Online measurement of laser ablation morphology and ablation feature recognition method

Abstract

The invention provides a method for online measurement of laser ablation morphology and recognition of ablation features, relates to the field of image processing, and solves the problems that the existing measurement method cannot adapt to the interference of tangential airflow, the strong contrast between strong and weak light leads to measurement distortion, and the ablation features cannot be obtained online. The method comprises: a fill light measurement process: using a high-speed camera to shoot a laser ablation morphology image, using a fill light source to irradiate the ablation area, suppressing the strong self-luminescence of the ablation area, and measuring the laser ablation morphology online; an image processing process: obtaining the calibration parameters of the high-speed camera, preprocessing the laser ablation morphology image, and correcting the image by perspective transformation, using background difference to segment the image, detecting the ablation area, measuring the size of the detected ablation area, and completing ablation feature recognition. The invention can effectively realize the online measurement of ablation morphology and the intelligent recognition of ablation morphology features.

Description

Technical Field

[0001] The invention relates to the field of image processing and is applied to a laser ablation profile measurement process, in particular to a laser ablation profile online measurement and ablation feature recognition method. Background Art

[0002] Due to its high energy density, low cost, fast response and precise pointing, high-intensity lasers have shown great potential in many fields such as precision manufacturing, material modification and military applications. The key to studying the ablation effect of high-intensity lasers is how to accurately obtain the changes in morphology and related parameters during the ablation process. However, current technical means are insufficient in monitoring the evolution of ablation morphology, which limits in-depth research in this field. At present, there are two main ways to collect ablation morphology:

[0003] Single measurement method: This method is usually carried out after the laser irradiation experiment is completed, and the surface image after ablation is taken by a camera to obtain the ablation morphology information. The main defects of this method are: since only a single measurement can be performed after the experiment, the dynamic changes in the ablation process cannot be captured, and thus the influence of time factors on the ablation effect cannot be studied; when measuring the key parameters of the ablation morphology, it relies on the subjective judgment of the operator, lacks consistency and standardization, and affects the objectivity of the ablation effect research.

[0004] Real-time measurement method: A high-speed camera is used to take multiple frames of images at equal intervals throughout the ablation process, and researchers subsequently analyze them frame by frame. Although this method can provide continuous image data, it also has the following problems: This method works well for a flat plate model without airflow under weak laser irradiation conditions, but under strong laser irradiation conditions, the image quality is reduced due to the self-luminescence caused by the strong laser and the light deflection caused by the high-speed airflow; for non-planar or complex curved samples, the light distribution is uneven, further reducing the usability of the image; the size parameters of the ablation area cannot be directly given, and manual intervention by researchers is still required, which not only increases the workload but also affects the accuracy of the results.

[0005] It can be seen that neither of the above two methods can fully meet the needs of strong laser ablation effect research. Specifically, the main shortcomings of the existing technology include: it is only applicable to static flat plate models irradiated by weak lasers. It is difficult to obtain accurate damage images for models with high-speed tangential airflow, strong laser irradiation, and curvature; it is impossible to obtain key size parameters of the damage image in real time, and the parameter acquisition judgment depends on the subjective judgment of the researchers, which affects the objectivity of the ablation effect research. Therefore, in view of the shortcomings and deficiencies of the existing technology, it is urgent to introduce new technologies to carry out further in-depth research. Summary of the invention

[0006] Based on the current status of the background technology, the purpose of the present invention is to solve the problems that the existing measurement methods cannot adapt to the interference of tangential airflow, the strong contrast between strong and weak light leads to measurement distortion, and the ablation characteristics cannot be obtained online. Therefore, a method for online measurement of laser ablation morphology and ablation feature recognition is proposed. The present invention uses a high-resolution and high-frequency camera to quantitatively measure the evolution of the ablation morphology of the laser irradiated target under the action of tangential airflow, uses pulsed laser fill light to eliminate local highlights caused by strong lasers, and combines camera calibration, mixed Gaussian model and background difference method to realize online measurement of ablation morphology and intelligent recognition of ablation morphology characteristics.

