Rectifier welding seam identification method, device and equipment

Abstract

The invention discloses a rectifier welding seam identification method, device and equipment, and the method comprises the steps: obtaining a plurality of original height data sets of a rectifier, and each original height data set comprises a plurality of original height data which are related to the rectifier and are sequentially arranged; for each original height data set, the following steps A-C are executed: step A, noise reduction processing is performed on the original height data according to the numerical value of the original height data to obtain a noise-reduced height data set; step B, performing second-order difference processing on the noise reduction height data set through a difference algorithm, and determining a height maximum value and a height minimum value in the noise reduction height data set; c, determining a first welding spot position coordinate corresponding to the height maximum value and a second welding spot position coordinate corresponding to the height minimum value according to the height maximum value and the height minimum value; and combining the first welding spot position coordinates and the second welding spot position coordinates in each original height data set to determine a welding seam area.

Description

Technical Field

[0001] The present application belongs to the field of welding detection, and in particular, relates to a rectifier weld identification method, device and equipment. Background Art

[0002] The manufacturing materials of aircraft engine blades have extremely high requirements and must have excellent heat resistance, strength, and toughness. When welding blades, the standards for process technology are also extremely strict, and it is necessary to ensure that the quality, reliability, and durability of the welded parts reach the highest level. In addition, the aviation industry is also extremely strict in formulating airworthiness certification standards after the welding of engine blades is completed to ensure that the blades can work stably for a long time in extreme environments. Therefore, it is necessary to realize automated welding technology for three-dimensional complex curved surface welds such as aircraft engines. It can greatly liberate manpower, realize the rational use of resources, improve work efficiency, reduce errors, and reduce failure rates, thereby greatly improving the production efficiency of aircraft engines.

[0003] In the process of identifying the weld position of an aircraft engine rectifier, factors such as natural light reflection, the reflective effect of the weld surface, and the unevenness of the weld surface will inevitably interfere with the identification of the weld position, so that the obtained weld position data is not ideal.

[0004] At present, the confirmation of the weld position in China is mainly to determine the position of the weld on the workpiece by adjusting the workpiece and the sensor to a preset position. Among them, the patent with the publication number CN111203676A proposes a laser vision-based method for identifying the intersection line weld of the front bracket of an aircraft engine, which is applied to the intersection line welding of the front bracket barrel wall and the connecting pipe. The image is collected by a ring-shaped laser vision sensor, and the deflection degree and cross-sectional center of the pipe are identified after image segmentation, and the inclination angle is calculated. The servo turntable is adjusted to make the flip surface of the positioning rod parallel to the positive projection plane of the connecting pipe, the flip mechanism is adjusted to make the positioning rod parallel to the connecting pipe, the positioning rod is moved to be located directly above the connecting pipe, and the telescopic mechanism extends the positioning rod and aligns it with the connecting pipe. The ring-shaped vision sensor scans the weld groove, fits and smoothes the scan signal, and obtains the intersection line weld. This method has a simple algorithm, can quickly and accurately identify the inclination angle of the connecting pipe and the weld position, has fewer steps to adjust the inclination angle of the welding gun positioning rod, has small welding errors, and is highly efficient.

[0005] The above patent determines the position of the weld by adjusting the position of the servo turntable and the flipping mechanism. It can quickly and accurately identify the inclination angle of the connecting pipe and the weld position, avoiding complex mathematical model calculations. However, this is limited by the structure of the servo turntable and the flipping mechanism and is not applicable to all scenarios for determining the weld position. Summary of the invention

[0006] The embodiments of the present application provide a rectifier weld identification method, device and equipment, which improve the accuracy of rectifier weld position identification.

[0007] According to a first aspect of the present application, an embodiment of the present application provides a rectifier weld identification method, which may include:

[0008] Acquire multiple original height data sets of the rectifier, each original height data set includes multiple original height data related to the rectifier arranged in sequence;

[0009] For each original height data set, perform the following steps A to C respectively:

[0010] Step A, performing noise reduction processing on the original height data according to the numerical value of the original height data to obtain a noise reduction height data set;

[0011] Step B, performing second-order difference processing on the denoised height data set by using a difference algorithm to determine the maximum and minimum height values ​​in the denoised height data set;

[0012] Step C, determining the first welding point position coordinates corresponding to the maximum height value and the second welding point position coordinates corresponding to the minimum height value according to the maximum height value and the minimum height value;

[0013] The first weld point position coordinates and the second weld point position coordinates in each original height data set are combined to determine the weld area.

[0014] Optionally, the original height data is subjected to denoising according to the numerical value of the original height data to obtain a denoised height data set, including:

[0015] According to the size relationship between each original height data in the original height data set and the height threshold interval, determining that the original height data in the original height data set that does not exist in the height threshold interval is noise height data;

[0016] Obtain target height data adjacent to the noise height data in the original height data set;

[0017] When the target height data exists in the height threshold range, fitting height data is generated according to the curvature fitting of the target height data;

[0018] The noisy height data in the original height data set is replaced with the fitted height data to obtain the denoised height data set.

