Method and device for determining position of ending lineation of single crystal rod and electronic equipment

Abstract

The invention provides a method and a device for determining the position of a single crystal rod ending lineation and electronic equipment. The method comprises the following steps: acquiring a lineation area image corresponding to the single crystal rod ending lineation; inputting the lineation area image into a pre-trained lineation identification model, extracting lineation local features, and determining a lineation pose corresponding to the ending lineation of the single crystal rod according to the lineation local features; determining a deviation value between the scribing pose and a preset target pose, and correcting the pose of the single crystal rod according to the deviation value until the scribing pose is matched with the preset target pose; and displaying an identification result corresponding to the ending lineation and the lineation pose to a user through a preset interactive interface. And high-precision detection of the scribing position of the single crystal rod can be realized.

Description

Technical Field

[0001] The present disclosure relates to the field of industrial automation technology, and in particular to a method, device and electronic equipment for determining the position of a finishing line of a single crystal rod. Background Art

[0002] In the semiconductor and photovoltaic industries, the quality and size of single crystal rods directly affect the performance and production efficiency of downstream products. Therefore, it is particularly important to perform high-precision inspection and marking of single crystal rods during their processing. Traditional single crystal rod marking position detection usually relies on manual operation or simple mechanical detection methods. These methods are not only inefficient, but the detection accuracy is also limited by the operator's experience and subjective judgment. In addition, changes in ambient lighting conditions will also have a significant impact on the detection results, resulting in deviations in the marking position, which in turn affects product quality. Therefore, how to develop an intelligent visual inspection solution that can perform high-precision detection of the marking position of single crystal rods has become an urgent problem to be solved in the industry. Summary of the invention

[0003] The embodiments of the present disclosure at least provide a method, device and electronic equipment for determining the position of the end marking line of a single crystal rod, which can realize high-precision detection of the marking position of the single crystal rod.

[0004] The present disclosure provides a method for determining the position of a finishing line of a single crystal rod, comprising:

[0005] Collect the image of the marking area corresponding to the final marking at the end of the single crystal rod;

[0006] Inputting the scribing area image into a pre-trained scribing recognition model, extracting local features of the scribing, and determining the scribing posture corresponding to the finishing scribing of the single crystal rod according to the local features of the scribing;

[0007] Determining a deviation value between the scribing posture and a preset target posture, and correcting the posture of the single crystal rod according to the deviation value until the scribing posture matches the preset target posture;

[0008] The recognition result corresponding to the ending stroke and the stroke posture are displayed to the user through a preset interactive interface.

[0009] In an optional implementation manner, collecting a line area image corresponding to the end line of the single crystal rod, specifically includes:

[0010] Obtain the diameter information corresponding to the single crystal rod;

[0011] Determine the target focal length corresponding to capturing the image of the lined area according to the diameter information;

[0012] Collecting the shooting distance between the current camera lens and the single crystal rod, and determining the displacement adjustment amount corresponding to the camera lens according to the target focal length and the shooting distance;

[0013] The camera lens is driven to move along the optical axis according to the displacement adjustment amount until the target focal length is reached.

[0014] In an optional implementation, the line recognition model is trained based on the following steps:

[0015] Constructing the line recognition model with a convolutional neural network as the base layer;

[0016] Acquire a sample single crystal rod image having a plurality of material types, material colors, and surface textures, wherein the single crystal rod in the sample single crystal rod image is provided with a plurality of end markings of marking colors and marking sizes;

[0017] The sample single crystal rod image is used as sample data, and the corresponding ending stroke in each sample single crystal rod image is used as a data label, which is input into the stroke recognition model for model training until the stroke recognition model has the ability to accurately recognize the stroke posture.

[0018] In an optional implementation manner, determining the deviation value between the marking posture and the preset target posture specifically includes:

[0019] The position change of the single crystal rod is monitored in real time through the preset displacement monitoring sensor;

[0020] Update the stroke pose according to the pose change;

[0021] The deviation value between the updated line drawing posture and the preset target posture is calculated.

[0022] In an optional implementation manner, correcting the posture of the single crystal rod according to the deviation value until the scribing posture matches the preset target posture specifically includes:

[0023] Determining whether the deviation value is greater than a preset tolerance threshold;

[0024] When the deviation value is greater than the preset tolerance threshold, driving a preset mechanical correction component to adjust the posture corresponding to the single crystal rod according to the deviation value;

[0025] In the process of adjusting the corresponding posture of the single crystal rod, a closed-loop control strategy is adopted to provide real-time feedback on the current posture of the single crystal rod until the scribing posture corresponding to the adjusted single crystal rod matches the preset target posture.

