effect diagram of the upper and lower boundaries of the celadon glaze layer, and (a)-(f) correspond to (a)-(f) in Figure 3 in turn.

Claims (7)

1. a kind of celadon glazed thickness method for automatic measurement based on SD-OCT image, which comprises the following steps:

Step 1) acquires the SD-OCT image of different types of celadon glaze layer；

Step 2) carries out noise reduction to the celadon glaze layer SD-OCT image that step 1) acquires；

Glaze layer coboundary in the celadon glaze layer SD-OCT image of step 3) positioning acquisition；

Celadon glaze layer SD-OCT image is carried out background separation and image flaky process by step 4)；

Glaze layer lower boundary in the celadon glaze layer SD-OCT image of step 5) positioning acquisition；

Step 6) calibrates the pixel axial resolution of different types of celadon glaze layer SD-OCT image；

Step 7) calculates glazed thickness.

2. a kind of celadon glazed thickness measurement method based on SD-OCT image according to claim 1, it is characterised in that: The step 2) specifically: the template for the use of size being 3 × 5 carries out intermediate value to the celadon glaze layer SD-OCT image that step 1) acquires Filtering.

3. a kind of celadon glazed thickness method for automatic measurement based on SD-OCT image according to claim 1, feature It is: the step 3) specifically:

3.1) the binarization threshold t that celadon glaze layer SD-OCT image is calculated using OSTU method, using binarization threshold t as canny The threshold value of operator carries out edge detection to the SD-OCT image after step 2) noise reduction process, obtains bianry image Bw；

3.2) the circular configuration member se that a radius is 5 length in pixels is defined, shape is carried out to bianry image Bw using structural elements se State closed operation obtains image fc, calculates as follows:

Fc=(Bw ⊕ se) Θ se

In formula, symbol ⊕ indicates morphological dilation, completes the object in " growth " or " roughening " bianry image Bw；Symbol Θ It indicates morphological erosion operation, completes the object in " contraction " or " refinement " bianry image Bw；

3.3) each pixel listed first existing gray value and be 1 in image fc is searched for from top to bottom and records this conduct To match pixel point, and the line position to match pixel point for recording each column sets Top (i), and wherein i is indicated to match pixel point institute Column position；

3.4) each adjacent pixel is searched with eight connection domain modes and form connected domain, obtain all connected domains in image；

3.5) using be located in the most connected domain of number of pixels to match pixel point as the coboundary pixel of celadon glaze layer；If The coboundary pixel of celadon glaze layer is not belonging to match pixel point, then updating using Lagrange's interpolation should be to match pixel point Line position set the value of Top (i), calculate as follows:

Top (i)=2 × Top (x)-Top (y)

Wherein, i indicates the column position to match pixel point；

If the column of i-th column the right and left, there are the coboundary pixel of celadon glaze layer, x indicates that distance i-th arranges in nearest presence The left column of boundary pixel, y indicate that distance i-th arranges the right column of nearest glaze layer coboundary pixel；If there are upper for the column on the i-th column left side Coboundary pixel is not present in boundary pixel, the column on the right, then it is nearest there are the left column of coboundary pixel to indicate that distance i-th arranges by x, Y indicates that distance i-th arranges that time close there are the left columns of coboundary pixel；If glaze layer coboundary pixel is not present in the column on the i-th column left side, The column on the right are there are glaze layer coboundary pixel, then x indicates that distance i-th arranges that nearest there are the right column of glaze layer coboundary pixel, y tables Show that distance i-th arranges that time close there are the right column of glaze layer coboundary pixel.

4. a kind of celadon glazed thickness method for automatic measurement based on SD-OCT image according to claim 3, feature It is: the step 4) specifically:

4.1) for image after step 2) median filtering, column line position each in image is set to the gray value of the pixel less than Top (i) It resets, separate picture background；

4.2) using the minimum value in Top (i) as the coboundary benchmark of celadon glaze layer, by the position for translating up each column pixel It sets, sets the line position of all coboundary pixels and be in same horizontal line, to obtain flattening image fl, while by Top (i) position U of the minimum value as glaze layer coboundary in, the measurement for celadon glazed thickness.

5. a kind of celadon glazed thickness method for automatic measurement based on SD-OCT image according to claim 1, feature It is: the step 5) specifically:

5.1) using the grey-scale contrast of contrast limited adaptive histogram equalization method enhancing flattening image fl, ash is obtained The enhanced image fa of contrast is spent, morphology opening operation then is carried out to image fa using structural elements se, obtains image fo, shape The calculating of state opening operation is as follows:

Fo=(fa Θ se) ⊕ se

In formula, symbol ⊕ indicates that morphological dilation, symbol Θ indicate morphological erosion operation；

5.2) using flattening image fl as mask, expansive working is carried out to image fo, completion morphology image reconstruction obtains figure As fr；

5.3) the circular configuration member that a radius is 3 length in pixels is defined, form is carried out to image fr using this circular configuration member Expansive working is learned, image fd is obtained；

5.4) binary conversion treatment is carried out to image fd using OSTU method, with behavior unit from the image fd Jing Guo binary conversion treatment The bottommost of glaze layer is begun stepping through, and traverses every a line from the bottom to top, until the row traversed does not include the picture of glaze layer region Element then stops traversing and recording this behavior stopping traversal row, counts in each column and set most positioned at the line position for stopping traversing row lower section Line number where small pixel, and the average value B of line number in all column is found out, B is following as the glaze layer of SD-OCT image The line position on boundary is set, the measurement for celadon glazed thickness.

6. a kind of celadon glazed thickness method for automatic measurement based on SD-OCT image according to claim 1, feature It is: the step 6) specifically:

6.1) the celadon fragment of difference scan different types, obtains the section SD-OCT image of celadon fragment；

6.2) using the thickness T of glaze layer in the section SD-OCT image of the calliper to measure celadon fragment of SD-OCT system_x；

6.3) the celadon glaze layer SD-OCT image with section SD-OCT image the same area of celadon fragment is scanned, while according to step It is rapid 2)-up-and-down boundary of the method for step 5) positioning celadon glaze layer and find out the pixel distance D between up-and-down boundary, calculate such as Under:

D=B-U

Wherein, B is glaze layer lower boundary, and U is glaze layer coboundary；

6.4) the pixel axial resolution Pr of different types of celadon glaze layer SD-OCT image is found out respectively using following formula_x:

7. a kind of celadon glazed thickness method for automatic measurement based on SD-OCT image according to claim 1, feature It is: the step 7) specifically: calculated in the celadon glaze layer SD-OCT image of step 1) scanning according to the following formula in glaze layer Pixel distance D between lower boundary:

D=B-U

Wherein, B is glaze layer lower boundary, and U is glaze layer coboundary；

The pixel axis of celadon glaze layer SD-OCT image identical with the celadon glaze channel type that step 1) scans is chosen in step 6) To resolution ratio Pr_x, then it is calculated using the following equation different types of celadon glazed thickness T_x:

Wherein, Pr_xFor step 1) scanning celadon glaze layer SD-OCT image pixel axial resolution, D be glaze layer up-and-down boundary it Between pixel distance D.