A method for acquiring images at the moment of load peak in fatigue test

Abstract

The application belongs to the field of crack image recognition in fatigue test, and particularly relates to a method for acquiring images at the moment of load peak value in fatigue test. The method comprises the following steps: collecting crack images, selecting reference frame images and target frame images; determining crack near-field regions of the reference frame images and the target frame images; extracting feature points of the crack near-field regions of the reference frame images and the target frame images, and performing feature point screening; calculating the reference feature point centroid and the target feature point centroid; calculating the offset of the target feature point centroid relative to the reference feature point centroid in the fatigue load loading direction; and selecting the target frame image corresponding to the maximum offset as the crack image corresponding to the moment of fatigue load peak value. The method for acquiring images at the moment of load peak value in fatigue test can independently and effectively recognize the image of the maximum crack opening state at the moment of fatigue load peak value.

Description

Technical Field

[0001] This application belongs to the field of crack image recognition in fatigue testing, and specifically relates to a method for acquiring images at the peak load moment in fatigue testing. Background Technology

[0002] During service, aircraft are subjected to fatigue loads, which can cause fatigue cracks to form in their structures. The propagation of fatigue cracks can significantly reduce the strength of the structure. In order to effectively evaluate the damage tolerance performance of aircraft structures, it is necessary to conduct targeted structural damage tolerance tests. During the test, continuously acquiring digital images of cracks in the structure and obtaining the characteristic parameters of the cracks through image analysis is an important technical means. The crack image at the peak of the fatigue load corresponds to the maximum crack opening state and is a very important crack state image in the crack analysis process.

[0003] Currently, crack images at the peak of fatigue load are obtained by communicating with the testing machine to trigger camera acquisition. However, effective acquisition cannot be completed when the testing machine interface is closed, which affects the work of analyzing structural fatigue characteristics based on digital images.

[0004] Therefore, it is desirable to have a technical solution to overcome or at least mitigate one of the aforementioned defects of the prior art. Summary of the Invention

[0005] The purpose of this application is to provide a method for obtaining images of the peak load moment in fatigue testing, in order to solve at least one problem existing in the prior art.

[0006] The technical solution of this application is:

[0007] A method for acquiring images of the peak load moment in a fatigue test, comprising:

[0008] Step 1: Acquire crack images of time-series frames within the fatigue load loading cycle, and select reference frame images and multiple target frame images from the crack images;

[0009] Step 2: Determine the near-field region of the crack in the reference frame image and the target frame image respectively;

[0010] Step 3: Extract feature points from the crack near-field region of the reference frame image and the target frame image respectively, and filter feature points according to the feature point matching relationship between the reference frame image and the target frame image;

[0011] Step 4: Calculate the centroid of the reference feature points in the reference frame image and the centroid of the target feature points in the target frame image based on the selected feature points;

[0012] Step 5: Calculate the offset of the centroid of the target feature point relative to the centroid of the reference feature point in the fatigue load loading direction;

[0013] Step 6: Repeat steps 2 to 5 to calculate the offset of the centroid of the target feature point relative to the centroid of the reference feature point in the fatigue load loading direction for all target frame images, and select the target frame image corresponding to the maximum offset as the crack image corresponding to the peak fatigue load moment.

[0014] In at least one embodiment of this application, step one involves acquiring crack images of time-series frames within a fatigue load loading cycle, and selecting a reference frame image and multiple target frame images from the crack images, including:

[0015] S101. Obtain the fatigue load loading frequency and select a CCD camera with a frame rate greater than twice the fatigue load loading frequency.

[0016] S102. Capture crack images of time-series frames within at least one fatigue load loading cycle using a CCD camera;

[0017] S103. The first crack image in the crack images of the time series frames is determined as the reference frame image, and the crack images of other frames are determined as the target frame images.

[0018] In at least one embodiment of this application, step two, determining the crack near-field region of the reference frame image and the target frame image respectively, includes:

[0019] The crack in the reference frame image and the target frame image is located at the center of the crack near-field region, and the length of the crack near-field region is 1.5 times the crack length, and the width of the crack near-field region is 0.75 times the crack length.

