Defect detection scheme design method and system based on micro-pore channel environment analysis

Abstract

The present invention discloses a defect detection scheme design method and system based on micro-pore channel environmental analysis, comprising the following steps: first, the computer tomography data of the target workpiece is obtained, a three-dimensional model is constructed, and the distribution of the micro-pore channels is determined to generate a distribution map. Then, the channel distribution map is analyzed based on the depth-first search algorithm to determine the defect detection path. Next, standard defect image data and environmental-defect image data under different environmental characteristics are obtained, the influence of image geometric deviation on detection accuracy is analyzed, and impact data is obtained. Finally, defect detection strategies under different environments are formulated according to the impact data, and a complete defect detection scheme is constructed in combination with the detection path. The present invention improves the accuracy and efficiency of micro-pore channel defect detection.

Description

Technical Field

[0001] The present invention relates to the field of defect detection technology, and in particular to a defect detection scheme design method and system based on micropore channel environment analysis. Background Art

[0002] Microporous channels are widely used structures in mechanical and electronic equipment. They usually have complex geometric shapes and tiny sizes. In practical applications, they require high precision and reliability. However, microporous channels are prone to various defects such as holes, cracks, blockages, burrs, etc. during manufacturing and use. These defects can significantly affect the normal operation of the equipment and even cause serious safety problems. Therefore, effective defect detection of microporous channels is of great practical significance.

[0003] Traditional defect detection methods mainly rely on manual inspection or automated inspection technology based on image processing. Although manual inspection methods can handle some defects, they are limited by the resolution ability of the human eye and it is difficult to detect small and hidden defects. In addition, manual inspection is inefficient and cannot meet the needs of large-scale production. Automated image processing technology uses machine vision systems to scan micro-pore channels and detect defects, but existing methods often lack a full understanding of the complex geometric structure of micro-pore channels, are easily interfered by noise, and have poor detection accuracy and robustness.

[0004] With the rapid development of industrial computed tomography (CT) technology, CT technology can obtain high-precision three-dimensional data of workpieces, providing new possibilities for defect detection of micro-pore channels. However, micro-pore channels often present complex branching structures, and there may be multiple options for detection paths within the channel. When facing such complex channel structures, existing methods may encounter problems such as unreasonable detection paths, redundant paths, or incomplete coverage, resulting in missed detection or low detection efficiency. In addition, traditional methods usually ignore the optimization problem in the detection path.

[0005] In order to solve the above problems, this paper proposes a defect detection scheme design method and system based on micro-pore channel environment analysis. By combining the depth-first search algorithm, the full convolutional network and the Canny edge detection algorithm, defect detection path planning and detection strategy optimization are performed according to different environmental characteristics, which can effectively improve the detection accuracy and efficiency. The present invention is not only suitable for various types of micro-pore channels, but also can work stably in complex environments, providing reliable technical support for defect detection of micro-pore channels. Summary of the invention

[0006] In order to solve at least one of the above technical problems, the present invention proposes a defect detection scheme design method and system based on micropore channel environment analysis.

[0007] The first aspect of the present invention provides a defect detection scheme design method based on micropore channel environment analysis, comprising:

[0008] Acquire computer tomography data of a target workpiece, construct a three-dimensional model of the target workpiece according to the computer tomography data, determine the distribution of micropore channels of the target workpiece according to the three-dimensional model of the target workpiece, and construct a micropore channel distribution map;

[0009] Analyzing the micro-hole channel distribution map based on a depth-first search algorithm to determine a defect detection path for the micro-hole channel of the target workpiece;

[0010] Obtaining standard defect image data of a micropore channel and environment-defect image data of a micropore channel defect under each channel environment feature, and determining a defect image geometric deviation under each environment feature according to the standard defect image data and the environment-defect image data;

[0011] Determine the influence of various environmental characteristics on the detection accuracy of micro-hole channel defects according to the defect image geometric deviation, and obtain influence data;

[0012] Determining a defect detection strategy for micropore channels under different environmental characteristics according to the influencing data;

[0013] A defect detection scheme for the micro-pore channel of the target workpiece is constructed according to the defect detection path and defect detection strategy.

[0014] In this solution, the computer tomography data of the target workpiece is obtained, a three-dimensional model of the target workpiece is constructed according to the computer tomography data, the distribution of micropore channels of the target workpiece is determined according to the three-dimensional model of the target workpiece, and a micropore channel distribution map is constructed, specifically:

[0015] Scanning the target workpiece based on industrial CT to obtain computer tomography data of the target workpiece, and constructing a three-dimensional model of the target workpiece according to the computer tomography data;

[0016] Obtaining density values ​​at various positions in the three-dimensional model, and constructing a density histogram of the three-dimensional model according to the density values ​​at various positions;

[0017] Obtaining density values ​​of preset points of micropore channels in the three-dimensional model, and determining a micropore channel density threshold for model segmentation according to the density values ​​of the preset points;

[0018] Segmenting the three-dimensional model according to a micropore channel density threshold and a density histogram of the three-dimensional model, and extracting micropore channel edge position information in the three-dimensional model;

[0019] According to the edge position information of the micropore channel, the micropore channel part in the three-dimensional model is converted into a line network based on a distance transformation method, and the micropore channel is re-voxelized according to the line network to obtain a three-dimensional model of the micropore channel;

[0020] The three-dimensional model of the micropore channel is used to determine the micropore channel distribution of the target workpiece and construct a micropore channel distribution map.

