A method and apparatus for extracting feature points of curved surface lap joint bevel

Abstract

This application discloses a method and apparatus for extracting feature points of curved surface overlap bevels. The method includes: using a 3D profilometer to scan the overlap bevel and obtain overlap bevel contour depth data; performing multi-dimensional interpolation filtering on the overlap bevel contour depth data to remove noisy contour data; setting ROI region segmentation to remove redundant data in the overlap bevel cross-section contour depth data; using the PELT variable point detection method to process the data within the ROI region and extract multiple sets of candidate feature points; and using dynamic threshold optimization to filter the multiple sets of candidate feature points to obtain target feature points.

Description

Technical Field

[0001] This application relates to the field of intelligent welding technology, and more specifically, to a method and apparatus for extracting feature points of curved surface lap joint bevels. Background Technology

[0002] Intelligent welding is developing rapidly and is commonly used in industrial applications requiring high strength and good sealing, such as automotive roof and body welding. Currently, most lap welds rely on robots with pre-set welding trajectories, and welding is performed according to these trajectories. This method not only increases labor and time costs but also lacks flexibility and stability. Furthermore, current methods for extracting bevel features are based on the condition that two plates are parallel and overlapping, but the effectiveness is greatly reduced if the plates are bent.

[0003] However, the lap weld of the roof is not the most common type of two plates overlapping parallel, but rather an overlap with a curve. However, most current positioning methods are based on the lap weld type of parallel overlapping plates, which has limited applicability. Summary of the Invention

[0004] To address at least one deficiency or improvement need in the existing technology, this invention provides a method and apparatus for extracting feature points of curved lap joint bevels. It uses a 3D line laser profilometer to collect data, extracts the bevel cross-sectional profile height information, and then utilizes a feature point extraction algorithm to calculate bevel feature information. This effectively solves the problem of extracting feature information of lap weld bevels for various types of plates, improves adaptability to different plates, and achieves the goal of enhancing welding automation and intelligence.

[0005] To achieve the above objectives, according to a first aspect of the present invention, a method for extracting feature points of a curved surface overlap bevel is provided. The method includes: using a 3D profilometer to scan the overlap bevel and obtain overlap bevel contour depth data; performing multi-dimensional interpolation filtering on the overlap bevel contour depth data to remove noisy contour data; setting a Region of Interest (ROI) segmentation to eliminate redundant data in the overlap bevel cross-sectional contour depth data; using the PELT variable point detection method to process the data within the ROI region and extract multiple sets of candidate feature points; and using dynamic threshold optimization to filter the multiple sets of candidate feature points to obtain target feature points.

[0006] In an exemplary embodiment, the step of performing multi-dimensional interpolation filtering on the overlap bevel contour depth data to remove noise contour data includes: obtaining the current contour data point, traversing the contour data point, and querying whether noise contour data exists; if noise contour data exists, performing forward interpolation in the Y direction on the noise contour data; if the current contour is the starting contour or the current contour and the previous contour both have noise points at the same position, performing linear interpolation in the X and Z directions on the current noise contour data.

[0007] In an exemplary embodiment, the step of processing data within the ROI region using the PELT change point detection method to extract multiple sets of candidate feature points includes: using the PELT change point detection method to perform change point statistics on the data, and obtaining a set of change point positions (t0, t1, ..., t) that satisfy the minimum value of the total cost function. k+1 The minimum value of the total cost function is specifically, Where t0=0, t k+1 =n, where n is the length of the ROI region data. For the variable point t i to t i+1 A piece of data, Let be the cost function, β be the penalty term coefficient, and k be the number of variables.

[0008] In an exemplary embodiment, the step of using dynamic threshold optimization to filter the multiple sets of candidate feature points to obtain target feature points includes: calculating the set of local slopes for each segment of change intervals; determining a dynamic threshold by calculating a local slope benchmark value and a local standard deviation; wherein the benchmark value is the median of the local slopes of all change intervals, and the local standard deviation is used to characterize the degree of fluctuation within the data of each segment of change intervals.