[0019] Optionally, after acquiring the target height data adjacent to the noise height data in the original height data set, the method further includes:

[0020] When the target height data does not exist in the height threshold interval, obtaining valid height data closest to both sides of the noise height data, and the valid height data exists in the height threshold interval;

[0021] The noisy height data in the original height data set is replaced with the average value of the valid height data to obtain the denoised height data set.

[0022] Optionally, before performing second-order difference processing on the denoised height data set by a difference algorithm to determine the height maximum and height minimum in the denoised height data set, the method further includes:

[0023] The noise reduction height data set is smoothed by using the local weighted regression scatter point smoothing filtering technique to obtain a smoothed height data set.

[0024] The denoised height dataset is processed by second-order difference algorithm to determine the maximum and minimum height values ​​in the denoised height dataset, including:

[0025] The second-order difference processing of the smoothed height data set is performed by the difference algorithm to determine the maximum and minimum height values ​​in the smoothed height data set.

[0026] Optionally, performing second-order difference processing on the denoised height dataset by a difference algorithm to determine the maximum and minimum heights in the denoised height dataset includes:

[0027] Generate a height data waveform image according to the denoised height data in the denoised height data set and the arrangement order of each denoised height data;

[0028] Performing second-order difference processing on the height data waveform image through a difference algorithm to generate a height difference image;

[0029] Performing filtering on the height difference image to obtain a filtered height data image;

[0030] The filtered height data image is traversed to determine the maximum and minimum heights of the denoised height data in the filtered height data image.

[0031] Optionally, performing second-order difference processing on the height data waveform image by a difference algorithm to generate a height difference image includes:

[0032] Calculate the difference between adjacent denoised height data in the height data waveform image to obtain a plurality of first-order adjacent data difference values;

[0033] Calculate the difference between a plurality of first-order adjacent data difference values ​​to obtain a plurality of second-order adjacent data difference values;

[0034] The height difference image is determined based on the second-order adjacent data difference.

[0035] Optionally, the filtering process includes at least one of limiting filtering and Gaussian filtering.

[0036] Optionally, combining the first weld point position coordinates and the second weld point position coordinates in each height data set to determine the weld area includes:

[0037] sequentially connecting adjacent first welding point position coordinates and second welding point position coordinates in each original height data set to generate a closed weld edge curve;

[0038] The area enclosed by the weld edge curve is determined as the weld area.

[0039] According to a second aspect of the present application, a rectifier weld identification device is provided, which may include:

[0040] An acquisition module, used for acquiring a plurality of original height data sets of the rectifier, each original height data set including a plurality of original height data related to the rectifier;

[0041] An execution module is used for each original height data set, which includes:

[0042] A denoising unit, used for performing denoising processing on the original height data according to the numerical value of the original height data to obtain a denoised height data set;

[0043] A second-order difference processing unit, used for performing second-order difference processing on the denoised height data set by using a difference algorithm, and determining a maximum height value and a minimum height value in the denoised height data set;

[0044] A determination unit, used to determine the first welding point position coordinates and the second welding point position coordinates corresponding to the height maximum value according to the height maximum value and the height minimum value respectively;

[0045] The combination module is used to combine the first welding point position coordinates and the second welding point position coordinates in each height data set to determine the weld area.

[0046] According to a third aspect of the present application, an embodiment of the present application provides a rectifier weld identification device, which includes: a processor and a memory storing computer program instructions;

[0047] When the processor executes the computer program instructions, the method for identifying a weld seam in a rectifier as described in any one of the first aspects is implemented.

[0048] The technical solution provided by the embodiments of the present application brings at least the following beneficial effects:

[0049] The embodiment of the present application provides a rectifier weld identification method, device, equipment and computer storage medium. After scanning the rectifier and obtaining the original height data sets of multiple rectifiers, the original height data in each original height data set is subjected to noise reduction and second-order difference processing, and then the height maximum and height minimum in each original height data set are determined. Thus, the first weld point position coordinates and the second weld point position coordinates can be determined according to the height maximum and height minimum, so that the first weld point coordinates and the second weld point coordinates in each original height data set can be combined, and the enclosed area is the weld area. Based on this, through key links such as data preprocessing, second-order difference processing, and weld point position determination, accurate division of the rectifier weld area is achieved. This method not only improves the accuracy and efficiency of detection, but also reduces the cost and error of manual intervention.

[0050] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the specification are used to explain the principles of the present application, and do not constitute an improper limitation on the present application.

