Workpiece appearance defect position automatic tracking method and system

Abstract

The invention discloses a workpiece appearance defect position automatic tracking method and system. The method comprises the steps that a plurality of observation point positions are arranged on a workpiece; adjusting a shooting angle according to the plurality of observation points, and obtaining a workpiece appearance detection image set of each processing node; setting original processing nodes, and dividing each processing node of the production line into an initial processing node set and a subsequent processing node set based on the original processing nodes; performing three-dimensional reconstruction on the workpiece appearance defect image data to obtain three-dimensional coordinate data of an appearance defect position; acquiring initial defect image data and defect change image data; and marking occurrence nodes of appearance defect positions of the workpiece based on the initial defect image data, and generating appearance defect tracking data. Defect position detection tracking analysis is carried out in the workpiece production process through the visual detection technology, the reliability of traceability detection of workpiece appearance machining defects by a production line is improved, meanwhile, the convenience of maintenance and adjustment of the production line is improved, and therefore the machining quality of the production line on the workpieces is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of workpiece visual inspection, and in particular to a method and system for automatically tracking the position of a workpiece appearance defect. Background Art

[0002] Automated production is the main development trend of mechanical processing at present. The processing efficiency of workpieces can be improved through fully automated production line operations. The current automated production lines generally obtain the processing yield rate of the production line by performing appearance defect detection on the workpieces after processing. This method can only detect the finished product status of the workpiece, and cannot immediately feedback the changes in the processing status of the workpiece in the production line. The processing quality of the workpiece on the production line needs to be improved, and since the production line cannot be accurately maintained according to the finished product status of the workpiece, the convenience of production line maintenance and adjustment is affected. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art. The present invention provides a method and system for automatically tracking the position of workpiece appearance defects. Through visual inspection technology, defect position detection, tracking and analysis are performed during the workpiece production process, thereby improving the reliability of the production line's tracing detection of workpiece appearance processing defects, and at the same time improving the convenience of production line maintenance and adjustment, thereby improving the production line's processing quality of the workpiece.

[0004] The present invention provides a method for automatically tracking the position of a workpiece appearance defect, the method comprising:

[0005] Obtain historical data on the appearance defects of workpieces processed on the production line, and set several observation points on the workpieces based on the historical data;

[0006] The workpiece is visually inspected at each processing node on the production line by using the shooting component, and the shooting angle is adjusted based on several observation points to obtain a set of workpiece appearance inspection images at each processing node;

[0007] Acquire processing nodes where workpiece appearance defects occur according to the workpiece appearance inspection image set, mark the processing nodes as original processing nodes, and divide each processing node of the production line into an initial processing node set and a subsequent processing node set based on the original processing nodes;

[0008] Extracting workpiece appearance defect image data from the workpiece appearance detection image set, performing three-dimensional reconstruction on the workpiece appearance defect image data, and obtaining three-dimensional coordinate data of the workpiece appearance defect position on the workpiece;

[0009] Detecting first workpiece image data corresponding to an initial processing node set according to the three-dimensional coordinate data of the appearance defect position to obtain initial defect image data;

[0010] Detecting second workpiece image data corresponding to a subsequent processing node set according to the three-dimensional coordinate data of the appearance defect position, and acquiring defect change image data;

[0011] The occurrence node of the appearance defect position of the workpiece is marked based on the initial defect image data, the defect degree of the appearance defect position of the workpiece is marked based on the defect change image data, and the appearance defect tracking detection data of the workpiece is generated by combining the occurrence node and the defect degree.

[0012] Furthermore, the acquisition of historical data of appearance defects of workpieces processed on the production line and setting a number of observation points on the workpieces according to the historical data include:

[0013] Obtain historical processing data of the production line in the database, and extract historical data on the appearance defects of workpieces processed in the production line;

[0014] Based on the historical data, the occurrence probabilities of different defect positions on the appearance of the workpiece are statistically analyzed to detect whether the occurrence probability of each defect position is greater than 50%. If so, the defect position corresponding to the occurrence probability is marked as an observation point.

[0015] Further, the obtaining of processing nodes where workpiece appearance defects occur according to the workpiece appearance detection image set, marking the processing nodes as original processing nodes, and dividing the processing nodes of the production line into an initial processing node set and a subsequent processing node set based on the original processing nodes includes:

[0016] Extracting a workpiece appearance standard image corresponding to each processing node from a database, performing a difference comparison between the workpiece appearance detection image of each processing node and the workpiece appearance standard image, obtaining image comparison data, and marking workpiece appearance defects based on the image comparison data;

[0017] According to the arrangement order of several processing nodes, the processing node where the appearance defect position of the workpiece is detected for the first time is marked as the original node.