[0009] In an exemplary embodiment, determining the dynamic threshold by calculating the baseline value and the local standard deviation includes: the dynamic threshold is expressed as dynamic_threshold = base_slope - K * local_std, where dynamic_threshold is the dynamic threshold, base_slope is the baseline value, local_std is the local standard deviation, and K is the sensitivity coefficient.

[0010] In an exemplary embodiment, the method further includes: if the overlap bevel type is left overlap, identifying the variable point interval where the local slope is less than the dynamic threshold as the target variable point interval, and extracting the target feature point from the target variable point interval; if the overlap bevel type is right overlap, identifying the variable point interval where the local slope is greater than the dynamic threshold as the target variable point interval, and extracting the target feature point from the target variable point interval.

[0011] According to a second aspect of the present invention, a feature point extraction device for a curved surface overlap bevel is also provided, comprising: an acquisition unit for acquiring overlap bevel contour depth data by scanning the overlap bevel using a 3D profilometer; a filtering unit for performing multi-dimensional interpolation filtering on the overlap bevel contour depth data to remove noisy contour data; a segmentation unit for setting ROI region segmentation to remove redundant data in the overlap bevel cross-sectional contour depth data; an extraction unit for processing the data within the ROI region using the PELT variable point detection method to extract multiple sets of candidate feature points; and a filtering unit for filtering the multiple sets of candidate feature points using dynamic threshold optimization to obtain target feature points.

[0012] According to a third aspect of the invention, a computer-readable storage medium is also provided, wherein a computer program is stored therein, wherein the computer program is configured to execute the above-described method for extracting feature points of surface overlap bevels during runtime.

[0013] According to a fourth aspect of the present invention, an electronic device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the above-described method for extracting feature points of curved surface bevels via the computer program.

[0014] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects:

[0015] (1) This invention provides a method for extracting feature points of curved lap joint grooves. By using a 3D profilometer to collect the cross-sectional height information of the lap weld, and based on noise filtering and ROI region selection, a PELT+dynamic threshold optimization method is further proposed for groove feature point extraction. This effectively solves the problem of extracting feature information of lap weld grooves for various types of plates, enhancing adaptability to different plates, making welding technology more automated and intelligent, saving manpower and time costs to a certain extent, simplifying the production process, and improving welding production efficiency. Utilizing multi-dimensional interpolation methods and ROI regions, the problem of noisy data generated by environmental and workpiece factors in the scanning data is effectively solved, while also addressing the problem of computational time consumption and memory overhead caused by data redundancy in analysis.

[0016] (2) In this invention, the mean mutation is proposed as a cost function in combination with the PELT method. In the complex trend of feature point change, multiple sets of candidate feature points can be effectively and accurately extracted. It has good recognition ability for both flat and curved plates. Then, dynamic threshold optimization is used to accurately locate the target point set, i.e. the bevel position, from the candidate feature point set. This greatly improves the accuracy and flexibility of bevel positioning and expands the application scenarios. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating an optional method for extracting feature points of curved surface overlap bevels provided in this application embodiment;

[0019] Figure 2 A schematic diagram illustrating the depth information of an optional overlap bevel section profile provided in an embodiment of this application;

[0020] Figure 3 An optional overlap bevel profile noise data diagram provided for an embodiment of this application;

[0021] Figure 4 A schematic diagram of the ROI region for an optional overlap bevel profile depth information provided in an embodiment of this application;

[0022] Figure 5 A schematic diagram of a set of change points found by an optional PELT change point algorithm provided in an embodiment of this application;

[0023] Figure 6 A schematic diagram of the local slope and dynamic threshold of an optional candidate variable point interval provided in an embodiment of this application;

[0024] Figure 7 A schematic diagram of an optional overlap bevel feature extraction result provided for an embodiment of this application;

[0025] Figure 8 A schematic diagram of an optional curved surface bevel feature point extraction device provided in an embodiment of this application;

[0026] Figure 9 This is a schematic diagram of an optional electronic device provided in an embodiment of this application. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0028] The terms "first," "second," "third," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0029] According to one aspect of the embodiments of this application, a method for extracting feature points of curved surface overlap bevels is provided. The following is in conjunction with... Figure 1 This application describes a method for extracting feature points of curved surface overlap bevels provided in its embodiments.

