Workpiece surface hole flaw detection method and system based on mechanical arm flaw detection equipment

Abstract

The invention discloses a workpiece surface hole flaw detection method and system based on mechanical arm flaw detection equipment, and the method comprises the steps: controlling a workpiece conveyor belt to convey a target workpiece to a position right below the visual field of a binocular vision camera; shooting the target workpiece to obtain a workpiece surface image of the target workpiece; coordinate information of each surface hole is obtained; generating a planned detection path according to the coordinate information of each surface hole of the target workpiece; a mechanical arm is controlled to drive an automatic coupling agent smearing device to conduct coupling agent smearing on each surface hole of the target workpiece; controlling a mechanical arm to drive an ultrasonic phased array probe to sequentially move to the mass center position of each surface hole of the target workpiece according to the planned detection path, and controlling the ultrasonic phased array probe to scan a first area corresponding to each surface hole so as to generate a corresponding ultrasonic image; and performing flaw detection on each surface hole of the target workpiece according to the ultrasonic image. Therefore, the accuracy and efficiency of flaw detection can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of flaw detection, and in particular to a method and system for flaw detection of workpiece surface holes based on a mechanical arm flaw detection device. Background Art

[0002] With the rapid development of industrial manufacturing, the demand for non-destructive testing of workpiece surface holes is increasing. During the use of workpieces, surface holes may be damaged due to repeated friction, impact, pressure and other factors. Defects in the surface holes of workpieces will affect the overall performance of the product, accelerate the expansion of hole wall damage, and even affect the service life of the equipment, so these defects need to be detected in a timely manner.

[0003] In the related technologies, non-destructive testing of workpiece surface holes mainly adopts eddy current testing, machine vision testing, magnetic particle testing, penetration testing and X-ray testing. However, these testing methods have low detection accuracy and low work efficiency. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a workpiece surface hole flaw detection method based on a robotic arm flaw detection device, which can accurately perform flaw detection on the surface holes of the target workpiece with high efficiency.

[0005] The technical solution adopted by the present invention is as follows: A method for flaw detection of workpiece surface holes based on a mechanical arm flaw detection device, wherein the mechanical arm flaw detection device comprises a fixed bracket, a binocular vision camera, a mechanical arm, an ultrasonic phased array probe, an automatic coupling agent coating device and a workpiece conveyor belt, wherein the binocular vision camera is fixed on the fixed bracket, and the binocular vision camera is arranged above the workpiece conveyor belt, the fixed bracket and the mechanical arm are respectively arranged on both sides of the workpiece conveyor belt, the ultrasonic phased array probe and the automatic coupling agent coating device are respectively fixed at the ends of the mechanical arm, wherein the method for flaw detection of workpiece surface holes comprises the following steps: S1, when a target workpiece is placed on the workpiece conveyor belt, controlling the workpiece conveyor belt to transport the target workpiece to directly below the field of view of the binocular vision camera; S2, controlling the binocular vision camera to conduct coupling agent coating on the target workpiece S3, calibrating each surface hole of the target workpiece according to the workpiece surface image, and obtaining the coordinate information of each surface hole; S4, planning the detection path of the robot arm according to the coordinate information of each surface hole of the target workpiece to generate a planned detection path; S5, controlling the robot arm to drive the automatic coupling agent application device to apply coupling agent to each surface hole of the target workpiece; S6, controlling the robot arm to drive the ultrasonic phased array probe to move to the centroid position of each surface hole of the target workpiece in turn according to the planned detection path, and controlling the ultrasonic phased array probe to scan the first area corresponding to each surface hole to generate a corresponding ultrasonic image; S7, performing flaw detection on each surface hole of the target workpiece according to the ultrasonic image.

[0006] In one embodiment of the present invention, step S4 specifically includes the following steps: S41, based on the crown porcupine algorithm, each initial detection path is regarded as a porcupine individual to initialize the porcupine group; wherein each porcupine individual represents a traversal order of the centroid of the surface holes, and the initial detection path is generated by random arrangement according to the coordinate information of each surface hole of the target workpiece; S42, the total travel distance of each of the initial detection paths is calculated to obtain the fitness value corresponding to each porcupine individual; S43, based on the defense mechanism, the path is optimized according to the fitness value corresponding to each porcupine individual to obtain the planned detection path.

