Automated Detection Method and System for Forming Dimensions of High-Reflectivity Bowl-Shaped Stamping Parts

Abstract

The present invention discloses an automated detection method and system for the forming dimensions of a highly reflective bowl-shaped stamping part, belonging to the field of metal plastic forming processes and detection. The present invention adopts the principle of non-contact laser measurement, and designs an inverse distance weighted interpolation algorithm considering curvature changes in the data processing stage, effectively suppressing the missing or distortion of measurement data caused by the highly reflective material of the workpiece. Therefore, during measurement, it is not affected by the highly reflective characteristics of the metal material. While stably achieving the improvement of measurement accuracy, the detection efficiency is also accelerated. The present invention obtains the information of the forming dimensions to be measured through automated online analysis, effectively solving the industrial production problem that it is difficult to complete batch detection of multi-size bowl-shaped stamping parts using customized measuring tools by traditional measurement methods.

Description

Technical Field

[0001] The present invention belongs to the field of metal plastic forming processes and detection, and more specifically, relates to an automated detection method and system for the forming dimensions of highly reflective bowl-shaped stamping parts. Background Art

[0002] As a seal to ensure that automotive parts meet the requirements of hot and cold processing technologies, the forming dimension reliability of bowl-shaped stamping parts usually requires a relatively high level. Its shape is simple, and it is difficult to control the forming accuracy in a single stamping process. Conducting a complete automated analysis of its overall forming dimensions (wall thickness, depth, taper, etc.) can provide basic measurement data for aspects such as product quality assurance, forming die repair, and reduction of the overall machine failure rate.

[0003] Currently, enterprises mainly rely on manual use of traditional mechanical measuring tools (depth gauges, vernier calipers, etc.) or customized inspection tools to detect the forming dimensions of bowl-shaped stamping parts. Although this type of contact detection method can complete the detection task with high precision and high stability, relatively speaking, these methods can only perform sampling detection on a small batch or key dimensions at limited positions, and cannot complete the three-dimensional measurement and accuracy analysis of the overall surface. On the other hand, the detection results dominated by manual participation are greatly affected by human factors, and there are problems such as poor consistency, high error rate, and low efficiency. These problems seriously restrict the improvement of the manufacturing accuracy of bowl-shaped stamping parts and pose potential safety hazards to the stable operation of automotive parts.

[0004] In the face of the current demand for the automated development of the high efficiency and accuracy of the quality inspection of automotive structural parts in the transportation field, some research institutions and enterprises have tried to use optical intelligent detection methods to ensure the detection accuracy and efficiency of workpiece processing errors. In actual detection conditions based on digital image processing technology, it is often necessary to ensure the high stability of data acquisition, the high accuracy of data stitching, and the high efficiency of data processing in order to meet the large batch detection requirements of enterprises. However, restricted by factors such as the diversity of the forming dimensions of the parts themselves and the high reflectivity of the metal material, it immediately leads to distorted data acquisition, and ultimately results in the detection accuracy not meeting the requirements of actual industrial production. Summary of the Invention

[0005] In view of the above-mentioned defects or improvement requirements of the prior art, the present invention provides an automated detection method and system for the forming dimensions of highly reflective bowl-shaped stamping parts, aiming to achieve the automated measurement of the forming dimensions of bowl-shaped stamping parts and solve the technical problem of measurement distortion caused by the highly reflective material of the parts.

[0006] To achieve the above object, according to one aspect of the present invention, there is provided an automated detection method for the forming dimensions of highly reflective bowl-shaped stamping parts, including:

[0007] S1. Scan the contour of the workpiece to be measured in a fixed posture by the first line laser sensor;

[0008] S2. Scan the contour of the workpiece to be measured in another posture by the second line laser sensor;

[0009] S3. Convert and merge the local measurement point clouds to obtain the initial point cloud of the complete contour of the workpiece to be measured;

[0010] S4. Preprocess the initial point cloud of the complete contour: Remove the outlier points and burr points, and correct the accuracy of the locally highly reflective measurement distortion points of the workpiece to be measured;

[0011] S5. Fit and reconstruct the preprocessed complete contour point cloud, and analyze to obtain the corresponding formed size information.

