Ultrasonic detection method, system and equipment applied to joint coating detection and medium

Abstract

The invention relates to an ultrasonic detection method, system and device applied to repaired mouth detection and a medium, and the method comprises the steps: obtaining first position information corresponding to a repaired mouth, and calculating a moving path of an ultrasonic detector corresponding to the first position information based on a preset first heuristic search algorithm; based on the moving path, acquiring a first ultrasonic image corresponding to the repaired mouth acquired by the ultrasonic detector; and joint coating identification information corresponding to the first ultrasonic image is obtained based on a trained identification model, and the joint coating identification information is abnormal judgment information or normal judgment information. According to the method, the ultrasonic detector can move according to the optimal moving path, the accuracy and efficiency of ultrasonic image recognition are improved, the standardization degree of repairing work of the repaired mouth and the repairing quality of the repaired mouth are improved, and efficient reinspection of the repaired mouth is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of ultrasonic detection of patching, and in particular to an ultrasonic detection method, system, equipment and medium applied to patching detection. Background Art

[0002] In the industrial and maintenance fields, ensuring structural integrity and safety is of vital importance. As a common method for connecting and repairing key structural components, the quality of joint bonding is directly related to the stability and service life of the entire structure. At present, ultrasonic testing technology is used to detect joint bonding. Ultrasonic testing technology is widely used in aviation, aerospace, shipbuilding, construction and other fields due to its high sensitivity, strong material penetration ability, and ability to detect deep defects.

[0003] The existing Chinese invention patent with announcement number CN101178386A discloses an ultrasonic automatic detection system for copper and stainless steel dissimilar metal vacuum brazing water joints, whose hardware consists of a water joint automatic detection device, a single crystal water immersion focusing probe, and a comparison test block; the detection device includes an industrial computer, a motor driver and a motor controlled by it, and an ultrasonic flaw detector connected to the single crystal water immersion focusing probe and the industrial computer respectively. A motor of the detection device drives a vertically moving transmission shaft, on which a probe angle adjuster is installed, and the single crystal water immersion focusing probe is installed at the lower end of the probe angle adjuster. Another motor drives a horizontal turntable equipped with a water joint to be tested, and the position of the probe in the water joint cavity is adjusted by adjusting the probe angle adjuster, and ultrasonic waves are emitted to detect the brazing quality of the water joint.

[0004] Although the above detection system can realize automatic control, the detection system has the disadvantage of not performing ultrasonic detection path planning during the actual patch detection process. Therefore, there is an urgent need for a solution for automatic patch detection that can automatically perform detection path planning. Summary of the invention

[0005] In order to facilitate ultrasonic detection path planning during the patch detection process, the present application provides an ultrasonic detection method, system, equipment and medium for patch detection.

[0006] In a first aspect, the present application provides an ultrasonic detection method for patch detection, comprising:

[0007] Acquire first position information corresponding to the patch, and calculate a moving path of the ultrasonic detector corresponding to the first position information based on a preset first heuristic search algorithm;

[0008] Based on the moving path, acquiring a first ultrasonic image corresponding to the repaired opening acquired by the ultrasonic detector;

[0009] Based on the trained recognition model, the incision repair recognition information corresponding to the first ultrasound image is obtained, and the incision repair recognition information is abnormal determination information or normal determination information.

[0010] The beneficial effects of the present application are as follows: during the patch detection process, the moving path of the ultrasonic detector is first planned. Through path planning, the ultrasonic detector can move according to the optimal moving path, reducing the possibility of unnecessary detours and repeated detections, thereby significantly improving the detection efficiency. The planning of the moving path can ensure that the ultrasonic detector can avoid obstacles and dangerous areas during movement, ensuring the safety and stability of the detection process. The planning of the moving path can ensure that the ultrasonic detector fully covers the patch that needs to be detected during the detection process, reducing the possibility of omissions and false detections.

[0011] Furthermore, if the patch identification information is abnormal determination information, the method further includes:

[0012] Obtain a target repair solution corresponding to the determined abnormal information, so as to repair the patch according to the target repair solution.

[0013] The beneficial effect of adopting the above further scheme is that when there is an abnormality in the patch, the patch can be repaired according to the target repair scheme corresponding to the abnormal information, thereby improving the standardization of the patch repair work and the quality of patch rework.

[0014] Further, the obtaining of the target repair solution corresponding to the determined abnormal information includes:

[0015] The current key feature corresponding to the abnormal information is obtained, and based on an established standard solution database, the target repair solution corresponding to the current key feature is queried, wherein the standard solution database includes repair solutions corresponding to different key features.

[0016] The beneficial effect of adopting the above further scheme is that through the standard scheme database, the target repair scheme corresponding to the abnormal information can be queried and determined, which improves the accuracy of the target repair scheme and ensures the repair effect of the patching.

[0017] Further, after obtaining the target repair solution corresponding to the determined abnormal information, the method further includes:

[0018] Acquire a second ultrasonic image corresponding to the patch when the repair is completed, which is captured by the ultrasonic detector;

[0019] determining whether the incision repair identification information corresponding to the second ultrasonic image is the abnormality determination information;

[0020] If so, the target repair scheme in the standard scheme database is corrected based on the second ultrasound image and the abnormality determination information.

[0021] The beneficial effect of adopting the above further scheme is: when the patch identification information corresponding to the second ultrasonic image is abnormal judgment information, there is a defect in the repaired patch, and the detected second ultrasonic image and the abnormal judgment information can be fed back to the standard scheme database to rewrite the standard scheme database to improve the reliability and effectiveness of the overall repair work.