[0030] Figure 1 This is a flowchart illustrating an optional method for extracting feature points of curved surface overlap bevels provided in an embodiment of this application, as shown below. Figure 1 As shown, the process of this method may include the following steps:

[0031] S102, use a 3D profiler to scan the overlap bevel and obtain the overlap bevel profile depth data;

[0032] S104, Perform multi-dimensional interpolation filtering on the overlap bevel contour depth data to remove noise contour data;

[0033] S106, Set ROI region segmentation to remove redundant data in the overlap bevel section contour depth data;

[0034] S108. The PELT change point detection method is used to process the data within the ROI region and extract multiple sets of candidate feature points.

[0035] S110, the multiple sets of candidate feature points are filtered using dynamic threshold optimization to obtain the target feature points.

[0036] The method for extracting feature points of curved surface lap joint bevels provided in this application can be applied to intelligent welding scenarios, such as welding of car roofs and bodies.

[0037] Optionally, in cases where the sheet material is curved or overlaps with curvature, to extract the feature information of the curved surface bevel, a robot-driven 3D profilometer scans the bevel to collect data, extracting the bevel cross-sectional profile height information. Multi-dimensional interpolation filtering is then applied to the extracted cross-sectional depth profile to process data noise. Furthermore, redundant data is eliminated based on the Region of Interest (ROI), and key areas are grouped to improve analysis efficiency. Finally, the feature point extraction algorithm of this invention is used to calculate the bevel feature information.

[0038] The robot drives a 3D profilometer to scan the point cloud data of the overlapping bevel. Because the point cloud data is large in volume, processing it incurs significant memory overhead, greatly impacting efficiency and causing high computational complexity. Therefore, during the scanning process, each frame of data scanned by the 3D profilometer is processed... Figure 2 As shown, the overlap bevel profile depth information is generated.

[0039] Through steps S102 to S110 above, the lap bevel contour depth data is obtained by scanning the lap bevel using a 3D profilometer; multi-dimensional interpolation filtering is performed on the lap bevel contour depth data to remove noise contour data; ROI region segmentation is set to remove redundant data in the lap bevel cross-sectional contour depth data; the data within the ROI region is processed using the Pruned Exact Linear Time (PELT) method to extract multiple sets of candidate feature points; dynamic threshold optimization is used to filter the multiple sets of candidate feature points to obtain target feature points. This effectively solves the problem of extracting feature information of lap weld bevels of various types of plates, improves the adaptability to plates, and achieves the goal of improving welding automation and intelligence.

[0040] In one exemplary embodiment, the step of performing multi-dimensional interpolation filtering to remove noise from the overlap bevel contour depth data includes:

[0041] S11, Obtain the current contour data point, traverse the contour data point, and query whether there is noisy contour data;

[0042] S12, If noise contour data exists, perform forward interpolation of the noise contour data in the Y direction.

[0043] S13. If the current contour is the starting contour or the current contour and the previous contour are both noise points, perform linear interpolation of the X and Z dimensions on the current noise contour data.

[0044] In this embodiment, the acquired contour depth information data is filtered to reduce the impact of noise data on the extraction process of bevel feature points, such as... Figure 3 The image shows the noise data encountered by the 3D profilometer during the scanning process.

[0045] For example, when performing multi-dimensional interpolation, the current contour data point is first obtained, and the contour data point is traversed to check whether there is noise data (i.e., the Z value is less than a certain set value).

[0046] If noisy data exists, first perform forward interpolation in the Y-axis of the noise contour. Specifically, record the current noise contour data points (X0, Z0), (X1, Z1), ..., (X... m Z m ), and obtain the data points (X0,Z0), (X1,Z1),...,(X) corresponding to the noise data points of the previous contour data point. n Z n If the Z value of the previous contour data point is not a noise point, then this value is assigned to the Z value of the noise data point at the current X position.