[0021] In this solution, the micro-pore channel distribution map is analyzed based on the depth-first search algorithm to determine the defect detection path of the micro-pore channel of the target workpiece, specifically:

[0022] Mapping the micropore channel distribution map onto a two-dimensional plane to obtain a micropore channel plane distribution map;

[0023] The channel intersection points and channel end points in the micropore channel plane distribution map are regarded as nodes, the channels between the nodes are regarded as edges, and the connection relationship between each node and the edge is determined according to the micropore channel plane distribution map;

[0024] Determine the length of the edge between each node according to the micropore channel distribution graph, and construct an undirected weighted graph of the micropore channels according to the nodes, edges, edge lengths, and connection relationships between nodes and edges;

[0025] Obtaining a starting node for micropore channel defect detection, determining whether the undirected weighted graph has a ring structure, and if so, disconnecting the ring node that is the farthest from the starting node in the ring structure and has the longest edge connected to the ring node at the ring node to form an undirected acyclic weighted graph;

[0026] Taking the starting node as the root node, constructing a weighted dendrogram according to the undirected acyclic weighted graph, finding the longest path of the dendrogram based on the depth-first search algorithm, marking the edges in the longest path as visited, removing the visited edges in the undirected acyclic weighted graph, updating the undirected acyclic weighted graph, and determining the end node of the longest path;

[0027] The node closest to the end node in the updated undirected acyclic weighted graph is used as the root node of the dendrogram to construct the dendrogram according to the longest path, and the end node and the undirected acyclic weighted graph are updated. The dendrogram is constructed repeatedly to find the longest path and update the end node and the undirected acyclic weighted graph until all edges in the undirected acyclic weighted graph are visited;

[0028] The longest path obtained each time is connected according to the acquisition time to obtain the defect detection path of the micro-hole channel of the target workpiece.

[0029] In this solution, the standard defect image data of the micropore channel and the environment-defect image data of the micropore channel defect under each channel environment feature are obtained, and the defect image geometric deviation under each environment feature is determined according to the standard defect image data and the environment-defect image data, specifically:

[0030] Obtaining target workpiece production samples with different defect types in micro-pore channels, and obtaining standard defect image data of different defect types in micro-pore channels according to the production samples, wherein the different defect types include holes, cracks, blockages, and burrs;

[0031] Performing experimental operations on the production sample, constructing different environmental features in the microporous channel of the production sample, and obtaining environmental-defect image data of different defect types of the microporous channel under different environmental features, wherein the environmental features include the degree of moisture, dust accumulation, and chemical corrosion in the microporous channel;

[0032] Extracting defect edge features of the standard defect image data and the environment-feature defect image data based on the Canny edge detection algorithm, determining defect geometric features according to the defect edge features, and obtaining standard defect geometric features and environment-defect geometric features;

[0033] The environment-defect geometric features are compared with standard defect geometric features to determine the defect image geometric deviation under each environment feature.

[0034] In this solution, the influence of the defect image geometric deviation on the micro-hole channel defect detection accuracy under various environmental characteristics is determined to obtain the influence data, which is specifically:

[0035] Building a defect detection model based on a fully convolutional network, and importing the standard defect geometric features into the defect detection model for training;

[0036] Importing the environment-defect geometric features into the defect detection model to perform defect detection and obtain defect detection results;

[0037] Determining the defect detection accuracy of the microporous channel under different environmental characteristics according to the defect detection results;

[0038] The defect image geometric deviation between the environment-defect geometric feature and the standard defect geometric feature is jointly analyzed with the defect detection accuracy under the different environmental features to determine the impact of each environmental feature on the micro-hole channel defect detection accuracy and obtain impact data.