[0026] In an optional implementation manner, the preset interactive interface is provided with a status overview area, a real-time monitoring area, a task control area, and a log recording area;

[0027] The status overview area is used to display the equipment operation status information corresponding to the single crystal rod production line;

[0028] The real-time monitoring area is used to display the recognition result corresponding to the ending stroke;

[0029] The task control area is used to respond to production process control operations and production parameter adjustment operations implemented by users;

[0030] The log recording area is used to record production log information and abnormal alarm information.

[0031] The present disclosure also provides a device for determining the position of a single crystal rod tail marking line, comprising:

[0032] An image acquisition module is used to acquire an image of a marking area corresponding to the final marking at the end of the single crystal rod;

[0033] A scribing posture detection module is used to input the scribing area image into a pre-trained scribing recognition model, extract the local features of the scribing, and determine the scribing posture corresponding to the finishing scribing of the single crystal rod according to the local features of the scribing;

[0034] A posture adjustment module, used to determine a deviation value between the scribing posture and a preset target posture, and correct the posture of the single crystal rod according to the deviation value until the scribing posture matches the preset target posture;

[0035] The information display module is used to display the recognition result corresponding to the end stroke and the stroke posture to the user through a preset interactive interface.

[0036] In an optional embodiment, the device is also used for:

[0037] Constructing the line recognition model with a convolutional neural network as the base layer;

[0038] Acquire a sample single crystal rod image having a plurality of material types, material colors, and surface textures, wherein the single crystal rod in the sample single crystal rod image is provided with a plurality of end markings of marking colors and marking sizes;

[0039] The sample single crystal rod image is used as sample data, and the corresponding ending stroke in each sample single crystal rod image is used as a data label, which is input into the stroke recognition model for model training until the stroke recognition model has the ability to accurately recognize the stroke posture.

[0040] The present disclosure also provides an electronic device, including: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the memory via the bus, and when the machine-readable instructions are executed by the processor, the method for determining the position of the end mark of the single crystal rod, or steps in any possible implementation of the method for determining the position of the end mark of the single crystal rod, are executed.

[0041] The embodiments of the present disclosure also provide a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for determining the position of the end mark of a single crystal rod or the steps in any possible implementation manner of the method for determining the position of the end mark of a single crystal rod is executed.

[0042] The embodiments of the present disclosure also provide a computer program product, including a computer program / instructions, which, when executed by a processor, implement the above-mentioned method for determining the position of the end mark of a single crystal rod, or the steps in any possible implementation of the above-mentioned method for determining the position of the end mark of a single crystal rod.

[0043] The disclosed embodiments provide a method, device and electronic device for determining the position of the end-marking of a single crystal rod, which collects the image of the marking area corresponding to the end-marking of the single crystal rod; inputs the image of the marking area into a pre-trained marking recognition model, extracts the local features of the marking, and determines the marking posture corresponding to the end-marking of the single crystal rod according to the local features of the marking; determines the deviation value between the marking posture and the preset target posture, and corrects the posture of the single crystal rod according to the deviation value until the marking posture matches the preset target posture; and displays the recognition result corresponding to the end-marking and the marking posture to the user through a preset interactive interface. High-precision detection of the marking position of the single crystal rod can be achieved.

[0044] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following is a brief introduction to the drawings required for use in the embodiments. The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used together with the specification to illustrate the technical solutions of the present disclosure. It should be understood that the following drawings only illustrate certain embodiments of the present disclosure and should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can also be obtained based on these drawings without creative work.

[0046] Figure 1 A flow chart of a method for determining the position of a finishing line of a single crystal rod provided by an embodiment of the present disclosure is shown;

[0047] Figure 2 A flow chart of a method for training a line recognition model provided by an embodiment of the present disclosure is shown;

[0048] Figure 3 A schematic diagram of a device for determining the position of a single crystal rod tail marking line provided by an embodiment of the present disclosure is shown;

[0049] Figure 4 A schematic diagram of an electronic device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical scheme and advantages of the embodiments of the present disclosure clearer, the technical scheme in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The components of the embodiments of the present disclosure generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure provided in the drawings is not intended to limit the scope of the present disclosure for protection, but merely represents the selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present disclosure.