[0007] The present invention adopts the following technical solutions to achieve the purpose:

[0008] A method for online measurement of laser ablation morphology and recognition of ablation features, comprising:

[0009] Fill-in light measurement process: A high-speed camera is used to capture the laser ablation morphology image. A narrow-band filter is pre-installed on the high-speed camera lens before capturing. While capturing, a fill-in light source is used to illuminate the ablation area to suppress the strong self-luminescence of the ablation area, measure the laser ablation morphology online, and obtain the laser ablation morphology image in real time.

[0010] Image processing process: For the laser ablation morphology image, ablation feature recognition is performed, the calibration parameters of the high-speed camera are obtained, the laser ablation morphology image is preprocessed, and the image is corrected by perspective transformation; the background difference method is used for image segmentation and the ablation area is detected; the size of the detected ablation area is measured to obtain the morphological feature changes of the ablation area during the ablation process, and the ablation feature recognition is completed.

[0011] Specifically, for the fill light measurement process, the laser to be measured is used to illuminate the irradiated target to produce an ablation area, and a high-speed camera is used to capture the damage process of the irradiated target under the action of the laser to be measured and the laser ablation morphology image formed in the corresponding ablation area; while shooting, a fill light source is used to emit a strong pulsed laser with a preset wavelength to the ablation area, thereby suppressing the strong self-luminescence of the ablation area.

[0012] Preferably, the fill light source used is a CAVILUX HF810 laser light source. While shooting, the pulse time of the strong pulse laser of the CAVILUX HF810 laser light source is coupled with the exposure time of the high-speed camera through the frame synchronization function and the corresponding TTL excitation signal.

[0013] Specifically, during the image processing process, the calibration parameters of the high-speed camera include external parameters and internal parameters; the external parameters are the posture and shooting distance of the high-speed camera relative to the irradiation plane where the irradiation target is located, and the internal parameters include the principal point coordinates, equivalent focal length and image distortion coefficient.

[0014] Preferably, the high-speed camera is calibrated using Zhang Zhengyou's plane calibration method, the external parameters and internal parameters of the high-speed camera are separated and the optimal solutions are gradually obtained respectively. The calibration operation method is as follows:

[0015] S1. Use a high-speed camera to photograph the same planar target from at least two different directions;

[0016] S2. Initialize the image distortion coefficient in the internal parameter to 0, select a point at the center of the captured image whose distortion is less than a threshold, and substitute the world coordinate value and image coordinate value of the point into the preset linear equation as known conditions;

[0017] S3, solving the calibration parameters of the high-speed camera by processing the linear equations through the optimization algorithm;

[0018] S4. The solved calibration parameters are used as initial values, all points on the plane target are selected to establish an optimization model, and the optimal solutions of the external parameters and internal parameters of the high-speed camera are solved through the optimization model.

[0019] Preferably, in the image processing process, the preprocessing method adopted in the preprocessing process of the laser ablation topography image includes Gaussian blurring and image equalization.

[0020] Specifically, in the image processing process, image correction is performed by perspective transformation, specifically: for the laser ablation morphology image taken by a high-speed camera, when it appears as a distorted image, the coordinates of the four corner points of the distorted image are obtained, which are recorded as the first set of coordinates; the coordinates of the four corner points of the corrected target image are determined, which are recorded as the second set of coordinates; based on the position difference relationship between the first set of coordinates and the second set of coordinates, the transformation matrix used for perspective transformation is calculated; the transformation operation corresponding to the transformation matrix is ​​performed on the laser ablation morphology image presented as a distorted image, and a front view of the target image after image correction is obtained.

[0021] Specifically, during the image processing process, when the background difference method is used for image segmentation, when the background change amplitude of the laser ablation topography image is lower than a preset threshold, the difference between the foreground and background of the laser ablation topography image is directly used to detect the change movement of the ablation area, and the ablation area used as the basis for size measurement is separated from the laser ablation topography image to realize the detection of the ablation area.

