A piston detection method

Abstract

The present application relates to a kind of piston detection methods, including detecting the defect of piston, specifically: step 1, construct CT image segmentation model, and the CT image segmentation model constructed is trained, and the CT image segmentation model after training is completed is obtained;Step 2, S piece of piston is placed on fixed tooling and is detected, each layer of fixed tooling can place N piece of piston, S and N are all positive integers, the inner cooling oil channel top, middle and bottom of the piston placed in each layer of fixed tooling are scanned once respectively, obtain multiple CT images, and all CT images obtain double-channel image, and the double-channel image obtained is input into the CT image segmentation model after training is completed, i.e. obtain predicted defect.The advantage is that: the method improves detection accuracy and detection efficiency.

Description

A piston detection method Technical Field

[0001] This invention relates to the field of detection technology, and in particular to a piston detection method. Background Technology

[0002] The piston is a core component of the engine, operating under high temperature and high pressure conditions, and its quality directly affects the engine's performance. The piston's internal cooling oil passages are generally formed using a water-soluble salt core method. By injecting engine oil, the oil vibrates inside the passages, absorbing heat before flowing out. This effectively reduces the temperature and thermal deformation of important parts such as the piston head and ring grooves, but it also correspondingly weakens the strength of the piston head structure.

[0003] Significant axial or radial misalignment of the piston's internal cooling oil passages can lead to the corresponding misaligned side being easily punctured during operation, potentially causing piston failure. Furthermore, casting residues (aluminum inclusions, residual iron, residual salt) within the internal cooling oil passages can cause blockages, preventing timely heat dissipation from the piston. Casting defects on the piston body (porosity, voids, inclusions, cracks) can also reduce the piston's mechanical properties, posing significant safety hazards.

[0004] In industrial applications, industrial CT inspection technology can effectively inspect the internal quality of pistons. During industrial CT inspection, the piston is first subjected to digital X-ray inspection, and the axial displacement of the internal cooling oil passages is measured based on the digital X-ray projection. Then, tomographic scans are performed at the top, middle, and bottom of the internal cooling oil passages to determine the presence of defects. Finally, the radial displacement of the internal cooling oil passages is measured based on the tomographic image of the middle section. The inspection process requires multiple workers to collaborate in measuring, recording relevant data, and evaluating the internal quality of the piston. However, due to the large production batches of pistons, manual evaluation is inefficient and costly, and the evaluation results are limited by the experience of the inspectors, making them prone to measurement errors, false positives, and missed detections. Furthermore, traditional image processing techniques are also susceptible to errors due to noise and structural artifacts. Therefore, further improvements to existing technologies are needed. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a piston detection method that can improve detection accuracy and detection efficiency, in contrast to the above-mentioned prior art.

[0006] The technical solution adopted by this invention to solve the above-mentioned technical problems is as follows: a piston detection method, characterized by including the detection of defects in the piston, specifically the following steps:

[0007] Step 1: Construct a CT image segmentation model and train the constructed CT image segmentation model to obtain the trained CT image segmentation model;

[0008] The specific process of training the CT image segmentation model is as follows:

[0009] Step 1-1: Acquire a certain number of CT images of the pistons and identify the piston outline in each CT image;

[0010] Steps 1-2: Perform the following cropping operation on each CT image:

[0011] Using the center of the piston circle identified in the current CT image as the center of the piston image corresponding to the current CT image, and the piston diameter as the length of the piston image corresponding to the current CT image, a square piston image corresponding to the current CT image is obtained by cutting it.

[0012] Steps 1-3: Enhance the local contrast of the piston image obtained in Step 1-2 to obtain a piston grayscale image;

[0013] Steps 1-4: Construct the distance feature matrix for each piston image, assign the pixel value of any point in the matrix to the Euclidean distance from that point to the center of the piston image, and uniformly map the distance feature matrix onto a grayscale space of 0 to 255 to obtain the distance feature grayscale image.

[0014] Steps 1-5: Merge the piston grayscale image and the distance feature grayscale image into a dual-channel image;

[0015] Steps 1-6: Input the dual-channel image into the constructed CT image segmentation model to obtain the prediction result, and perform defect annotation on the piston image obtained in Steps 1-2 to obtain the piston defect mask image; calculate the loss function using the prediction result and the piston defect mask image, and then use the loss function to update the parameters of the constructed CT image segmentation model to obtain the CT image segmentation model after training.