[0020] In at least one embodiment of this application, step three involves extracting feature points from the crack near-field region of the reference frame image and the target frame image, respectively, and filtering feature points based on the feature point matching relationship between the reference frame image and the target frame image, including:

[0021] S301. Extract SIFT feature points of the crack near-field region of the reference frame image and the target frame image respectively;

[0022] S302. Calculate the matching point pairs between the reference frame image and the target frame image based on the SIFT feature points;

[0023] S303. Construct a homography matrix constraint model between the reference frame image and the target frame image;

[0024] S304. Based on the homography matrix constraint model, select matching point pairs between the reference frame image and the target frame image that satisfy the constraint conditions.

[0025] In at least one embodiment of this application, in step S302, calculating the matching point pair corresponding to the reference frame image and the target frame image based on the SIFT feature points includes:

[0026] Select SIFT feature point A in the reference frame image, and find SIFT feature point B with the smallest Euclidean distance and SIFT feature point C with the second smallest Euclidean distance from SIFT feature point A in the target frame image. If the following conditions are met:

[0027] Min / SecMin < 1.0

[0028] Where Min is the Euclidean distance between SIFT feature point A and SIFT feature point B, and SecMin is the Euclidean distance between SIFT feature point A and SIFT feature point C.

[0029] Then SIFT feature point A and SIFT feature point B are defined as a matching point pair.

[0030] In at least one embodiment of this application, in step S303, constructing a homography matrix constraint model between the reference frame image and the target frame image includes:

[0031] The homography matrix constraint model between the reference frame image and the target frame image is as follows:

[0032] ;

[0033] ;

[0034] ;

[0035] ;

[0036] ;

[0037] in, H It is a homography matrix. The coordinates of feature points in the reference frame image. The coordinates of the feature points in the target frame image. .

[0038] In at least one embodiment of this application, step S304, selecting matching point pairs between the reference frame image and the target frame image that satisfy the constraint conditions according to the homography matrix constraint model, includes:

[0039] S3041. Randomly select 4 sets of matching point pairs from the matching point pairs corresponding to the reference frame image and the target frame image, and calculate the homography matrix according to the homography matrix constraint model. H ;

[0040] S3042. Select points from the 4 sets of matched point pairs that satisfy the Euclidean distance. L Matching point pairs < 6 pixels are selected as those satisfying the homography matrix. H Define the interior points of the constraints and record the number of interior points;

[0041] S3043, Repeat steps S3041-S3042 multiple times, selecting the homography matrix corresponding to the maximum number of interior points. H Interior points of the constraint.

[0042] In at least one embodiment of this application, step four, calculating the centroid of the reference feature points of the reference frame image and the centroid of the target feature points of the target frame image based on the selected feature points, includes:

[0043] The centroid of the reference feature points of the reference frame image for:

[0044] ;

[0045] ;

[0046] The centroid of the target feature points in the target frame image for:

[0047] ;

[0048] ;

[0049] in, N This represents the number of interior points.

[0050] In at least one embodiment of this application, step five, calculating the offset of the centroid of the target feature point relative to the centroid of the reference feature point in the fatigue load loading direction, includes:

[0051] ;

[0052] ;

[0053] in, The offset of the centroid of the target feature point relative to the centroid of the reference feature point in the fatigue load loading direction.

[0054] The invention has at least the following beneficial technical effects:

[0055] The fatigue test image acquisition method of this application can independently and effectively identify the image of the maximum crack opening state at the peak fatigue load moment, which helps to improve the accuracy of crack analysis and monitoring and improve the digital level of damage tolerance test. Attached Figure Description

[0056] Figure 1 This is a flowchart of a method for acquiring images of the peak load time in a fatigue test according to one embodiment of this application;

[0057] Figure 2 This is a schematic diagram of the selection of the near-field region of a crack according to one embodiment of this application;

[0058] Figure 3 This is a schematic diagram of the feature point coordinates of one embodiment of this application. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be described in more detail below with reference to the accompanying drawings. In the drawings, the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The described embodiments are some, but not all, embodiments of this application. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. The embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0060] In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting the scope of protection of this application.

[0061] The following is in conjunction with the appendix Figures 1 to 3 This application will be described in further detail.