[0047] If the current contour is the starting contour (i.e., there is no previous contour) or if the current contour and the previous contour both contain noise points at the same positions, then linear interpolation is performed on the current noise contour in the X and Z dimensions. Similarly, the data points of the currently existing noise contour (X0, Z0), (X1, Z1), ..., (X...) are recorded. m Z m ), iterate through these noisy data points, for example (X i Z i ), obtain the non-noise data points (X) at the corresponding positions of each noise data point. i-1 Z i-1 ) and (X i+1 Z i+1 If the current noise data point Z is... i for:

[0048]

[0049] The above steps are repeated during the contour feature information extraction process to complete the filtering of noisy contours during the extraction process.

[0050] In one exemplary embodiment, the step of processing data within the ROI region using the PELT change point detection method to extract multiple sets of candidate feature points includes:

[0051] S21, Use the PELT change point detection method to count abrupt changes in the data and obtain a set of abrupt change point locations (t0, t1, ..., t) that satisfy the minimum value of the total cost function. k+1 );

[0052] S22, the minimum value of the total cost function is specifically... Where t0=0, t k+1 =n, where n is the length of the ROI region data. For the variable point t i to t i+1 A piece of data, Let be the cost function, β be the penalty term coefficient, and k be the number of variables.

[0053] For example, after filtering and denoising the original contour data, since the bevel information of the overall contour depth occupies a relatively small amount of data, a region of interest of appropriate size is set to facilitate the location of the bevel, thereby reducing interference introduced by irrelevant regions and mitigating the problems caused by the inaccuracy of target analysis. This also reduces the computational time and memory overhead caused by the increase of irrelevant regions. Figure 4 As shown, the internal region defined by the two red dashed lines is the ROI region.

[0054] Furthermore, PELT is used to statistically analyze mutation points in the data. A set of mutation point locations t0, t1, ..., t k+1 Divide the X direction into k+1 segments to minimize the total cost function:

[0055]

[0056] The above t0=0, t k+1 =n (where n is the length of the ROI region data) is a fixed boundary; For the variable point t i to t i+1 A piece of data, Let β be the cost function, and β be the penalty term coefficient.

[0057] For example, such as Figure 5 As shown, the PELT algorithm is used to select the set of candidate feature points (variable points) as (t1, t2, ..., t7).

[0058] For the complex trend data within the aforementioned ROI region, the PELT change point detection method is used to find multiple sets of candidate feature points. Optionally, the model is first initialized, i.e., the cost function is defined. Since the data trend around the feature point is flat → suddenly drops → flat again, the mean abrupt change is chosen as the cost function, which is defined as follows: in, Secondly, the sparsity of the number of variable points k is controlled by the penalty term coefficient β. This parameter is used to balance the number of variable points to avoid overfitting and underfitting.

[0059] In an exemplary embodiment, the step of using dynamic threshold optimization to filter the multiple sets of candidate feature points to obtain target feature points includes:

[0060] S31, calculate the set of local slopes for each segment of the variable point interval;

[0061] S32, determine the dynamic threshold by calculating the local slope benchmark value and the local standard deviation; wherein, the benchmark value is the median of the local slope of all variable point intervals, and the local standard deviation is used to characterize the degree of fluctuation within each variable point interval data.

[0062] In this embodiment of the application, in order to extract the final feature points from the candidate feature point set (t1,t2,...,t7), dynamic threshold optimization can be used to filter out the variable points, filter out false variable points, and select the final result. At the same time, it ensures that the global fixed threshold cannot adapt to the fluctuation characteristics of different intervals.

[0063] In an optional example, first, calculate the set of local slopes for each interval of change points. Secondly, the baseline and local standard deviation are calculated to determine the dynamic threshold. It should be noted that the baseline value is the median of the local slope of all variable point intervals, while the local standard deviation is used to obtain the degree of fluctuation within each variable point interval.

[0064] In one exemplary embodiment, determining the dynamic threshold by calculating the baseline value and the local standard deviation includes:

[0065] S41, the dynamic threshold is expressed as, dynamic_threshold = base_slope - K * local_std,

[0066] Where dynamic_threshold is the dynamic threshold, base_slope is the baseline value, local_std is the local standard deviation, and K is the sensitivity coefficient.