[0052] Figure 1 is a flow chart of a rectifier weld identification method according to an exemplary embodiment;

[0053] Figure 2 is a structural schematic diagram of a rectifier weld identification device according to an exemplary embodiment;

[0054] Figure 3 The figure is a structural block diagram of a rectifier weld identification device according to an exemplary embodiment. DETAILED DESCRIPTION

[0055] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0056] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0057] As described in the background technology, during the weld position identification process of an aircraft engine rectifier, factors such as natural light reflection, the reflective effect of the weld surface, and the unevenness of the weld surface will inevitably interfere with the identification of the weld position, so that the obtained weld position data is not ideal.

[0058] Based on this, the present application provides a rectifier weld identification method, device and equipment. The following first introduces the rectifier weld identification method provided in the embodiment of the present application.

[0059] Embodiment 1;

[0060] Figure 1 A schematic diagram of a process flow of a rectifier weld identification method provided by an embodiment of the present application is shown. Figure 1 As shown, it may include the following steps:

[0061] S101, acquiring a plurality of original height data sets of a rectifier, each original height data set comprising a plurality of original height data related to the rectifier arranged in sequence;

[0062] S102: for each original height data set, respectively perform the following operations of step A to step C:

[0063] Step A, performing noise reduction processing on the original height data according to the numerical value of the original height data to obtain a noise reduction height data set;

[0064] Step B, performing second-order difference processing on the denoised height data set by using a difference algorithm to determine the maximum and minimum height values ​​in the denoised height data set;

[0065] Step C, determining the first welding point position coordinates corresponding to the maximum height value and the second welding point position coordinates corresponding to the minimum height value according to the maximum height value and the minimum height value;

[0066] S103, combining the first welding point position coordinates and the second welding point position coordinates in each original height data set to determine a welding seam area.

[0067] Based on the above embodiment, after scanning the rectifier and obtaining the original height data sets of multiple rectifiers, the original height data in each original height data set is subjected to noise reduction and second-order difference processing, and then the height maximum and height minimum in each original height data set are determined, thereby determining the first welding point position coordinates and the second welding point position coordinates according to the height maximum and height minimum, so that the first welding point coordinates and the second welding point coordinates in each original height data set can be combined, and the enclosed area is the weld area. Based on this, through key links such as data preprocessing, second-order difference processing, and weld point position determination, accurate division of the rectifier weld area is achieved. This method not only improves the accuracy and efficiency of detection, but also reduces the cost and error of manual intervention.

[0068] In the above S101, by binding the laser scanner to the robot arm, the path driving trajectory is set for the controller, and the controller controls the robot arm to perform uniform linear motion along the x-axis direction according to the set program. At the same time, the laser of the linear laser scanner is perpendicular to the surface of the rectifier to scan the rectifier, so as to obtain the original height data set through the scanner. Each time the scanner scans, its x-direction and y-direction are fixed, and the scanning is performed along the z-direction. Therefore, the data obtained from each scan is a set of multiple original height data of the rectifier, and in the original height data set, each original height data is arranged in the order of acquisition time, that is, arranged in the scanning order. After completing the scan of the entire rectifier, multiple original height data sets are obtained.

[0069] In the above S102, for each set of original height data sets, the following operations of step A to step C are performed;

[0070] In the above step A, since the scanner scans the entire rectifier, and during the scanning process, it is affected by factors such as natural light reflection, reflective effect of the weld surface and unevenness of the weld surface, it is inevitable that interference will be caused to the original height data collection. Therefore, there will be a lot of noise data in the original height data set of the original height data. There is a difference in data value between the noise data and the normal data. Therefore, the noise data and the normal data in the original height data can be distinguished according to the numerical value of the original height data, so as to realize denoising processing of the original height data in the original height data set, thereby obtaining the denoised height data after denoising. The collection of denoised height data is the denoised height data set.

[0071] In order to further improve the accuracy of the data in the noise reduction height data set, Example 1 of the present application also provides another implementation method of the rectifier weld identification method.

[0072] Specifically, the above step A may also include:

[0073] Step A01, determining that the original height data in the original height data set that does not exist in the height threshold interval is noise height data according to the size relationship between each original height data in the original height data set and the height threshold interval;

[0074] Step A02, obtaining target height data adjacent to the noise height data in the original height data set;

[0075] Step A03, generating fitting height data according to the curvature fitting of the target height data when the target height data exists in the height threshold interval;

[0076] Step A04: Replace the noisy height data in the original height data set with the fitted height data to obtain a denoised height data set.

[0077] Based on the above embodiment, by comparing the original height data with the preset height threshold interval, the original height data that is not in the height threshold interval can be regarded as noise data, thereby fitting the target data adjacent to the noise data to generate fitting height data, and replacing the noise data with the fitting height data. Since the target data satisfies the height threshold interval, the fitting height data generated by fitting also satisfies the height threshold interval. Therefore, after the fitting height data replaces the noise height data, the data in the newly generated denoised height data set all exist in the noise threshold interval, thereby improving the accuracy of the data in the denoised height data set.