[0022] Preferably, when the background change amplitude of the laser ablation topography image is higher than or equal to a preset threshold, the background is dynamically updated based on a mixed Gaussian background model, that is: determine the central value of the laser ablation topography image at the current time, and obtain the deviation of the value of each pixel point in the laser ablation topography image at the current time; if the degree to which the value of a certain pixel point deviates from the central value is higher than or equal to the separation threshold, the pixel point corresponding to the value is the foreground; if the degree to which the value of a certain pixel point deviates from the central value is lower than the separation threshold, the pixel point corresponding to the value is the background; through the application of the mixed Gaussian background model, the foreground and background of the laser ablation topography image are separated during the background change process, thereby realizing the detection of the ablation area.

[0023] Specifically, when measuring the size of the detected ablation area, the target image where the ablation area is located is subjected to adaptive image binarization processing, morphological processing, and spot detection processing in sequence, and then the circumscribed rectangle of the ablation area is obtained to obtain the pixel length and width of the ablation area. and ;

[0024] The pixel length and width of the ablation area and Equivalent focal length in the calibration parameters of high-speed cameras and shooting distance Combined with the above, the actual physical dimensions of the ablation area, length and width, are calculated using the following formula:

[0025]

[0026]

[0027] In the formula, is the actual physical length of the ablated area, is the actual physical size width of the ablation area; based on the real-time acquired laser ablation morphology image, the actual physical size of the ablation area in each frame of the image is detected to obtain the morphological feature changes of the ablation area during the entire online measurement process, and complete the ablation feature recognition.

[0028] In summary, due to the adoption of this technical solution, the beneficial effects of the present invention are as follows:

[0029] The present invention realizes the online measurement of ablation morphology, overcoming the measurement distortion problem caused by the interference of tangential airflow and the strong contrast between strong and weak light in the traditional method; and uses a high-resolution high-frequency camera to capture the dynamic changes of the ablation process, ensuring the continuity and accuracy of data acquisition. At the same time, the present invention effectively eliminates the local highlight phenomenon caused by strong laser through pulse laser supplementary light technology, thereby obtaining a clearer ablation morphology image; this improvement significantly improves the image quality and lays a solid foundation for subsequent feature extraction and analysis.

[0030] Combining camera calibration, mixed Gaussian model, background difference and other technical means, the present invention realizes intelligent recognition of ablation morphological features; this method can automatically distinguish the ablation area from the background, accurately extract ablation features, and greatly improve the analysis efficiency and accuracy. Without affecting the measurement process, the present invention can calculate the key parameters of the ablation image, such as the ablation size, in real time; this immediacy not only saves time, but also ensures the timeliness and reliability of data analysis.

[0031] In the complex environment of strong laser high-speed flow field and curvature model, the present invention can still maintain good fill-in light effect and image clarity; precisely because the influence of local high brightness caused by strong laser on the measurement result is eliminated, the consistency and validity of the measurement data are ensured. Through rigorous experimental verification, the effectiveness and practicality of the present invention are also proved; the fill-in light technology and intelligent recognition method have shown excellent performance under various test conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic diagram of the overall process of the method of the present invention;

[0033] Figure 2 Schematic diagram of the hardware working principle of high-speed camera coupled fill light exposure;

[0034] Figure 3 Schematic diagram of the calibration checkerboard used for high-speed camera calibration;

[0035] Figure 4 Schematic diagram of a calibration checkerboard image set taken by a high-speed camera;

[0036] Figure 5 A schematic diagram showing the effect of extracting corner points on a calibration chessboard;

[0037] Figure 6 This is a three-dimensional schematic diagram of the external parameters displayed after the high-speed camera is calibrated;

[0038] Figure 7 This is a schematic diagram of the corner projection error analysis of a high-speed camera;

[0039] Figure 8 This is a schematic diagram of the effect of the image equalization process, where Figure 8 (a) is a schematic diagram of the effect before equalization. Figure 8 (b) is a schematic diagram of the effect after equalization;

[0040] Fig. 9 Schematic diagram of the principle of correcting the image perspective transformation process;

[0041] Fig.10This is the experimental effect diagram of the image perspective transformation process, where Fig.10 (a) is a schematic diagram showing the comparison of distorted images during shooting and data preprocessing. Fig.10 (b) is a schematic diagram of the corresponding front view image after correction;