[0051] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0052] The term "and / or" herein only describes an association relationship, indicating that three relationships may exist. For example, A and / or B may represent the following three situations: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C may represent including any one or more elements selected from the set consisting of A, B, and C.

[0053] Research has found that in the semiconductor and photovoltaic industries, the quality and size of single crystal rods directly affect the performance and production efficiency of downstream products. Therefore, in the processing of single crystal rods, it is particularly important to perform high-precision detection and marking. Traditional single crystal rod scribing position detection usually relies on manual operation or simple mechanical detection methods. These methods are not only inefficient, but the detection accuracy is limited by the operator's experience and subjective judgment. In addition, changes in ambient lighting conditions will also have a significant impact on the detection results, resulting in deviations in the scribing position, which in turn affects product quality. Therefore, how to develop an intelligent visual detection solution that can perform high-precision detection of the scribing position of single crystal rods has become an urgent problem to be solved in the industry.

[0054] Based on the above research, the present disclosure provides a method, device and electronic device for determining the position of the end-marking of a single crystal rod, by collecting the image of the marking area corresponding to the end-marking of the single crystal rod; inputting the image of the marking area into a pre-trained marking recognition model, extracting the local features of the marking, and determining the marking posture corresponding to the end-marking of the single crystal rod according to the local features of the marking; determining the deviation value between the marking posture and the preset target posture, correcting the posture of the single crystal rod according to the deviation value until the marking posture matches the preset target posture; and displaying the recognition result corresponding to the end-marking and the marking posture to the user through a preset interactive interface. High-precision detection of the marking position of the single crystal rod can be achieved.

[0055] To facilitate understanding of this embodiment, a method for determining the position of a single crystal rod tail marking line disclosed in the embodiment of the present disclosure is first introduced in detail. The execution subject of the method for determining the position of a single crystal rod tail marking line provided in the embodiment of the present disclosure is generally a computer device with certain computing capabilities, and the computer device includes, for example: a terminal device or a server or other processing device, and the terminal device can be a user equipment (User Equipment, UE), a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (Personal Digital Assistant, PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. In some possible implementations, the method for determining the position of a single crystal rod tail marking line can be implemented by a processor calling a computer-readable instruction stored in a memory.

[0056] See also Figure 1 FIG. 1 is a flow chart of a method for determining the position of a single crystal rod tail marking line provided by an embodiment of the present disclosure, the method comprising steps S101 to S104, wherein:

[0057] S101, collecting a marking area image corresponding to the final marking at the end of the single crystal rod.

[0058] In a specific implementation, an adaptive optics and focusing system may be used to capture an image of a marking area corresponding to the final marking at the end of a single crystal rod, so as to quickly and accurately obtain a clear marking image on single crystal rods of different diameters and materials.

[0059] Here, the adaptive optics and focusing system can capture images by combining a high dynamic range (HDR) camera with a ring-shaped LED light source, and can capture clear line images even under complex conditions such as reflection on the surface of the single crystal rod and shadow obstruction.

[0060] Specifically, the diameter information corresponding to the single crystal rod is obtained; the target focal length corresponding to the image of the marked area is determined according to the diameter information; the shooting distance between the current camera lens and the single crystal rod is collected, and the displacement adjustment amount corresponding to the camera lens is determined according to the target focal length and the shooting distance; the camera lens is driven to move along the optical axis according to the displacement adjustment amount until the target focal length is reached.

[0061] Here, the diameter of the single crystal rod determines the size of the scribing area and the choice of the shooting focal length. The diameter information of the single crystal rod is obtained through an integrated measuring device or sensor, or the information is extracted from the single crystal rod specification data known during the production process. In order to ensure that the camera focuses accurately and obtains a clear image of the scribing area, the optimal shooting focal length is calculated based on the diameter information of the single crystal rod using a preset optical model or lookup table (LUT) to ensure clear imaging of the scribing area.

[0062] Furthermore, a laser rangefinder, ultrasonic sensor or other distance measuring device is used to measure the distance between the camera lens and the single crystal rod in real time. The measured actual shooting distance is compared with the target focal length, and the distance the camera needs to move is calculated to align with the best shooting position.

[0063] Here, the camera lens is driven to move along the optical axis by an electric slide or precision motor until the camera reaches the target focal length. During the movement, the camera provides real-time feedback of its displacement status to ensure accurate focusing, so that the lens can accurately focus on the marked area.

[0064] During the shooting process, the lighting conditions (such as the brightness and angle of the LED light source) are adjusted to capture the image of the marked area at the end of the single crystal rod to ensure that the image has sufficient contrast and clarity.