[0078] In the above steps A01-A04, the height threshold interval is a calibration value set by the experimenter after the experimenter has verified and analyzed whether each data is noise height data by counting a large amount of original height data, and can be directly obtained. For the height data that is not in the height threshold interval, it can be identified as noise data. Noise height data needs to be eliminated from the original height data set, but if the noise height data is simply eliminated from the original height data set, a large number of data gaps will appear in the data in the original height data set, resulting in the discontinuity of the data in the original height data set, so it is necessary to refill new height data at the position of the noise height data, and in order to ensure the coherence of the data in the original height data set, the target height data adjacent to the noise height data can be obtained. If the target height data all meet the height threshold interval, the target height data can be fitted according to the change curvature between the target height data, so as to generate a fitting height data that can also meet the height threshold interval, so that the original noise height data can be replaced with the fitting height data to obtain a new height data set, that is, the noise reduction height data set.

[0079] It should be noted that

[0080] In order to further improve the accuracy of the data in the noise reduction height data set, Example 1 of the present application also provides another implementation method of the rectifier weld identification method.

[0081] More specifically, after the above step A02, the rectifier weld identification method may further include:

[0082] Step A05, when the target height data does not exist in the height threshold interval, obtaining valid height data closest to both sides of the noise height data, and the valid height data exists in the height threshold interval;

[0083] Step A06, replacing the noisy height data in the original height data set with the average value of the valid height data to obtain a noise-reduced height data set.

[0084] Based on the above embodiment, when the adjacent target height data also does not meet the height threshold interval, the noise height data is replaced by obtaining the valid height data closest to the noise height data and calculating the average value of the valid height data, so that the denoised height data in the new denoised height data set is still complete and continuous.

[0085] In the above step A05, if the height data most adjacent to the noise height data on both sides still does not exist in the height threshold interval, then according to the position of the noise height data, the target height data closest to the noise height data is obtained, wherein the target height data refers to the height data existing in the height threshold interval.

[0086] In the above step A06, the average value of the valid height data adjacent to the noise height data is calculated, and the original noise height data is replaced by the average value to form a new height data set, namely, the noise reduction height data set.

[0087] It should be noted that, in the present scheme, the target height data adjacent to the noise height data refers to two height data adjacent to the noise height data in the original height data set and arranged in front of and behind (left and right) the noise height data. In the case where the noise height data is the edge data of the original height data set, since the number of original height data in the original height data set is very large, the noise height data at the edge position can be directly ignored, or replaced by the height data located in the height threshold interval that is closest to the noise height data at the edge position.

[0088] In the above step B, the second-order difference is a form of differential operation, which is usually used in time series analysis, signal processing and numerical analysis. The basic idea of ​​differential operation is to reveal the changing trend of data by calculating the difference between adjacent data points. Therefore, by performing second-order difference processing on the denoised height data in the denoised height data set, the maximum and minimum height values ​​in the denoised height data set can be more intuitively seen.

[0089] It should be noted that, since the external shapes of the welding point and the rectifier on the weld are different, the maximum height value and the minimum height value in the original height data set refer to the height of the welding point in the original height data set.

[0090] In the above step C, since the original height data is obtained in the process, what changes in each original height data set is the height (z), and what remains unchanged are the length (x) and width (y), so after determining each height maximum and height minimum, the height maximum, height minimum and the length and width corresponding to the original height data set can be combined to determine the first weld point position coordinates corresponding to the height maximum in this original height data set, and the second weld point position coordinates corresponding to the height minimum.

[0091] In the above S103, after respectively determining the first weld point position coordinates and the second weld point position coordinates in each original height data set, the first weld point position coordinates and the second weld point position coordinates in each original height data set are sequentially connected in the three-dimensional coordinate system to form an area, which is the welding area.

[0092] In order to accurately obtain the weld area, the present application also provides another implementation of the rectifier weld identification method.

[0093] The above S103 may also include:

[0094] S1031, sequentially connecting adjacent first welding point position coordinates and second welding point position coordinates in each original height data set to generate a closed weld edge curve;

[0095] S1032, determining that the area enclosed by the weld edge curve is the weld area.

[0096] In the above embodiment, by sequentially connecting the first welding point coordinates and the second welding point coordinates in each original height data set, the welding points in each scanning process can be connected, thereby enclosing a welding seam area, thereby improving the accuracy of forming the welding seam area.

[0097] In the above S1031, in the three-dimensional coordinate system, the first weld point position coordinates and the second weld point position coordinates in each original height data set are marked, and the first weld point position coordinates and the second weld point position coordinates in each original height data set are respectively connected to each other, and the first weld point position coordinates of the edge are connected to the second weld point position coordinates of the edge.