[0007] In one embodiment of the present invention, step S7 specifically includes: S71, acquiring historical ultrasonic images and annotating each of the historical ultrasonic images; S72, preprocessing the annotated images to obtain a target training set; S73, configuring a YOLO-v8 network, and training the YOLO-v8 network according to the target training set to generate a YOLO-v8 model; S74, performing flaw detection on each surface hole of the target workpiece according to the ultrasonic image based on the YOLO-v8 model.

[0008] A workpiece surface hole flaw detection system based on a robotic arm flaw detection device, the robotic arm flaw detection device comprising a fixed bracket, a binocular vision camera, a robotic arm, an ultrasonic phased array probe, an automatic coupling agent application device and a workpiece conveyor belt, wherein the binocular vision camera is fixed on the fixed bracket, and the binocular vision camera is arranged above the workpiece conveyor belt, the fixed bracket and the robotic arm are respectively arranged on both sides of the workpiece conveyor belt, the ultrasonic phased array probe and the automatic coupling agent application device are respectively fixed at the ends of the robotic arm, wherein the workpiece surface hole flaw detection system comprises: a first control module, the first control module is used to control the workpiece conveyor belt to transport the target workpiece to the directly below the field of view of the binocular vision camera when the target workpiece is placed on the workpiece conveyor belt; a second control module, the second control module is used to control the binocular vision camera to shoot the target workpiece to obtain the workpiece surface of the target workpiece image; an acquisition module, the acquisition module is used to calibrate each surface hole of the target workpiece according to the workpiece surface image, and obtain the coordinate information of each surface hole; a generation module, the generation module is used to plan the detection path of the robot arm according to the coordinate information of each surface hole of the target workpiece, so as to generate a planned detection path; a third control module, the third control module is used to control the robot arm to drive the coupling agent automatic application device to apply coupling agent to each surface hole of the target workpiece; a fourth control module, the fourth control module is used to control the robot arm to drive the ultrasonic phased array probe to move to the centroid position of each surface hole of the target workpiece in turn according to the planned detection path, and control the ultrasonic phased array probe to scan the first area corresponding to each surface hole to generate a corresponding ultrasonic image; a detection module, the detection module is used to perform flaw detection on each surface hole of the target workpiece according to the ultrasonic image.

[0009] In one embodiment of the present invention, the generation module is specifically used to: regard each initial detection path as a porcupine individual based on the crown porcupine algorithm to initialize the porcupine group; wherein each porcupine individual represents a traversal order of the centroid of a surface hole, and the initial detection path is generated by random arrangement according to the coordinate information of each surface hole of the target workpiece; calculate the total travel distance of each of the initial detection paths to obtain the fitness value corresponding to each porcupine individual; perform path optimization based on the fitness value corresponding to each porcupine individual based on the defense mechanism to obtain the planned detection path.

[0010] In one embodiment of the present invention, the detection module is specifically used to: acquire historical ultrasonic images and annotate each of the historical ultrasonic images; preprocess the annotated images to obtain a target training set; configure a YOLO-v8 network and train the YOLO-v8 network according to the target training set to generate a YOLO-v8 model; and perform flaw detection on each surface hole of the target workpiece according to the ultrasonic image based on the YOLO-v8 model.

[0011] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above-mentioned workpiece surface hole flaw detection method based on a robotic arm flaw detection device is implemented.

[0012] A non-temporary computer-readable storage medium stores a computer program, which, when executed by a processor, implements the above-mentioned workpiece surface hole flaw detection method based on a robotic arm flaw detection device.

[0013] Beneficial effects of the present invention: The present invention can accurately perform flaw detection on the surface holes of a target workpiece with high efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a flow chart of a workpiece surface hole flaw detection method based on a mechanical arm flaw detection device according to an embodiment of the present invention; Figure 2 A schematic diagram of the structure of a mechanical arm flaw detection device according to an embodiment of the present invention; Figure 3 It is a block diagram of a workpiece surface hole flaw detection system based on a robotic arm flaw detection device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] Figure 1 The figure is a flow chart of a workpiece surface hole flaw detection method based on a robotic arm flaw detection device.