[0018] Further, the obtaining of processing nodes where workpiece appearance defects occur according to the workpiece appearance detection image set, marking the processing nodes as original processing nodes, and dividing the processing nodes of the production line into an initial processing node set and a subsequent processing node set based on the original processing nodes also includes:

[0019] According to the arrangement position of the original processing node in the processing steps of the production line, several processing nodes located before the processing step of the original processing node are organized into an initial processing node set, and several processing nodes located after the processing step of the original processing node are organized into a subsequent processing node set.

[0020] Further, extracting workpiece appearance defect image data from the workpiece appearance detection image set, performing three-dimensional reconstruction on the workpiece appearance defect image data, and obtaining three-dimensional coordinate data of the workpiece appearance defect position on the workpiece includes:

[0021] Extracting the workpiece appearance standard image data corresponding to the processing node from the database, and performing a difference comparison between the image data of the workpiece appearance detection image set and the workpiece appearance standard image data;

[0022] The defect position contour of the workpiece appearance inspection image is obtained based on the difference comparison result, thereby obtaining the workpiece appearance defect image data.

[0023] Furthermore, extracting workpiece appearance defect image data from the workpiece appearance detection image set, performing three-dimensional reconstruction on the workpiece appearance defect image data, and obtaining three-dimensional coordinate data of the workpiece appearance defect position on the workpiece also includes:

[0024] Processing the appearance defect image data based on projection matrix analysis to obtain the distance between each contour point of the workpiece and the shooting component, thereby constructing a three-dimensional model of the workpiece;

[0025] Based on the distance between the workpiece appearance defect position and each contour point of the workpiece, the relative position of the workpiece appearance defect position in the workpiece three-dimensional model is obtained.

[0026] Furthermore, constructing multiple groups of appearance defect image data of the same processing node into a matrix array, obtaining the distance between each contour point of the workpiece and the shooting component based on the array, and thus constructing a three-dimensional model of the workpiece includes:

[0027] The processing of appearance defect image data based on the projection matrix includes:

[0028] f=P*R T ;

[0029] S D-L =(A D-M *B M-L ) / f;

[0030] Where P is the projection matrix, R T is the decomposition matrix, D is the three-dimensional pixel space, M is the virtual pixel space, L is the image dimension space, f is the matrix decomposition image corresponding to the workpiece appearance defect image, A is the basic matrix, B is the coefficient matrix, S D-L Pixel bands for 3D reconstruction of matrix decomposed images.

[0031] Furthermore, constructing multiple groups of appearance defect image data of the same processing node into a matrix array, obtaining the distance between each contour point of the workpiece and the shooting component based on the array, and thus constructing a three-dimensional model of the workpiece includes:

[0032] The step of obtaining the relative position of the workpiece appearance defect position in the workpiece three-dimensional model based on the distance between the workpiece appearance defect position and each contour point of the workpiece includes:

[0033]

[0034] Where d is the distance between the workpiece defect position and the workpiece contour point, k is the number of calculations, Q is the rotation matrix, S D-L is the pixel band for 3D reconstruction of the matrix decomposition image, θ is the rotation angle, n1 is the variable coefficient of the polar equation of the workpiece appearance defect image, and n2 is the variable coefficient of the polar equation of the workpiece 3D model data.

[0035] Further, the marking of the occurrence node of the appearance defect position of the workpiece based on the initial defect image data, the marking of the defect degree of the appearance defect position of the workpiece based on the defect change image data, and the generation of the appearance defect tracking detection data of the workpiece in combination with the occurrence node and the defect degree include:

[0036] A transmission path of appearance processing defects of the workpiece during the production line processing is constructed according to the occurrence node, and tracking detection data of the workpiece appearance defects is generated in combination with the degree of the workpiece appearance defects.

[0037] The present invention also provides an automatic tracking system for the position of a workpiece appearance defect, the automatic tracking system is used to execute the automatic tracking method for the position of a workpiece appearance defect, and the automatic tracking system comprises:

[0038] Shooting component: It is set at each processing node of the production line to shoot the processing image of the workpiece at each processing node of the production line;

[0039] Database: used to record workpiece processing and visual inspection data;

[0040] Main control system: used to obtain the workpiece detection image data transmitted by the shooting component, and generate tracking detection data of workpiece appearance defects in combination with the data in the database.