[0039] In this solution, the defect detection strategy of the micropore channel under different environmental characteristics is determined according to the influencing data, specifically:

[0040] According to the influence data, the environmental characteristics of the micro-pore channel defect detection accuracy less than the preset influence value are marked to obtain a type of environmental characteristics, and when the micro-pore channel environment is a type of environmental characteristics, the micro-pore channel is defect-detected according to the defect detection model;

[0041] A second-class defect detection model is constructed based on a fully convolutional network, and the environment-defect geometric features formed by the non-first-class environmental features are introduced into the second-class defect detection model for training;

[0042] Extracting defect geometric features under each non-class I environmental feature and importing them into the class II defect detection model for defect detection, marking the non-class I environmental features corresponding to the defect geometric features with a defect detection accuracy higher than a preset value as class II environmental features, and marking the remaining environmental features as class III environmental features, and when the micro-pore channel environment is a class II environmental feature, performing defect detection on the micro-pore channel according to the class II defect detection model;

[0043] The environment-defect geometric features of three types of environmental characteristics are obtained, and the environment-defect geometric features of the three types of environmental characteristics are compared with the standard defect geometric features to determine the correction parameters of the environment-defect geometric features of the three types of environmental characteristics. When the micropore channel environment is the three types of environmental characteristics, the defect geometric features obtained under the three types of environmental characteristics are corrected according to the correction parameters, and the corrected defect geometric features are imported into the defect detection model for defect detection, so as to obtain the defect detection strategy of the micropore channel under different environmental characteristics.

[0044] In this solution, the defect detection solution for the micro-hole channel of the target workpiece is constructed according to the defect detection path and defect detection strategy, specifically:

[0045] Constructing a defect detection scheme for a micro-pore channel of a target workpiece according to the defect detection path and defect detection strategy;

[0046] Defect detection is performed on the micro-hole channel of the target workpiece according to the defect detection path, and the environmental characteristics of the micro-hole channel are determined, and defect detection is performed on defect geometric characteristics of different environmental characteristics according to the defect detection strategy.

[0047] The second aspect of the present invention further provides a defect detection scheme design system based on micropore channel environment analysis, the system comprising: a memory, a processor, the memory comprising a defect detection scheme design method program based on micropore channel environment analysis, the defect detection scheme design method program based on micropore channel environment analysis when executed by the processor, implements the following steps:

[0048] Acquire computer tomography data of a target workpiece, construct a three-dimensional model of the target workpiece according to the computer tomography data, determine the distribution of micropore channels of the target workpiece according to the three-dimensional model of the target workpiece, and construct a micropore channel distribution map;

[0049] Analyzing the micro-hole channel distribution map based on a depth-first search algorithm to determine a defect detection path for the micro-hole channel of the target workpiece;

[0050] Obtaining standard defect image data of a micropore channel and environment-defect image data of a micropore channel defect under each channel environment feature, and determining a defect image geometric deviation under each environment feature according to the standard defect image data and the environment-defect image data;

[0051] Determine the influence of various environmental characteristics on the detection accuracy of micro-hole channel defects according to the defect image geometric deviation, and obtain influence data;

[0052] Determining a defect detection strategy for micropore channels under different environmental characteristics according to the influencing data;

[0053] A defect detection scheme for the micro-pore channel of the target workpiece is constructed according to the defect detection path and defect detection strategy.

[0054] The present invention discloses a defect detection scheme design method and system based on micro-pore channel environmental analysis, comprising the following steps: first, obtaining computer tomography data of the target workpiece, constructing a three-dimensional model, and determining the distribution of micro-pore channels to generate a distribution map. Then, the channel distribution map is analyzed based on a depth-first search algorithm to determine the defect detection path. Next, standard defect image data and environmental-defect image data under different environmental characteristics are obtained, and the influence of image geometric deviation on detection accuracy is analyzed to obtain impact data. Finally, defect detection strategies under different environments are formulated based on the impact data, and a complete defect detection scheme is constructed in combination with the detection path. The present invention improves the accuracy and efficiency of micro-pore channel defect detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A flow chart of a defect detection scheme design method based on micropore channel environment analysis of the present invention is shown;

[0056] Figure 2 A flow chart of obtaining influence data according to the present invention is shown;

[0057] Figure 3 A flow chart showing a defect detection scheme for constructing a micro-hole channel of a target workpiece according to the present invention is shown;

[0058] Figure 4 A block diagram of a defect detection scheme design system based on micropore channel environment analysis of the present invention is shown. DETAILED DESCRIPTION

[0059] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0060] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0061] Figure 1 A flow chart of a defect detection scheme design method based on micropore channel environment analysis of the present invention is shown.

[0062] like Figure 1 As shown, the first aspect of the present invention provides a defect detection scheme design method based on micropore channel environment analysis, comprising:

[0063] S102, acquiring computer tomography data of a target workpiece, constructing a three-dimensional model of the target workpiece according to the computer tomography data, determining a micropore channel distribution of the target workpiece according to the three-dimensional model of the target workpiece, and constructing a micropore channel distribution map;

[0064] S104, analyzing the micro-hole channel distribution map based on a depth-first search algorithm to determine a defect detection path of the micro-hole channel of the target workpiece;

[0065] S106, acquiring standard defect image data of a micropore channel and environment-defect image data of a micropore channel defect under each channel environment feature, and determining a defect image geometric deviation under each environment feature according to the standard defect image data and the environment-defect image data;

[0066] S108, determining the influence of various environmental features on the detection accuracy of micro-hole channel defects according to the defect image geometric deviation, and obtaining influence data;

[0067] S110, determining a defect detection strategy for a micropore channel under different environmental characteristics according to the influencing data;

[0068] S112, constructing a defect detection scheme for the micro-hole channel of the target workpiece according to the defect detection path and the defect detection strategy.