[0017] Among them, in one embodiment of the present invention, Figure 2 As shown, the robotic arm flaw detection equipment 1 includes a fixed bracket 10, a binocular vision camera 20, a robotic arm 30, an ultrasonic phased array probe 40, an automatic coupling agent application device 50 and a workpiece conveyor belt 60, wherein the binocular vision camera 20 is fixed on the fixed bracket 10, and the binocular vision camera 20 is arranged above the workpiece conveyor belt 60, the fixed bracket 10 and the robotic arm 30 are respectively arranged on both sides of the workpiece conveyor belt 60, and the ultrasonic phased array probe 40 and the automatic coupling agent application device 50 are respectively fixed at the end of the robotic arm 30.

[0018] like Figure 1 As shown, the workpiece surface hole flaw detection method according to the embodiment of the present invention may include the following steps: S1, when the target workpiece is placed on the workpiece conveyor belt, the workpiece conveyor belt is controlled to transport the target workpiece to the position directly below the field of view of the binocular vision camera.

[0019] S2, controlling the binocular vision camera to shoot the target workpiece to obtain a workpiece surface image of the target workpiece.

[0020] S3, calibrating each surface hole of the target workpiece according to the workpiece surface image, and obtaining coordinate information of each surface hole.

[0021] In one embodiment of the present invention, step S3 specifically includes the following steps: S31, binocular vision camera calibration. The intrinsic and extrinsic parameters of the binocular vision camera are obtained through the checkerboard camera calibration method, so that the coordinate system conversion problem of the image, pixel, camera and world can be handled.

[0022] S32, distortion correction. By using the OpenCV (cross-platform computer vision library) toolkit, the radial distortion coefficient and tangential distortion coefficient of the binocular vision camera can be obtained, so as to correct the distortion of the image, solve the nonlinear distortion phenomenon generated by the binocular vision camera during the imaging process, improve the image geometric accuracy and visual quality, and make subsequent image processing more accurate and reliable.

[0023] S23, extracting the characteristic points of the target workpiece surface hole by a method based on centroid fitting. The centroid refers to the mass center of an object, which is the average position of the accumulated mass inside the object and is unique. The present invention accurately describes the key characteristic points of the target workpiece surface hole according to a characteristic point extraction method based on centroid fitting, and then accurately segments the contour of the target workpiece surface hole through morphological processing and connected domain analysis, and finally obtains the centroid of the target workpiece surface hole according to the Hough circle detection method.

[0024] S24, arranging the extracted centroids. The present invention adopts a pixel coordinate position sorting method, and sorts the centroids of the holes on the surface of the target workpiece in ascending order of the horizontal coordinates. If the pixels are the same, the centroids are sorted in ascending order of the vertical coordinates. The final points will be arranged from left to right and from top to bottom, and the centroids of the holes on the surface of the target workpiece after sorting are marked as stereo matching primitives.

[0025] S25, stereo matching of binocular vision. In order to transform the actual binocular vision platform into an ideal parallel structure, it is necessary to perform stereo correction on it so that the images captured by the left and right cameras are corrected to a parallel state and the row coordinates of the left and right images are aligned. The geometric description between the left and right images is established through epipolar geometry, and the search for matching pixels is limited to a certain horizontal line to achieve row alignment of the left and right images, and the binocular images are corrected to the images collected by the standard parallel binocular camera. Finally, the binocular images are stereo corrected through the Bouguet algorithm, and the rotation matrix R and translation matrix T of the right camera relative to the left camera are used to minimize the number of reprojections of each left and right image, and the binocular camera posture is corrected. Finally, the three-dimensional coordinates of the center of mass of the hole on the surface of the target workpiece are obtained.

[0026] S4, planning a detection path of the robot arm according to the coordinate information of each surface hole of the target workpiece to generate a planned detection path.