[0098] In the above S1032, after all the first welding point coordinates and the second welding point coordinates in all the original height data sets are connected in sequence, an area formed in the three-dimensional coordinate system is the weld area.

[0099] Embodiment 2;

[0100] Based on Example 1, in order to obtain more accurate maximum and minimum height values, this application also provides another implementation method of the rectifier weld identification method.

[0101] Before the above step B, the method may further include:

[0102] Step D, smoothing the noise reduction height data set by using a local weighted regression scatter point smoothing filtering technique to obtain a smoothed height data set;

[0103] The above step B may also include:

[0104] Step B1, performing second-order difference processing on the smoothed height data set by using a difference algorithm to determine the maximum and minimum height values ​​in the smoothed height data set.

[0105] Based on the above embodiment, before performing the second-order difference processing, the denoised height data set is subjected to smoothing filtering processing to obtain a new smoothed height data set, and then the smoothed height data set is subjected to second-order difference processing to obtain the height maximum and height minimum. Based on this, by adopting the smoothing filtering processing, the data distortion problem caused by burrs on the weldment surface can be effectively dealt with, thereby obtaining a more accurate and smooth smoothed height data set, thereby making the height maximum and height minimum obtained in the end more accurate.

[0106] In the above step D, the local weighted regression scatter point smoothing filtering technique (also known as Lowess filtering) smoothes the weldment joint data. The main idea of ​​the local weighted regression scatter point smoothing filtering is to select a point x as the center, select a set of frace points on its left and right sides, and perform weighted linear regression on these set points through the weight function ω, so that is its center value, is the corresponding value of the curve after fitting. There are several points in the set of points, and several regression lines can be drawn. The lines between them constitute the Lowess curve of this set of points, and the points on the Lowess curve constitute a smooth height data set.

[0107] The data of the weld joint is averaged and regressed using formula (1), and weighted smoothing is performed using formula (2). The Lowess estimate is calculated as:

[0108] f ∧ (x) = Ave (y i ∣x i ∈N k x i ) (1)

[0109]

[0110] Among them, K λ (x0,x i )=D(|x0-xi| / λ), D(t)=0.75(1-t 2 ), when |t|<1, λ is the window width.

[0111] f ∧ (x0)=∑ i ω i (x0y i ) (3)

[0112] In the above formula (3): i (x0) is the weight. Each point in the scatter plot and a certain number of points around it are called a window λ. Weighted straight line fitting is performed inside each window λ. The weight is related to two factors: the distance from the current point and the distance between the actual point and the fitting point during the last fitting. Repeated fitting can get the best curve. The smoothness of the curve depends on the set λ: the smaller the λ, the less smooth the fitting, and vice versa. However, excessive smoothness will lead to the loss of the original morphological features.

[0113] In the above step B1, after obtaining a smoothed height data set, the smoothed height data set is further processed by second-order difference to obtain a processed curve, and the maximum height and minimum height in the smoothed height data set are selected from the curve. The maximum height and minimum height are the maximum height and minimum height in the original height data set.

[0114] The other features, working principles and beneficial effects of this embodiment are consistent with those of Embodiment 1 and will not be described in detail here.

[0115] Embodiment 3;

[0116] Based on Example 1, in order to obtain more accurate maximum and minimum height values, this application also provides another implementation method of the rectifier weld identification method.

[0117] The above step B may include:

[0118] Step B2: Generate a height data waveform image according to the denoised height data in the denoised height data set and the arrangement order of each denoised height data;

[0119] Step B3: Performing second-order difference processing on the height data waveform image by using a difference algorithm to generate a height difference image;

[0120] Step B4: filtering the height difference image to obtain a filtered height data image;

[0121] Step B5: traverse the filtered height data image to determine the maximum and minimum heights of the denoised height data in the filtered height data image.

[0122] Based on the above embodiment, by generating a height data waveform image, the noise-reduced height data is displayed in a graphical manner, making the data change trend and characteristics more intuitive and easy to understand. Therefore, the height data waveform graph generated by performing second-order difference processing on the height data waveform graph can highlight the change points in the data, and then the height difference image is filtered to remove the impulse noise interference and other interference signals that follow the normal distribution law in the height difference image, so that the height maximum and minimum values ​​can be obtained more accurately from the filtered height data image.

[0123] In the above step B2, according to the arrangement order of each denoised height data in the denoised height data set, each denoised height data is connected in sequence to form a height data waveform graph, which is a curve graph that can reflect the change pattern of the height data.