[0012] Further, step S4 is specifically as follows,

[0013] If the Z coordinate component of a certain point exceeds the preset reference limit Then this point is identified as an outlier point and removed; if the depth change Δz between adjacent data points exceeds the depth change critical value ΔZ, then this point is identified as a burr point and removed

[0014] According to the workpiece CAD design model Z 0 (x i , y i ), obtain the curvature information ρ of the missing points at the highly reflective surface i_0 ;

[0015] Construct the minimum neighborhood from the projection coordinate information of the missing points in the measurement plane, calculate the distance d from the missing points in the neighborhood to all points i , and obtain its weight λ i :

[0016]

[0017] n is the number of measured points in the neighborhood, w is the adjustment coefficient, and the value range is 0 - 1;

[0018] Calculate the depth information of the missing points:

[0019]

[0020] Calculate and extract the curvature value ρ at the compensation point from the compensated point cloud information i , when ρ i ≤ρ i_0 is satisfied, then it is regarded as a valid compensation point, otherwise readjust w until the above judgment criterion is satisfied.

[0021] Further, when scanning, the measurement light field of the line laser sensor covers the complete contour of the workpiece.

[0022] The present invention also provides an automatic detection system for the forming dimensions of a highly reflective bowl-shaped stamping part, including: a data acquisition module, a motion control module, and a data processing module;

[0023] The data acquisition module includes a first line laser sensor and a second line laser sensor; the motion control module includes a first conveyor belt and a second conveyor belt; the first line laser sensor is arranged above the first conveyor belt, and the second line laser sensor is arranged above the second conveyor belt; the measurement plane of each line laser sensor is parallel to the motion plane of the corresponding conveyor belt, and the line laser light strip emitted by the line laser sensor is perpendicular to the motion direction of the corresponding conveyor belt, ensuring that the measurement light field of the line laser sensor can cover the complete contour of the workpiece;

[0024] The first conveyor belt and the second conveyor belt are respectively used to realize the single and orderly conveyance of the workpiece to be measured in different postures;

[0025] The first line laser sensor and the second line laser sensor are respectively used to obtain the contour data of the front and back sides of the workpiece;

[0026] The data processing module is used to analyze the contour images of the workpiece to be measured collected by the first line laser sensor and the second line laser sensor in their respective local coordinate systems into three-dimensional data, and perform precision correction on the local highly reflective measurement distortion points of the workpiece to complete the reconstruction of the three-dimensional model contour of the workpiece to be measured, and further realize the effective extraction of the detected dimensions.

[0027] Further, the specific process of calibrating the global coordinate system of the detection system is as follows:

[0028] Use the line laser sensor to scan the target image at a specified position;

[0029] Extract the geometric information of the feature points from the collected image, and then, based on the principle of the consistency of the rigid transformation of the feature points in the three-dimensional space, complete the calibration of the sensor camera by constructing the matching relationship between the point pairs;

[0030] Using the camera calibration result and combining the position information of the center point of the collected target image, complete the calibration of the laser plane and the motion direction of the line laser sensor by plane fitting, and finally achieve the global unification of the local measurement coordinate systems within the measurement system.

[0031] Further, the specific implementation process of the data processing module is as follows:

[0032] Analyze the measurement point cloud of each line laser sensor in the local coordinate system;

[0033] Based on the global coordinate system calibration result, obtain the initial point cloud of the inner and outer contours of the workpiece to be measured by converting and splicing the local measurement point cloud;

[0034] Remove the burr points and outliers from the initial point cloud: If the Z - coordinate component of a certain point exceeds the preset reference limit then this point is identified as an outlier and is removed; if the depth change Δz between adjacent data points exceeds the depth - change critical value ΔZ, then this point is identified as a burr point and is removed;

[0035] According to the Z 0 (x i ,y i ) of the workpiece CAD design model, obtain the curvature information ρ of the missing points at the highly reflective areas on the surface i_0 ;

[0036] Construct the minimum neighborhood from the projection coordinate information of the missing points on the measurement plane, and calculate the distance d from the missing points in the neighborhood to all points i , and obtain its weight λ i :

[0037]

[0038] n is the number of measured points in the neighborhood, w is an adjustment coefficient, and its value ranges from 0 to 1;

[0039] Calculate the depth information of the missing points:

[0040]

[0041] Calculate and extract the curvature value ρ at the compensation points from the compensated point - cloud information i , when ρ i ≤ρ i_0 is satisfied, then it is regarded as a valid compensation point, otherwise readjust w until the above - mentioned judgment criterion is met.

[0042] Furthermore, the detection system further includes a real - time display module; the real - time display module is connected to the image - processing module and is used to display the automated detection results of the forming dimensions of the bowl - shaped stamping part.

[0043] Generally speaking, compared with the prior art by the above - mentioned technical solution conceived by the present invention, the following beneficial effects can be achieved.