[0041] The present invention provides a method and system for automatically tracking the position of appearance defects of workpieces. The method and system can obtain the position of appearance defects of workpieces through visual recognition during production line processing of the workpieces, and perform traceability detection and tracking detection based on the position of appearance defects of the workpieces to obtain the processing conditions of appearance processing defects on the workpieces during production line processing. The method can improve the monitoring reliability of workpiece processing production, perform real-time monitoring and adjustment of the production line according to the workpiece processing appearance defect conditions, improve the convenience of production line maintenance, and thus improve the quality of workpiece processing on the production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a flow chart of a method for automatically tracking the position of a workpiece appearance defect in an embodiment of the present invention;

[0043] Figure 2 is an operation flow chart of a method for automatically tracking the position of a workpiece appearance defect in an embodiment of the present invention;

[0044] Figure 3 Schematic diagram of an automatic tracking system for the position of appearance defects of a workpiece in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Embodiment 1:

[0047] Figure 1 A flow chart of an automatic tracking method for the position of an appearance defect of a workpiece according to an embodiment of the present invention is shown. The automatic tracking method comprises:

[0048] S11: Obtain historical data on the appearance defects of workpieces processed on the production line, and set a number of observation points on the workpieces according to the historical data.

[0049] The historical processing data of the production line is extracted from the database, and the historical data of the appearance defects of the workpiece processing are extracted from the historical processing data of the production line to determine the position distribution of defects in the workpieces processed on the production line, and a number of observation points are set on the workpiece according to the position distribution of the defects. Based on the several observation points, the main inspection positions of the visual inspection and tracking in the production line are determined, thereby improving the reliability and accuracy of the visual inspection system of the production line in observing the defect positions of the workpieces.

[0050] S12: Perform visual inspection of the workpiece at each processing node on the production line through the shooting component, adjust the shooting angle based on a number of observation points, and obtain a set of workpiece appearance inspection images for each processing node.

[0051] A shooting component is configured at each processing node position of the production line, and the shooting angle of each shooting component is adjusted based on a number of set observation points, so that any of the shooting components can cover more observation points on the corresponding processing node, thereby ensuring the reliability of the shooting component in detecting the position of appearance defects of the workpiece.

[0052] Furthermore, the workpiece is photographed at different angles based on the photographing component at each processing node to obtain multiple photographed images of the workpiece, and a set of appearance inspection images of the workpiece is constructed according to the processing sequence of the workpiece at different processing nodes of the production line.

[0053] Furthermore, the transmission and processing of the workpiece in the production line and the construction of the appearance inspection image set of the workpiece can be carried out simultaneously, so that the shooting component can perform real-time detection and tracking analysis of the location of appearance defects during the transmission and processing of the workpiece.

[0054] S13: acquiring processing nodes where workpiece appearance defects occur according to the workpiece appearance inspection image set, marking the processing nodes as original processing nodes, and dividing the processing nodes of the production line into an initial processing node set and a subsequent processing node set based on the original processing nodes.

[0055] Extracting a workpiece appearance standard image corresponding to each processing node from a database, performing a difference comparison between the workpiece appearance detection image of each processing node and the workpiece appearance standard image, obtaining image comparison data, and marking workpiece appearance defects based on the image comparison data;

[0056] According to the arrangement order of several processing nodes, the processing node where the appearance defect position of the workpiece is detected for the first time is marked as the original node.

[0057] According to the arrangement position of the original processing node in the processing steps of the production line, several processing nodes located before the processing step of the original processing node are organized into an initial processing node set, and several processing nodes located after the processing step of the original processing node are organized into a subsequent processing node set.

[0058] S14: extracting workpiece appearance defect image data from the workpiece appearance inspection image set, performing three-dimensional reconstruction on the workpiece appearance defect image data, and obtaining three-dimensional coordinate data of the workpiece appearance defect position on the workpiece.

[0059] Determine the processing node where the defect occurs, extract the image data of the workpiece in different positions of the workpiece by the camera component in the processing node, and the image data includes the defect position of the workpiece. Perform three-dimensional reconstruction on the image data by matrix decomposition to obtain the three-dimensional coordinates of the defect position on the workpiece.