[0067] It should be noted that the sensitivity coefficient is usually between 1.5 and 3.0, and the larger the value, the more stringent the threshold.

[0068] In one exemplary embodiment, the method further includes:

[0069] S51, if the overlap bevel type is left overlap, the variable point interval where the local slope is less than the dynamic threshold is identified as the target variable point interval, and the target feature point is extracted from the target variable point interval.

[0070] S52, if the overlap bevel type is right overlap, the variable point interval where the local slope is greater than the dynamic threshold is identified as the target variable point interval, and the target feature point is extracted from the target variable point interval.

[0071] In this embodiment of the application, the condition for determining whether it is the final change point interval is:

[0072] If the overlap bevel type is left overlap (the overlapped part is located on the operator's left side): final variable point range

[0073] If the bevel type is right-side overlap (the overlapped portion is located to the operator's left): final variable point range

[0074] For example, such as Figure 6 As shown, if the overlap bevel type described in the above steps is a left overlap bevel type, the final change point interval condition is used to determine the final local slope. The conditions are met, so the interval between the changing points is (t3:t4), meaning the final feature points (changing points) are t3 and t4. Following the steps above, as follows... Figure 7 As shown, the final result is the variable point interval composed of H and L, where H and L are the final extracted feature points (i.e., the feature points of the overlapping bevel).

[0075] This embodiment accurately locates the target point set, i.e., the bevel position, from the candidate feature point set, greatly improving the accuracy and flexibility of bevel positioning and expanding application scenarios.

[0076] According to another aspect of the embodiments of this application, a feature point extraction device is also provided for implementing the above-described method for extracting feature points of curved surface overlap bevels. Figure 8 This is a schematic diagram of an optional curved surface overlap bevel feature point extraction device according to an embodiment of this application, as shown below. Figure 8 As shown, the device may include:

[0077] The acquisition unit 802 is used to scan the overlap bevel using a 3D profiler to obtain the overlap bevel profile depth data.

[0078] Filtering unit 804 is used to perform multi-dimensional interpolation filtering on the overlap bevel contour depth data to remove noise contour data.

[0079] The segmentation unit 806 is used to set the ROI region segmentation to remove redundant data in the overlap bevel section contour depth data;

[0080] Extraction unit 808 is used to process data within the ROI region using the PELT change point detection method to extract multiple sets of candidate feature points;

[0081] The filtering unit 810 is used to filter the multiple sets of candidate feature points using dynamic threshold optimization to obtain target feature points.

[0082] It should be noted that the acquisition unit 802 in this embodiment can be used to perform the above step S102, the filtering unit 804 in this embodiment can be used to perform the above step S104, the segmentation unit 806 in this embodiment can be used to perform the above step S106, the extraction unit 808 in this embodiment can be used to perform the above step S108, and the filtering unit 810 in this embodiment can be used to perform the above step S110.

[0083] Through the above modules, the lap bevel contour depth data is obtained by scanning the lap bevel using a 3D profilometer; multi-dimensional interpolation filtering is performed on the lap bevel contour depth data to remove noise contour data; ROI region segmentation is set to remove redundant data in the lap bevel cross-sectional contour depth data; the PELT variable point detection method is used to process the data within the ROI region to extract multiple sets of candidate feature points; dynamic threshold optimization is used to filter the multiple sets of candidate feature points to obtain target feature points. This effectively solves the problem of extracting feature information of lap weld bevels of various types of plates, improves the adaptability to plates, and achieves the goal of improving welding automation and intelligence.

[0084] In an exemplary embodiment, the filtering unit includes: a first acquisition module, configured to acquire current contour data points, traverse the contour data points, and query whether noise contour data exists; a first interpolation module, configured to perform forward interpolation of the noise contour data in the Y direction if noise contour data exists; and a second interpolation module, configured to perform linear interpolation of the current noise contour data in the X and Z directions if the current contour is the starting contour or if the current contour and the previous contour both have noise points at the same position.