[0044] (1) The present invention adopts the non - contact laser measurement principle and designs the inverse - distance weighted interpolation algorithm considering curvature changes in the data - processing stage, effectively suppressing the missing or distorted measurement data caused by the highly reflective material of the workpiece. Therefore, during measurement, it is not affected by the highly reflective characteristics of the metal material. While stably achieving the improvement of measurement accuracy, the detection efficiency is also increased.

[0045] (2) The present invention sets the scanning process line laser sensor in a stationary state and relies on the translation of the conveyor belt, which effectively reduces the system measurement error caused by multiple splicing of point clouds involved in multiple local coordinate systems under the multi-posture viewing angle measurement of the sensor.

[0046] (3) The present invention obtains the information of the formed dimensions to be measured through automated online analysis, which effectively solves the industrial production problem that it is difficult to use customized measuring tools to complete batch inspection of bowl-shaped stamping parts of multiple sizes using traditional measurement methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a flow chart of a measurement method constructed according to a preferred embodiment of the present invention;

[0048] Figure 2 is a schematic structural diagram of a measuring device constructed according to a preferred embodiment of the present invention;

[0049] Figure 3 is a schematic diagram of a bowl-shaped stamping part to be tested according to a preferred embodiment of the present invention;

[0050] Figure 4 It is a schematic diagram of solving the forming dimensions of a bowl-shaped stamping part to be tested constructed according to a preferred embodiment of the present invention;

[0051] Throughout the drawings, the same reference numerals are used to denote the same elements or structures, wherein:

[0052] 1 is a frame, 2 and 5 are conveying mechanisms, 3 and 6 are sensor supporting racks, 4 and 7 are auxiliary workpiece turning mechanisms, 8 and 10 are line laser sensors, 11 is a workpiece forming quality qualified sorter, 12 is a qualified workpiece container, and 13 is a unqualified workpiece container. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0054] refer to Figure 1 The present invention provides an automatic detection method for the forming dimensions of a highly reflective bowl-shaped stamping part, which is characterized by comprising:

[0055] S1. Scanning the contour of the workpiece under fixed posture by the first line laser sensor;

[0056] S2. Scan the contour of the workpiece to be measured in another posture by the second line laser sensor;

[0057] S3. Convert and splice the local measurement point clouds to obtain the initial point cloud of the complete contour of the workpiece to be measured;

[0058] S4. Preprocess the initial point cloud of the complete contour: Remove outlier points and burr points, and perform accuracy correction on the locally highly reflective measurement distortion points of the workpiece to be measured;

[0059] S5. Fit and reconstruct the preprocessed complete contour point cloud, and analyze to obtain the corresponding forming dimension information.

[0060] To implement the above method, an embodiment of the present invention also provides Figure 2 The automatic detection system for the forming dimensions of a highly reflective bowl-shaped stamping part shown in the figure, including a motion control module, a data acquisition module, a data processing module, and a real-time display module, wherein:

[0061] The motion control module preferentially completes the single-drive diversion of the workpiece to be measured, so that it can move smoothly in an equidistant order and keep the workpiece in an upward / downward opening state.

[0062] The data acquisition module consists of 2 line laser sensors, which are cascaded above the conveyor belt. During measurement, the workpiece is at the center of the sensor measurement light field. Specifically, during implementation, the sensors are in a stationary state, and the workpiece is driven by the logic motion control module and pulled by the conveyor belt to the turntable to complete the scan, thereby batch obtaining the overall contour data of the workpiece.

[0063] The data processing module mainly analyzes the contour image of the workpiece to be measured collected by the line laser sensor into three-dimensional data and analyzes it to obtain the detection result. The implementation steps of the data processing module are as follows:

[0064] 1) First, on the basis of completing the calibration work of the line laser sensor, analyze the contour images collected by the sensor units in their respective local coordinate systems into three-dimensional data;

[0065] 2) Then, according to the global coordinate system calibration result, complete the coordinate unification of the local scan three-dimensional data points, thereby obtaining the initial point cloud of the complete three-dimensional contour of the workpiece;

[0066] 3) Utilize the distribution characteristics of the measurement point clouds in the depth direction (Z-axis direction) to preferentially remove outlier points and burr points. On this basis, obtain the contour surface curvature change information from the CAD design model of the workpiece, and design an inverse distance weighted interpolation algorithm considering the curvature change to perform accuracy correction on the locally highly reflective measurement distortion points of the workpiece;

[0067] 4) Combining with the specific definitions of the dimensions to be detected of the workpiece (such as wall thickness, depth, taper, etc.), and by means of the data fitting algorithm, on the basis of completing the three-dimensional model contour reconstruction of the workpiece to be detected, the effective extraction of the detection dimensions is realized.