[0016] Step 2: Place S pistons on a fixed fixture for inspection. Each layer of the fixture can hold N pistons, where S and N are both positive integers. Perform a CT scan on the top, middle, and bottom of the internal cooling oil passage of each piston placed on the fixed fixture to obtain multiple CT images. Then, acquire dual-channel images of all CT images in the same way as in steps 1-1 to 1-5. Input the acquired dual-channel images into the trained CT image segmentation model to obtain the predicted defects.

[0017] Preferably, in step 1-1, Hough circle transform is performed on all CT images to identify the piston contour within each CT image.

[0018] Furthermore, it also includes calculating the axial offset value of the piston's internal cooling oil passage, specifically:

[0019] A digital ray image segmentation model is constructed by acquiring a certain number of digital ray images of pistons and annotating the oil passage contours and piston cavity contours within the digital ray images of pistons to form a dataset. The constructed digital ray image segmentation model is trained using the dataset to obtain the trained digital ray image segmentation model.

[0020] Digital ray scanning is performed on the pin holes of multiple pistons to be tested in the direction facing each other to obtain piston digital ray images. The piston digital ray images are then input into the trained digital ray image segmentation model to obtain the oil passage contour and piston inner cavity contour in the piston digital ray images.

[0021] The oil passage contour in the piston digital ray image is analyzed using a connected component detection method to obtain 2S connected components. The center coordinates of each connected component are calculated, and the 2S connected components are numbered 1 to 2S according to their center coordinates. The relationship between the center coordinates is as follows:

[0022] when hour:

[0023] x1 <x2<…<x 2S

[0024] Where x1 is the x-coordinate of the center of the first connected component, x2 is the x-coordinate of the center of the second connected component, and x... 2S The x-coordinate of the center of the 2Sth connected component;

[0025] when hour:

[0026]

[0027] Where j is a positive integer greater than 1. is a universal quantifier; k∈{0,1,2…j-1}, (x1,y1) is the center coordinate of the first connected component;

[0028] Calculate the axial offset of the internal cooling oil passages for each piston from 1 to S:

[0029] d 轴向偏移 (1)=|y max (1)-y max (2)|

[0030] d 轴向偏移 (2)=|y max (3)-y max (4)|

[0031] ...

[0032] d 轴向偏移 (S)=|ymax (2S - 1)-y max (2S)|

[0033] where d 轴向偏移 (1) is the axial offset value of the internal cooling oil passage of the first piston, y max (1) represents the maximum value of the ordinate within the first connected domain, y max (2) represents the maximum value of the ordinate within the second connected domain, d 轴向偏移 (2) is the axial offset value of the internal cooling oil passage of the second piston, y max (3) represents the maximum value of the ordinate within the third connected domain, y max (4) represents the maximum value of the ordinate within the fourth connected domain, d 轴向偏移 (S) is the axial offset value of the internal cooling oil passage of the S - th piston, y max (2S - 1) represents the maximum value of the ordinate within the (2S - 1)-th connected domain, y max (2S) represents the maximum value of the ordinate within the 2S - th connected domain.

[0034] Furthermore, it also includes verifying defects near the central region of the piston inner cavity. The specific verification steps are as follows:

[0035] Step a: According to the piston inner cavity contour in the obtained digital radiography image of the piston, obtain the cumulative distribution function P(x) of the piston inner cavity part between (x1, x2), and based on the piston CT tomographic scan height H and the tomographic scan slice thickness d, obtain the coordinates on the height of the piston inner cavity

[0036] Step b: For the rectangle formed by the four points and calculate the cumulative gray value of each column within the rectangle:

[0037] where ρ1 and ρ2 respectively represent the gray values of the cavity part and the piston matrix in the CT image;

[0038]

[0039] where ρ1 and ρ2 respectively represent the gray values of the cavity part and the piston matrix in the CT image;

[0040] Step c: Calculate the average gray value of each column as

[0041] Step d: Taking the axis region of the piston as the abscissa and the piston center as the coordinate center, construct a coordinate system on the piston CT cut image, and extract the gray value change curve S′(x) in the y = 0 direction of the defect region, with the range x1′ < x < x2′, where the identified defect range (x1′, x2′) ∈ (x1, x2);

[0042] Step e, calculate S′(x) and The similarity SSIM is calculated using the following formula:

[0043]

[0044] in, express The average value within (x1′, x2′), μ S′(x) Let S′(x) be the average value of S′(x) in (x1′, x2′). express The covariance of S′(x) in (x1′, x2′), where C1 and C2 represent preset constants. express The square of the standard deviation within (x1′, x2′), σ S′(x) 2 Let S'(x) represent the square of the standard deviation of S'(x) in (x1', x2').