[0042] Fig.11 is a schematic diagram of image background difference, where Fig.11 (a) is the laser ablation morphology image captured by a high-speed camera during the ablation process. Fig.11 (b) is the corresponding background image before ablation begins. Fig.11 (c) is the foreground image of the ablation area obtained by subtracting the background image from the laser ablation topography image;

[0043] Fig.12 Schematic diagram of the ablation process when the background changes;

[0044] Fig.13 This is a schematic diagram of the further processing flow after completing the image background difference, where Fig.13 (a) is an image that has been subjected to adaptive image binarization processing. Fig.13 (b) is the image after morphological processing. Fig.13 (c) is the image processed with spot detection;

[0045] Fig.14 Schematic diagram of different sequence image processing results for size measurement, where Fig.14 (a) is the image of the ablation area after the above image processing is completed at time t=0.2 seconds. Fig.14 (b) to Fig.14 (f) corresponds to the ablation area images after the above image processing is completed at time t=0.5, t=0.8, t=1.1, t=1.4 and t=1.7 seconds respectively;

[0046] Fig.15 A schematic diagram of specific measurement parameters for dimensional measurement;

[0047] Fig.16 This is a schematic diagram of the dimensional curve report obtained after dimensional measurement, where Fig.16 (a) shows the ablation area Schematic diagram of the change of morphological characteristics in the length direction over time. Fig.16 (b) shows the ablation area Schematic diagram of the change of morphological features in the width direction over time. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the parts of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0049] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] Example

[0051] A method for online measurement of laser ablation morphology and recognition of ablation features, comprising:

[0052] Fill-in light measurement process: A high-speed camera is used to capture the laser ablation morphology image. A narrow-band filter is pre-installed on the high-speed camera lens before capturing. While capturing, a fill-in light source is used to illuminate the ablation area to suppress the strong self-luminescence of the ablation area, measure the laser ablation morphology online, and obtain the laser ablation morphology image in real time.

[0053] Image processing process: For the laser ablation morphology image, ablation feature recognition is performed, the calibration parameters of the high-speed camera are obtained, the laser ablation morphology image is preprocessed, and the image is corrected by perspective transformation; the background difference method is used for image segmentation and the ablation area is detected; the size of the detected ablation area is measured to obtain the morphological feature changes of the ablation area during the ablation process, and the ablation feature recognition is completed.

[0054] Part of the processing flow of this method can be found in Figure 1 For illustration, this embodiment will introduce the details of each part of the method in detail in a certain order.

[0055] 1. Fill-light measurement solution

[0056] When a high-speed camera directly captures the damage process of a laser irradiated target, the strong laser will generate thermal radiation light, which will cover the visible light spectrum and cause the image to be overexposed. At the same time, the tangential airflow will produce a more complex flow field structure, resulting in light deflection and scattering. If the irradiated target has a large curvature, a local over-bright area will be formed under the ambient light. These factors will cause the high-speed camera to shoot unclearly, and in severe cases, it may even be impossible to obtain the laser damage morphology. In this embodiment, for the fill light measurement process, the measured laser is used to illuminate the irradiated target to generate an ablation area, and a high-speed camera is used to capture the damage process of the irradiated target under the action of the measured laser and the laser ablation morphology image formed in the corresponding ablation area. While shooting, a fill light source is used to emit a strong pulsed laser with a preset wavelength to the ablation area, thereby suppressing the strong self-luminescence of the ablation area.