[0060] Furthermore, the appearance defect image data is processed based on projection matrix analysis to obtain the distance between each contour point of the workpiece and the shooting component, thereby constructing a three-dimensional model of the workpiece;

[0061] Based on the distance between the workpiece appearance defect position and each contour point of the workpiece, the relative position of the workpiece appearance defect position in the workpiece three-dimensional model is obtained. A three-dimensional coordinate system is constructed on the workpiece three-dimensional model, and based on the relative position of the workpiece appearance defect position in the workpiece three-dimensional model, the three-dimensional coordinates of the defect position on the workpiece are determined.

[0062] S15: Detecting first workpiece image data corresponding to an initial processing node set according to the three-dimensional coordinate data of the appearance defect position to obtain initial defect image data.

[0063] Specifically, in the captured image data of the initial processing node set, the corresponding position of the workpiece in the captured image data is marked according to the three-dimensional coordinate data of the appearance defect position, and the marked captured image data is compared with the appearance defect image data to detect the appearance defect situation of the marked position corresponding to the captured image data of the initial processing node set, so as to obtain the first workpiece image data.

[0064] Furthermore, the captured image data with the marked position extracted from the initial processing node set is subjected to difference processing with the appearance defect image data to detect the contour change of the marked position. The contour change of the workpiece marked position is obtained based on the processing sequence of the workpiece in the initial processing node set, and the captured image data with contour change is marked as the initial defect image data.

[0065] S16: Detecting second workpiece image data corresponding to a subsequent processing node set according to the three-dimensional coordinate data of the appearance defect position, and acquiring defect change image data.

[0066] Specifically, in the captured image data of the subsequent processing node set, the corresponding position of the workpiece in the captured image data is marked according to the three-dimensional coordinate data of the appearance defect position, and the marked captured image data is compared with the appearance defect image data to detect the appearance defect situation of the marked position corresponding to the captured image data of the subsequent processing node set, so as to obtain the second workpiece image data.

[0067] Furthermore, the captured image data with the marked position extracted from the subsequent processing node set is subjected to difference processing with the appearance defect image data to detect the contour change of the appearance defect position. The contour change of the workpiece marked position is obtained based on the processing sequence of the workpiece in the subsequent processing node set, and the captured image data with contour change is marked as defect change image data.

[0068] S17: Marking the occurrence node of the appearance defect position of the workpiece based on the initial defect image data, marking the defect degree of the appearance defect position of the workpiece based on the defect change image data, and generating appearance defect tracking detection data of the workpiece in combination with the occurrence node and the defect degree.

[0069] The image data of the processing node where the defect first occurs in the workpiece during the processing is acquired according to the initial defect image data, and the processing node is marked as the occurrence node of the appearance defect position.

[0070] After the appearance defect is obtained in the defect change image data, the change state of the corresponding appearance defect position of the workpiece during the processing is obtained, and the change degree of the appearance defect position of the workpiece is obtained in the defect change image according to the processing sequence.

[0071] Furthermore, according to the appearance defects of the workpieces during the processing of the production line and the degree of the appearance defects of the workpieces, the appearance defect tracking detection data of the workpieces is generated, so that the workpiece personnel can accurately adjust the processing equipment of the production line according to the appearance defect tracking detection data of the workpieces, thereby improving the convenience of production line maintenance and thus improving the processing quality of the production line for the workpieces.

[0072] An embodiment of the present invention provides a method for automatically tracking the position of workpiece appearance defects. The method performs visual inspection of workpieces at various processing positions of a production line through a shooting component, and performs traceability analysis and subsequent processing tracking inspection based on the appearance defects of the workpieces. The method can analyze and obtain the appearance processing defects of the workpieces in real time, thereby accurately maintaining and adjusting relevant processing equipment of the production line, thereby improving the processing efficiency and processing quality of the production line for the workpieces.

[0073] Embodiment 2:

[0074] Figure 2 The following is an operation flow chart of a method for automatically tracking the position of a workpiece appearance defect according to an embodiment of the present invention. The method includes:

[0075] S101: Setting observation points.

[0076] A shooting component is set at each processing node of the production line, and visual inspection is performed on the workpiece based on the shooting component to obtain the appearance change state of the workpiece after processing at different processing nodes, so as to perform real-time detection and tracking of the processing state of the workpiece on the production line, so as to obtain the appearance defects of the workpiece.