[0085] In one exemplary embodiment, the extraction unit includes: a change point detection module, used to perform change point statistics on the data using the PELT change point detection method, and obtain a set of change point positions (t0, t1, ..., t2) that satisfy the minimum value of the total cost function. k+1 The minimum value of the total cost function is specifically, Where t0=0, t k+1 =n, where n is the length of the ROI region data. For the variable point t i to t i+1 A piece of data, Let be the cost function, β be the penalty term coefficient, and k be the number of variables.

[0086] In an exemplary embodiment, the filtering unit includes: a calculation module for calculating the set of local slopes for each segment of the change point interval; and a determination module for determining a dynamic threshold by calculating a local slope benchmark value and a local standard deviation; wherein the benchmark value is the median of the local slopes of all change point intervals, and the local standard deviation is used to characterize the degree of fluctuation within the data of each segment of the change point interval.

[0087] In an exemplary embodiment, the determining module includes a dynamic threshold submodule, wherein the dynamic threshold is expressed as dynamic_threshold = base_slope - K * local_std, where dynamic_threshold is the dynamic threshold, base_slope is the baseline value, local_std is the local standard deviation, and K is the sensitivity coefficient.

[0088] In one exemplary embodiment, the apparatus further includes: a first determining unit, configured to, if the overlap bevel type is left overlap, identify the variable point interval where the local slope is less than the dynamic threshold as a target variable point interval, and extract the target feature point from the target variable point interval; and a second determining unit, configured to, if the overlap bevel type is right overlap, identify the variable point interval where the local slope is greater than the dynamic threshold as a target variable point interval, and extract the target feature point from the target variable point interval.

[0089] It should be noted that the examples and scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should be noted that the above modules, as part of the device, can run in a hardware environment and can be implemented by software or hardware. The hardware environment includes a network environment.

[0090] According to another aspect of the embodiments of this application, a storage medium is also provided. Optionally, in this embodiment, the storage medium can be used to execute the program code of any of the surface overlap bevel feature point extraction methods described above in the embodiments of this application.

[0091] Optionally, in this embodiment, the storage medium is configured to store program code for performing the following steps:

[0092] S1. Use a 3D profilometer to scan the overlap bevel and obtain the overlap bevel profile depth data.

[0093] S2, perform multi-dimensional interpolation filtering on the overlap bevel contour depth data to remove noise contour data;

[0094] S3, Set ROI region segmentation to remove redundant data in the overlap bevel section contour depth data;

[0095] S4. The PELT change point detection method is used to process the data within the ROI region and extract multiple sets of candidate feature points.

[0096] S5, use dynamic threshold optimization to filter the multiple sets of candidate feature points to obtain target feature points.

[0097] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments, and will not be repeated in this embodiment.

[0098] The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, as well as magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0099] According to another aspect of the embodiments of this application, an electronic device for implementing the above-described method for extracting feature points of curved surface bevel joints is also provided. The electronic device may be a server, a terminal, or a combination thereof.

[0100] Figure 9 This is a schematic diagram of the structure of an optional electronic device according to an embodiment of this application, such as... Figure 9 As shown, it includes a processor 902, a communication interface 904, a memory 906, and a communication bus 908. The processor 902, communication interface 904, and memory 906 communicate with each other via the communication bus 908.

[0101] Memory 906 is used to store computer programs;

[0102] When processor 902 executes a computer program stored in memory 906, it performs the following steps:

[0103] S1. Use a 3D profilometer to scan the overlap bevel and obtain the overlap bevel profile depth data.

[0104] S2, perform multi-dimensional interpolation filtering on the overlap bevel contour depth data to remove noise contour data;

[0105] S3, Set ROI region segmentation to remove redundant data in the overlap bevel section contour depth data;

[0106] S4. The PELT change point detection method is used to process the data within the ROI region and extract multiple sets of candidate feature points.

[0107] S5, use dynamic threshold optimization to filter the multiple sets of candidate feature points to obtain target feature points.

[0108] Optionally, the communication bus can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 9 The symbol is represented by a single thick line, but this does not indicate that there is only one bus or one type of bus. The communication interface is used for communication between the aforementioned electronic device and other devices.