[0057] Specifically, first, a narrowband filter is installed at the lens of the high-speed camera to filter out part of the thermal radiation light. The filter allows light signals to pass through within a preset band, while light signals on both sides that deviate from this band are blocked. However, since the local area irradiated by the laser is still in a high-temperature liquid metal state, the self-luminescence will still cause the ablation area of ​​the high-speed shooting to be in an overexposed image state. For this reason, the present embodiment uses the CAVILUX HF810 laser light source as the fill light source to emit a strong pulsed laser with a preset wavelength to the ablation area, thereby suppressing the strong self-luminescence of the ablation area, helping the high-speed camera to penetrate the strong self-luminescence, and shoot the high-speed ablation process of the metal; in actual application, the parameters of the pulse fill light source that matches the high-speed camera can be selected through repeated tests and adjustments. At the same time, in order to better eliminate local highlights, compared with ordinary continuous light sources, the present embodiment uses the frame synchronization function and the corresponding TTL excitation signal to couple the pulse time of the strong pulsed laser of the CAVILUX HF810 laser light source with the exposure time of the high-speed camera. The hardware working principle of exposure shooting can be found in Figure 2 's hint.

[0058] 2. High-speed camera calibration

[0059] To achieve accurate measurement of ablation morphology, the spatial position relationship between the high-speed camera and the ablation target must be known. Therefore, it is necessary to calibrate the high-speed camera to obtain the calibration parameters of the high-speed camera, namely the external parameters and internal parameters. The external parameters are the posture and shooting distance of the high-speed camera relative to the irradiation plane where the irradiation target is located, and the internal parameters include the principal point coordinates, equivalent focal length and image distortion coefficient.

[0060] In this embodiment, the Zhang Zhengyou plane calibration method is used to calibrate the high-speed camera. This method is a calibration method between the traditional calibration method and the self-calibration method. It avoids the shortcomings of the traditional method such as high equipment requirements and cumbersome calibration process, and has higher accuracy and better stability than the self-calibration method. The details of this method are well known to those skilled in the art and will not be described in detail here. Only some key features required by this embodiment are described as follows:

[0061] Zhang Zhengyou's plane calibration method is a typical two-step method, which separates the external parameters and internal parameters of the high-speed camera and gradually obtains the optimal solution. The calibration operation is as follows:

[0062] S1. Use a high-speed camera to photograph the same planar target from at least two different directions;

[0063] S2. Initialize the image distortion coefficient in the internal parameter to 0, select a point at the center of the captured image whose distortion is less than a threshold, and substitute the world coordinate value and image coordinate value of the point into the preset linear equation as known conditions;

[0064] S3, solving the calibration parameters of the high-speed camera by processing the linear equations through the optimization algorithm;

[0065] S4. The solved calibration parameters are used as initial values, all points on the plane target are selected to establish an optimization model, and the optimal solutions of the external parameters and internal parameters of the high-speed camera are solved through the optimization model.

[0066] In this calibration method, the high-speed camera is required to shoot the same planar target at two or more different orientations. Both the high-speed camera and the planar target can move freely without knowing the motion parameters. During the calibration process, it is assumed that the internal parameters of the high-speed camera remain unchanged, that is, no matter from which angle the high-speed camera shoots the template, its internal parameters are constants, and only the external parameters change.

[0067] This embodiment uses a Prosilica_GC_1380 industrial camera as a high-speed camera. The resolution of the industrial camera is 1360×1024, the CCD pixel size is 6.45μm×6.45μm, and the maximum focal length is 35mm. The calibration is implemented as follows:

[0068] (1) Make a calibration plate. The size of the calibration plate is 390 mm × 270 mm, and it consists of 13 × 9 squares of 30 mm × 30 mm. Figure 3 shown.

[0069] (2) Image acquisition: According to Zhang Zhengyou's calibration theory, each camera must have at least three calibration images. To ensure higher accuracy, this embodiment uses the left and right cameras to simultaneously capture 16 images each. Figure 4 shown.

[0070] (3) Corner point extraction: Harris corner point detector is used for corner point extraction. The results are as follows: Figure 5 shown.

[0071] (4) High-speed camera calibration, after completing corner point extraction, the calibration results are as follows:

[0072] Internal parameters :

[0073]

[0074] Distortion Vector :

[0075]

[0076] The three-dimensional display of external parameters is as follows Figure 6 shown.

[0077] (5) Corner point projection error analysis: To verify the calibration results, the projection error of the extracted corner points is analyzed using the internal and external parameters of the calibrated high-speed camera, such as Figure 7 As shown in the figure, the color representation of 16 different calibration images is covered, and each "+" symbol represents a corner point on the calibration image. Figure 7 The units of the horizontal and vertical coordinates are both pixels. It can be seen that the errors of all corner points do not exceed 1 pixel and are mainly concentrated near the 0 point, which means the results are good.