[0077] The historical processing data of the production line is extracted from the database, and the historical data of the appearance defects of the workpiece processing are extracted from the historical processing data of the production line to determine the position distribution of defects in the workpieces processed on the production line, and a number of observation points are set on the workpiece according to the position distribution of the defects. Based on the several observation points, the main inspection positions of the visual inspection and tracking in the production line are determined, thereby improving the reliability and accuracy of the visual inspection system of the production line in observing the defect positions of the workpieces.

[0078] Specifically, historical processing data of the production line is obtained from the database, and historical data of appearance defects of workpieces processed on the production line are extracted;

[0079] Based on the historical data, the occurrence probabilities of different defect positions on the appearance of the workpiece are statistically analyzed to detect whether the occurrence probability of each defect position is greater than 50%. If so, the defect position corresponding to the occurrence probability is marked as an observation point.

[0080] S102: Adjust the shooting angle.

[0081] According to the distribution positions of several observation points of the workpiece, several shooting planes are divided on the workpiece, and a single shooting plane can cover more than one observation point, so as to improve the convenience of the shooting component in shooting and detecting the appearance defect position of the workpiece.

[0082] Multiple groups of shooting cameras can be set up in the same processing node to cover the shooting requirements of different angles and positions of the workpiece. Based on the multi-camera shooting method, the workpiece shooting at the processing node can meet the visual inspection requirements of the workpiece appearance defects.

[0083] Furthermore, the transmission and processing of the workpiece in the production line and the construction of the appearance inspection image set of the workpiece can be carried out simultaneously, so that the shooting component can perform real-time detection and tracking analysis of the location of appearance defects during the transmission and processing of the workpiece.

[0084] S103: Detect appearance defects.

[0085] Specifically, a standard image of the workpiece appearance corresponding to each processing node is extracted from the database, a difference comparison is performed between the workpiece appearance detection image of each processing node and the standard image of the workpiece appearance to obtain image comparison data, and workpiece appearance defects are marked based on the image comparison data.

[0086] Furthermore, standard image data of the workpiece at each processing node is extracted from the database, and the detection image data is compared with the standard image data to obtain the position of the appearance defect of the workpiece.

[0087] The detection image data of the workpiece and the standard image data of the workpiece are preprocessed, and the difference between the detection image data of the workpiece and the standard image data of the workpiece is calculated through filtering, noise reduction and gray value processing, so as to obtain the distribution of the position of the appearance defect of the workpiece.

[0088] S104: Mark the original processing node.

[0089] According to the arrangement order of several processing nodes, the processing node where the appearance defect position of the workpiece is detected for the first time is marked as the original processing node.

[0090] S105: construct a three-dimensional model of the workpiece.

[0091] Determine the processing node where the defect occurs, extract the image data of the workpiece in different positions of the workpiece by the camera component in the processing node, and the image data includes the defect position of the workpiece. Perform three-dimensional reconstruction on the image data by matrix decomposition to obtain the three-dimensional coordinates of the defect position on the workpiece.

[0092] Furthermore, the appearance defect image data is processed based on projection matrix analysis to obtain the distance between each contour point of the workpiece and the shooting component, thereby constructing a three-dimensional model of the workpiece;

[0093] Based on the distance between the workpiece appearance defect position and each contour point of the workpiece, the relative position of the workpiece appearance defect position in the workpiece three-dimensional model is obtained. A three-dimensional coordinate system is constructed on the workpiece three-dimensional model, and based on the relative position of the workpiece appearance defect position in the workpiece three-dimensional model, the three-dimensional coordinates of the defect position on the workpiece are determined.

[0094] Specifically, the processing of appearance defect image data based on the projection matrix includes:

[0095] f=P*R T ;

[0096] S D-L =(A D-M *B M-L ) / f;

[0097] Among them, P is the projection matrix, R T is the decomposition matrix, D is the three-dimensional pixel space, M is the virtual pixel space, L is the image dimension space, f is the matrix decomposition image corresponding to the workpiece appearance defect image, A is the basic matrix, B is the coefficient matrix, S D-L Pixel bands for 3D reconstruction of matrix decomposed images.

[0098] Furthermore, by decomposing the matrix R T The projection matrix P is decomposed and converted into the internal parameter matrix, rotation matrix and translation vector matrix of the shooting component. The pixel conversion band of the image data from two-dimensional to three-dimensional model is calculated based on the decomposed matrix data.

[0099] Furthermore, based on matrix decomposition, the image data of the appearance defects of the workpiece is processed into a matrix decomposition image, and the pixel conversion between the two-dimensional image data and the three-dimensional model data is realized by using the virtual pixel space as a transfer, thereby realizing the three-dimensional model reconstruction operation of the workpiece.