[0065] S102, inputting the scribing area image into a pre-trained scribing recognition model, extracting local features of the scribing, and determining the scribing posture corresponding to the finishing scribing of the single crystal rod according to the local features of the scribing.

[0066] In the specific implementation, the collected image of the marked area is first preprocessed to ensure that the quality of the input image meets the model requirements. The process may include converting the color image into a grayscale image to reduce the computational complexity, using filtering techniques (such as Gaussian filtering, median filtering) to remove noise in the image, and normalizing the image pixel values ​​to a fixed range (such as 0 to 1) so that the model converges faster.

[0067] Here, the image data is converted into a tensor format that the model can process, and then input into a pre-trained line recognition model to extract key features related to the line from the input image, such as the edge, texture, width, color, etc. of the line.

[0068] Among them, multiple convolution kernels (filters) are used to perform convolution operations on the image to extract features of different scales and directions. The nonlinear activation function such as ReLU is used to enhance the feature expression ability. The maximum pooling or average pooling operation is used to reduce the dimension of the feature map, retain the main features, and reduce the amount of calculation. The batch normalization layer is used to improve the model training speed and stability. The extracted features are integrated into a multi-dimensional feature map, including the local features of the line.

[0069] Optionally, further processing is performed to optimize the extracted features so that they more accurately reflect the position and shape of the line. The attention mechanism is used to strengthen the line feature area. The high-dimensional features are compressed into low-dimensional representations through fully connected layers or global pooling layers to facilitate pose calculation.

[0070] Furthermore, the model outputs the positions of key points of the marking, such as the starting point, end point or intermediate feature point of the marking. According to the spatial coordinates of the key points, the geometric method or regression model is used to estimate the position and orientation of the marking line, including the position (X, Y, Z coordinates) and direction (angle). Combined with the geometric information of the single crystal rod, the marking position and orientation are further corrected to ensure its accuracy.

[0071] As a possible implementation, see Figure 2 FIG. 2 is a flowchart of a method for training a line recognition model provided by an embodiment of the present disclosure, wherein the method includes steps S201 to S203, wherein:

[0072] S201, constructing the line recognition model with a convolutional neural network as the base layer.

[0073] S202, acquiring a sample single crystal rod image having a plurality of material types, material colors, and surface textures, wherein the single crystal rod in the sample single crystal rod image is provided with a plurality of end markings of marking colors and marking sizes.

[0074] S203, using the sample single crystal rod image as sample data and the corresponding end stroke in each sample single crystal rod image as a data label, inputting them into the stroke recognition model for model training until the stroke recognition model has the ability to accurately recognize the stroke posture.

[0075] In the specific implementation, we first build a line recognition model that is highly sensitive to small features in a complex background. The network architecture of the line recognition model includes a convolutional neural network (CNN) base layer, residual connections, and an attention mechanism. Standardize and normalize the original image to eliminate the deviation caused by uneven illumination. Use GPU to accelerate the training process, save model checkpoints regularly, monitor changes in the loss function, and adjust the learning rate in a timely manner. Verify the model performance on an independent test set, analyze the type of recognition errors, and feedback to adjust the model architecture or parameters. Integrate the trained model into the visual system, process the single crystal rod images on the production line in real time, and output the coordinates of the line position.

[0076] Here, the convolutional neural network extracts local features from the image, such as edges, textures, etc., through convolution operations. Each convolutional layer contains multiple filters, each of which focuses on identifying specific image features. The size of the feature map is reduced through maximum pooling or average pooling operations, reducing the computational complexity while retaining key information. Nonlinear activation functions such as ReLU (Rectified Linear Unit) are used to increase the nonlinear expression ability of the model. The features extracted by the convolutional layer and the pooling layer are mapped to the output layer for the final classification or regression task. Depending on the specific task, such as the pose estimation of the line drawing, the output layer can be a multidimensional regression output representing the position and angle.

[0077] The input layer receives the image of the line area, the multi-layer convolution and pooling layers extract the local features of the line in the image, and the fully connected layer converts the extracted features into the pose information of the line. Introducing residual blocks in deeper network layers can alleviate the gradient vanishing problem, accelerate the training process and improve the recognition accuracy of the model. By introducing channel attention and spatial attention mechanisms, the network can focus on the line area and ignore background noise and other non-critical information.