[0022] Furthermore, after the repair of the patching is completed based on the target repair solution, the method further includes:

[0023] If the repaired patch needs to be re-inspected, obtaining the first position coordinates corresponding to the ultrasonic detector and the second position coordinates corresponding to the patch;

[0024] Calculate the recheck path corresponding to the first position coordinate and the second position coordinate based on a preset second heuristic search algorithm;

[0025] Acquire a third ultrasonic image corresponding to the repaired patch collected by the ultrasonic detector during re-inspection according to the re-inspection path;

[0026] The third ultrasonic image is compared with a preset ultrasonic image corresponding to the repaired patch, and based on the comparison result, it is determined whether the re-inspection position corresponding to the patch is accurate.

[0027] The beneficial effect of adopting the above further solution is to ensure that the ultrasonic detector can be quickly and accurately moved to the position of the repaired patch along the re-inspection path, thereby achieving efficient re-inspection.

[0028] Further, the obtaining of first position information corresponding to the patching includes:

[0029] Sa: Based on the search area, a group of particles are initialized to obtain a particle set, wherein the second position information of each particle represents a hypothetical position corresponding to the patch, and an initial weight is assigned to each particle, wherein the initial weights corresponding to each particle are equal, and the weight corresponding to each particle is a parameter representing the accuracy of the hypothetical position;

[0030] Sb: predicting the next step position of each particle in the particle set based on the static model, and updating the second position information of the corresponding particle according to each predicted position;

[0031] Sc: for each particle, based on the updated second position information of the particle, obtain the patch feature matching degree corresponding to the particle, and based on the patch feature matching degree, obtain the update weight corresponding to the particle, the update weight is used to replace the initial weight as the new weight corresponding to the particle;

[0032] Sd: resampling based on the update weights corresponding to each of the particles to obtain a new set of particles;

[0033] Se: Based on the new particle set, repeat steps Sb to Sd until a preset end condition is reached, and output the final first position information based on the updated second position information of the particles in the new particle set and the updated weight.

[0034] The beneficial effect of adopting the above further solution is: achieving the feature extraction and accurate identification of the spatial position corresponding to the patch to be detected, providing a reliable data basis for the subsequent optimization of the moving path.

[0035] Furthermore, the training method of the recognition model includes:

[0036] Acquiring a wound repair ultrasound image dataset, preprocessing the wound repair ultrasound image dataset, and dividing the preprocessed wound repair ultrasound image dataset into a training set and a validation set;

[0037] Annotating each ultrasonic image data in the training set and the validation set to determine the incision recognition information corresponding to each ultrasonic image data;

[0038] The initial model is trained based on the training set, and the trained initial model is verified through the verification set to obtain the trained recognition model.

[0039] The beneficial effect of adopting the above further solution is that the well-trained recognition model can automatically identify the patch recognition information corresponding to the first ultrasonic image, thereby improving the accuracy and efficiency of ultrasonic image recognition.

[0040] In a second aspect, the present application provides an ultrasonic detection system for joint patching detection, comprising:

[0041] A path planning unit, used for acquiring first position information corresponding to the patching, and calculating a moving path of the ultrasonic detector corresponding to the first position information based on a preset first heuristic search algorithm;

[0042] an acquisition unit, configured to acquire, based on the movement path, a first ultrasonic image corresponding to the repaired opening acquired by the ultrasonic detector;

[0043] The abnormality identification unit is used to obtain the repairing identification information corresponding to the first ultrasonic image based on the trained recognition model, and the repairing identification information is abnormality determination information or normal determination information.

[0044] In a third aspect, the present application provides an electronic device, including a processor and a memory, wherein the processor is coupled to the memory;

[0045] The processor is used to execute the computer program stored in the memory so that the electronic device performs the method as described in any one of the first aspects.

[0046] In a fourth aspect, the present application provides a computer-readable storage medium, comprising a computer program or instructions, which, when executed on a computer, enables the computer to execute the method as described in any one of the first aspects. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic flow chart of an ultrasonic detection method applied to patch detection according to an embodiment of the present application;

[0048] Figure 2 This is a flowchart of a method for determining first location information according to an embodiment of the present application;

[0049] Figure 3 A schematic diagram of a flow chart of a method for determining a moving path according to an embodiment of the present application;

[0050] Figure 4 This is a structural block diagram of an ultrasonic detection system used for patch detection in an embodiment of the present application;

[0051] Figure 5 This is an overall block diagram of an ultrasonic detection system used for patch detection in an embodiment of the present application;

[0052] Figure 6 This is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0053] The present application is further described in detail below in conjunction with the accompanying drawings.

[0054] The embodiment of the present application provides an ultrasonic detection method for patch detection, which can be executed by a device, which can be a server or a terminal device, wherein the server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a desktop computer, etc., but is not limited thereto.

[0055] like Figure 1As shown, an ultrasonic detection method for patch detection is performed by an electronic device. The main process of the method is described as follows (steps S101 to S103):

[0056] Step S101: obtaining first position information corresponding to the patch, and calculating a moving path of the ultrasonic detector corresponding to the first position information based on a preset first heuristic search algorithm.

[0057] In this embodiment, the first position information may be a position area corresponding to the patch. It is easy to understand that the ultrasonic detector is a device that uses ultrasonic detection technology to detect the adhesion of the patch. The moving path is a path that guides the ultrasonic detector to reach the patch for detection, and the first heuristic search algorithm may be an A* heuristic search algorithm.