[0027] In one embodiment of the present invention, step S4 specifically includes the following steps: S41, based on the crown porcupine algorithm, each initial detection path is regarded as a porcupine individual to initialize the porcupine group; wherein each porcupine individual represents a traversal order of the centroid of the surface holes, and the initial detection path is generated by random arrangement according to the coordinate information of each surface hole of the target workpiece.

[0028] S42, calculating the total travel distance of each initial detection path to obtain the fitness value corresponding to each porcupine individual.

[0029] Among them, the total travel distance can be used to represent the fitness value corresponding to the corresponding porcupine individual. The shorter the total travel distance, the higher the corresponding fitness value.

[0030] S43, based on the defense mechanism, path optimization is performed according to the fitness value corresponding to each porcupine individual to obtain a planned detection path.

[0031] Specifically, the four defense mechanisms of the crested porcupine algorithm may include visual defense (wide-area exploration), sound defense (diversity preservation), odor defense (local development), and physical attack defense (deep optimization). The four defense mechanisms of the crested porcupine algorithm can be specifically applied to path optimization in the following ways: Visual defense: Porcupine individuals will explore in a large area and randomly or semi-randomly adjust the path order to cover a larger solution space. In the path planning problem of the present invention, visual defense can be achieved by randomly exchanging nodes in the path. For example, the positions of two centroids in the path are randomly exchanged to generate a new path to increase the coverage of the solution space.

[0032] The sound defense mechanism encourages individual paths to maintain differences from each other, thereby preventing multiple paths from falling into the same local optimal solution. In the path planning problem of the present invention, the diversity of paths can be ensured by calculating the similarity between paths and "perturbing" or "reorganizing" paths that are too similar. For example, for similar paths, a portion of nodes can be randomly rearranged to avoid excessive concentration of groups.

[0033] Odor defense In a local area, porcupine individuals with higher path quality will mark their own positions and conduct more detailed development. In the path planning problem of the present invention, when a shorter path is found, local adjustments (such as exchange and reversal operations) can be made near it to further optimize the current path. For example, two node positions in the path are selected for exchange, or a section of the path is reversed, and a detailed search is performed near the current high-quality solution.

[0034] Physical attack defense will be further developed in the optimal path area to ensure that the best path is not missed. This can be achieved through the "partial path fixation" strategy, that is, retaining part of the structure in the high-quality path and adjusting the other parts. This prevents the path from changing drastically while fine-tuning the quality of the local solution.

[0035] In each iteration of path optimization, the path is adjusted according to the defense strategy, the fitness value of the path individual is updated, the optimal path is retained and continuously optimized. At the same time, the iteration stop condition is set, for example, reaching a specific number of times, the total path distance change is less than a threshold, etc. Usually, the iteration stops after the path converges or a better solution is reached.

[0036] At the end of the algorithm, the porcupine individual with the best fitness is the optimal path, which represents the order in which the robot arm traverses the centroids of all surface holes in the shortest distance. The robot arm can traverse point by point in this order to achieve time optimization of the shortest path.

[0037] Therefore, the present invention adopts the crown porcupine algorithm to solve the path optimization problem of the robot arm, so that the path of the robot arm traversing the centroid of each target workpiece surface hole is the shortest, thereby improving the efficiency of automation.

[0038] S5, controlling the robot arm to drive the coupling agent automatic coating device to coat each surface hole of the target workpiece with coupling agent.

[0039] Specifically, the robot arm aims the coupling agent nozzle of the automatic coupling agent application device at the center of the surface hole, takes the longest radius of the surface hole as the initial spraying radius of the coupling agent nozzle, and after spraying the first circle around the hole, increases the radius by half on the basis of the initial spraying radius for the second spraying. Applying the coupling agent evenly and appropriately can effectively eliminate the air layer between the probe and the workpiece surface, ensuring the stable transmission of the ultrasonic signal.

[0040] S6, according to the planned detection path, the robotic arm is controlled to drive the ultrasonic phased array probe to move to the centroid position of each surface hole of the target workpiece in sequence, and the ultrasonic phased array probe is controlled to scan the first area corresponding to each surface hole to generate a corresponding ultrasonic image.