[0069] It should be noted that by determining the micropore channel distribution map and determining the defect detection path of the micropore channel based on the depth-first search algorithm, the depth-first search algorithm can traverse all nodes and edges in the micropore channel distribution map to ensure that the detection path can cover the entire micropore channel structure and avoid missed detection. This is especially important for complex and branched micropore channels, which can ensure that all channels are fully detected; the defect detection strategy of the micropore channel is determined by determining the influence of different environmental characteristics on the defect detection accuracy of the micropore channel. Different environmental characteristics may have different effects on the geometric characteristics of the defect image. By analyzing these impact data, the detection strategy can be automatically adjusted to adapt to environmental changes. Setting specific detection strategies for different environmental characteristics helps to reduce false detection or missed detection problems caused by environmental changes, thereby improving the overall detection accuracy; the micropore channel of the target workpiece is an intricately connected channel, in which precision components can be installed or liquid can be added for heat dissipation. In order for the precision components to be accurately placed in the micropore channel, the micropore channel needs to be processed and inspected for quality to avoid processing defects that cause deviations in the installation position of the precision components or insufficient rigidity of the installation connection position, resulting in equipment performance failure.

[0070] According to an embodiment of the present invention, the step of acquiring computer tomography data of a target workpiece, constructing a three-dimensional model of the target workpiece according to the computer tomography data, determining the distribution of micropore channels of the target workpiece according to the three-dimensional model of the target workpiece, and constructing a micropore channel distribution map is specifically as follows:

[0071] Scanning the target workpiece based on industrial CT to obtain computer tomography data of the target workpiece, and constructing a three-dimensional model of the target workpiece according to the computer tomography data;

[0072] Obtaining density values ​​at various positions in the three-dimensional model, and constructing a density histogram of the three-dimensional model according to the density values ​​at various positions;

[0073] Obtaining density values ​​of preset points of micropore channels in the three-dimensional model, and determining a micropore channel density threshold for model segmentation according to the density values ​​of the preset points;

[0074] Segmenting the three-dimensional model according to a micropore channel density threshold and a density histogram of the three-dimensional model, and extracting micropore channel edge position information in the three-dimensional model;

[0075] According to the edge position information of the micropore channel, the micropore channel part in the three-dimensional model is converted into a line network based on a distance transformation method, and the micropore channel is re-voxelized according to the line network to obtain a three-dimensional model of the micropore channel;

[0076] The three-dimensional model of the micropore channel is used to determine the micropore channel distribution of the target workpiece and construct a micropore channel distribution map.

[0077] According to an embodiment of the present invention, the micro-hole channel distribution map is analyzed based on a depth-first search algorithm to determine a defect detection path for the micro-hole channel of the target workpiece, specifically:

[0078] Mapping the micropore channel distribution map onto a two-dimensional plane to obtain a micropore channel plane distribution map;

[0079] The channel intersection points and channel end points in the micropore channel plane distribution map are regarded as nodes, the channels between the nodes are regarded as edges, and the connection relationship between each node and the edge is determined according to the micropore channel plane distribution map;

[0080] Determine the length of the edge between each node according to the micropore channel distribution graph, and construct an undirected weighted graph of the micropore channels according to the nodes, edges, edge lengths, and connection relationships between nodes and edges;

[0081] Obtaining a starting node for micropore channel defect detection, determining whether the undirected weighted graph has a ring structure, and if so, disconnecting the ring node that is the farthest from the starting node in the ring structure and has the longest edge connected to the ring node at the ring node to form an undirected acyclic weighted graph;

[0082] Taking the starting node as the root node, constructing a weighted dendrogram according to the undirected acyclic weighted graph, finding the longest path of the dendrogram based on the depth-first search algorithm, marking the edges in the longest path as visited, removing the visited edges in the undirected acyclic weighted graph, updating the undirected acyclic weighted graph, and determining the end node of the longest path;

[0083] The node closest to the end node in the updated undirected acyclic weighted graph is used as the root node of the dendrogram to construct the dendrogram according to the longest path, and the end node and the undirected acyclic weighted graph are updated. The dendrogram is constructed repeatedly to find the longest path and update the end node and the undirected acyclic weighted graph until all edges in the undirected acyclic weighted graph are visited;

[0084] The longest path obtained each time is connected according to the acquisition time to obtain the defect detection path of the micro-hole channel of the target workpiece.