[0078] Furthermore, high-quality images covering a variety of single crystal rod materials, different colors and surface states, and a variety of line colors and widths are collected as diverse sample single crystal rod images. A variety of material types, colors, and textures can enhance the generalization ability of the model, enabling it to adapt to different types of single crystal rods.

[0079] Here, single crystal rods made of different materials, such as silicon, germanium, etc., are collected. The colors include single crystal rods from light to dark colors, ensuring that the model has a good recognition effect on lines of various colors. The surface texture includes smooth, rough, striped and other surface textures to increase the adaptability of the model to different surface features.

[0080] Among them, a variety of marking colors and sizes are set for each type of single crystal rod, including red, blue, green and other colors to ensure that the model can accurately identify markings of different colors. The marking sizes range from thin lines to thick lines to ensure that the model has good recognition capabilities for various marking widths.

[0081] Here, in the sample data preparation stage, a high-resolution camera is used to capture images of scribing on various single crystal rods. The images are standardized, including cropping, grayscale, denoising, etc. The data labels are scribing pose labels, including the starting point, end point, direction, angle, and position information of the scribing segment. The labels can be obtained through manual labeling or semi-automatic labeling tools.

[0082] Optionally, you can use image augmentation techniques (rotation, scaling, brightness adjustment, etc.) to expand the dataset and enhance the generalization ability of the model.

[0083] Furthermore, the preprocessed sample images and the corresponding line position labels are input into the line recognition model. A weighted cross entropy loss function is used to assign different penalty weights to different types of errors to ensure that the model accurately predicts the line position. At the same time, the Adam optimizer is used in combination with the learning rate decay strategy to balance the convergence speed and the generalization performance of the model.

[0084] Here, during the training process, the image is forward propagated through the layers of the line recognition model to generate the predicted output. The gradient of the loss function is calculated using the backpropagation algorithm. The network weights are updated using an optimization algorithm (such as Adam or SGD) so that the model gradually converges to the optimal solution.

[0085] Here, during the training process, the validation set is used to monitor the model performance to avoid overfitting. Performance indicators can include average error, recognition accuracy, etc. The model performance is optimized by adjusting hyperparameters such as learning rate, number of convolution kernels, number of layers, etc.

[0086] Furthermore, in the process of model verification and tuning, the cross-validation method is used to evaluate the model performance, and the recognition effect is measured by indicators such as precision, recall rate, F1 score, etc. The hyperparameters such as the number of network layers, number of nodes, activation function, etc. are adjusted and fine-tuned to achieve the best recognition accuracy.

[0087] After that, the non-maximum suppression (NMS) technique is applied to filter the overlapping prediction boxes to ensure that each stroke corresponds to only one prediction result. The threshold strategy is set to filter out low-confidence predictions and improve the accuracy and reliability of the final recognition.

[0088] S103, determining a deviation value between the marking posture and a preset target posture, and correcting the posture of the single crystal rod according to the deviation value until the marking posture matches the preset target posture.

[0089] In the specific implementation, in order to ensure that the single crystal rod finishing marking positioning system can work continuously and accurately during the production process, the position deviation of the single crystal rod is monitored and corrected in real time by integrating high-precision sensors and advanced control algorithms. The preset displacement monitoring sensor monitors the corresponding posture change of the single crystal rod in real time; updates the marking posture according to the posture change; and calculates the deviation value between the updated marking posture and the preset target posture.

[0090] Here, highly sensitive position sensors (such as photoelectric encoders, laser rangefinders, etc.) are deployed to monitor the position and posture changes of the single crystal rod in real time during the transmission process. Inertial sensors such as accelerometers and gyroscopes are installed to detect the vibration and offset of the equipment itself. The sensor data is continuously collected and transmitted to the central processing unit (CPU or embedded controller). The processing unit uses filtering algorithms (such as Kalman filtering) to process the raw data, remove noise, and extract effective information, such as the deviation between the actual position of the single crystal rod and the ideal position.

[0091] After that, the deviation calculated based on the sensor data is judged whether it exceeds the preset tolerance range. If there is a deviation, the necessary correction actions (such as lateral / longitudinal displacement of the mobile platform, rotation angle adjustment, etc.) and the corresponding servo motor drive instructions are calculated.

[0092] After receiving the command, the servo control system quickly drives the mechanical components (such as precision guide rails and rotating tables) to perform precise displacement or rotation adjustments according to the calculated parameters. During the execution of the action, the closed-loop control strategy continuously feeds back the current position information to ensure the accurate execution of the action until the deviation is corrected to an acceptable range.