[0124] In the above step B3, a differential algorithm is used to perform second-order differential processing on the height data waveform image. The basic idea of ​​differential operation is to reveal the changing trend of data by calculating the difference between adjacent data points. The first-order difference is to calculate the difference between adjacent data points, and the second-order difference is to perform a differential operation on the result of the first-order difference again. By calculating the second-order difference of the signal, it is easier to identify the mutation point or turning point in the signal, which can make it easier to find the height maximum and minimum values ​​later.

[0125] Specifically, the above step B3 may include:

[0126] Calculate the difference between adjacent denoised height data in the height data waveform image to obtain a plurality of first-order adjacent data difference values;

[0127] Calculate the difference between a plurality of first-order adjacent data difference values ​​to obtain a plurality of second-order adjacent data difference values;

[0128] The height difference image is determined based on the second-order adjacent data difference.

[0129] In the above step B4, since the height difference image may be distorted to a certain extent, this will inevitably have a certain impact on the subsequent feature point extraction task. In order to solve the impulse noise interference and other interference signals that follow the normal distribution law in the height difference image, it is necessary to perform targeted filtering on the height difference image to obtain a filtered height data image.

[0130] The filtering process may include: limiting filtering process and Gaussian filtering process.

[0131] The principle of the limiting filtering method is shown in formula (4), which is mainly used to suppress the pulse interference effect caused by sudden abnormal signals.

[0132]

[0133] Among them, y n With y n-1 is the sampling value, and ΔT is the threshold value.

[0134] Gaussian filtering achieves a series of goals such as removing noise, smoothing images, protecting edges, and reducing high-frequency components by performing weighted averaging on data in the spatial domain or frequency domain, thereby improving image quality and facilitating subsequent processing and analysis. The expression of the Gaussian function is:

[0135]

[0136] The height difference image can be filtered by executing the formula (5) above through a Gaussian filter. If the σ parameter value is increased in the Gaussian filter, it means that the "window" or the influence range of the filter will be expanded. In this way, the result of the data processing will show a higher smoothness and the noise component will be effectively suppressed. However, at the same time, some important feature information in the original data may be weakened or lost. In order to find the most suitable σ parameter value, a series of experiments are conducted to adjust and verify its effect. The optimal parameter value will vary according to different application scenarios and conditions.

[0137] Although various external factors may cause irregularities or local unevenness on the weld surface, the basic shape of the joint can still be identified. After smoothing the local peaks on the weld surface using filtering technology, the overall joint inclination information is effectively retained. This process ensures that the feasibility of extracting feature points based on the differential method is also ensured to a certain extent even when the weld surface is uneven.

[0138] In the above step B5, after the filtered height data image is obtained, since the interference signal and other factors in the original height difference image have been removed, the maximum height value and the minimum height value can be directly found from the filtered height data image.

[0139] The other features, working principles and beneficial effects of this embodiment are consistent with those of Embodiment 1 and will not be described in detail here.

[0140] Embodiment 4;

[0141] Based on the same inventive concept, this embodiment further provides a rectifier weld identification device 200 on the basis of Embodiment 1, specifically as follows Figure 2 As shown, it includes:

[0142] An acquisition module 210 is used to acquire a plurality of original height data sets of the rectifier, each original height data set including a plurality of original height data related to the rectifier;

[0143] The execution module 220 is used for each original height data set, which may include:

[0144] A denoising unit 221 is used to perform denoising on the original height data according to the numerical value of the original height data to obtain a denoised height data set;

[0145] A second-order difference processing unit 222 is used to perform second-order difference processing on the denoised height data set by a difference algorithm to determine a maximum height value and a minimum height value in the denoised height data set;

[0146] A first determining unit 223, configured to determine the first welding point position coordinates and the second welding point position coordinates corresponding to the height maximum value according to the height maximum value and the height minimum value;

[0147] The combining module 230 is used to combine the first welding point position coordinates and the second welding point position coordinates in each height data set to determine the weld area.

[0148] Based on the above embodiment, after the acquisition module 210 acquires the original height data sets of multiple rectifiers by scanning the rectifier, the noise reduction unit 221 and the second-order difference processing unit 222 respectively perform noise reduction and second-order difference processing on the original height data in each original height data set, and the first determination unit 223 determines the height maximum and height minimum in each original height data set, thereby determining the first welding point position coordinates and the second welding point position coordinates according to the height maximum and height minimum, so that the combination module 230 can combine the first welding point coordinates and the second welding point coordinates in each original height data set, and the enclosed area is the weld area. Based on this, through key links such as data preprocessing, second-order difference processing, and weld point position determination, the accurate division of the rectifier weld area is achieved. This method not only improves the accuracy and efficiency of detection, but also reduces the cost and error of manual intervention.