[0062] This application provides a method for obtaining images of the peak load moment in fatigue testing, such as... Figure 1 As shown, it includes the following steps:

[0063] Step 1: Acquire crack images of time-series frames within the fatigue load loading cycle, and select reference frame images and multiple target frame images from the crack images;

[0064] Step 2: Determine the near-field region of the crack in the reference frame image and the target frame image respectively;

[0065] Step 3: Extract feature points from the crack near-field region of the reference frame image and the target frame image respectively, and filter feature points according to the feature point matching relationship between the reference frame image and the target frame image;

[0066] Step 4: Calculate the centroid of the reference feature points in the reference frame image and the centroid of the target feature points in the target frame image based on the selected feature points.

[0067] Step 5: Calculate the offset of the centroid of the target feature point relative to the centroid of the reference feature point in the fatigue load loading direction;

[0068] Step 6: Repeat steps 2 to 5 to calculate the offset of the centroid of the target feature point relative to the centroid of the reference feature point in the fatigue load loading direction for all target frame images. Select the target frame image with the largest offset as the crack image corresponding to the peak fatigue load moment.

[0069] The method for acquiring images at the peak load time in fatigue testing according to this application firstly involves acquiring crack images of time-series frames within the fatigue load loading period, and selecting reference frame images and multiple target frame images from the crack images. Specifically, this includes:

[0070] S101. Obtain the fatigue load loading frequency and select a CCD camera with a frame rate greater than twice the fatigue load loading frequency.

[0071] S102. Capture crack images of time-series frames within at least one fatigue load loading cycle using a CCD camera;

[0072] S103. The first crack image in the crack images of the time series frames is determined as the reference frame image, and the crack images of other frames are determined as the target frame images.

[0073] When taking pictures with a CCD camera, first place the specimen in the clamp of the testing machine to ensure that the specimen is vertical, and adjust the height and object distance of the CCD camera so that the shooting area is in the center of the field of view and the camera line of sight is perpendicular to the plane of the specimen. If the light at the test site is insufficient, artificial lighting is required.

[0074] In the fatigue test image acquisition method of this application, after acquiring the crack image, the near-field region of the crack in the image is determined based on the length and location of the crack in the reference frame image and the target frame image. Figure 2As shown, the crack in the reference frame image and the target frame image is located at the center of the crack near-field region, and the length of the crack near-field region is 1.5 times the crack length, and the width of the crack near-field region is 0.75 times the crack length.

[0075] In the fatigue test image acquisition method of this application, step three involves extracting feature points from the crack near-field region of the reference frame image and the target frame image, respectively, and filtering feature points based on the feature point matching relationship between the reference frame image and the target frame image. Specifically, this includes:

[0076] S301. Extract SIFT feature points of the crack near-field region from the reference frame image and the target frame image respectively;

[0077] Based on the crack near-field region obtained in step two, feature points of the crack near-field region in the reference frame image and the target frame image are obtained. In this embodiment, firstly, a size space pyramid is obtained by convolving the image with a scale-variable Gaussian function composed of different scale spatial factors. Then, the principal direction of the feature points is calculated and SIFT feature descriptors are generated. Finally, SIFT feature points are calculated. The calculation of SIFT feature points can use the corresponding functions in the open-source vision library OpenCV.

[0078] S302. Calculate the matching point pairs between the reference frame image and the target frame image based on the SIFT feature points;

[0079] In this embodiment, the feature point matching relationship between the reference frame image and the target frame image is established in the following way:

[0080] Select SIFT feature point A in the reference frame image, and find SIFT feature point B with the smallest Euclidean distance and SIFT feature point C with the second smallest Euclidean distance from SIFT feature point A in the target frame image. If the following conditions are met:

[0081] Min / SecMin < 1.0

[0082] Where Min is the Euclidean distance between SIFT feature point A and SIFT feature point B, and SecMin is the Euclidean distance between SIFT feature point A and SIFT feature point C.

[0083] Then SIFT feature point A and SIFT feature point B are defined as a matching point pair.