[0093] Optionally, after the correction action is completed, the system will continue to monitor the position status of the single crystal rod and prepare for the next round of deviation detection and correction. It should be noted that the entire process is completed in a very short time, ensuring the continuity of the production process and the processing accuracy of the product.

[0094] In this way, real-time dynamic correction technology can quickly and accurately adjust the tiny position deviations of the single crystal rod on the production line, thereby ensuring high precision of marking recognition and cutting, and improving the overall manufacturing quality and efficiency.

[0095] As a possible implementation method, it is determined whether the deviation value is greater than a preset tolerance threshold; when the deviation value is greater than the preset tolerance threshold, the preset mechanical correction component is driven to adjust the corresponding posture of the single crystal rod according to the deviation value; in the process of adjusting the corresponding posture of the single crystal rod, a closed-loop control strategy is used to provide real-time feedback on the current posture of the single crystal rod until the marking posture corresponding to the adjusted single crystal rod matches the preset target posture.

[0096] The sensor captures the position and angle changes of the single crystal rod in three-dimensional space in real time. High-precision sensors such as laser displacement sensors, optical encoders or three-dimensional posture sensors are used to ensure the accuracy of posture data. Before starting monitoring, the sensor records the initial posture of the single crystal rod, including the initial position and direction. The sensor continuously collects the posture data of the single crystal rod, capturing changes in position (X, Y, Z) and direction (pitch angle, yaw angle, roll angle).

[0097] Here, the displacement vector is calculated by comparing the currently collected position data with the initial position data. The angle change is calculated by comparing the current direction angle with the initial direction angle. According to the position change provided by the sensor, the line position is updated in real time. The current line position is determined by the initial position plus the displacement vector, and the current line direction is determined by the initial direction plus the angle change.

[0098] Among them, the three-dimensional space distance between the updated line-marking position and the target position is calculated through the difference between the updated line-marking posture and the preset target posture, and the angle difference between the updated line-marking direction and the target direction is calculated.

[0099] It should be noted that at every moment in the entire process, the posture is monitored and updated in real time. The calculated deviation value is used for real-time adjustment to ensure that the single crystal rod gradually approaches the target posture.

[0100] S104: Display the recognition result corresponding to the ending stroke and the stroke posture to the user through a preset interactive interface.

[0101] In the specific implementation, an intuitive operation interface is provided to display key information such as the line recognition status and position coordinates, while allowing the operator to make necessary interventions and parameter adjustments to ensure the flexibility and controllability of the system. The operation interface integrates multiple modules such as hardware control, data analysis, and user interaction, and adopts a distributed architecture design to ensure efficient and stable operation.

[0102] Here, the interactive interface is provided with a status overview area, a real-time monitoring area, a task control area and a log recording area; the status overview area is used to display the equipment operation status information corresponding to the single crystal rod production line; the real-time monitoring area is used to display the recognition result corresponding to the finishing line; the task control area is used to respond to the production process control operations and production parameter adjustment operations implemented by the user; the log recording area is used to record production log information and abnormal alarm information.

[0103] The position, shape, color and other information of the analyzed line are obtained from the line recognition model. The recognition results are converted into a format that is easy for users to understand, such as image overlay, numerical display, annotation, etc. The overall status of the single crystal rod production line is displayed, including equipment operation status, current production stage, single crystal rod processing status, etc. The recognition results, such as the position, size and color of the line, are superimposed on the real-time image of the line area of ​​the single crystal rod. Next to the image, the specific parameters of the line posture, such as displacement, angle, etc., are displayed in numerical form.

[0104] Optionally, the current posture of the single crystal rod is displayed in the form of a three-dimensional model, and the current position of the line and its deviation relative to the target position are marked, and the deviation value between the current posture and the target posture is displayed in a chart or numerical value, so that the user can quickly understand the posture correction status.

[0105] Furthermore, users can perform operations in the task control area, such as adjusting correction parameters, starting / stopping processing tasks, etc. The system will update the display content in real time to ensure that the status changes after the user's operation are intuitively presented. At the same time, the function of manually adjusting the posture is provided, allowing users to make fine adjustments according to the system's suggestions. Each posture adjustment and line recognition result will be automatically recorded, including timestamp, operation content, adjustment results, etc. If the recognition result deviation exceeds the preset threshold or the equipment operates abnormally, the system will issue an alarm prompt in the log recording area and retain relevant records for subsequent analysis.