[0068] In the present invention, by setting the line laser sensor in a stationary state during the scanning process and relying on the translational movement of the conveyor belt, the system measurement error caused by the multiple point cloud mosaics involved in the unification of multiple local coordinate systems under the multi-attitude perspective measurement of the sensor is effectively reduced. And in the data processing stage, an inverse distance weighted interpolation algorithm considering the curvature change is designed to effectively suppress the missing or distortion of the measurement data caused by the highly reflective material of the workpiece.

[0069] The workpiece to be detected is as Figure 3 shown. The working process of the detection system is as follows: When the mechanism 7 loading the workpiece guides a single workpiece to slide down along the X-axis one by one to the conveying mechanism 5 through mechanical vibration, at this time, the line laser sensor 8 is used to dynamically complete the scanning and sampling work of the single-sided contour image of the workpiece. Then, the mechanism 4 completes the flipping of the workpiece. At this time, with the assistance of the conveying mechanism 2, the line laser sensor 10 samples the other side contour of the workpiece. After the sampling is completed, it enters the forming quality qualified selector 11 of the workpiece. At this time, if it is qualified, it enters the container 13 for unqualified detected workpieces, otherwise, it enters the container 12 for qualified detected workpieces. By repeating the above steps, the automatic batch detection of the forming dimensions of the bowl-shaped stamping parts is effectively realized.

[0070] After the scanning and sampling work of the overall contour image of the workpiece obtained by the line laser sensors 8 and 10 in the above steps, at this time, the upper computer PC terminal completes the image analysis and integrates and transforms it into the corresponding three-dimensional spatial point cloud. In this process, the complete point cloud of the workpiece is obtained by each line laser sensor in the measurement system. The specific implementation steps are as follows:

[0071] (1) First, the line laser sensor scans the planar target for calibration, and extracts and sorts the centers of the circles of the images collected by each sensor in the system, so as to establish the corresponding points of the center coordinates of the camera calibration board. Here, the calibration target used is made of ceramic or marble material with less influence by temperature change.

[0072] (2) Then, according to the internal parameters of the sensor, calculate the three-dimensional coordinates of the center of the calibration board circle in the local coordinate system of each line laser sensor. Secondly, according to the three-dimensional coordinates of the center of the calibration board circle in the global coordinate system, complete the calibration of the laser plane of the sensor, and thus realize the unification of the local coordinate system and the global coordinate system.

[0073] (3) Using the calibration result of the coordinate system, register and fuse the inner and outer contour point clouds of the workpiece to be detected extracted by the line laser sensors in the system respectively, and obtain the initial point cloud of the complete contour of the workpiece.

[0074] Figure 4It is a schematic diagram for solving the forming dimensions of a bowl-shaped stamping part constructed according to the preferred embodiment of the present invention. The specific implementation steps of this process are as follows:

[0075] (1) First, as shown in (a) of Figure 4 , consider the distribution characteristics of the target point cloud in the depth direction (Z-axis direction), and set the upper and lower reference limit values and the critical depth change value ΔZ. If the Z coordinate component of a certain point exceeds the preset reference limit , then this point is identified as an outlier and removed; further consider the depth change Δz between adjacent data points. When ΔZ < Δz is satisfied, then this point is identified as a burr point and removed.

[0076] (2) Secondly, according to the workpiece CAD design model Z 0 (x i , y i ), obtain the curvature information ρ i_0 of the missing points at the highly reflective surface. Further, construct the smallest neighborhood from the projection coordinate information of the missing points on the measurement plane (such as the XOY plane), calculate the distance d i from the missing points in the neighborhood to all points, and obtain its weight λ i , as shown in (b) of Figure 4 :

[0077]

[0078] n is the number of measured points in the neighborhood. w is an adjustment coefficient, and its value ranges from 0 to 1.

[0079] Thus, the depth information of the missing points is obtained:

[0080]

[0081] Finally, calculate and extract the curvature value ρ i at the compensation point from the compensated point cloud information. When ρ i ≤ρ i_0 is satisfied, it is regarded as a valid compensation point; otherwise, readjust w in formula (1) until the above judgment criterion is met.

[0082] (3) On this basis, as shown in (c) of Figure 4 , based on the least squares fitting, use the extracted contour outer ring data point cloud to complete the fitting construction of the workpiece outer circle contour equation, and thus solve for the outer contour diameter D.