[0109] The memory may include RAM, or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0110] As an example, the memory 906 described above may include, but is not limited to, the acquisition unit 802, filtering unit 804, segmentation unit 806, extraction unit 808, and screening unit 810 from the surface overlap bevel feature point extraction device described above. Furthermore, it may include, but is not limited to, other module units from the surface overlap bevel feature point extraction device described above, which will not be elaborated upon in this example.

[0111] The processor mentioned above can be a general-purpose processor, including but not limited to: CPU (Central Processing Unit), NP (Network Processor), etc.; it can also be DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0112] Optionally, specific examples in this embodiment can refer to the examples described in the above embodiments, and will not be repeated here.

[0113] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0114] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0115] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0116] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0117] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0118] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0119] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0120] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of embodiments of this disclosure upon considering the specification and practicing the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described herein. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

[0121] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0122] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for extracting feature points of a curved surface overlap bevel, characterized in that, include: Use a 3D profilometer to scan the overlap bevel and obtain the overlap bevel profile depth data; The overlap bevel contour depth data is subjected to multi-dimensional interpolation filtering to remove noise contour data; Set up ROI region segmentation to remove redundant data from the overlap bevel section contour depth data; The PELT change point detection method is used to process the data within the ROI region and extract multiple sets of candidate feature points; The target feature points are obtained by filtering the multiple sets of candidate feature points using dynamic threshold optimization. The process of using dynamic threshold optimization to filter the multiple sets of candidate feature points to obtain target feature points includes: Calculate the set of local slopes for each interval of change points; The dynamic threshold is determined by calculating the local slope baseline value and the local standard deviation; wherein, the baseline value is the median of the local slope of all variable point intervals, and the local standard deviation is used to characterize the degree of fluctuation within each variable point interval data segment. The determination of the dynamic threshold by calculating the local slope benchmark value and the local standard deviation includes: The dynamic threshold is expressed as follows: ,in, For dynamic thresholds, As the baseline value, For local standard deviation, This is the sensitivity coefficient; The method further includes: If the overlap bevel type is left overlap, the variable point interval where the local slope is less than the dynamic threshold is identified as the target variable point interval, and the target feature point is extracted from the target variable point interval. If the overlap bevel type is right overlap, the variable point interval where the local slope is greater than the dynamic threshold is identified as the target variable point interval, and the target feature point is extracted from the target variable point interval.

2. The method for extracting feature points of curved surface overlap bevels as described in claim 1, characterized in that, The step of performing multi-dimensional interpolation filtering to remove noise from the overlap bevel contour depth data includes: Obtain the current contour data point, traverse the contour data point, and query whether there is noisy contour data; If noisy contour data exists, perform forward interpolation of the noise contour data in the Y direction. If the current contour is the starting contour or both the current contour and the previous contour are noise points, perform linear interpolation of the X and Z dimensions on the current noise contour data.

3. A device for extracting feature points of a curved surface overlap bevel, used to perform the method for extracting feature points of a curved surface overlap bevel as described in claim 1, characterized in that, include: The acquisition unit is used to scan the overlap bevel using a 3D profilometer to obtain the overlap bevel profile depth data. The filtering unit is used to perform multi-dimensional interpolation filtering on the overlap bevel contour depth data to remove noise contour data. The segmentation unit is used to set the ROI region segmentation and remove redundant data from the overlap bevel section contour depth data; The extraction unit is used to process the data within the ROI region using the PELT change point detection method and extract multiple sets of candidate feature points. The filtering unit is used to filter the multiple sets of candidate feature points using dynamic threshold optimization to obtain target feature points.

4. The feature point extraction device for curved surface overlap bevel as described in claim 3, characterized in that, The filtering unit includes: The first acquisition module is used to acquire the current contour data points, traverse the contour data points, and query whether there is noisy contour data. The first interpolation module is used to perform forward interpolation of the noise contour data in the Y direction if noise contour data exists. The second interpolation module is used to perform linear interpolation of the X and Z dimensions of the current noise contour data if the current contour is the starting contour or if the current contour and the previous contour are both noise points at the same position.

5. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program, when executed, performs the method of any one of claims 1 to 2.

6. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to execute the method of any one of claims 1 to 2 through the computer program.

Images