[0106] The disclosed embodiment provides a method for determining the position of the end-marking of a single crystal rod, which collects the image of the marking area corresponding to the end-marking of the single crystal rod; inputs the image of the marking area into a pre-trained marking recognition model, extracts the local features of the marking, and determines the marking posture corresponding to the end-marking of the single crystal rod according to the local features of the marking; determines the deviation value between the marking posture and the preset target posture, and corrects the posture of the single crystal rod according to the deviation value until the marking posture matches the preset target posture; and displays the recognition result corresponding to the end-marking and the marking posture to the user through a preset interactive interface. High-precision detection of the marking position of the single crystal rod can be achieved.

[0107] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.

[0108] Based on the same inventive concept, the embodiment of the present disclosure also provides a device for determining the position of the tail line of a single crystal rod corresponding to the method for determining the position of the tail line of a single crystal rod. Since the principle of solving the problem by the device in the embodiment of the present disclosure is similar to the method for determining the position of the tail line of a single crystal rod in the above-mentioned embodiment of the present disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0109] See also Figure 3 , Figure 3 A schematic diagram of a device for determining the position of a single crystal rod tail marking line provided in an embodiment of the present disclosure. Figure 3 As shown in FIG. , the device 300 for determining the position of the end marking line of a single crystal rod provided in an embodiment of the present disclosure includes:

[0110] The image acquisition module 310 is used to acquire an image of a marking area corresponding to the final marking at the end of the single crystal rod.

[0111] The scribing posture detection module 320 is used to input the scribing area image into a pre-trained scribing recognition model, extract the local features of the scribing, and determine the scribing posture corresponding to the finishing scribing of the single crystal rod according to the local features of the scribing.

[0112] The posture adjustment module 330 is used to determine the deviation value between the marking posture and the preset target posture, and correct the posture of the single crystal rod according to the deviation value until the marking posture matches the preset target posture.

[0113] The information display module 340 is used to display the recognition result corresponding to the end stroke and the stroke posture to the user through a preset interactive interface.

[0114] For descriptions of the processing flow of each module in the device and the interaction flow between each module, reference may be made to the relevant descriptions in the above method embodiment, which will not be described in detail here.

[0115] The disclosed embodiment provides a device for determining the position of the end-marking of a single crystal rod, which collects the image of the marking area corresponding to the end-marking of the single crystal rod; inputs the image of the marking area into a pre-trained marking recognition model, extracts the local features of the marking, and determines the marking posture corresponding to the end-marking of the single crystal rod according to the local features of the marking; determines the deviation value between the marking posture and the preset target posture, and corrects the posture of the single crystal rod according to the deviation value until the marking posture matches the preset target posture; and displays the recognition result corresponding to the end-marking and the marking posture to the user through a preset interactive interface. High-precision detection of the marking position of the single crystal rod can be achieved.

[0116] Corresponds to Figure 1 and Figure 2 The present disclosure also provides an electronic device 400, such as Figure 4 FIG. 4 is a schematic diagram of the structure of an electronic device 400 provided in an embodiment of the present disclosure, including:

[0117] Processor 41, memory 42, and bus 43; memory 42 is used to store execution instructions, including memory 421 and external memory 422; memory 421 here is also called internal memory, which is used to temporarily store operation data in processor 41 and data exchanged with external memory 422 such as hard disk. Processor 41 exchanges data with external memory 422 through memory 421. When the electronic device 400 is running, the processor 41 communicates with the memory 42 through bus 43, so that the processor 41 executes Figure 1 and Figure 2 The steps of the method for determining the position of the finishing line of the single crystal rod.

[0118] The embodiment of the present disclosure also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method for determining the position of the single crystal rod tail marking line described in the above method embodiment are executed. The storage medium can be a volatile or non-volatile computer-readable storage medium.

[0119] The embodiments of the present disclosure also provide a computer program product, which includes computer instructions. When the computer instructions are executed by a processor, the steps of the method for determining the position of the tail marking line of the single crystal rod described in the above method embodiment can be executed. For details, please refer to the above method embodiment, which will not be repeated here.

[0120] The computer program product may be implemented in hardware, software or a combination thereof. In one optional embodiment, the computer program product is implemented as a computer storage medium. In another optional embodiment, the computer program product is implemented as a software product, such as a software development kit (SDK).

[0121] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the several embodiments provided in the present disclosure, it should be understood that the disclosed device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms.

[0122] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0123] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0124] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as 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.