[0085] It should be noted that by constructing an undirected weighted graph, the connection relationship between channel intersections and the weights (such as length) between channels can be effectively represented. Each node represents the intersection or end point of a channel, and the edge represents the channel between these nodes. By modeling these connection relationships, path planning and optimization can be performed more easily; the ring structure will lead to repeated traversal of the path. In defect detection, repeatedly passing through the same channel or path in the ring structure not only wastes detection time, but also may increase the wear of the detection equipment. By disconnecting the longest edge in the ring structure, the path can be prevented from returning to the visited node, thereby reducing repeated detection paths. The depth-first search algorithm may fall into an infinite loop in a graph containing a ring structure and cannot effectively complete the path search. By disconnecting the ring structure to form a graph without a loop, the path planning problem can be simplified and the efficiency of the algorithm can be improved; the longest path found by the depth-first search algorithm can maximize the coverage of important channel areas during defect detection. The selection of the longest path helps to optimize the allocation of detection resources, making the detection process more efficient and comprehensive. By repeatedly using the depth-first search algorithm to find the longest path of the dendrogram constructed each time, the repeated detection paths in the micropore detection process can be greatly reduced, the detection efficiency can be improved, and the redundant detection content can be reduced; after finding a longest path and marking it as visited, the undirected acyclic weighted graph is updated, and the dendrogram is rebuilt with the new end node as the starting point. This phased path planning method can gradually cover all nodes and edges. Each updated graph ensures that the remaining unvisited nodes and edges can be included in the subsequent path planning; there can be multiple channel end points.

[0086] According to an embodiment of the present invention, the acquisition of standard defect image data of a micropore channel and environment-defect image data of a micropore channel defect under each channel environment feature, and the determination of the defect image geometric deviation under each environment feature according to the standard defect image data and the environment-defect image data are specifically as follows:

[0087] Obtaining target workpiece production samples with different defect types in micro-pore channels, and obtaining standard defect image data of different defect types in micro-pore channels according to the production samples, wherein the different defect types include holes, cracks, blockages, and burrs;

[0088] Performing experimental operations on the production sample, constructing different environmental features in the microporous channel of the production sample, and obtaining environmental-defect image data of different defect types of the microporous channel under different environmental features, wherein the environmental features include the degree of moisture, dust accumulation, and chemical corrosion in the microporous channel;

[0089] Extracting defect edge features of the standard defect image data and the environment-feature defect image data based on the Canny edge detection algorithm, determining defect geometric features according to the defect edge features, and obtaining standard defect geometric features and environment-defect geometric features;

[0090] The environment-defect geometric features are compared with standard defect geometric features to determine the defect image geometric deviation under each environment feature.

[0091] It should be noted that a comprehensive defect library can be established by acquiring standard defect image data with different defect types (such as holes, cracks, blockages, and burrs). These standard images provide a reliable benchmark for identifying and classifying different types of defects; by extracting the geometric features of the defects through the Canny edge detection algorithm and comparing the environment-defect image data with the standard defect image data, the geometric deviations of the defects under different environmental conditions can be identified and quantified; understanding the geometric deviations of defect images can help formulate optimized detection strategies for different environmental characteristics. By analyzing the geometric deviations under different environments, the parameters of the detection algorithm can be adjusted to improve the detection accuracy and sensitivity under various environmental conditions; the defect geometric features include texture features, area features, shape features, texture direction features, and boundary features; there are multiple environment-defect geometric features, including environment-defect geometric features formed under different environmental conditions.

[0092] Figure 2 A flow chart of obtaining influence data according to the present invention is shown.

[0093] According to an embodiment of the present invention, the influence of the defect image geometric deviation on the micro-hole channel defect detection accuracy under various environmental characteristics is determined to obtain the influence data, specifically:

[0094] S202, constructing a defect detection model based on a fully convolutional network, and importing the standard defect geometric features into the defect detection model for training;

[0095] S204, importing the environment-defect geometric features into the defect detection model to perform defect detection and obtain defect detection results;

[0096] S206, determining the defect detection accuracy of the microporous channel under different environmental characteristics according to the defect detection result;

[0097] S208, jointly analyzing the defect image geometric deviation between the environment-defect geometric feature and the standard defect geometric feature and the defect detection accuracy under the different environmental features, determining the impact of each environmental feature on the micro-hole channel defect detection accuracy, and obtaining impact data.

[0098] It should be noted that by obtaining the impact data, it is possible to identify which environmental characteristics or under what parameter conditions of the environmental characteristics have a negative impact on the detection accuracy; understanding the impact data under different environmental conditions can help system designers effectively suppress environmental interference during the detection process, thereby improving the robustness and stability of the system.

[0099] According to an embodiment of the present invention, the defect detection strategy of the micro-pore channel under different environmental characteristics is determined according to the influencing data, specifically:

[0100] According to the influence data, the environmental characteristics of the micro-pore channel defect detection accuracy less than the preset influence value are marked to obtain a type of environmental characteristics, and when the micro-pore channel environment is a type of environmental characteristics, the micro-pore channel is defect-detected according to the defect detection model;

[0101] It should be noted that by calibrating the environmental characteristics for micro-pore channel defect detection accuracy less than the preset impact value as a type of environmental characteristics, the defect geometric characteristics extracted under this type of environmental characteristics can be considered to be unaffected by the environmental characteristics, and the defect detection model can perform good defect identification on the defect geometric characteristics extracted under a type of environmental characteristics.