[0083] (4) Similarly, as shown in (d) of Figure 4 , select the middle section of the workpiece as the reference section for solving other parameters, and complete the extraction of the middle section point cloud data according to the Z component information of the scanned point cloud data.

[0084] (5) On the basis of step (3), based on the least squares method, the upper and lower contour section curve equations f 1 (x) and f 2 (x) of the middle section are respectively fitted; thus, the formed wall thickness △h is obtained from the difference information of the highest points of the upper and lower contour curves of this section.

[0085] (6) Further, from the intersection information of the upper contour curve f 1 (x) and the Y-axis of the section, the formed height h of the workpiece can be solved.

[0086] (7) Then, the tangent equation F(x) of the upper contour curve f 1 (x) at the intersection with the Y-axis is preferentially solved, and thus the taper λ of the workpiece is solved from the tangent slope.

[0087] (8) Finally, on the above basis, the calculated results are compared with the standard design values. When all the formed dimension information to be detected is within the deviation range, the formed quality of the workpiece is qualified; otherwise, it is determined to be unqualified.

[0088] In the present invention, intelligent detection is implemented by combining the automatic scanning of a line laser sensor. The detection process is mainly divided into the following steps: the inner and outer complete contours of the bowl-shaped stamping part are scanned by multiple line laser sensors, and a weight interpolation algorithm considering the curvature change of the measurement surface is designed. While suppressing the measurement distortion caused by the highly reflective metal surface of the workpiece, the three-dimensional point cloud information of the complete contour of the workpiece to be measured is obtained; then, based on the section contour fitting, the automatic extraction of the formed dimensions of the highly reflective bowl-shaped stamping part is completed. The present invention not only realizes the automatic extraction of the formed dimensions (wall thickness, taper, depth, diameter, etc.) of the workpiece, but also further obtains the specific formed quality information of the workpiece by using the extraction results, and finally improves the overall service performance of the workpiece. In short, the present invention realizes the rapid detection of the formed dimensions of the bowl-shaped stamping part, has the advantages of high efficiency, strong adaptability, high detection result accuracy, etc. The present invention is particularly suitable for the intelligent extraction of the contour dimensions of circular ring parts.

[0089] Existing non-destructive testing technologies are not suitable for use considering the detection environment, accuracy, testing equipment, production batch, etc. of the overall contour morphology characteristics of bowl-shaped stamped parts. For example: coordinate measuring machines (when using contact probes, there are defects such as difficult scanning path planning and slow measurement speed during measurement, which directly leads to inability to measure some positions and inability to achieve corresponding dense sampling detection), ultrasonic methods (easily affected by on-site environmental noise, so the accuracy cannot be guaranteed), ray detection methods (such as X-rays, there is a problem of too large ray spot, so the sampling accuracy cannot meet the measurement requirements), machine vision methods (require high requirements for scanning equipment, directly resulting in an increase in detection costs. In addition, it is also affected by the high reflectivity characteristics of the metal material on the workpiece surface, and it is difficult to obtain high-precision detection results), etc. In this example, a line laser sensor is used to automatically extract the data points of the complete contour morphology of the bowl-shaped stamped part by generating speckle stripes. Based on the non-contact characteristics of laser optics, non-destructive measurement sampling can be achieved, and an independently designed data compensation and correction algorithm can effectively solve problems such as measurement distortion caused by the high-reflectivity material of the workpiece. Finally, with the help of online assisted analysis by the data fitting and reconstruction algorithm, the overall forming size distribution of the workpiece is obtained in real time, and then compared with the standard design value to obtain the forming quality detection result.

[0090] Those skilled in the art can easily understand that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An automated detection method for the forming dimensions of a highly reflective bowl-shaped stamping part, characterized in that, it includes: S1. Scanning the contour of the workpiece to be measured in a fixed posture by a first line laser sensor; S2. Scanning the contour of the workpiece to be measured in another posture by a second line laser sensor; S3. Converting and stitching the local measurement point clouds to obtain the initial point cloud of the complete contour of the workpiece to be measured; S4. Preprocessing the initial point cloud of the complete contour: removing outlier points and burr points, and performing precision correction on the locally highly reflective measurement distortion points of the workpiece to be measured; S5. Fitting and reconstructing the preprocessed complete contour point cloud, and analyzing to obtain the corresponding forming dimension information; Specifically, step S4 is If the Z - coordinate component of a certain point exceeds the preset reference limit then this point is identified as an outlier and removed; if the depth change Δz between adjacent data points exceeds the depth - change critical value ΔZ, then this point is identified as a spurious point and removed; According to the workpiece CAD design model Z 0 (x i , y i ), obtain the curvature information ρ of the missing points at the highly reflective surface i_0 ; Construct the smallest neighborhood from the projection coordinate information of the missing points on the measurement plane, and calculate the distance d from the missing points in the neighborhood to all points i , and obtain its weight λ i : n is the number of measured points in the neighborhood, w is the adjustment coefficient, and the value range is 0-1; Calculating the depth information of the missing points: Calculate and extract the curvature value ρ at the compensation point from the compensated point cloud information i , when ρ i ≤ρ i_0 is satisfied, it is regarded as a valid compensation point; otherwise, w is adjusted again until the above judgment criterion is met.