[0149] Optionally, the noise reduction unit 221 may include:

[0150] A first determining subunit is used to determine that the original height data in the original height data set that does not exist in the height threshold interval is noise height data according to the size relationship between each original height data in the original height data set and the height threshold interval;

[0151] A first acquisition subunit is used to acquire target height data adjacent to the noise height data in the original height data set;

[0152] A fitting subunit, for generating fitting height data according to a curvature fitting of the target height data when the target height data exists within a height threshold range;

[0153] The first replacement subunit is used to replace the noisy height data in the original height data set with the fitted height data to obtain a denoised height data set.

[0154] Optionally, the noise reduction unit 221 may further include:

[0155] A second acquisition subunit is used to acquire valid height data closest to both sides of the noise height data when the target height data does not exist in the height threshold interval, and the valid height data exists in the height threshold interval;

[0156] The second replacement subunit is used to replace the noisy height data in the original height data set with the average value of the valid height data to obtain a noise-reduced height data set.

[0157] Optionally, the execution module 220 may further include:

[0158] A smoothing processing unit is used to smooth the noise reduction height data set by using a local weighted regression scatter point smoothing filtering technique to obtain a smoothed height data set;

[0159] The second-order difference processing unit 222 may include:

[0160] The first second-order difference processing subunit is used to perform second-order difference processing on the smoothed height data set through a difference algorithm to determine the maximum height value and the minimum height value in the smoothed height data set.

[0161] Optionally, the second-order difference processing unit 222 may include:

[0162] A sorting subunit, used for generating a height data waveform image according to the noise reduction height data in the noise reduction height data set and the arrangement order of each noise reduction height data;

[0163] A second second-order difference processing subunit is used to perform second-order difference processing on the height data waveform image by using a difference algorithm to generate a height difference image;

[0164] A filtering processing subunit is used to filter the height difference image to obtain a filtered height data image;

[0165] The second determining subunit is used to traverse the filtered height data image and determine the maximum and minimum height values ​​of the denoised height data in the filtered height data image.

[0166] Optionally, the combining module 230 may include:

[0167] A connection unit, used to sequentially connect the adjacent first welding point position coordinates and the second welding point position coordinates in each original height data set to generate a closed weld edge curve;

[0168] The second determining unit is used to determine that the area enclosed by the weld edge curve is the weld area.

[0169] In Example 4, each process of the above-mentioned Examples 1-3 can be implemented and the same technical effect can be achieved. To avoid repetition, it will not be described again here.

[0170] Figure 3 A schematic diagram of the hardware structure of a rectifier weld identification device provided by an embodiment of the present invention is shown.

[0171] The rectifier weld identification device may include a processor 301 and a memory 302 storing computer program instructions.

[0172] Specifically, the processor 301 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiment of the present invention.

[0173] The memory 302 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 302 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 302 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 302 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 32 is a non-volatile solid-state memory.

[0174] In certain embodiments, the memory 302 may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk storage media device, an optical storage media device, a flash memory device, an electrical, optical or other physical / tangible memory storage device. Thus, typically, the memory 302 includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors 301), it is operable to perform the operations described with reference to the method according to an aspect of the present application.

[0175] The processor 301 reads and executes the computer program instructions stored in the memory 302 to implement any one of the rectifier weld identification methods in the above embodiments.

[0176] In one example, the rectifier weld identification device may further include a communication interface 303 and a bus 304. As shown in the figure, the processor 301, the memory 302, and the communication interface 303 are connected through the bus 304 and communicate with each other.

[0177] The communication interface 303 is mainly used to implement the communication between the modules, devices, units and / or equipment in the embodiment of the present invention.

[0178] The bus 304 includes hardware, software or both. For example and not limitation, the bus 304 may include an accelerated graphics port (AGP) or other graphics bus, an enhanced industry standard architecture (EISA) bus, a front-side bus (FSB), a hypertransport (HT) interconnect, an industry standard architecture (ISA) bus, a wireless bandwidth interconnect, a low pin count (LPC) bus, a memory bus, a microchannel architecture (MCA) bus, a peripheral control interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a serial advanced technology attachment (SATA) bus, a video electronics standard association local (VLB) bus or other suitable bus or a combination of two or more of these. Where appropriate, the bus 304 may include one or more buses 304. Although the present application embodiment describes and shows a specific bus 304, the present application considers any suitable bus 304 or interconnection.

[0179] The rectifier weld seam identification device can be based on the current rectifier weld seam identification method, thereby realizing the combination Figure 1 , 2 A rectifier weld seam identification method and a control module are described.

[0180] In addition, the embodiment of the present application also provides a computer program product, including computer program instructions, and when the computer program product is executed by the processor 301, the steps and corresponding contents of the aforementioned method embodiment can be implemented.

[0181] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0182] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0183] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0184] Aspects of the present disclosure are described above with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, or a processor of other programmable rectifier weld identification methods, devices and equipment to produce a machine so that these instructions executed by the processor of the computer or other programmable rectifier weld identification methods, devices and equipment enable the implementation of the functions / actions specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0185] The above is only a specific implementation of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.