[0102] A second-class defect detection model is constructed based on a fully convolutional network, and the environment-defect geometric features formed by the non-first-class environmental features are introduced into the second-class defect detection model for training;

[0103] Extracting defect geometric features under each non-class I environmental feature and importing them into the class II defect detection model for defect detection, marking the non-class I environmental features corresponding to the defect geometric features with a defect detection accuracy higher than a preset value as class II environmental features, and marking the remaining environmental features as class III environmental features, and when the micro-pore channel environment is a class II environmental feature, performing defect detection on the micro-pore channel according to the class II defect detection model;

[0104] It should be noted that, the defect geometric features extracted under various non-class I environmental features are used to construct a class II defect detection model to perform defect detection on the defect geometric features extracted under non-class I environmental features. The class II defect detection model learns the defect geometric features formed under non-class I environmental features, and therefore can detect the defect geometric features extracted under various non-class I environmental features. The non-class I environmental features corresponding to the defect geometric features whose defect detection accuracy is higher than the preset value are marked as class II environmental features. When the micropore channel environment is a class II environmental feature, defect detection is performed on the micropore channel according to the class II defect detection model.

[0105] The environment-defect geometric features of three types of environmental characteristics are obtained, and the environment-defect geometric features of the three types of environmental characteristics are compared with the standard defect geometric features to determine the correction parameters of the environment-defect geometric features of the three types of environmental characteristics. When the micropore channel environment is the three types of environmental characteristics, the defect geometric features obtained under the three types of environmental characteristics are corrected according to the correction parameters, and the corrected defect geometric features are imported into the defect detection model for defect detection, so as to obtain the defect detection strategy of the micropore channel under different environmental characteristics.

[0106] It should be noted that since some of the non-class I environmental features are too harsh, the extracted defect features are too fuzzy, and the class II defect detection model cannot learn this type of defect features, and calibrates this type of environmental features as class III environmental features; therefore, at this time, by making special corrections to the defect geometric features obtained from the class III environmental features, the geometric feature deviations caused by harsh environmental conditions can be effectively corrected; there are multiple environment-defect geometric features, including environment-defect geometric features formed under different environmental conditions.

[0107] Figure 3 A flow chart of a defect detection scheme for constructing a micro-hole channel of a target workpiece according to the present invention is shown.

[0108] According to an embodiment of the present invention, the defect detection scheme for the micro-hole channel of the target workpiece is constructed according to the defect detection path and the defect detection strategy, specifically:

[0109] S302, constructing a defect detection scheme for a micro-hole channel of a target workpiece according to the defect detection path and defect detection strategy;

[0110] S304, performing defect detection on the micro-hole channel of the target workpiece according to the defect detection path, determining the environmental characteristics of the micro-hole channel, and performing defect detection on defect geometric characteristics of different environmental characteristics according to the defect detection strategy.

[0111] Figure 4 A block diagram of a defect detection scheme design system based on micropore channel environment analysis of the present invention is shown.

[0112] The second aspect of the present invention further provides a defect detection scheme design system 4 based on micropore channel environment analysis, the system comprising: a memory 41, a processor 42, the memory comprising a defect detection scheme design method program based on micropore channel environment analysis, the defect detection scheme design method program based on micropore channel environment analysis when executed by the processor, implements the following steps:

[0113] Acquire computer tomography data of a target workpiece, construct a three-dimensional model of the target workpiece according to the computer tomography data, determine the distribution of micropore channels of the target workpiece according to the three-dimensional model of the target workpiece, and construct a micropore channel distribution map;

[0114] Analyzing the micro-hole channel distribution map based on a depth-first search algorithm to determine a defect detection path for the micro-hole channel of the target workpiece;

[0115] Obtaining standard defect image data of a micropore channel and environment-defect image data of a micropore channel defect under each channel environment feature, and determining a defect image geometric deviation under each environment feature according to the standard defect image data and the environment-defect image data;

[0116] Determine the influence of various environmental characteristics on the detection accuracy of micro-hole channel defects according to the defect image geometric deviation, and obtain influence data;

[0117] Determining a defect detection strategy for micropore channels under different environmental characteristics according to the influencing data;

[0118] A defect detection scheme for the micro-pore channel of the target workpiece is constructed according to the defect detection path and defect detection strategy.

[0119] The present invention discloses a defect detection scheme design method and system based on micro-pore channel environmental analysis, comprising the following steps: first, obtaining computer tomography data of the target workpiece, constructing a three-dimensional model, and determining the distribution of micro-pore channels to generate a distribution map. Then, the channel distribution map is analyzed based on a depth-first search algorithm to determine the defect detection path. Next, standard defect image data and environmental-defect image data under different environmental characteristics are obtained, and the influence of image geometric deviation on detection accuracy is analyzed to obtain impact data. Finally, defect detection strategies under different environments are formulated based on the impact data, and a complete defect detection scheme is constructed in combination with the detection path. The present invention improves the accuracy and efficiency of micro-pore channel defect detection.