[0045] Step f: Determine whether SSIM is less than the preset constant C3. If yes, it is considered that there is a defect near the center area of ​​the piston cavity; otherwise, it is considered that there is no defect in the center area of ​​the piston cavity.

[0046] Furthermore, it also includes calculating the radial offset value of the piston's internal cooling oil passage, specifically:

[0047] The piston image is obtained by cutting it out in the manner described in steps 1-2, and the CT tomographic image of the middle part of the internal cooling oil passage of the piston image is extracted.

[0048] A circular fit was performed on the piston outer wall within the CT tomographic image of the middle section of the internal cooling oil passage, and the coordinates of the fitted circle's center were calculated.

[0049] Calculate the coordinates of the fitted circle center The Euclidean distance between the piston center and the center of the piston in steps 1-2 is the radial offset value of the piston's internal cooling oil passage.

[0050] Compared with existing technologies, the advantages of this invention are as follows: By constructing a CT image segmentation model, training the constructed CT image segmentation model, and predicting piston defects using the trained CT image segmentation model, this method utilizes image processing technology to detect pistons without requiring multiple personnel to collaborate in measuring, recording relevant data, and evaluating the internal quality of the piston. The detection results are not limited by the experience of the inspectors. Furthermore, by merging the piston grayscale image and the distance feature grayscale image into a dual-channel image, interference from noise and structural artifacts can be effectively avoided, thus improving detection accuracy and efficiency. On the other hand, by placing multiple pistons on a fixed fixture for detection, multiple pistons can be detected at once, further improving detection efficiency. Attached Figure Description

[0051] Figure 1 is a piston image in an embodiment of the present invention;

[0052] Figure 2 is the piston defect mask image corresponding to the piston image in Figure 1;

[0053] Figure 3 is a grayscale image of the distance feature in an embodiment of the present invention;

[0054] Figure 4 is a digital ray image of the piston in an embodiment of the present invention;

[0055] Figure 5 shows the oil passage profile obtained from Figure 4;

[0056] Figure 6 is a CT tomographic image of the middle part of the internal cooling oil passage in an embodiment of the present invention. Detailed Implementation

[0057] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0058] The piston detection method in this embodiment includes detecting defects in the piston, calculating the axial offset value of the piston's internal cooling oil passage, verifying defects near the center region of the piston's internal cavity, and calculating the radial offset value of the piston's internal cooling oil passage.

[0059] In this embodiment, the specific steps for detecting defects in the piston are as follows:

[0060] Step 1: Construct a CT image segmentation model and train the constructed CT image segmentation model to obtain the trained CT image segmentation model;

[0061] The CT image segmentation model in this embodiment can use a deep learning network commonly used in the prior art, which will not be elaborated here; for example: U-Net network;

[0062] The specific process of training the CT image segmentation model is as follows:

[0063] Step 1-1: Acquire a certain number of CT images of the piston and identify the piston outline in each CT image; In this embodiment, perform Hough circle transform on all CT images to identify the piston outline in each CT image.

[0064] Steps 1-2: Perform the following cropping operation on each CT image:

[0065] Using the center of the piston circle identified in the current CT image as the center of the piston image corresponding to the current CT image, and the piston diameter as the length of the piston image corresponding to the current CT image, a square piston image corresponding to the current CT image is obtained by cutting it.

[0066] Steps 1-3: Enhance the local contrast of the piston image obtained in Step 1-2 to obtain a piston grayscale image;

[0067] In this embodiment, contrast-limited adaptive histogram equalization is used to enhance the local contrast of the piston image.

[0068] Steps 1-4: Construct the distance feature matrix for each piston image, assign the pixel value of any point in the matrix to the Euclidean distance from that point to the center of the piston image, and uniformly map the distance feature matrix onto a grayscale space of 0 to 255 to obtain the distance feature grayscale image; the distance feature grayscale image is shown in Figure 3.