Claims

1. A rectifier weld identification method, characterized in that: The method comprises: Acquire a plurality of original height data sets of the rectifier, each of the original height data sets comprising a plurality of original height data related to the rectifier arranged in sequence; For each of the original height data sets, the following steps A to C are performed respectively: Step A, performing noise reduction processing on the original height data according to the numerical value of the original height data to obtain a noise reduction height data set; Step B, performing second-order difference processing on the de-noised height data set by using a difference algorithm to determine the height maximum and height minimum in the de-noised height data set; Step C, determining the first welding point position coordinates corresponding to the maximum height value and the second welding point position coordinates corresponding to the minimum height value according to the maximum height value and the minimum height value; The first welding point position coordinates and the second welding point position coordinates in each of the original height data sets are combined to determine a welding seam area.

2. The method according to claim 1, characterized in that The step of performing noise reduction processing on the original height data according to the numerical value of the original height data to obtain a noise reduction height data set includes: According to the size relationship between each of the original height data in the original height data set and the height threshold interval, determining that the original height data in the original height data set that does not exist in the height threshold interval is noise height data; Acquire target height data adjacent to the noise height data in the original height data set; In the case where the target height data exists in the height threshold interval, generating fitting height data according to the curvature fitting of the target height data; The noisy height data in the original height data set is replaced with the fitted height data to obtain a denoised height data set.

3. The method according to claim 2, characterized in that After acquiring the target height data adjacent to the noise height data in the original height data set, the method further includes: When the target height data does not exist in the height threshold interval, acquiring valid height data closest to both sides of the noise height data, the valid height data existing in the height threshold interval; The noise height data in the original height data set is replaced with the average value of the valid height data to obtain the noise-reduced height data set.

4. The method according to claim 1, characterized in that Before performing second-order difference processing on the denoised height data set by using a difference algorithm to determine the height maximum and height minimum in the denoised height data set, the method further includes: Smoothing the noise-reduced height data set by using a local weighted regression scatter point smoothing filtering technique to obtain a smoothed height data set; The performing second-order difference processing on the de-noised height data set by using a difference algorithm to determine a maximum height value and a minimum height value in the de-noised height data set includes: The smoothed height data set is subjected to second-order difference processing by using a difference algorithm to determine the maximum height value and the minimum height value in the smoothed height data set.

5. The method according to claim 1, characterized in that The performing second-order difference processing on the de-noised height data set by using a difference algorithm to determine a maximum height value and a minimum height value in the de-noised height data set includes: Generate a height data waveform image according to the denoised height data in the denoised height data set and the arrangement order of each of the denoised height data; Performing second-order difference processing on the height data waveform image by using the difference algorithm to generate a height difference image; Performing filtering processing on the height difference image to obtain a filtered height data image; The filtered height data image is traversed to determine the maximum height value and the minimum height value of the denoised height data in the filtered height data image.

6. The method according to claim 5, characterized in that The step of performing second-order difference processing on the height data waveform image by using the difference algorithm to generate a height difference image comprises: Calculate the difference between adjacent noise-reduced height data in the height data waveform image to obtain a plurality of first-order adjacent data difference values; Calculating the difference between a plurality of the first-order adjacent data difference values ​​to obtain a plurality of second-order adjacent data difference values; The height difference image is determined according to the second-order adjacent data difference.

7. The method according to claim 5, characterized in that The filtering process includes at least one of a limiting filtering process and a Gaussian filtering process.

8. The method according to any one of claims 1 to 7, characterized in that The combining the first welding point position coordinates and the second welding point position coordinates in each of the original height data sets to determine the welding seam area includes: sequentially connecting adjacent first welding point position coordinates and second welding point position coordinates in each of the original height data sets to generate a closed weld edge curve; The area enclosed by the weld edge curve is determined as the weld area.

9. A rectifier weld identification device, characterized in that: include: An acquisition module, used for acquiring a plurality of original height data sets of the rectifier, each of the original height data sets comprising a plurality of original height data related to the rectifier; An execution module is used for each of the original height data sets, comprising: A denoising unit, configured to perform denoising on the original height data according to the numerical value of the original height data to obtain a denoised height data set; A second-order difference processing unit, used for performing second-order difference processing on the denoised height data set by a difference algorithm to determine a maximum height value and a minimum height value in the denoised height data set; A determining unit, configured to determine, according to the maximum height value and the minimum height value, the first welding point position coordinates and the second welding point position coordinates corresponding to the maximum height value, respectively; A combination module is used to combine the first welding point position coordinates and the second welding point position coordinates in each of the height data sets to determine the weld area.

10. A rectifier weld identification device, characterized in that: The device comprises: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, the rectifier weld identification method as described in any one of claims 1-6 is implemented.

Images