[0100] Specifically, obtaining the relative position of the workpiece appearance defect position in the workpiece three-dimensional model based on the distance between the workpiece appearance defect position and each contour point of the workpiece includes:

[0101]

[0102] Where d is the distance between the workpiece defect position and the workpiece contour point, k is the number of calculations, Q is the rotation matrix, S D-L is the pixel band for 3D reconstruction of the matrix decomposition image, θ is the rotation angle, n1 is the variable coefficient of the polar equation of the workpiece appearance defect image, and n2 is the variable coefficient of the polar equation of the workpiece 3D model data. The n1 and n2 are set according to the shooting angle and parameters of the shooting component.

[0103] S106: Divide the detection image set.

[0104] According to the arrangement position of the original processing node in the processing steps of the production line, several processing nodes located before the processing step of the original processing node are organized into an initial processing node set, and several processing nodes located after the processing step of the original processing node are organized into a subsequent processing node set.

[0105] S107: Obtain the appearance defect occurrence node.

[0106] Specifically, in the captured image data of the initial processing node set, the corresponding position of the workpiece in the captured image data is marked according to the three-dimensional coordinate data of the appearance defect position, and the marked captured image data is compared with the appearance defect image data to detect the appearance defect situation of the marked position corresponding to the captured image data of the initial processing node set, so as to obtain the first workpiece image data.

[0107] Furthermore, the captured image data with the marked position extracted from the initial processing node set is subjected to difference processing with the appearance defect image data to detect the contour change of the marked position. The contour change of the workpiece marked position is obtained based on the processing sequence of the workpiece in the initial processing node set, and the captured image data with contour change is marked as the initial defect image data.

[0108] The image data of the processing node where the defect first occurs in the workpiece during the processing is acquired according to the initial defect image data, and the processing node is marked as the occurrence node of the appearance defect position.

[0109] S108: Obtain the degree of appearance defects of the workpiece.

[0110] In the captured image data of the subsequent processing node set, the corresponding position of the workpiece in the captured image data is marked according to the three-dimensional coordinate data of the appearance defect position, and the marked captured image data is compared with the appearance defect image data to detect the appearance defect situation of the marked position corresponding to the captured image data of the subsequent processing node set, so as to obtain the second workpiece image data.

[0111] Furthermore, the captured image data with the marked position extracted from the subsequent processing node set is subjected to difference processing with the appearance defect image data to detect the contour change of the appearance defect position. The contour change of the workpiece marked position is obtained based on the processing sequence of the workpiece in the subsequent processing node set, and the captured image data with contour change is marked as defect change image data.

[0112] After the appearance defect is obtained in the defect change image data, the change state of the corresponding appearance defect position of the workpiece during the processing is obtained, and the degree of the appearance defect of the workpiece is obtained in the defect change image according to the processing sequence.

[0113] S109: Output tracking detection data.

[0114] Mark the nodes where appearance defects occur during workpiece processing, as well as the last processing nodes of the workpiece appearance defects that can be obtained based on image processing, build the transmission path of the workpiece appearance defects during production line processing, and generate tracking detection data for workpiece appearance defects based on the degree of workpiece appearance defects.

[0115] An embodiment of the present invention provides a method for automatically tracking the position of appearance defects of workpieces. During the production line processing of the workpieces, the position of appearance defects of the workpieces is obtained through visual recognition, and traceability detection and tracking detection are performed based on the position of appearance defects of the workpieces. The processing conditions of appearance processing defects of the workpieces during the production line processing are obtained, which can improve the monitoring reliability of workpiece processing production, and perform real-time monitoring and adjustment of the production line according to the workpiece processing appearance defect conditions, thereby improving the convenience of production line maintenance and thus improving the quality of workpiece processing on the production line.

[0116] Embodiment three:

[0117] Figure 3 The schematic diagram of the structure of the automatic tracking system for the appearance defects of workpieces in an embodiment of the present invention is shown, and the automatic tracking system comprises:

[0118] The photographing component 10 is arranged at each processing node of the production line, and is used to photograph the processing image of the workpiece at each processing node of the production line.

[0119] Database 20: used to record workpiece processing and visual inspection data;

[0120] The main control system 30 is used to obtain the workpiece detection image data transmitted by the shooting component 10, and generate tracking detection data of the workpiece appearance defects in combination with the data in the database 20.