[0084] S303. Construct a homography matrix constraint model between the reference frame image and the target frame image;

[0085] In this embodiment, the homography matrix constraint model between the reference frame image and the target frame image is as follows:

[0086] ;

[0087] ;

[0088] ;

[0089] ;

[0090] ;

[0091] in, H It is a homography matrix. The coordinates of feature points in the reference frame image. The coordinates of the feature points in the target frame image. .

[0092] S304. Based on the homography matrix constraint model, select matching point pairs between the reference frame image and the target frame image that satisfy the constraint conditions.

[0093] In this embodiment, the matching point pairs that meet the constraints are filtered as follows:

[0094] S3041. Randomly select 4 sets of matching point pairs from the matching point pairs corresponding to the reference frame image and the target frame image, and calculate the homography matrix according to the homography matrix constraint model. H ;

[0095] S3042. Select points from the 4 sets of matched point pairs that satisfy the Euclidean distance. L Matching point pairs < 6 pixels are selected as those satisfying the homography matrix. H Define the interior points of the constraints and record the number of interior points;

[0096] S3043, Repeat steps S3041-S3042 multiple times, selecting the homography matrix corresponding to the maximum number of interior points. H Interior points of the constraint.

[0097] In this embodiment, steps S3041-S3042 are repeated 20 times to obtain the homography matrix corresponding to the maximum number of interior points. H The interior points constrained are used as the final interior points, and those that do not satisfy the Euclidean distance are eliminated using the above method. L Matching point pairs of less than 6 pixels.

[0098] In the fatigue test image acquisition method of this application, step four involves calculating the centroid of the reference feature points in the reference frame image and the centroid of the target feature points in the target frame image based on the matching point pairs selected in step three.

[0099] Centroid of reference feature points in the reference frame image for:

[0100] ;

[0101] ;

[0102] Centroid of target feature points in the target frame image for:

[0103] ;

[0104] ;

[0105] in, N This represents the number of interior points.

[0106] In the fatigue test image acquisition method of this application, step five involves calculating the offset of the centroid of the target feature point relative to the centroid of the reference feature point in the fatigue load loading direction, including:

[0107] ;

[0108] ;

[0109] in, The offset of the centroid of the target feature point relative to the centroid of the reference feature point in the fatigue load loading direction.

[0110] In the fatigue test image acquisition method of this application, step six involves repeating steps two through five to calculate the offset of the centroid of the target feature point relative to the centroid of the reference feature point in the fatigue load loading direction for all target frame images, and comparing the offsets corresponding to each target frame image. Select the maximum offset The corresponding target frame image is the crack image corresponding to the peak moment of fatigue load. This crack image is the image corresponding to the maximum crack opening state within one cycle.

[0111] The method for acquiring images at the peak load moment in fatigue testing, as described in this application, is applicable to the identification of crack images at the peak fatigue load moment in crack propagation tests of components and structures. This application can obtain images of the crack at its maximum opening without interacting with the testing machine, and these images can serve as crucial input for subsequent precise crack analysis. Secondly, this application offers fast analysis speed and high reliability, ensuring real-time acquisition of peak load crack images. Furthermore, this application is an important component of crack image feature detection technology for components and structures in damage tolerance tests, contributing to improved digital crack detection capabilities and possessing significant engineering value.

[0112] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for acquiring images of the peak load time in a fatigue test, characterized in that, include: Step 1: Acquire crack images of time-series frames within the fatigue load loading cycle, and select reference frame images and multiple target frame images from the crack images; Step 2: Determine the near-field region of the crack in the reference frame image and the target frame image respectively; Step 3: Extract feature points from the crack near-field region of the reference frame image and the target frame image respectively, and filter feature points according to the feature point matching relationship between the reference frame image and the target frame image; Step 4: Calculate the centroid of the reference feature points in the reference frame image and the centroid of the target feature points in the target frame image based on the selected feature points; Step 5: Calculate the offset of the centroid of the target feature point relative to the centroid of the reference feature point in the fatigue load loading direction; Step 6: Repeat steps 2 to 5 to calculate the offset of the centroid of the target feature point relative to the centroid of the reference feature point in the fatigue load loading direction for all target frame images, and select the target frame image corresponding to the maximum offset as the crack image corresponding to the peak fatigue load moment.