[0041] Specifically, after applying the coupling agent, the robotic arm is controlled to drive the ultrasonic phased array probe to move to the centroid position of the first surface hole according to the planned detection path, and the ultrasonic phased array probe is aimed at the first area corresponding to the surface hole, so that the scanning beam of the ultrasonic phased array probe covers the potential damage area around the surface hole. In order to ensure full coverage of the detection area, the ultrasonic phased array probe can be used to perform a small-range circular path scan around the surface hole to ensure that every angle can be covered by the sound wave. The ultrasonic phased array probe receives the echo signal reflected from the damage around the surface hole, combines the reflection information received by the beam at different angles, and uses C scanning to generate a plan view of the damaged area (i.e., an ultrasonic image) to display the distribution of damage around the surface hole. When the first surface hole is scanned, the robotic arm continues to detect the next surface hole according to the planned detection path.

[0042] Therefore, compared with a single probe, the use of an ultrasonic phased array probe can adjust the beam direction to cover a larger detection area, thereby reducing the probe movement frequency.

[0043] S7, performing flaw detection on each surface hole of the target workpiece according to the ultrasonic image.

[0044] In one embodiment of the present invention, step S7 specifically includes the following steps: S71, obtaining historical ultrasonic images, and labeling each historical ultrasonic image.

[0045] Specifically, sufficient historical ultrasonic images can be extracted from the database to ensure that all possible defect situations for target detection and classification are included in the images. Then, a labeling tool is used to label the detection target for each historical ultrasonic image.

[0046] S72, preprocessing the labeled images to obtain a target training set.

[0047] Specifically, the annotated images are normalized and scaled, that is, the ultrasound images are scaled to the input size suitable for the YOLO-v8 network, while maintaining the original aspect ratio to avoid distortion of the target shape caused by scaling. Data expansion is performed by rotation, flipping, blurring, and noise addition to increase the robustness of the model.

[0048] S73, configure a YOLO-v8 network, and train the YOLO-v8 network according to the target training set to generate a YOLO-v8 model.

[0049] Specifically, the configuration file of the YOLO-v8 network is adjusted according to the actual needs of ultrasonic detection. The configuration file of the YOLO-v8 network may include parameters such as model architecture, input size, and number of categories. The number of categories of the YOLO-v8 network is set to the actual number of categories in the ultrasonic image detection task so that the model can correctly classify the target.

[0050] Second, set the training parameters, such as learning rate, batch size, and number of training rounds, to ensure that the model reaches optimal results in a reasonable amount of time.

[0051] Then, the model is tested on the validation set, and indicators such as mAP (Mean Average Precision) and F1 Score (F1 score, the harmonic mean of precision and recall) are calculated to measure the accuracy and effect of the model. Each defect category in the ultrasound image is evaluated to check whether there are problems such as category imbalance and insufficient detection.

[0052] S74, performing flaw detection on each surface hole of the target workpiece according to the ultrasonic image based on the YOLO-v8 model.

[0053] In summary, according to the workpiece surface hole flaw detection method based on the mechanical arm flaw detection device of the embodiment of the present invention, when the target workpiece is placed on the workpiece conveyor belt, the workpiece conveyor belt is controlled to transport the target workpiece to the position directly below the field of view of the binocular vision camera, and the binocular vision camera is controlled to shoot the target workpiece to obtain the workpiece surface image of the target workpiece, and each surface hole of the target workpiece is calibrated according to the workpiece surface image, and the coordinate information of each surface hole is obtained, and the detection path of the mechanical arm is planned according to the coordinate information of each surface hole of the target workpiece to generate a planned detection path, and the mechanical arm is controlled to drive the coupling agent automatic application device to apply coupling agent to each surface hole of the target workpiece, and the mechanical arm is controlled to drive the ultrasonic phased array probe to move to the centroid position of each surface hole of the target workpiece in sequence according to the planned detection path, and the ultrasonic phased array probe is controlled to scan the first area corresponding to each surface hole to generate a corresponding ultrasonic image, and the surface holes of the target workpiece are flawed according to the ultrasonic image. Thus, the surface holes of the target workpiece can be accurately flawed and the efficiency is high.