[0121] Specifically, historical processing data of the production line is obtained in the database 20, and historical data on the appearance defects of workpiece processing of the production line is extracted. The occurrence probabilities of different defect positions on the appearance of the workpiece are statistically analyzed based on the historical data, and whether the occurrence probability of each defect position is greater than 50% is detected. If so, the defect position corresponding to the occurrence probability is marked as an observation point.

[0122] The main control system 30 adjusts the shooting angle of the shooting component according to the set observation point, so that the shooting component 10 can accurately shoot and detect the workpiece, thereby improving the accuracy of the detection of the appearance defect position of the workpiece during the production line processing.

[0123] Furthermore, the shooting component 10 is connected to the main control system 30 by signal. Based on the image detection data of the workpiece at different processing nodes in the production line by the shooting component 10, and the image detection data is sent to the main control system 30, the main control system 30 performs real-time analysis on the image detection data, thereby tracking and detecting the location of appearance defects of the workpiece.

[0124] The present invention provides an automatic tracking system for the position of appearance defects of workpieces. A shooting component 10 is set at each processing node position of the production line, and image capture and detection of the workpieces on the production line is performed based on the shooting component 10. The position of the appearance defects of the workpieces during the production line processing is tracked and detected based on the main control system 30, thereby improving the reliability of the production line's traceability detection of the position of the appearance defects of the workpieces, and improving the convenience and accuracy of the adjustment and maintenance of the production line processing equipment, thereby improving the processing efficiency of the production line and the processing quality of the workpieces on the production line.

[0125] A person skilled in the art may understand that all or part of the steps in the various methods of the above embodiments may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0126] In addition, the embodiments of the present invention are described in detail above. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for automatically tracking the position of a workpiece appearance defect, characterized in that: The automatic tracking method comprises: Obtain historical data on the appearance defects of workpieces processed on the production line, and set several observation points on the workpieces based on the historical data; The workpiece is visually inspected at each processing node on the production line by using the shooting component, and the shooting angle is adjusted based on several observation points to obtain a set of workpiece appearance inspection images at each processing node; Acquire processing nodes where workpiece appearance defects occur according to the workpiece appearance inspection image set, mark the processing nodes as original processing nodes, and divide each processing node of the production line into an initial processing node set and a subsequent processing node set based on the original processing nodes; Extracting workpiece appearance defect image data from the workpiece appearance detection image set, performing three-dimensional reconstruction on the workpiece appearance defect image data, and obtaining three-dimensional coordinate data of the workpiece appearance defect position on the workpiece; Detecting first workpiece image data corresponding to an initial processing node set according to the three-dimensional coordinate data of the appearance defect position to obtain initial defect image data; Detecting second workpiece image data corresponding to a subsequent processing node set according to the three-dimensional coordinate data of the appearance defect position, and acquiring defect change image data; The occurrence node of the appearance defect position of the workpiece is marked based on the initial defect image data, the defect degree of the appearance defect position of the workpiece is marked based on the defect change image data, and the appearance defect tracking detection data of the workpiece is generated by combining the occurrence node and the defect degree.

2. The method for automatically tracking the position of a workpiece appearance defect according to claim 1, characterized in that: The obtaining of historical data of appearance defects of workpieces processed on the production line and setting a number of observation points on the workpieces according to the historical data include: Obtain historical processing data of the production line in the database, and extract historical data on the appearance defects of workpieces processed in the production line; Based on the historical data, the occurrence probabilities of different defect positions on the appearance of the workpiece are statistically analyzed to detect whether the occurrence probability of each defect position is greater than 50%. If so, the defect position corresponding to the occurrence probability is marked as an observation point.

3. The method for automatically tracking the position of a workpiece appearance defect according to claim 1, characterized in that: The step of acquiring a processing node where a workpiece appearance defect occurs according to the workpiece appearance detection image set, marking the processing node as an original processing node, and dividing each processing node of the production line into an initial processing node set and a subsequent processing node set based on the original processing node comprises: Extracting a workpiece appearance standard image corresponding to each processing node from a database, performing a difference comparison between the workpiece appearance detection image of each processing node and the workpiece appearance standard image, obtaining image comparison data, and marking workpiece appearance defects based on the image comparison data; According to the arrangement order of several processing nodes, the processing node where the appearance defect position of the workpiece is detected for the first time is marked as the original node.