[0058] Step S102: Based on the moving path, a first ultrasonic image corresponding to the patch collected by the ultrasonic detector is acquired.

[0059] It is easy to understand that an ultrasound image is a non-invasive imaging image generated by using the reflection and scattering characteristics of ultrasound at the repair site, using an ultrasound probe to transmit and receive ultrasound, and then obtaining the internal structure and tissue information of the repair site.

[0060] Step S103: Based on the trained recognition model, the patch recognition information corresponding to the first ultrasound image is obtained, where the patch recognition information is abnormality determination information or normal determination information.

[0061] In this embodiment, the recognition model may be a deep neural network model. The ultrasound image corresponding to the patch is input into the trained deep neural network model. The deep neural network model may recognize the ultrasound image and output patch recognition information. The patch recognition information is a parameter characterizing the patch detection result. When the patch recognition information is abnormal judgment information, it indicates that the patch is abnormal. When the patch recognition information is normal judgment information, it indicates that the patch is not abnormal.

[0062] The abnormal information can include the abnormal type, abnormal distribution and abnormal degree. The abnormal type can include cracks, pores and inclusions. The abnormal distribution can include dispersed defects, local concentrated defects and widely distributed defects. The abnormal degree can include slight, moderate and severe. It is easy to understand that cracks refer to cracks in the patch, pores refer to cavities or bubbles in the patch, and inclusions refer to foreign matter in the patch. Dispersed defects refer to the random distribution of abnormalities in the patch, local concentrated defects refer to the high concentration of abnormalities in a local area of ​​the patch, and widely distributed defects refer to the widespread existence of abnormalities in the entire patch. Slight means that the abnormality has little effect on the performance of the patch, moderate means that the abnormality has a certain effect on the performance of the patch, and severe means that the abnormality has a serious effect on the performance of the patch.

[0063] The electronic device has preset correspondences between abnormality types, abnormality distribution conditions and abnormality degrees. For example, if the abnormality type is cracks and the abnormality distribution conditions are widely distributed, the corresponding abnormality degree may be severe; if the abnormality type is pores and the abnormality distribution conditions are dispersed, the corresponding abnormality degree may be slight; if the abnormality type is inclusions and the abnormality distribution conditions are locally concentrated, the corresponding abnormality degree may be medium.

[0064] In this embodiment, deep learning technology is used, and the trained deep neural network model can automatically identify the patch identification information corresponding to the first ultrasound image, thereby improving the accuracy and efficiency of ultrasound image recognition.

[0065] The electronic device can feed back the patch identification information to the user end, so that the user can connect to the current patch detection status. During the patch detection process, the moving path of the ultrasonic detector is first planned. Through path planning, the ultrasonic detector can move according to the optimal moving path, reducing the possibility of unnecessary detours and repeated detections, thereby significantly improving the detection efficiency. The planning of the moving path can ensure that the ultrasonic detector can avoid obstacles and dangerous areas during movement, ensuring the safety and stability of the detection process. The planning of the moving path can ensure that the ultrasonic detector fully covers the patches that need to be detected during the detection process, reducing the possibility of omissions and false detections.

[0066] In this embodiment, a positioning and identification strategy for identifying the first position information is stored in the electronic device, and the positioning and identification strategy specifically includes the following processing: collecting an ultrasonic image of the patch to be detected by a high-resolution device, and performing image denoising and image enhancement and other preprocessing on the ultrasonic image to optimize the ultrasonic image quality; using a convolutional neural network to extract the position and morphological features of the patch from the preprocessed ultrasonic image, and distinguishing the patch from the surrounding structure through image segmentation to obtain a search area corresponding to the patch to be detected; and using a positioning algorithm based on a probability model to automatically and accurately determine the first position information of the patch.

[0067] The positioning algorithm can adopt the Monte Carlo positioning algorithm. Through the positioning and recognition strategy, it can achieve accurate imaging of the patch to be detected, feature extraction corresponding to the patch and accurate recognition of the spatial position, providing a reliable data basis for the subsequent optimization of the moving path.

[0068] like Figure 2 As shown, in this embodiment, the positioning algorithm based on the probability model realizes automatic and accurate determination of the first position information of the patching, which specifically includes the following processing:

[0069] Sa: Based on the search area, a group of particles are initialized to obtain a particle set, the second position information of each particle represents a hypothetical position corresponding to the patch, and an initial weight is assigned to each particle, the initial weights corresponding to each particle are equal, and the weight corresponding to each particle is a parameter that characterizes the accuracy of the hypothetical position.

[0070] Initially, particles are randomly distributed in the search area, and each particle is assigned the same initial weight, indicating that all position hypotheses have the same probability at the beginning.

[0071] Sb: Based on the static model, predict the next step position of each particle in the particle set, and update the corresponding second position information of the particle according to each predicted position.

[0072] The static model predicts the possible position of each particle in the particle set at the next moment (or the next step), and updates the second position information of the particle based on the prediction. The static model is the prior art content in this field, and is not the inventive solution of the present application, and will not be described in detail here.

[0073] Sc: For each of the particles, based on the second position information of the particle after update, the patch feature matching degree corresponding to the particle is obtained, and based on the patch feature matching degree, the updated weight corresponding to the particle is obtained, and the updated weight is used to replace the initial weight as the new weight corresponding to the particle.