[0069] Steps 1-5: Merge the piston grayscale image and the distance feature grayscale image into a dual-channel image;

[0070] Steps 1-6: Input the dual-channel image into the constructed CT image segmentation model to obtain the prediction result, and annotate the piston image obtained in Steps 1-2 to obtain the piston defect mask image; calculate the loss function using the prediction result and the piston defect mask image, and then use the loss function to update the parameters of the constructed CT image segmentation model to obtain the CT image segmentation model after training; Figure 1 shows one of the piston images, and the piston defect mask image corresponding to this piston image is shown in Figure 2;

[0071] Step 2: Place S pistons on the fixed fixture for inspection. Each layer of the fixed fixture can hold N pistons, where S and N are both positive integers. Perform a CT scan on the top, middle, and bottom of the internal cooling oil passage of each piston placed on the fixed fixture to obtain multiple CT images. Obtain dual-channel images from all CT images in the same way as in steps 1-1 to 1-5. Input the obtained dual-channel images into the trained CT image segmentation model to obtain the predicted defect. In this embodiment, each layer of the fixed fixture can hold 3 pistons, and there are 4 layers in total, with 12 pistons placed at a time.

[0072] The axial offset of the piston's internal cooling oil passage is calculated as follows:

[0073] A digital ray image segmentation model is constructed by acquiring a certain number of digital ray images of pistons and annotating the oil passage contours and piston cavity contours within the digital ray images of pistons to form a dataset. The constructed digital ray image segmentation model is trained using the dataset to obtain the trained digital ray image segmentation model. The digital ray image segmentation model in this embodiment also adopts the deep learning network commonly used in the prior art, which will not be elaborated here.

[0074] Digital ray scanning is performed on the pin holes of multiple pistons to be tested in the direction of facing each other to obtain piston digital ray images. The piston digital ray images are then input into the trained digital ray image segmentation model to obtain the oil passage contour and piston inner cavity contour in the piston digital ray images. As shown in Figure 4, the piston digital ray images include digital ray images of 12 pistons. Figure 5 shows the oil passage contour obtained from Figure 4.

[0075] The oil passage contour in the piston digital ray image is analyzed using a connected component detection method to obtain 2S connected components. The center coordinates of each connected component are calculated, and the 2S connected components are numbered 1 to 2S according to their center coordinates. The relationship between the center coordinates is as follows:

[0076] when hour:

[0077] x1 <x2<…<x 2S

[0078] Where x1 is the x-coordinate of the center of the first connected component, x2 is the x-coordinate of the center of the second connected component, and x... 2S The x-coordinate of the center of the 2Sth connected component;

[0079] when hour:

[0080]

[0081] Where j is a positive integer greater than 1. is a universal quantifier; k∈{0,1,2…j-1}, (x1,y1) is the center coordinate of the first connected component;

[0082] Calculate the axial offset of the internal cooling oil passages for each piston from 1 to S:

[0083] d 轴向偏移 (1)=|y max (1)-y max (2)|

[0084] d轴向偏移 (2)=|y max (3)-y max (4)|

[0085] ...

[0086] d 轴向偏移 (S)=|y max (2S-1)-y max (2S)|

[0087] Where, d 轴向偏移 (1) is the axial offset value of the internal cooling oil passage of the first piston, y max (1) represents the maximum value of the ordinate in the first connected component, y max (2) represents the maximum value of the ordinate in the second connected component, d 轴向偏移 (2) is the axial offset value of the internal cooling oil passage of the second piston, y max (3) represents the maximum value of the ordinate in the third connected component, y max (4) represents the maximum value of the ordinate in the 4th connected component, d 轴向偏移 (S) represents the axial offset value of the internal cooling oil passage of the S-th piston, y max (2S-1) represents the maximum value of the ordinate in the (2S-1)th connected component, y max (2S) represents the maximum value of the ordinate in the 2Sth connected component.