4. The method for automatically tracking the position of a workpiece appearance defect according to claim 3, characterized in that: The step of acquiring a processing node where a workpiece appearance defect occurs according to the workpiece appearance detection image set, marking the processing node as an original processing node, and dividing each processing node of the production line into an initial processing node set and a subsequent processing node set based on the original processing node also includes: According to the arrangement position of the original processing node in the processing steps of the production line, several processing nodes located before the processing step of the original processing node are organized into an initial processing node set, and several processing nodes located after the processing step of the original processing node are organized into a subsequent processing node set.

5. The method for automatically tracking the position of a workpiece appearance defect according to claim 1, characterized in that: The extracting workpiece appearance defect image data from the workpiece appearance detection image set, performing three-dimensional reconstruction on the workpiece appearance defect image data, and obtaining three-dimensional coordinate data of the workpiece appearance defect position on the workpiece includes: Extracting the workpiece appearance standard image data corresponding to the processing node from the database, and performing a difference comparison between the image data of the workpiece appearance detection image set and the workpiece appearance standard image data; The defect position contour of the workpiece appearance inspection image is obtained based on the difference comparison result, thereby obtaining the workpiece appearance defect image data.

6. The method for automatically tracking the position of a workpiece appearance defect according to claim 5, characterized in that: The extracting workpiece appearance defect image data from the workpiece appearance detection image set, performing three-dimensional reconstruction on the workpiece appearance defect image data, and obtaining three-dimensional coordinate data of the workpiece appearance defect position on the workpiece also includes: Processing the appearance defect image data based on projection matrix analysis to obtain the distance between each contour point of the workpiece and the shooting component, thereby constructing a three-dimensional model of the workpiece; Based on the distance between the workpiece appearance defect position and each contour point of the workpiece, the relative position of the workpiece appearance defect position in the workpiece three-dimensional model is obtained.

7. The method for automatically tracking the position of a workpiece appearance defect according to claim 6, characterized in that: The method of constructing a plurality of groups of appearance defect image data of the same processing node into a matrix array, obtaining the distance between each contour point of the workpiece and the shooting component based on the array, and thus constructing a three-dimensional model of the workpiece includes: The processing of appearance defect image data based on the projection matrix includes: f=P*R T ; S D-L =(A D-M *B M-L ) / f; Among them, P is the projection matrix, R T is the decomposition matrix, D is the three-dimensional pixel space, M is the virtual pixel space, L is the image dimension space, f is the matrix decomposition image corresponding to the workpiece appearance defect image, A is the basic matrix, B is the coefficient matrix, S D-L Pixel bands for 3D reconstruction of matrix decomposed images.

8. The method for automatically tracking the position of a workpiece appearance defect according to claim 6, characterized in that: The method of constructing a plurality of groups of appearance defect image data of the same processing node into a matrix array, obtaining the distance between each contour point of the workpiece and the shooting component based on the array, and thus constructing a three-dimensional model of the workpiece includes: The step of obtaining the relative position of the workpiece appearance defect position in the workpiece three-dimensional model based on the distance between the workpiece appearance defect position and each contour point of the workpiece includes: Where d is the distance between the workpiece defect position and the workpiece contour point, k is the number of calculations, Q is the rotation matrix, S D-L is the pixel band for 3D reconstruction of the matrix decomposition image, θ is the rotation angle, n1 is the variable coefficient of the polar equation of the workpiece appearance defect image, and n2 is the variable coefficient of the polar equation of the workpiece 3D model data.

9. The method for automatically tracking the position of a workpiece appearance defect according to claim 1, characterized in that: The step of marking the occurrence node of the appearance defect position of the workpiece based on the initial defect image data, marking the defect degree of the appearance defect position of the workpiece based on the defect change image data, and combining the occurrence node and the defect degree to generate the appearance defect tracking detection data of the workpiece includes: A transmission path of appearance processing defects of the workpiece during the production line processing is constructed according to the occurrence node, and tracking detection data of the workpiece appearance defects is generated in combination with the degree of the workpiece appearance defects.

10. An automatic tracking system for the position of a workpiece appearance defect, characterized in that: The automatic tracking system is used to execute the automatic tracking method for the position of the appearance defect of a workpiece according to any one of claims 1 to 9, and the automatic tracking system comprises: Shooting component: It is set at each processing node of the production line to shoot the processing image of the workpiece at each processing node of the production line; Database: used to record workpiece processing and visual inspection data; Main control system: used to obtain the workpiece detection image data transmitted by the shooting component, and generate tracking detection data of workpiece appearance defects in combination with the data in the database.

Images