[0120] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0121] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0122] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0123] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks or optical disks, and other media that can store program codes.

[0124] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0125] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A defect detection scheme design method based on micropore channel environment analysis, characterized in that: The following steps are involved: Acquire computer tomography data of a target workpiece, construct a three-dimensional model of the target workpiece according to the computer tomography data, determine the distribution of micropore channels of the target workpiece according to the three-dimensional model of the target workpiece, and construct a micropore channel distribution map; Analyzing the micro-hole channel distribution map based on a depth-first search algorithm to determine a defect detection path for the micro-hole channel of the target workpiece; Obtaining standard defect image data of a micropore channel and environment-defect image data of a micropore channel defect under each channel environment feature, and determining a defect image geometric deviation under each environment feature according to the standard defect image data and the environment-defect image data; Determine the influence of various environmental characteristics on the detection accuracy of micro-hole channel defects according to the defect image geometric deviation, and obtain influence data; Determining a defect detection strategy for micropore channels under different environmental characteristics according to the influencing data; A defect detection scheme for the micro-pore channel of the target workpiece is constructed according to the defect detection path and defect detection strategy.

2. A defect detection scheme design method based on micropore channel environment analysis according to claim 1, characterized in that: The step of acquiring computer tomography data of a target workpiece, constructing a three-dimensional model of the target workpiece according to the computer tomography data, determining the distribution of micropore channels of the target workpiece according to the three-dimensional model of the target workpiece, and constructing a micropore channel distribution map is specifically as follows: Scanning the target workpiece based on industrial CT to obtain computer tomography data of the target workpiece, and constructing a three-dimensional model of the target workpiece according to the computer tomography data; Obtaining density values ​​at various positions in the three-dimensional model, and constructing a density histogram of the three-dimensional model according to the density values ​​at various positions; Obtaining density values ​​of preset points of micropore channels in the three-dimensional model, and determining a micropore channel density threshold for model segmentation according to the density values ​​of the preset points; Segmenting the three-dimensional model according to a micropore channel density threshold and a density histogram of the three-dimensional model, and extracting micropore channel edge position information in the three-dimensional model; According to the edge position information of the micropore channel, the micropore channel part in the three-dimensional model is converted into a line network based on a distance transformation method, and the micropore channel is re-voxelized according to the line network to obtain a three-dimensional model of the micropore channel; The three-dimensional model of the micropore channel is used to determine the micropore channel distribution of the target workpiece and construct a micropore channel distribution map.

3. The defect detection scheme design method based on micropore channel environment analysis according to claim 1 is characterized in that: The micro-hole channel distribution map is analyzed based on the depth-first search algorithm to determine the defect detection path of the micro-hole channel of the target workpiece, specifically: Mapping the micropore channel distribution map onto a two-dimensional plane to obtain a micropore channel plane distribution map; The channel intersection points and channel end points in the micropore channel plane distribution map are regarded as nodes, the channels between the nodes are regarded as edges, and the connection relationship between each node and the edge is determined according to the micropore channel plane distribution map; Determine the length of the edge between each node according to the micropore channel distribution graph, and construct an undirected weighted graph of the micropore channels according to the nodes, edges, edge lengths, and connection relationships between nodes and edges; Obtaining a starting node for micropore channel defect detection, determining whether the undirected weighted graph has a ring structure, and if so, disconnecting the ring node that is the farthest from the starting node in the ring structure and has the longest edge connected to the ring node at the ring node to form an undirected acyclic weighted graph; Taking the starting node as the root node, constructing a weighted dendrogram according to the undirected acyclic weighted graph, finding the longest path of the dendrogram based on the depth-first search algorithm, marking the edges in the longest path as visited, removing the visited edges in the undirected acyclic weighted graph, updating the undirected acyclic weighted graph, and determining the end node of the longest path; The node closest to the end node in the updated undirected acyclic weighted graph is used as the root node of the dendrogram to construct the dendrogram according to the longest path, and the end node and the undirected acyclic weighted graph are updated. The dendrogram is constructed repeatedly to find the longest path and update the end node and the undirected acyclic weighted graph until all edges in the undirected acyclic weighted graph are visited; The longest path obtained each time is connected according to the acquisition time to obtain the defect detection path of the micro-hole channel of the target workpiece.