[0074] After updating the second position information of the particle, the matching degree of the second position information of the particle and the patch feature (such as the edge and texture in the ultrasound image) can be calculated to obtain the patch feature matching degree corresponding to each particle, and then the corresponding particle is reassigned an update weight according to the patch feature matching degree. For example, a particle with a high patch feature matching degree will be assigned a higher update weight, indicating that the second position information of the particle is more likely to be the true position of the patch.

[0075] Sd: resampling based on the update weights corresponding to each of the particles to obtain a new set of particles.

[0076] As time goes by, the update weights of some particles will become very low, while the update weights of other particles will become very high. In order to avoid wasting computing resources on particles with low update weights, resampling is required. During the resampling process, particles with high update weights are more likely to be selected multiple times, while particles with low update weights can be ignored or selected only once. In this way, the new set of particles will be more concentrated near the actual position of the patch.

[0077] Se: Based on the new particle set, repeat steps Sb to Sd until a preset end condition is reached, and output the final first position information based on the updated second position information of the particles in the new particle set and the updated weight.

[0078] The preset end condition may be that the number of repeated executions reaches a preset number of iterations, or that the updated second position information of a preset number of particles in the new particle set reaches position estimation convergence.

[0079] Due to the randomness of particles and the dynamics of the system, the second position information of particles in the particle set will change in each update iteration. However, when the position estimation converges, the distribution of the particle set can stably reflect the true position of the patch, that is, the preset number of particles in the particle set can be concentrated near the true position of the patch, and this distribution no longer changes significantly.

[0080] In this embodiment, a weighted average position can be calculated based on the updated second position information and update weights of all particles, and the weighted average position can be used as the first position information of the patch, wherein particles with higher update weights contribute more to the final first position information.

[0081] like Figure 3 As shown, in this embodiment, taking the first heuristic search algorithm as the A* heuristic search algorithm as an example, step S101 specifically further includes the following processing:

[0082] S1011: Discretize the patching location area into a series of nodes, each node represents a potential monitoring point, create an open list and a closed list, and add the start node to the open list.

[0083] In the Open List (K i , K j ) and close list (G i , G j ), i, j = 0...N.

[0084] Nodes are the basic unit of the search algorithm. Two lists are created to manage the nodes during the search process. The open list can contain nodes to be evaluated, while the closed list can contain nodes that have been evaluated. The starting node (the initial position of the ultrasound detector) is added to the open list as the starting point of the search.

[0085] S1012: Constructing a reachability model between nodes according to the mobility of the ultrasonic detector.

[0086] S1013: Define an evaluation function, which can be expressed as:

[0087]

[0088] Among them, g(n) represents the distance from the starting node (K0, K0) to the current node The actual distance from the current node is h(n). To the target node The predicted distance of e(n) is the distance passing through the target node. The energy consumption of , α, β and γ represent weight coefficients, which are used to adjust the impact of each part on the total evaluation.

[0089] S1014: Perform a circular search.

[0090] S1014 specifically includes: when the open list is not empty, repeating steps f, g and h;

[0091] Step f: Select the one with the lowest The node of the value is the current node

[0092] Step g: Set the current node Move to the closed list;

[0093] Step h: For the current node Neighboring nodes Conduct an assessment.

[0094] Step h specifically includes:

[0095] If the adjacent nodes If the connection is in the closed list or is judged to be unreachable based on the reachability model, it will be skipped;

[0096] Otherwise, calculate from the starting node (K0, K0) through the current node Reaching adjacent nodes Path;

[0097] If the adjacent nodes If it is not in the open list, add it;

[0098] If the adjacent nodes If the new path is in the open list but is better, update the current node information;

[0099] Record the current node Neighboring nodes This allows you to rebuild the path later.

[0100] S1015: Reconstruct the moving path and perform simulated annealing optimization on the calculated moving path.

[0101] Target Node When it is added to the open list, it means that a path has been found through the adjacent nodes Information from the target node By tracing back to the starting node (K0, K0), the complete moving path can be reconstructed, and then the calculated moving path is optimized by simulated annealing. The moving path is converted into a control instruction to drive the ultrasonic detector to detect along the moving path. When all potential monitoring points of the patch are detected, the patch detection is completed.

[0102] In this embodiment, S1012 specifically includes the following processing:

[0103] The environment of the patching location area is converted into a discrete grid structure, the mobility of the ultrasonic detector is determined according to the physical characteristics of the ultrasonic detector, and the size and shape of the grid structure are determined according to the mobility of the ultrasonic detector. The mobility of the ultrasonic detector includes the moving direction, moving distance and steering limit.

[0104] The patching location area is converted into a discrete grid structure to facilitate modeling and calculation.

[0105] Clearly define the boundaries of the area to be inspected and mark obstacles in the grid structure that are not passable, including pipes, equipment and other fixed structures.

[0106] By setting specific values ​​at the positions of obstacles in the grid structure, the obstacle coordinates can be marked. By marking obstacles in impassable areas in the grid structure, it can be ensured that the generated moving path will not pass through these areas.

[0107] Create an r×r matrix, where r represents the total number of nodes in the environment. For each pair of nodes (p, q), p, q = 1...r, if node q is within the moving range of node p and there are no obstacles between nodes, the corresponding position in the matrix is ​​marked as reachable.

[0108] The matrix created can be used to record the reachability between nodes. If there are no obstacles between two nodes and the distance is within the moving range of the probe, they are considered reachable.