[0078] 3. Image Preprocessing

[0079] For the preprocessing of the laser ablation morphology image, after the target image is captured, the noise in the image is filtered out through preprocessing to improve the image quality. The preprocessing method adopted in this embodiment includes Gaussian blur and image equalization. Gaussian blur can remove the noise points in the image, and image equalization makes the grayscale distribution of the image more uniform and the details more prominent, thereby improving the resolution effect of the image. The effect of image equalization is shown in the figure below. Figure 8 As shown, Figure 8 (a) is before equalization, Figure 8 (b) After equalization.

[0080] 4. Perspective Transformation

[0081] Since the installation position of the high-speed camera is not necessarily completely facing the target, this embodiment also needs to use perspective transformation to correct the distorted image (the so-called oblique image) into a front view image. Perspective transformation is the process of projecting an image onto a new viewing plane, such as Fig. 9As shown, for the laser ablation topography image taken by a high-speed camera, when it appears as a distorted image, the coordinates of the four corner points of the distorted image are obtained, which are recorded as the first set of coordinates; the coordinates of the four corner points of the corrected target image are determined, which are recorded as the second set of coordinates; based on the position difference relationship between the first set of coordinates and the second set of coordinates, the transformation matrix for perspective transformation is calculated; the transformation operation corresponding to the transformation matrix is ​​performed on the laser ablation topography image presented as a distorted image, and a front view of the target image after image correction is obtained.

[0082] In actual operation, after the high-speed camera is fixed, an object containing a rectangular figure (such as a calibration plate, cardboard, etc.) is placed in the experimental area. The plane of the calibration plate is parallel to the plane of the observation window glass in the experimental area. After the high-speed camera takes an image, the above transformation matrix is ​​used to obtain a corrected front view image of the distorted image. An example of this process can be found in Fig.10 , Fig.10 (a) shows the comparison of distorted images during shooting and data preprocessing. Fig.10 (b) The corresponding front view image after correction; this process can be applied to the perspective transformation processing of the laser ablation morphology image.

[0083] 5. Background subtraction

[0084] The correct segmentation of the ablation area is the most critical step in obtaining the ablation morphology. Since the entire ablated object is in a fixed state when the high-speed camera is shooting, only the ablation area changes, so the background difference method can effectively realize the detection of the ablation area.

[0085] In this embodiment, when the background difference method is used for image segmentation, when the background change amplitude of the laser ablation topography image is lower than the preset threshold, the difference between the foreground and background of the laser ablation topography image is directly used to detect the change movement of the ablation area, and the ablation area used as the basis for size measurement is separated from the laser ablation topography image to achieve the detection of the ablation area. Fig.11 As shown, Fig.11 (a) is the laser ablation morphology image taken by a high-speed camera at a certain moment. Fig.11 (b) is the corresponding image before ablation begins, which is used as the background. Fig.11 (a) Subtract Fig.11 (b) can be obtained Fig.11 (c) Foreground image of the ablated area shown.

[0086] The background difference method can achieve good results for scenes with little background change, that is, when the background change amplitude is lower than the preset threshold, but when the background changes, the effect of this method will be affected, for example Fig.12In the laser irradiation ablation process shown in the figure, the background illumination changes significantly. Therefore, when the background change amplitude of the laser ablation topography image is higher than or equal to the preset threshold, the background is dynamically updated by using a method based on a mixed Gaussian background model.

[0087] In the mixed Gaussian background model, the value (or feature) of each pixel in the image is distributed within a certain distance around a central value in a short period of time. Usually, the central value can be replaced by the mean, and the distance can be replaced by the variance. This distribution is regular. According to statistical laws, if there are enough data points, these points can be considered to be normally distributed, that is, Gaussian distribution.