[0125] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The protection scope of the present disclosure is not limited thereto. Although the present disclosure is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed in the present disclosure, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A method for determining the position of a single crystal rod end marking line, characterized in that: include: Collect the image of the marking area corresponding to the final marking at the end of the single crystal rod; Inputting the scribing area image into a pre-trained scribing recognition model, extracting local features of the scribing, and determining the scribing posture corresponding to the finishing scribing of the single crystal rod according to the local features of the scribing; Determining a deviation value between the scribing posture and a preset target posture, and correcting the posture of the single crystal rod according to the deviation value until the scribing posture matches the preset target posture; The recognition result corresponding to the ending stroke and the stroke posture are displayed to the user through a preset interactive interface.

2. The method according to claim 1, characterized in that Collect the image of the marking area corresponding to the end marking of the single crystal rod, including: Obtain the diameter information corresponding to the single crystal rod; Determine the target focal length corresponding to capturing the image of the lined area according to the diameter information; Collecting the shooting distance between the current camera lens and the single crystal rod, and determining the displacement adjustment amount corresponding to the camera lens according to the target focal length and the shooting distance; The camera lens is driven to move along the optical axis according to the displacement adjustment amount until the target focal length is reached.

3. The method according to claim 1, characterized in that The line recognition model is trained based on the following steps: Constructing the line recognition model with a convolutional neural network as the base layer; Acquire a sample single crystal rod image having a plurality of material types, material colors, and surface textures, wherein the single crystal rod in the sample single crystal rod image is provided with a plurality of end markings of marking colors and marking sizes; The sample single crystal rod image is used as sample data, and the corresponding ending stroke in each sample single crystal rod image is used as a data label, which is input into the stroke recognition model for model training until the stroke recognition model has the ability to accurately recognize the stroke posture.

4. The method according to claim 1, characterized in that: Determining a deviation value between the marking posture and a preset target posture specifically includes: The position change of the single crystal rod is monitored in real time through the preset displacement monitoring sensor; Update the stroke pose according to the pose change; The deviation value between the updated line drawing posture and the preset target posture is calculated.

5. The method according to claim 1, characterized in that Correcting the posture of the single crystal rod according to the deviation value until the scribing posture matches the preset target posture specifically includes: Determining whether the deviation value is greater than a preset tolerance threshold; When the deviation value is greater than the preset tolerance threshold, driving a preset mechanical correction component to adjust the posture corresponding to the single crystal rod according to the deviation value; In the process of adjusting the corresponding posture of the single crystal rod, a closed-loop control strategy is adopted to provide real-time feedback on the current posture of the single crystal rod until the scribing posture corresponding to the adjusted single crystal rod matches the preset target posture.

6. The method according to claim 1, characterized in that The preset interactive interface is provided with a status overview area, a real-time monitoring area, a task control area and a log recording area; The status overview area is used to display the equipment operation status information corresponding to the single crystal rod production line; The real-time monitoring area is used to display the recognition result corresponding to the ending stroke; The task control area is used to respond to production process control operations and production parameter adjustment operations implemented by users; The log recording area is used to record production log information and abnormal alarm information.

7. A device for determining the position of a single crystal rod end marking, characterized in that: include: An image acquisition module is used to acquire an image of a marking area corresponding to the final marking at the end of the single crystal rod; A scribing posture detection module is used to input the scribing area image into a pre-trained scribing recognition model, extract the local features of the scribing, and determine the scribing posture corresponding to the finishing scribing of the single crystal rod according to the local features of the scribing; A posture adjustment module, used to determine a deviation value between the scribing posture and a preset target posture, and correct the posture of the single crystal rod according to the deviation value until the scribing posture matches the preset target posture; The information display module is used to display the recognition result corresponding to the end stroke and the stroke posture to the user through a preset interactive interface.

8. The device according to claim 7, characterized in that The device is also used for: Constructing the line recognition model with a convolutional neural network as the base layer; Acquire a sample single crystal rod image having a plurality of material types, material colors, and surface textures, wherein the single crystal rod in the sample single crystal rod image is provided with a plurality of end markings of marking colors and marking sizes; The sample single crystal rod image is used as sample data, and the corresponding ending stroke in each sample single crystal rod image is used as a data label, which is input into the stroke recognition model for model training until the stroke recognition model has the ability to accurately recognize the stroke posture.

9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the steps of the method for determining the position of the tail marking of a single crystal rod as described in any one of claims 1 to 6 are performed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for determining the position of the single crystal rod tail marking line according to any one of claims 1 to 6 are executed.

Images