[0054] Corresponding to the workpiece surface hole flaw detection method based on a robotic arm flaw detection device in the above embodiment, the present invention also proposes a workpiece surface hole flaw detection system based on a robotic arm flaw detection device.

[0055] Among them, Figure 2 As shown, the robotic arm flaw detection equipment 1 includes a fixed bracket 10, a binocular vision camera 20, a robotic arm 30, an ultrasonic phased array probe 40, an automatic coupling agent application device 50 and a workpiece conveyor belt 60, wherein the binocular vision camera 20 is fixed on the fixed bracket 10, and the binocular vision camera 20 is arranged above the workpiece conveyor belt 60, the fixed bracket 10 and the robotic arm 30 are respectively arranged on both sides of the workpiece conveyor belt 60, and the ultrasonic phased array probe 40 and the automatic coupling agent application device 50 are respectively fixed at the end of the robotic arm 30.

[0056] like Figure 3 As shown, the workpiece surface hole flaw detection system based on the robotic arm flaw detection equipment of an embodiment of the present invention may include: a first control module 100, a second control module 200, an acquisition module 300, a generation module 400, a third control module 500, a fourth control module 600 and a detection module 700.

[0057] Among them, the first control module 100 is used to control the workpiece conveyor belt to transport the target workpiece to the bottom of the field of view of the binocular vision camera when the target workpiece is placed on the workpiece conveyor belt; the second control module 200 is used to control the binocular vision camera to shoot the target workpiece to obtain the workpiece surface image of the target workpiece; the acquisition module 300 is used to calibrate each surface hole of the target workpiece according to the workpiece surface image, and obtain the coordinate information of each surface hole; the generation module 400 is used to plan the detection path of the robot arm according to the coordinate information of each surface hole of the target workpiece to generate a planned detection path; the third control module 500 is used to control the robot arm to drive the automatic coupling agent application device to apply coupling agent to each surface hole of the target workpiece; the fourth control module 600 is used to control the robot arm to drive the ultrasonic phased array probe to move to the centroid position of each surface hole of the target workpiece in turn according to the planned detection path, and control the ultrasonic phased array probe to scan the first area corresponding to each surface hole to generate a corresponding ultrasonic image; the detection module 700 is used to perform flaw detection on each surface hole of the target workpiece according to the ultrasonic image.

[0058] In one embodiment of the present invention, the generation module 400 is specifically used for: considering each initial detection path as a porcupine individual based on the crown porcupine algorithm to initialize the porcupine group; wherein each porcupine individual represents a traversal order of the centroid of a surface hole, and the initial detection path is generated by random arrangement according to the coordinate information of each surface hole of the target workpiece; calculating the total travel distance of each initial detection path to obtain the fitness value corresponding to each porcupine individual; optimizing the path according to the fitness value corresponding to each porcupine individual based on the defense mechanism to obtain the planned detection path. In one embodiment of the present invention, the detection module 700 is specifically used for: obtaining historical ultrasonic images and annotating each historical ultrasonic image; preprocessing the annotated images to obtain a target training set; configuring a YOLO-v8 network and training the YOLO-v8 network according to the target training set to generate a YOLO-v8 model; performing flaw detection on each surface hole of the target workpiece according to the ultrasonic image based on the YOLO-v8 model.

[0059] It should be noted that for the details not disclosed in the workpiece surface hole flaw detection system based on the robotic arm flaw detection equipment in the embodiment of the present invention, please refer to the details disclosed in the above-mentioned workpiece surface hole flaw detection method based on the robotic arm flaw detection equipment, and the details will not be described in detail here.