[0088] The specific steps for verifying defects near the center region of the piston cavity are as follows:

[0089] Step a: Based on the piston cavity contour in the obtained digital X-ray image of the piston, obtain the cumulative distribution function P(x) of the piston cavity portion between (x1, x2), and based on the piston CT scan height H and the scan slice thickness d, obtain the piston cavity... coordinates at altitude

[0090] Step b, for in and Within the rectangle formed by the four points,

[0091] Cumulative grayscale value for each column:

[0092]

[0093] Wherein, ρ1 and ρ2 represent the gray values ​​of the cavity portion and the piston base in the CT image, respectively;

[0094] Step c: Calculate the average grayscale value for each column.

[0095] Step d: Taking the axis region of the piston as the abscissa and the center of the piston as the coordinate center, a coordinate system is constructed on the piston CT cut image, and the gray value change curve S′(x) in the y = 0 direction of the defect region is extracted, with the range x1′ < x < x2′, where the identified defect range (x1′, x2′) ∈ (x1, x2);

[0096] Step e: Calculate the similarity SSIM between S′(x) and , and the calculation formula is:

[0097]

[0098] where, represents the average value within (x1′, x2′), μ represents the average value of S′(x) within (x1′, x2′), S′(x) represents the covariance between and S′(x) within (x1′, x2′), and C1 and C2 represent preset constants, represents the square of the standard deviation within (x1′, x2′), σ S′(x) 2 2

[0099] Step f: Determine whether SSIM is less than the preset constant C3. If so, it is considered that there is a defect near the center region of the piston inner cavity; if not, it is determined that there is no defect in the center region of the piston inner cavity. The specific value of C3 above is confirmed according to experience.

[0100] Calculate the radial offset value of the internal cooling oil channel of the piston, specifically:

[0101] Cut the piston image in the manner of Steps 1 - 2, and extract the CT tomographic image in the middle of the internal cooling oil channel of the piston image; [[ID=4O]]

[0102] Perform circular fitting on the outer wall of the piston in the CT tomographic image in the middle of the internal cooling oil channel, and calculate the center coordinates of the fitting as shown in Figure 6;

[0103] Calculate the center coordinates of the fitting The Euclidean distance between and the center of the piston in Steps 1 - 2, and this Euclidean distance is the radial offset value of the internal cooling oil channel of the piston.

[0104] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A piston detection method, characterized in that... The process includes detecting defects in pistons. The specific steps are as follows: Step 1: Construct a CT image segmentation model and train it to obtain the trained CT image segmentation model. The specific training process for the CT image segmentation model is as follows: Step 1-1: Acquire a certain number of CT images of pistons and identify the piston contour in each CT image; Step 1-2: Perform the following cutting operation on each CT image: using the center of the piston circle identified in the current CT image as the center of the piston image corresponding to the current CT image, and the piston diameter as the length of the piston image corresponding to the current CT image, cut to obtain a square piston image corresponding to the current CT image; Step 1-3: Enhance the local contrast of the piston image obtained in Step 1-2 to obtain a piston grayscale image; Step 1-4: Construct a distance feature matrix for each piston image, assign the pixel value of any point in the matrix to the Euclidean distance from that point to the center of the piston image, and uniformly map the distance feature matrix to 0~255. In the grayscale space, a distance feature grayscale image is obtained; Steps 1-5: The piston grayscale image and the distance feature grayscale image are merged into a dual-channel image; Steps 1-6: The dual-channel image is input into the constructed CT image segmentation model to obtain the prediction result, and the piston image obtained in Step 1-2 is labeled with defects to obtain a piston defect mask image; The loss function is calculated using the prediction result and the piston defect mask image, and then the parameters of the constructed CT image segmentation model are updated using the loss function to obtain the CT image segmentation model after training; Step 2: S pistons are placed on a fixed fixture for detection. N pistons are placed in each layer of the fixed fixture, where S and N are both positive integers. A CT tomographic scan is performed on the top, middle and bottom of the internal cooling oil passage of each piston placed in each layer of the fixed fixture to obtain multiple CT images. All CT images are used to obtain dual-channel images in the same way as in Steps 1-1 to 1-5, and the obtained dual-channel images are input into the CT image segmentation model after training to obtain the predicted defects.

2. The piston detection method according to claim 1, characterized in that: In step 1-1, Hough circle transform is performed on all CT images to identify the piston contour in each CT image.