2. The method for acquiring images of the peak load time in fatigue testing according to claim 1, characterized in that, In step one, crack images of time-series frames within the fatigue load loading period are acquired, and reference frame images and multiple target frame images are selected from the crack images, including: S101. Obtain the fatigue load loading frequency and select a CCD camera with a frame rate greater than twice the fatigue load loading frequency. S102. Capture crack images of time-series frames within at least one fatigue load loading cycle using a CCD camera; S103. The first crack image in the crack images of the time series frames is determined as the reference frame image, and the crack images of other frames are determined as the target frame images.

3. The method for acquiring images of the peak load time in fatigue testing according to claim 2, characterized in that, In step two, the crack near-field regions of the reference frame image and the target frame image are determined respectively, including: The crack in the reference frame image and the target frame image is located at the center of the crack near-field region, and the length of the crack near-field region is 1.5 times the crack length, and the width of the crack near-field region is 0.75 times the crack length.

4. The method for acquiring images of the peak load time in fatigue testing according to claim 3, characterized in that, In step three, feature points of the crack near-field region are extracted from the reference frame image and the target frame image, respectively, and feature point filtering is performed based on the feature point matching relationship between the reference frame image and the target frame image, including: S301. Extract SIFT feature points of the crack near-field region of the reference frame image and the target frame image respectively; S302. Calculate the matching point pairs between the reference frame image and the target frame image based on the SIFT feature points; S303. Construct a homography matrix constraint model between the reference frame image and the target frame image; S304. Based on the homography matrix constraint model, select matching point pairs between the reference frame image and the target frame image that satisfy the constraint conditions.

5. The method for acquiring images of the peak load time in fatigue testing according to claim 4, characterized in that, In S302, the matching point pairs corresponding to the reference frame image and the target frame image are calculated based on the SIFT feature points, including: Select SIFT feature point A in the reference frame image, and find SIFT feature point B with the smallest Euclidean distance and SIFT feature point C with the second smallest Euclidean distance from SIFT feature point A in the target frame image. If the following conditions are met: Min / SecMin < 1.0 Where Min is the Euclidean distance between SIFT feature point A and SIFT feature point B, and SecMin is the Euclidean distance between SIFT feature point A and SIFT feature point C. Then SIFT feature point A and SIFT feature point B are defined as a matching point pair.

6. The method for acquiring images of the peak load moment in fatigue testing according to claim 5, characterized in that, In S303, constructing a homography matrix constraint model between the reference frame image and the target frame image includes: The homography matrix constraint model between the reference frame image and the target frame image is as follows: ； ； ； ； ； in, H It is a homography matrix. The coordinates of feature points in the reference frame image. The coordinates of the feature points in the target frame image. .

7. The method for acquiring images of the peak load time in fatigue testing according to claim 6, characterized in that, In S304, matching point pairs between the reference frame image and the target frame image that satisfy the constraint conditions are selected according to the homography matrix constraint model, including: S3041. Randomly select 4 sets of matching point pairs from the matching point pairs corresponding to the reference frame image and the target frame image, and calculate the homography matrix according to the homography matrix constraint model. H ; S3042. Select points from the 4 sets of matched point pairs that satisfy the Euclidean distance. L Matching point pairs < 6 pixels are selected as those satisfying the homography matrix. H Define the interior points of the constraints and record the number of interior points; S3043, Repeat steps S3041-S3042 multiple times, selecting the homography matrix corresponding to the maximum number of interior points. H Interior points of the constraint.

8. The method for acquiring images of the peak load time in fatigue testing according to claim 7, characterized in that, In step four, the centroids of the reference feature points in the reference frame image and the centroids of the target feature points in the target frame image are calculated based on the selected feature points, including: The centroid of the reference feature points of the reference frame image for: ； ； The centroid of the target feature points in the target frame image for: ； ； in, N This represents the number of interior points.

9. The method for acquiring images of the peak load time in fatigue testing according to claim 8, characterized in that, Step five involves calculating the offset of the centroid of the target feature point relative to the centroid of the reference feature point in the fatigue load loading direction, including: ； ； in, The offset of the centroid of the target feature point relative to the centroid of the reference feature point in the fatigue load loading direction.

Images