[0109] According to the reachability information between nodes, a graph T = (V, E) is constructed, where V is a set of nodes, indicating that each node corresponds to a specific location in the physical space; E is a set of edges, indicating the connection relationship between nodes. For each pair of nodes in the graph, if the distance between the nodes is less than or equal to the distance of one movement of the probe, and there are no obstacles on the line between the nodes, then an edge is added between the nodes.

[0110] Based on the reachability information between nodes, a graph is constructed to represent the connection relationship between nodes. Each edge in the graph represents a possible movement path between nodes.

[0111] The energy and time consumption between nodes are calculated, the cost information is included in the critical matrix, and a cost value is assigned to each reachable node.

[0112] Assign a cost value to each reachable node. These cost values ​​can include factors such as energy consumption and time consumption to evaluate the quality of the path. The critical matrix is ​​used to store the reachability information between nodes. When checking whether the nodes are reachable, the critical matrix can be used to find the entries in the critical matrix. When it is necessary to check whether the nodes are reachable, the corresponding entry can be found in the critical matrix according to the node index (or identifier).

[0113] By calculating the moving path of the ultrasonic detector through the A* heuristic search algorithm, the entire patching location area can be covered efficiently and comprehensively, the detection efficiency and coverage rate are improved, and the accuracy and reliability of the detection process are ensured.

[0114] In this embodiment, the training method of the recognition model specifically includes the following processing:

[0115] Acquire a patched ultrasound image dataset, preprocess the patched ultrasound image dataset, and divide the preprocessed patched ultrasound image dataset into a training set and a validation set; annotate each ultrasound image data in the training set and the validation set to determine the patched recognition information corresponding to each ultrasound image data; train an initial model based on the training set, and verify the trained initial model through the validation set to obtain the trained recognition model.

[0116] The deep neural network model is trained with a large amount of ultrasound image data of incision repair, so that the deep neural network model can automatically learn and recognize the characteristic patterns of ultrasound image data of incision repair. The preprocessing of ultrasound image data set of incision repair includes resizing, normalizing pixel values ​​and enhancing contrast, so as to reduce the computational burden of model training and improve recognition accuracy.

[0117] After the incision repair ultrasound image dataset is preprocessed, the ultrasound image data in the ultrasound image dataset needs to be labeled to clarify the incision repair identification information corresponding to each ultrasound image data.

[0118] The initial model is trained using the labeled training set. The ultrasound image features are extracted from the input ultrasound image data and mapped to the correct patch recognition information. The weights in the network are adjusted through the back propagation algorithm. The cross entropy loss function is used to measure the classification performance index. The RMSprop algorithm is used to optimize the parameters of the deep neural network model. The formula of the RMSprop algorithm can be expressed as:

[0119]

[0120] Among them, v t represents the exponentially weighted moving average of the square of the gradient, ω represents the attenuation factor, which is usually set to 0.9, and g t represents the gradient at time step t, θ t represents the parameter model at time step t, η represents the learning rate, and ε represents a constant to avoid division by zero.

[0121] The performance of the deep neural network model was evaluated using the ultrasound image data in the validation set, and the deep neural network model structure, hyperparameters, and training strategies were adjusted to improve the generalization ability of the deep neural network model.

[0122] In this embodiment, if the patch identification information is abnormal determination information, the method further includes: acquiring a target repair solution corresponding to the abnormal determination information, so as to repair the patch according to the target repair solution.

[0123] The repair plan is a plan for repairing the patch. When there is an abnormality in the patch, the patch can be repaired according to the target repair plan corresponding to the abnormal information, thereby improving the standardization of the patch repair work and the quality of the patch rework.

[0124] The obtaining of the target repair solution corresponding to the determined abnormal information specifically includes the following processing: obtaining the current key feature corresponding to the determined abnormal information, and querying the target repair solution corresponding to the current key feature based on an established standard solution database, wherein the standard solution database includes repair solutions corresponding to different key features.

[0125] In this embodiment, the specific contents of establishing the standard solution database include:

[0126] Determine the data types, query types and user groups that the standard solution database needs to support, and identify the data sources;

[0127] Collect and organize historical restoration cases, including restoration plans, material types used, and restoration results, and obtain industry standards and specification documents for joint bonding quality;

[0128] Perform data cleaning on the collected historical restoration cases, extract key features from complex data, and convert the data into a unified format for storage and query. Data cleaning includes removing inconsistent, erroneous, and incomplete data. Key features may include material type, anomaly type, anomaly distribution, anomaly degree, and restoration plan.

[0129] Design the entity-relationship model of the standard solution database, including entities, attributes and relationships, set the data storage structure as a relational database, set hash indexes to improve query efficiency, and set data access permissions to read-only, read-write and management based on user roles.

[0130] Data types may include structured data, semi-structured data, and unstructured data. Structured data may include tables, numbers, and text; semi-structured data may include XML and JSON; and unstructured data may include images, audio, and video.

[0131] Query types can include simple query, complex query, full-text search and real-time query. User groups can include testing engineers, maintenance personnel, R&D personnel and supervisors. Data sources can include internal data sources and external data sources.

[0132] Through the correspondence between key features and repair solutions in the standard solution database, the target repair solution corresponding to the current key feature can be queried.

[0133] In this embodiment, after obtaining the target repair scheme corresponding to the determined abnormal information, it also includes: obtaining a second ultrasonic image corresponding to the patch when the repair is completed, which is collected by the ultrasonic detector; determining whether the patch identification information corresponding to the second ultrasonic image is the determined abnormal information; if so, correcting the target repair scheme in the standard scheme database based on the second ultrasonic image and the determined abnormal information.