2. An automated detection method for the forming dimensions of a highly reflective bowl-shaped stamping part according to claim 1, characterized in that, During scanning, the measurement light field of the line laser sensor covers the complete contour of the workpiece.

3. An automated detection system for the forming dimensions of a highly reflective bowl-shaped stamping part, characterized in that, it includes: A data acquisition module, a motion control module and a data processing module; The data acquisition module includes a first line laser sensor and a second line laser sensor; The motion control module includes a first conveyor belt and a second conveyor belt; the first line laser sensor is arranged above the first conveyor belt, and the second line laser sensor is arranged above the second conveyor belt; the measurement plane of each line laser sensor is parallel to the motion plane of the corresponding conveyor belt, and the line laser light strip emitted by the line laser sensor is perpendicular to the motion direction of the corresponding conveyor belt, ensuring that the measurement light field of the line laser sensor can cover the complete contour of the workpiece; The first conveyor belt and the second conveyor belt are respectively used to realize the single-order transfer of the workpiece to be measured in different postures; The first line laser sensor and the second line laser sensor are respectively used to obtain the contour data of the front and back sides of the workpiece; The data processing module is used to analyze the contour images of the workpiece to be measured collected by the first line laser sensor and the second line laser sensor in their respective local coordinate systems into three-dimensional data, and perform precision correction on the locally highly reflective measurement distortion points of the workpiece to complete the reconstruction of the three-dimensional model contour of the workpiece to be measured, and then realize the effective extraction of the detected dimensions; The specific implementation process of the data processing module is: Analyzing the measurement point clouds of each line laser sensor in the local coordinate system; Based on the calibration result of the global coordinate system, by converting and stitching the local measurement point clouds, obtaining the initial point clouds of the inner and outer contours of the workpiece to be measured; Remove the burr points and outliers from the initial point cloud: If the Z - coordinate component of a certain point exceeds the preset reference limit then this point is identified as an outlier and is removed; If the depth change Δz between adjacent data points exceeds the depth - change critical value ΔZ, then this point is identified as a burr point and is removed; According to the workpiece CAD design model Z 0 (x i , y i ), obtain the curvature information ρ of the missing points at the highly reflective surface i_0 ; Construct the minimum neighborhood from the projection coordinate information of the missing points in the measurement plane, and calculate the distance d from the missing points in the neighborhood to all points i , and obtain its weight λ i : n is the number of measured points in the neighborhood, w is the adjustment coefficient, and the value range is 0-1; Calculating the depth information of the missing points: Calculate and extract the curvature value ρ at the compensation point from the compensated point cloud information i , when ρ i ≤ρ i_0 is satisfied, it is regarded as a valid compensation point; otherwise, w is adjusted again until the above judgment criterion is met.

4. An automated detection system for the forming dimensions of a highly reflective bowl-shaped stamping part according to claim 3, characterized in that, The specific process of calibrating the global coordinate system of the detection system is: Using the line laser sensor to scan the target image at a specified position; Extracting the geometric information of the feature points from the collected image, and then based on the principle of the consistency of the three-dimensional spatial rigid transformation of the feature points, by constructing the matching relationship between point pairs, completing the calibration of the sensor camera. Using the camera calibration results and combining with the position information of the center point of the collected target image, the calibration of the laser plane and the motion direction of the line laser sensor is completed by plane fitting, and finally the global unification of each local measurement coordinate system in the measurement system is achieved.

5. An automatic detection system for the forming size of a highly reflective bowl-shaped stamping part according to claim 3 or 4, characterized in that, the detection system further includes a real-time display module; the real-time display module is connected to the data processing module and is used to display the automatic detection results of the forming size of the bowl-shaped stamping part.

Images