[0060] According to the workpiece surface hole flaw detection system based on the mechanical arm flaw detection device of the embodiment of the present invention, when the target workpiece is placed on the workpiece conveyor belt, the first control module controls the workpiece conveyor belt to convey the target workpiece to the right below the field of view of the binocular vision camera, and the second control module controls the binocular vision camera to shoot the target workpiece to obtain the workpiece surface image of the target workpiece, and the acquisition module calibrates each surface hole of the target workpiece according to the workpiece surface image, and obtains the coordinate information of each surface hole, and the generation module plans the detection path of the mechanical arm according to the coordinate information of each surface hole of the target workpiece to generate the planned detection path, and the third control module controls the mechanical arm to drive the coupling agent automatic application device to apply the coupling agent to each surface hole of the target workpiece, and the fourth control module controls the mechanical arm to drive the ultrasonic phased array probe to move to the centroid position of each surface hole of the target workpiece in turn according to the planned detection path, and controls the ultrasonic phased array probe to scan the first area corresponding to each surface hole to generate a corresponding ultrasonic image, and the detection module performs flaw detection on each surface hole of the target workpiece according to the ultrasonic image. Thus, the surface holes of the target workpiece can be accurately detected and the efficiency is high.

[0061] Corresponding to the above embodiment, the present invention also provides a computer device.

[0062] The computer device of the embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the workpiece surface hole flaw detection method based on the robotic arm flaw detection device of the above embodiment is implemented.

[0063] The computer device according to the embodiment of the present invention can accurately perform flaw detection on the surface holes of the target workpiece with high efficiency.

[0064] Corresponding to the above embodiment, the present invention also proposes a non-transitory computer-readable storage medium.

[0065] The non-transitory computer-readable storage medium of the embodiment of the present invention stores a computer program, and when the program is executed by a processor, the above-mentioned workpiece surface hole flaw detection method based on the robot arm flaw detection equipment is implemented.

[0066] According to the non-transitory computer-readable storage medium of the embodiment of the present invention, it is possible to accurately perform flaw detection on the surface holes of the target workpiece with high efficiency.

[0067] In the description of the present invention, 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 one or more of the features. "Multiple" means two or more, unless otherwise clearly and specifically defined.

[0068] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0069] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0070] 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 invention. 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, unless they are contradictory.

[0071] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0072] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A method for detecting flaws on a workpiece surface hole based on a mechanical arm flaw detection device, characterized in that: The robotic arm flaw detection equipment comprises a fixed bracket, a binocular vision camera, a robotic arm, an ultrasonic phased array probe, an automatic coupling agent application device and a workpiece conveyor belt, wherein the binocular vision camera is fixed on the fixed bracket, and the binocular vision camera is arranged above the workpiece conveyor belt, the fixed bracket and the robotic arm are respectively arranged on both sides of the workpiece conveyor belt, the ultrasonic phased array probe and the automatic coupling agent application device are respectively fixed at the end of the robotic arm, wherein the workpiece surface hole flaw detection detection method comprises the following steps: S1, when a target workpiece is placed on the workpiece conveyor belt, controlling the workpiece conveyor belt to convey the target workpiece to a position directly below the field of view of the binocular vision camera; S2, controlling the binocular vision camera to shoot the target workpiece to obtain a workpiece surface image of the target workpiece; S3, calibrating each surface hole of the target workpiece according to the workpiece surface image, and obtaining coordinate information of each surface hole; S4, planning a detection path of the robot arm according to the coordinate information of each surface hole of the target workpiece to generate a planned detection path; S5, controlling the robot arm to drive the coupling agent automatic application device to apply coupling agent to each surface hole of the target workpiece; S6, controlling the robot arm to drive the ultrasonic phased array probe to move to the centroid position of each surface hole of the target workpiece in sequence according to the planned detection path, and controlling the ultrasonic phased array probe to scan the first area corresponding to each surface hole to generate a corresponding ultrasonic image; S7, performing flaw detection on each surface hole of the target workpiece according to the ultrasonic image.

2. The workpiece surface hole flaw detection method based on a mechanical arm flaw detection device according to claim 1 is characterized in that: Step S4 specifically includes the following steps: S41, based on the crown porcupine algorithm, each initial detection path is regarded as a porcupine individual to initialize the porcupine group; wherein each porcupine individual represents a traversal order of the centroid of the surface holes, and the initial detection path is generated by random arrangement according to the coordinate information of each surface hole of the target workpiece; S42, calculating the total travel distance of each of the initial detection paths to obtain the fitness value corresponding to each porcupine individual; S43, performing path optimization according to the fitness value corresponding to each porcupine individual based on the defense mechanism to obtain the planned detection path.