[0134] During the repair process, ultrasonic testing equipment is used to collect test data. Based on the actual ultrasonic test results, it is determined whether the repair quality of the patch meets the industry standards for patch bonding quality and the predetermined standards in the specification documents. If the predetermined standards are met, the patch identification information corresponding to the second ultrasonic image is judged as normal information. If the predetermined standards are not met, the patch identification information corresponding to the second ultrasonic image is judged as abnormal information.

[0135] When the patch identification information corresponding to the second ultrasonic image is abnormality determination information, the repaired patch has defects. The detected second ultrasonic image and the abnormality determination information can be fed back to the standard scheme database to rewrite the standard scheme database to improve the reliability and effectiveness of the overall repair work.

[0136] In this embodiment, after the patch is repaired based on the target repair scheme, the following processing is also included: if the repaired patch needs to be re-inspected, the first position coordinates corresponding to the ultrasonic detector and the second position coordinates corresponding to the patch are obtained; based on a preset second heuristic search algorithm, the re-inspection path corresponding to the first position coordinates and the second position coordinates are calculated; a third ultrasonic image corresponding to the repaired patch, which is collected by the ultrasonic detector during the re-inspection according to the re-inspection path, is obtained; the third ultrasonic image is compared with the preset ultrasonic image corresponding to the repaired patch, and based on the comparison result, it is determined whether the re-inspection position corresponding to the patch is accurate.

[0137] According to the second position coordinates of the corrected patch and the first position coordinates of the ultrasonic detector, a fast path planning algorithm can be used to calculate the shortest and obstacle-free optimal re-inspection path to ensure that the ultrasonic detector can move quickly and accurately to the position of the repaired patch along the re-inspection path to achieve efficient re-inspection.

[0138] The second heuristic search algorithm may also be an A* heuristic search algorithm. Based on the preset second heuristic search algorithm, calculating the corresponding recheck path between the first position coordinate and the second position coordinate specifically includes:

[0139] Get the first position coordinates (x c ,y c ) and all the second position coordinates (x i ,y i ), all the second position coordinates that need to be rechecked are taken as the target coordinates;

[0140] Use the A* heuristic search algorithm to calculate the shortest path by constructing a node network, where each node represents a location point, the edge between nodes represents the feasibility of movement, and the weight of the edge represents the cost of movement;

[0141] Define the evaluation function, which can be expressed as:

[0142] f(n)=g(n)+h(n)

[0143] Among them, g(n) is the actual distance from the starting point to the current node, and Manhattan distance is used as the movement cost, then g(n) = d(start,n), h(n) is the estimated distance from the current node to the target node, based on heuristic estimation, using the Euclidean distance definition, then x n ,y n Respectively represent the horizontal and vertical coordinates of the current node n in the two-dimensional space, x goal ,y goalRespectively represent the horizontal and vertical coordinates of the target node in the two-dimensional space;

[0144] Execute the A* heuristic search algorithm and select the node with the lowest f(n) value as the next visited node until the target node is reached, generating a recheck path.

[0145] The entire search area is abstracted as a node network, where each node represents a location point (which can be a starting node, a target node, or any point on the path). The edge between nodes represents the feasibility of moving from one node (location) to another node, and the weight of the edge represents the cost (or price) of moving from one node to another. In this embodiment, Manhattan distance is used to calculate the actual movement cost.

[0146] Manhattan distance, which represents the distance from one point to another when moving only along the grid lines. Euclidean distance, which provides an estimate of the straight-line distance between two points.

[0147] The electronic device is equipped with an image guidance strategy, which includes obtaining a real-time third ultrasonic image of the repair patch position through an ultrasonic detector, and performing pre-processing such as denoising and enhancement on the third ultrasonic image to improve the clarity of the third ultrasonic image.

[0148] The electronic device stores a preset ultrasound image corresponding to the repaired patch, and uses the preset ultrasound image as a standard ultrasound image. The Canny edge detection algorithm can be used to identify the contour of the patch in the third ultrasound image, and the third ultrasound image is compared with the preset ultrasound image, and the structural similarity index is used to verify the accuracy of the re-inspection position corresponding to the patch.

[0149] Through the image guidance strategy, it is possible to ensure that the ultrasonic detector is efficiently and accurately moved to the position of the patch that needs to be re-inspected after repair during the re-inspection process, and to achieve high-precision re-inspection verification. Among them, the Canny edge detection algorithm and the structural similarity index are the existing technical content in this field, and are not the creative solution of this application, and will not be described here.

[0150] Based on the same technical concept, the present application also provides an ultrasonic detection system for patch detection, such as Figure 4 As shown, the ultrasonic detection system 200 applied to patch detection mainly includes:

[0151] A path planning unit 201 is used to obtain first position information corresponding to the patching, and calculate a moving path of the ultrasonic detector corresponding to the first position information based on a preset first heuristic search algorithm;

[0152] An acquisition unit 202 is configured to acquire, based on the moving path, a first ultrasonic image corresponding to the repaired opening acquired by the ultrasonic detector;

[0153] The abnormality identification unit 203 is used to obtain the incision repair identification information corresponding to the first ultrasonic image based on the trained recognition model, and the incision repair identification information is abnormality determination information or normal determination information.