[0088] Based on the above characteristics, this embodiment first determines the central value of the laser ablation topography image at the current time, and obtains the deviation of the value of each pixel point in the laser ablation topography image at the current time; if the degree to which the value of a certain pixel point deviates from the central value is higher than or equal to the separation threshold, then the pixel point corresponding to the value is the foreground; if the degree to which the value of a certain pixel point deviates from the central value is lower than the separation threshold, then the pixel point corresponding to the value is the background; through the application of the mixed Gaussian background model, the foreground and background of the laser ablation topography image are separated during the background change process, thereby realizing the detection of the ablation area.

[0089] 6. Dimension measurement

[0090] In this embodiment, when measuring the size of the detected ablation area, the target image where the ablation area is located is subjected to adaptive image binarization processing, morphological processing and spot detection processing in sequence, and then the circumscribed rectangle of the ablation area is obtained to obtain the pixel length and width of the ablation area. and .

[0091] Fig.13 Shows the Fig.11 (c) represents the process of further image processing of the foreground image (i.e., the target image where the ablation area is located). Fig.13 (a) Adaptive image binarization processing is performed. Fig.13 (b) Morphological processing was performed. Fig.13 (c) Spot detection processing was performed. Fig.14 (a) to Fig.14 (f) corresponds to six images of the ablation area sequence after the above image processing content is completed during the ablation process from 0.2 seconds to 1.7 seconds. After obtaining their respective circumscribed rectangles, the change process of the ablation area is reflected. Fig.15 is the pixel length and width of the ablation area and Annotations on the image.

[0092] Therefore, the pixel length and width of the ablation area are obtained and Then, the pixel length and width of the ablation area are and Equivalent focal length in the calibration parameters of high-speed cameras and shooting distance Combined with the above, the actual physical dimensions of the ablation area, length and width, are calculated using the following formula:

[0093]

[0094]

[0095] In the formula, is the actual physical length of the ablated area, is the actual physical width of the ablated area.

[0096] 7. Generate Report

[0097] After completing the above six parts, the actual physical size of the ablation area in each frame of the laser ablation image is detected based on the real-time laser ablation morphology image, and the morphological feature changes of the ablation area during the entire online measurement process are obtained, which can be recorded as follows: Fig.16 The dimensional curve shown, where Fig.16 (a) shows the ablation area The changes of morphological characteristics in the length direction over time, Fig.16 (b) shows the ablation area The morphological characteristics of the width direction change over time. The size curve can be used to prepare relevant requirement documents such as the measured laser damage report, thereby completing the ablation feature identification.

[0098] In summary, the present invention not only provides a reliable online measurement technology for laser ablation morphology, but also opens up new possibilities for research and development in related fields. By improving the measurement accuracy and automation level, the present invention is expected to promote the application of laser ablation technology in industrial manufacturing, materials science and other fields.