3. The workpiece surface hole flaw detection method based on a mechanical arm flaw detection device according to claim 1 is characterized in that: Step S7 specifically includes: S71, acquiring historical ultrasonic images, and marking each of the historical ultrasonic images; S72, preprocessing the labeled images to obtain a target training set; S73, configuring a YOLO-v8 network, and training the YOLO-v8 network according to the target training set to generate a YOLO-v8 model; S74, performing flaw detection on each surface hole of the target workpiece according to the ultrasonic image based on the YOLO-v8 model.

4. A workpiece surface hole flaw detection system based on a mechanical arm flaw detection device, characterized in that: The mechanical arm flaw detection equipment includes a fixed bracket, a binocular vision camera, a mechanical arm, an ultrasonic phased array probe, an automatic coupling agent coating device and a workpiece conveyor belt, wherein the binocular vision camera is fixed on the fixed bracket, and the binocular vision camera is arranged above the workpiece conveyor belt, the fixed bracket and the mechanical arm are respectively arranged on both sides of the workpiece conveyor belt, the ultrasonic phased array probe and the automatic coupling agent coating device are respectively fixed at the end of the mechanical arm, wherein the workpiece surface hole flaw detection detection system includes: A first control module, wherein the first control module is used to control the workpiece conveyor belt to convey the target workpiece to a position directly below the field of view of the binocular vision camera when the target workpiece is placed on the workpiece conveyor belt; A second control module, the second control module is used to control the binocular vision camera to shoot the target workpiece to obtain a workpiece surface image of the target workpiece; An acquisition module, the acquisition module is used to calibrate each surface hole of the target workpiece according to the workpiece surface image, and acquire coordinate information of each surface hole; A generation module, the generation module is used to plan the detection path of the robot arm according to the coordinate information of each surface hole of the target workpiece, so as to generate a planned detection path; A third control module, the third control module is used to control the mechanical arm to drive the coupling agent automatic application device to apply coupling agent to each surface hole of the target workpiece; a fourth control module, the fourth control module being used to control the robot arm to drive the ultrasonic phased array probe to move to the centroid position of each surface hole of the target workpiece in sequence according to the planned detection path, and control the ultrasonic phased array probe to scan the first area corresponding to each surface hole to generate a corresponding ultrasonic image; A detection module is used to perform flaw detection on each surface hole of the target workpiece according to the ultrasonic image.

5. The workpiece surface hole flaw detection system based on the mechanical arm flaw detection equipment according to claim 4 is characterized in that: The generation module is specifically used for: Based on the crown porcupine algorithm, each initial detection path is regarded as a porcupine individual to initialize the porcupine group; wherein each porcupine individual represents a traversal order of the centroid of a surface hole, and the initial detection path is generated by random arrangement according to the coordinate information of each surface hole of the target workpiece; Calculate the total travel distance of each of the initial detection paths to obtain the fitness value corresponding to each porcupine individual; Based on the defense mechanism, path optimization is performed according to the fitness value corresponding to each porcupine individual to obtain the planned detection path.

6. The workpiece surface hole flaw detection system based on the mechanical arm flaw detection equipment according to claim 4 is characterized in that: The detection module is specifically used for: Acquire historical ultrasonic images, and annotate each of the historical ultrasonic images; Preprocess the annotated images to obtain the target training set; Configure a YOLO-v8 network, and train the YOLO-v8 network according to the target training set to generate a YOLO-v8 model; Based on the YOLO-v8 model, flaw detection is performed on each surface hole of the target workpiece according to the ultrasonic image.

7. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the workpiece surface hole flaw detection method based on a robotic arm flaw detection device according to any one of claims 1 to 3 is implemented.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, a workpiece surface hole flaw detection method based on a robotic arm flaw detection device according to any one of claims 1 to 3 is implemented.

Images