[0154] like Figure 4 and Figure 5 As shown, optionally, the path planning unit 201 includes a target positioning unit and a path optimization unit;

[0155] The target positioning unit includes:

[0156] An initial submodule is used to initialize a group of particles based on the search area to obtain a particle set, wherein the second position information of each particle represents a hypothetical position corresponding to the patch, and an initial weight is assigned to each particle, wherein the initial weights corresponding to the particles are equal, and the weight corresponding to each particle is a parameter representing the accuracy of the hypothetical position;

[0157] A prediction and updating submodule, used for predicting the next step position of each particle in the particle set based on a static model, and updating the second position information of the corresponding particle according to each predicted position;

[0158] A weight updating submodule, for obtaining, for each particle, a patch feature matching degree corresponding to the particle based on the updated second position information of the particle, and obtaining an updated weight corresponding to the particle based on the patch feature matching degree, wherein the updated weight is used to replace the initial weight as a new weight corresponding to the particle;

[0159] A resampling submodule, used for resampling based on the update weights corresponding to the respective particles to obtain a new set of particles;

[0160] The iterative submodule is used to repeatedly execute the processing from the prediction update submodule to the resampling submodule based on the new particle set until a preset end condition is reached, and output the final first position information based on the updated second position information of the particles in the new particle set and the updated weight.

[0161] Optionally, the first heuristic search algorithm may be an A* heuristic search algorithm, and the path optimization unit is used to calculate the moving path based on the A* heuristic search algorithm.

[0162] Optionally, the abnormality identification unit includes an abnormality recognition unit; the recognition model may be a deep neural network model. The abnormality recognition unit is used to obtain the patch recognition information based on the trained deep neural network model.

[0163] The training device of the recognition model comprises:

[0164] An acquisition processing module is used to acquire a wound repair ultrasound image dataset, preprocess the wound repair ultrasound image dataset, and divide the preprocessed wound repair ultrasound image dataset into a training set and a validation set;

[0165] A labeling module, used to label each ultrasonic image data in the training set and the validation set, and determine the incision recognition information corresponding to each ultrasonic image data;

[0166] The training and verification module is used to train the initial model based on the training set, and verify the trained initial model through the verification set to obtain the trained recognition model.

[0167] Optionally, if the patch identification information is abnormal determination information, the system further includes a rework strategy unit and a quality monitoring unit;

[0168] A repair strategy unit, used for obtaining a target repair scheme corresponding to the determined abnormal information, so as to repair the patch according to the target repair scheme;

[0169] The Rework Strategy Unit also includes:

[0170] The query submodule is used to obtain the current key feature corresponding to the abnormal information, and query the target repair solution corresponding to the current key feature based on the established standard solution database, wherein the standard solution database includes repair solutions corresponding to different key features.

[0171] The quality control unit also includes:

[0172] A first acquisition submodule is used to acquire a second ultrasonic image corresponding to the repaired wound when the repair is completed, which is collected by the ultrasonic detector;

[0173] A judgment submodule, used to judge whether the incision repair identification information corresponding to the second ultrasonic image is the determined abnormal information; if so, transfer to the repair submodule;

[0174] A repair submodule is used to correct the target repair scheme in the standard scheme database based on the second ultrasound image and the determined abnormality information.

[0175] Optionally, after the patching is repaired based on the target repair scheme, the system further includes a re-inspection positioning unit, which includes a re-inspection planning unit and a re-inspection execution unit;

[0176] Re-examination plan unit, including:

[0177] A second acquisition submodule is used to acquire the first position coordinates corresponding to the ultrasonic detector and the second position coordinates corresponding to the patch when the patch needs to be re-inspected after repair;

[0178] The calculation submodule is used to calculate the corresponding re-inspection path between the first position coordinates and the second position coordinates based on a preset second heuristic search algorithm.

[0179] The re-inspection execution unit comprises:

[0180] A third acquisition submodule is used to acquire a third ultrasonic image corresponding to the repaired patch collected by the ultrasonic detector during the re-inspection according to the re-inspection path;

[0181] The comparison and judgment submodule is used to compare the third ultrasonic image with the preset ultrasonic image corresponding to the repaired patch, and based on the comparison result, judge whether the re-inspection position corresponding to the patch is accurate.

[0182] In one example, the module in any of the above systems can be one or more integrated circuits configured to implement the above methods, such as: one or more application specific integrated circuits (ASICs), or, one or more digital signal processors (DSPs), or, one or more field programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0183] For another example, when the modules in the system can be implemented in the form of a processing element scheduling program, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call a program. For another example, these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0184] Various objects such as various messages / information / equipment / network elements / systems / devices / actions / operations / processes / concepts that may appear in this application are named. It can be understood that these specific names do not constitute a limitation on the relevant objects. The names assigned may change with factors such as scenarios, contexts or usage habits. The understanding of the technical meaning of the technical terms in this application should be mainly determined from the functions and technical effects embodied / executed in the technical scheme.

[0185] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0186] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0187] Based on the same technical concept, the present application also provides an electronic device, such as Figure 6 As shown, the electronic device 300 includes a processor 301 and a memory 302 , and may further include an information input / information output (I / O) interface 303 , one or more of a communication component 304 , and a communication bus 305 .

[0188] The processor 301 is used to control the overall operation of the electronic device 300 to complete all or part of the steps in the ultrasonic detection method for patch detection; the memory 302 is used to store various types of data to support the operation of the electronic device 300, and these data may include, for example, instructions for any application or method operated on the electronic device 300, and data related to the application. The memory 302 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0189] The I / O interface 303 provides an interface between the processor 301 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 304 is used to test the wired or wireless communication between the electronic device 300 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 304 can include: Wi-Fi components, Bluetooth components, NFC components.