Claims

1. A method for online measurement of laser ablation morphology and identification of ablation features, characterized in that: include: Fill-in light measurement process: A high-speed camera is used to capture the laser ablation morphology image. A narrow-band filter is pre-installed on the high-speed camera lens before capturing. While capturing, a fill-in light source is used to illuminate the ablation area to suppress the strong self-luminescence of the ablation area, measure the laser ablation morphology online, and obtain the laser ablation morphology image in real time. Image processing process: for laser ablation morphology images, ablation feature recognition is performed, calibration parameters of high-speed cameras are obtained, laser ablation morphology images are preprocessed, and image correction is performed by perspective transformation; image segmentation is performed by background difference method to detect ablation areas; the size of the detected ablation areas is measured to obtain the changes in the morphological features of the ablation areas during the ablation process, and ablation feature recognition is completed; For the fill-in light measurement process, the laser to be measured is used to illuminate the target to generate an ablation area, and a high-speed camera is used to capture the damage process of the irradiated target under the action of the laser to be measured and the laser ablation morphology image formed in the corresponding ablation area; while capturing the image, a fill-in light source is used to emit a strong pulsed laser with a preset wavelength to the ablation area, thereby suppressing the strong self-luminescence of the ablation area; In the image processing process, the calibration parameters of the high-speed camera include external parameters and internal parameters; the external parameters are the posture and shooting distance of the high-speed camera relative to the irradiation plane where the irradiation target is located, and the internal parameters include the principal point coordinates, equivalent focal length and image distortion coefficient; In the image processing process, when the background difference method is used for image segmentation, when the background change amplitude of the laser ablation topography image is lower than the preset threshold, the difference between the foreground and background of the laser ablation topography image is directly used to detect the change movement of the ablation area, and the ablation area as the basis for size measurement is separated from the laser ablation topography image to realize the detection of the ablation area; When the background change amplitude of the laser ablation topography image is higher than or equal to the preset threshold, the background is dynamically updated based on the mixed Gaussian background model, that is: determine the central value of the laser ablation topography image at the current time, and obtain the deviation of the value of each pixel in the laser ablation topography image at the current time; if the degree of deviation of the value of a certain pixel from the central value is higher than or equal to the separation threshold, the pixel corresponding to the value is the foreground; if the degree of deviation of the value of a certain pixel from the central value is lower than the separation threshold, the pixel corresponding to the value is the background; through the application of the mixed Gaussian background model, the foreground and background of the laser ablation topography image are separated in the process of background change, so as to realize the detection of the ablation area; When measuring the size of the detected ablation area, the target image where the ablation area is located is processed by adaptive image binarization, morphological processing and spot detection in sequence, and then the circumscribed rectangle of the ablation area is obtained to obtain the pixel length and width of the ablation area. and ; The pixel length and width of the ablation area and Equivalent focal length in the calibration parameters of high-speed cameras and shooting distance Combined with the above, the actual physical dimensions of the ablation area, length and width, are calculated using the following formula: In the formula, is the actual physical length of the ablated area, is the actual physical size width of the ablation area; based on the real-time acquired laser ablation morphology image, the actual physical size of the ablation area in each frame of the image is detected to obtain the morphological feature changes of the ablation area during the entire online measurement process, and complete the ablation feature recognition.

2. The method for online measurement of laser ablation morphology and identification of ablation features according to claim 1, characterized in that: The fill light source used is CAVILUX HF810 laser light source. While shooting, the pulse time of the strong pulse laser of CAVILUX HF810 laser light source is coupled with the exposure time of the high-speed camera through the frame synchronization function and the corresponding TTL excitation signal.

3. The method for online measurement of laser ablation morphology and identification of ablation features according to claim 1, characterized in that: The high-speed camera is calibrated using Zhang Zhengyou's plane calibration method. The external and internal parameters of the high-speed camera are separated and the optimal solutions are gradually obtained. The calibration operation is as follows: S1. Use a high-speed camera to photograph the same planar target from at least two different directions; S2. Initialize the image distortion coefficient in the internal parameter to 0, select a point at the center of the captured image whose distortion is less than a threshold, and substitute the world coordinate value and image coordinate value of the point into the preset linear equation as known conditions; S3, solving the calibration parameters of the high-speed camera by processing the linear equations through the optimization algorithm; S4. The solved calibration parameters are used as initial values, all points on the plane target are selected to establish an optimization model, and the optimal solutions of the external parameters and internal parameters of the high-speed camera are solved through the optimization model.

4. The method for online measurement of laser ablation morphology and identification of ablation features according to claim 1, characterized in that: In the image processing process, the preprocessing methods used in the preprocessing process of the laser ablation morphology image include Gaussian blur and image equalization.

5. The method for online measurement of laser ablation morphology and identification of ablation features according to claim 1, characterized in that: In the process of image processing, image correction is performed by perspective transformation, specifically: for the laser ablation morphology image taken by a high-speed camera, when it appears as a distorted image, the coordinates of the four corner points of the distorted image are obtained, which are recorded as the first set of coordinates; the coordinates of the four corner points of the corrected target image are determined, which are recorded as the second set of coordinates; based on the position difference relationship between the first set of coordinates and the second set of coordinates, the transformation matrix used for perspective transformation is calculated; the transformation operation corresponding to the transformation matrix is ​​performed on the laser ablation morphology image presented as a distorted image, and a front view of the target image after image correction is obtained.

Images