4. The defect detection scheme design method based on micropore channel environment analysis according to claim 1 is characterized in that: The step of acquiring standard defect image data of the micropore channel and environment-defect image data of the micropore channel defect under each channel environment feature, and determining the defect image geometric deviation under each environment feature according to the standard defect image data and the environment-defect image data is specifically as follows: Obtaining target workpiece production samples with different defect types in micro-pore channels, and obtaining standard defect image data of different defect types in micro-pore channels according to the production samples, wherein the different defect types include holes, cracks, blockages, and burrs; Performing experimental operations on the production sample, constructing different environmental features in the microporous channel of the production sample, and obtaining environmental-defect image data of different defect types of the microporous channel under different environmental features, wherein the environmental features include humidity, dust accumulation, and chemical corrosion in the microporous channel; Extracting defect edge features of the standard defect image data and the environment-feature defect image data based on the Canny edge detection algorithm, determining defect geometric features according to the defect edge features, and obtaining standard defect geometric features and environment-defect geometric features; The environment-defect geometric features are compared with standard defect geometric features to determine the defect image geometric deviation under each environment feature.

5. A defect detection scheme design method based on micropore channel environment analysis according to claim 4, characterized in that: The influence of the defect image geometric deviation on the micro-hole channel defect detection accuracy under various environmental characteristics is determined to obtain the influence data, which is specifically: Building a defect detection model based on a fully convolutional network, and importing the standard defect geometric features into the defect detection model for training; Importing the environment-defect geometric features into the defect detection model to perform defect detection and obtain defect detection results; Determining the defect detection accuracy of the microporous channel under different environmental characteristics according to the defect detection results; The defect image geometric deviation between the environment-defect geometric feature and the standard defect geometric feature is jointly analyzed with the defect detection accuracy under the different environmental features to determine the impact of each environmental feature on the micro-hole channel defect detection accuracy and obtain impact data.

6. The defect detection scheme design method based on micropore channel environment analysis according to claim 1 is characterized in that: The defect detection strategy of the micropore channel under different environmental characteristics is determined according to the influencing data, specifically: According to the influence data, the environmental characteristics of the micro-pore channel defect detection accuracy less than the preset influence value are marked to obtain a type of environmental characteristics, and when the micro-pore channel environment is a type of environmental characteristics, the micro-pore channel is defect-detected according to the defect detection model; A second-class defect detection model is constructed based on a fully convolutional network, and the environment-defect geometric features formed by the non-first-class environmental features are introduced into the second-class defect detection model for training; Extracting defect geometric features under each non-class I environmental feature and importing them into the class II defect detection model for defect detection, marking the non-class I environmental features corresponding to the defect geometric features with a defect detection accuracy higher than a preset value as class II environmental features, and marking the remaining environmental features as class III environmental features, and when the micro-pore channel environment is a class II environmental feature, performing defect detection on the micro-pore channel according to the class II defect detection model; The environment-defect geometric features of three types of environmental characteristics are obtained, and the environment-defect geometric features of the three types of environmental characteristics are compared with the standard defect geometric features to determine the correction parameters of the environment-defect geometric features of the three types of environmental characteristics. When the micropore channel environment is the three types of environmental characteristics, the defect geometric features obtained under the three types of environmental characteristics are corrected according to the correction parameters, and the corrected defect geometric features are imported into the defect detection model for defect detection, so as to obtain the defect detection strategy of the micropore channel under different environmental characteristics.

7. The defect detection scheme design method based on micropore channel environment analysis according to claim 1 is characterized in that: The defect detection scheme for the micro-hole channel of the target workpiece is constructed according to the defect detection path and the defect detection strategy, specifically: Constructing a defect detection scheme for a micro-pore channel of a target workpiece according to the defect detection path and defect detection strategy; Defect detection is performed on the micro-hole channel of the target workpiece according to the defect detection path, and the environmental characteristics of the micro-hole channel are determined, and defect detection is performed on defect geometric characteristics of different environmental characteristics according to the defect detection strategy.

8. A defect detection scheme design system based on micropore channel environment analysis, characterized in that: The defect detection scheme design system based on micropore channel environment analysis includes a storage device and a processor. The storage device includes a defect detection scheme design method program based on micropore channel environment analysis. When the defect detection scheme design method program based on micropore channel environment analysis is executed by the processor, the following steps are implemented: Acquire computer tomography data of a target workpiece, construct a three-dimensional model of the target workpiece according to the computer tomography data, determine the distribution of micropore channels of the target workpiece according to the three-dimensional model of the target workpiece, and construct a micropore channel distribution map; Analyzing the micro-hole channel distribution map based on a depth-first search algorithm to determine a defect detection path for the micro-hole channel of the target workpiece; Obtaining standard defect image data of a micropore channel and environment-defect image data of a micropore channel defect under each channel environment feature, and determining a defect image geometric deviation under each environment feature according to the standard defect image data and the environment-defect image data; Determine the influence of various environmental characteristics on the detection accuracy of micro-hole channel defects according to the defect image geometric deviation, and obtain influence data; Determining a defect detection strategy for micropore channels under different environmental characteristics according to the influencing data; A defect detection scheme for the micro-pore channel of the target workpiece is constructed according to the defect detection path and defect detection strategy.

Images