[0190] The communication bus 305 may include a path to transmit information between the above components. The communication bus 305 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 305 may be divided into an address bus, a data bus, a control bus, etc.

[0191] The electronic device 300 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the ultrasonic detection method for patching detection given in the above embodiment.

[0192] The electronic device 300 may include but is not limited to mobile terminals such as digital broadcast receivers, PDAs (personal digital assistants), PMPs (portable multimedia players), etc., and fixed terminals such as digital TVs, desktop computers, etc., and may also be servers, etc.

[0193] Based on the same technical concept, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned ultrasonic detection method for patching detection are implemented.

[0194] The computer-readable storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0195] The terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or also includes elements inherent to such process, method, article, or apparatus.

[0196] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0197] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0198] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. An ultrasonic testing method for joint patching detection, It is characterized in that include: Acquire first position information corresponding to the patch, and calculate a moving path of the ultrasonic detector corresponding to the first position information based on a preset first heuristic search algorithm; Based on the moving path, acquiring a first ultrasonic image corresponding to the repaired opening acquired by the ultrasonic detector; Based on the trained recognition model, the incision repair recognition information corresponding to the first ultrasound image is obtained, and the incision repair recognition information is abnormal determination information or normal determination information.

2. According to claim 1, an ultrasonic detection method for patch detection, It is characterized in that If the patch identification information is abnormal determination information, the method further includes: Obtain a target repair solution corresponding to the determined abnormal information, so as to repair the patch according to the target repair solution.

3. The ultrasonic detection method for patching detection according to claim 2, It is characterized in that The obtaining of a target repair solution corresponding to the determined abnormal information includes: The current key feature corresponding to the abnormal information is obtained, and based on an established standard solution database, the target repair solution corresponding to the current key feature is queried, wherein the standard solution database includes repair solutions corresponding to different key features.

4. The ultrasonic detection method for patching detection according to claim 3, It is characterized in that After obtaining the target repair solution corresponding to the abnormal information, the method further includes: Acquire a second ultrasonic image corresponding to the patch when the repair is completed, which is captured by the ultrasonic detector; determining whether the incision repair identification information corresponding to the second ultrasonic image is the abnormality determination information; If so, the target repair scheme in the standard scheme database is corrected based on the second ultrasound image and the abnormality determination information.

5. An ultrasonic detection method for patch detection according to any one of claims 2 to 4, It is characterized in that After the patching is completed based on the target repair solution, the method further includes: If the repaired patch needs to be re-inspected, obtaining the first position coordinates corresponding to the ultrasonic detector and the second position coordinates corresponding to the patch; Calculate the recheck path corresponding to the first position coordinate and the second position coordinate based on a preset second heuristic search algorithm; Acquire a third ultrasonic image corresponding to the repaired patch collected by the ultrasonic detector during re-inspection according to the re-inspection path; The third ultrasonic image is compared with a preset ultrasonic image corresponding to the repaired patch, and based on the comparison result, it is determined whether the re-inspection position corresponding to the patch is accurate.

6. The ultrasonic detection method for patching detection according to claim 1, It is characterized in that The obtaining of first position information corresponding to the patching includes: Sa: Based on the search area, a group of particles are initialized to obtain a particle set, wherein the second position information of each particle represents a hypothetical position corresponding to the patch, and an initial weight is assigned to each particle, wherein the initial weights corresponding to each particle are equal, and the weight corresponding to each particle is a parameter representing the accuracy of the hypothetical position; Sb: predicting the next step position of each particle in the particle set based on the static model, and updating the second position information of the corresponding particle according to each predicted position; Sc: for each particle, based on the updated second position information of the particle, obtaining the patch feature matching degree corresponding to the particle, and based on the patch feature matching degree, obtaining the update weight corresponding to the particle, the update weight is used to replace the initial weight as the new weight corresponding to the particle; Sd: resampling based on the update weights corresponding to each of the particles to obtain a new set of particles; Se: Based on the new particle set, repeat steps Sb to Sd until a preset end condition is reached, and output the final first position information based on the updated second position information of the particles in the new particle set and the updated weight.

7. The ultrasonic detection method for patch detection according to claim 1, It is characterized in that The training method of the recognition model comprises: Acquiring a wound repair ultrasound image dataset, preprocessing the wound repair ultrasound image dataset, and dividing the preprocessed wound repair ultrasound image dataset into a training set and a validation set; Annotating each ultrasonic image data in the training set and the validation set to determine the incision recognition information corresponding to each ultrasonic image data; The initial model is trained based on the training set, and the trained initial model is verified through the verification set to obtain the trained recognition model.

8. An ultrasonic testing system for joint detection, It is characterized in that include: A path planning unit, used for acquiring first position information corresponding to the patching, and calculating a moving path of the ultrasonic detector corresponding to the first position information based on a preset first heuristic search algorithm; an acquisition unit, configured to acquire, based on the movement path, a first ultrasonic image corresponding to the repaired opening acquired by the ultrasonic detector; The abnormality identification unit is used to obtain the repairing identification information corresponding to the first ultrasonic image based on the trained recognition model, and the repairing identification information is abnormality determination information or normal determination information.

9. An electronic device, It is characterized in that comprising a processor and a memory, wherein the processor is coupled to the memory; The processor is configured to execute a computer program stored in the memory, so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, It is characterized in that The method comprises a computer program or an instruction, which, when executed on a computer, causes the computer to execute the method according to any one of claims 1 to 7.

Images