3. The piston detection method according to claim 1, characterized in that: This also includes calculating the axial offset value of the piston's internal cooling oil passage. Specifically: A digital ray image segmentation model is constructed; a certain number of digital ray images of the piston are acquired, and the oil passage contours and piston cavity contours within the digital ray images are labeled to form a dataset; the constructed digital ray image segmentation model is trained using the dataset to obtain the trained digital ray image segmentation model; digital ray scanning is performed on the pin holes of multiple pistons to be detected in the direction directly opposite to them to obtain piston digital ray images, and these images are input into the trained digital ray image segmentation model to obtain the oil passage contours and piston cavity contours within the piston digital ray images; the oil passage contours in the piston digital ray images are analyzed using a connected component detection method to obtain 2S connected components; the center coordinates of each connected component are calculated, and for the 2S connected components, they are numbered 1 to 2S according to their center coordinates. The relationship between the center coordinates is: when... Time: x1 < x2 < ... < x 2S Where x1 is the x-coordinate of the center of the first connected component, x2 is the x-coordinate of the center of the second connected component, and x... 2S Let x be the x-coordinate of the center of the 2Sth connected component; when hour: Where j is a positive integer greater than 1. The quantifier is a universal quantifier; k∈{0,1,2…j-1}, (x1,y1) is the center coordinate of the first connected domain; calculate the axial offset value of the internal cooling oil passage for each piston 1~S: d 轴向偏移 (1)=|y max (1)-y max (2)|d 轴向偏移 (2)=|y max (3)-y max (4)|……d 轴向偏移 (S)=|y max (2S-1)-y max (2S)|wherein, d 轴向偏移 (1) is the axial offset value of the internal cooling oil passage of the first piston, y max (1) represents the maximum value of the ordinate in the first connected component, y max (2) represents the maximum value of the ordinate in the second connected component, d 轴向偏移 (2) is the axial offset value of the internal cooling oil passage of the second piston, y max (3) represents the maximum value of the ordinate in the third connected component, y max (4) represents the maximum value of the ordinate in the 4th connected component, d 轴向偏移 (S) represents the axial offset value of the internal cooling oil passage of the S-th piston, y max (2s-1) represents the maximum value of the ordinate in the (2s-1)th connected component, y max (2S) represents the maximum value of the ordinate in the 2Sth connected component.

4. The piston detection method according to claim 3, characterized in that: It also includes verifying defects near the center region of the piston cavity. The specific verification steps are as follows: Step a: Based on the piston cavity contour in the obtained digital X-ray image of the piston, obtain the cumulative distribution function P(x) of the piston cavity portion between (x1, x2), and obtain the piston cavity based on the piston CT scan height H and the scan slice thickness d. coordinates at altitude Step b, for in and The cumulative grayscale value of each column within the rectangle formed by the four points: Where ρ1 and ρ2 represent the grayscale values ​​of the cavity portion and piston base in the CT image, respectively; step c, calculate the average grayscale value of each column as follows: Step d: Construct a coordinate system on the piston CT cut image with the piston's axial region as the abscissa and the piston's center as the coordinate center. Extract the gray value change curve S′(x) of the defect area in the y=0 direction, with a range of x1′<x<x2′, where the identified defect range (x1′,x2′)∈(x1,x2); Step e: Calculate S′(x) and The similarity SSIM is calculated using the following formula: in, express The average value within (x1′, x2′), μ S′(x) Let S′(x) be the average value of S′(x) in (x1′, x2′). express The covariance of S′(x) in (x1′, x2′), where C1 and C2 represent preset constants. express The square of the standard deviation within (x1′, x2′), σ s′(x) 2 This represents the square of the standard deviation of S′(x) within (x1′, x2′); Step f: Determine whether SSIM is less than the preset constant C3. If so, it is considered that there is a defect near the center region of the piston cavity; if not, it is considered that there is no defect in the center region of the piston cavity.

5. The piston detection method according to any one of claims 1 to 4, characterized in that: This also includes calculating the radial offset value of the piston's internal cooling oil passage. Specifically, the piston image is obtained by cutting it out as described in steps 1-2, and the CT tomographic image of the middle part of the internal cooling oil passage in the piston image is extracted; a circle is fitted to the piston outer wall in the CT tomographic image of the middle part of the internal cooling oil passage, and the coordinates of the fitted circle center are calculated. Calculate the coordinates of the fitted circle center The Euclidean distance between the piston center and the center of the piston in steps 1-2 is the radial offset value of the piston's internal cooling oil passage.

Images