A high-precision material edge deviation correction method and controller

Abstract

The application relates to a high-precision material edge deviation correction method, which comprises the following steps: S1. deploying a machine vision system; S2. deploying a sensor network on a material conveying path; S3. pre-processing collected material edge images, performing edge detection on the pre-processed images, and extracting the accurate contour of the material edge; accurately aligning and fusing the edge detection result with the sensor data in time and space to obtain a multidimensional data set; S4. establishing a physical model of material movement according to material characteristics and transmission system dynamics, which is used for predicting the future position of the material edge; S5. training the physical model by taking the multidimensional data set as the input of the physical model; S6. predicting the material edge position by using the trained physical model, comparing the predicted material edge position with the real-time detected material edge position, calculating the deviation amount, and generating specific control instructions based on the deviation amount to correct the deviation amount.

Description

Technical Field

[0001] This invention relates to the field of automation technology, and more specifically, to a high-precision material edge correction method and controller. Background Technology

[0002] In modern industrial automation, material handling and positioning are crucial links in the production line, and their accuracy and efficiency directly affect product quality and production efficiency. Especially in high-speed, high-precision production environments, accurate edge correction of materials is paramount. However, traditional material correction methods often rely on mechanical structures, fixed-parameter control algorithms, or simple sensor detection. These methods often fall short when faced with complex and varied material shapes, conveying speeds, and environmental conditions.

[0003] First, the diversity of material shapes poses a significant challenge to the accuracy of the web guiding mechanism. Different materials may have different edge characteristics, such as straight lines, curves, unevenness, etc. Traditional mechanical structures are difficult to adapt flexibly to these changes, resulting in limited web guiding accuracy.

[0004] Secondly, changes in transmission speed also increase the difficulty of correction. During high-speed transmission, the position and shape of the material edge may change rapidly due to factors such as inertia and vibration, making it difficult for traditional fixed-parameter control algorithms to track and accurately adjust the correction strategy in real time. Summary of the Invention

[0005] The purpose of this invention is to provide a high-precision material edge correction technology based on multimodal fusion and real-time feedback, aiming to solve the problems of insufficient accuracy and poor adaptability of existing material correction methods when facing complex and ever-changing material shapes, transmission speeds, and environmental conditions. By integrating machine vision, sensor data, and real-time physical models, high-precision, real-time detection and correction of material edges can be achieved, improving the overall efficiency of the production line and product quality.

[0006] To achieve the above-mentioned objectives, the technical solution adopted is as follows:

[0007] A high-precision material edge correction method includes the following steps:

[0008] S1. Deploy a machine vision system to acquire images of material edges;

[0009] S2. Deploy a sensor network along the material transport path to collect data on the physical position, velocity, and acceleration of the material; the data collected by the sensor network is collectively referred to as sensor data.

[0010] S3. Preprocess the acquired material edge image, perform edge detection on the preprocessed image, and extract the precise contour of the material edge; precisely align and fuse the edge detection result with the sensor data in time and space to obtain a multidimensional dataset;

[0011] S4. Based on the material properties and the dynamics of the transport system, establish a physical model of the material movement to predict the future position of the material edge;

[0012] S5. Use the multidimensional dataset as input to train the physical model. The training process is as follows:

[0013] The physical model outputs the predicted material edge position; the predicted material edge position is compared with the detected material edge position, the deviation is calculated, and the parameters of the physical model are dynamically adjusted based on the deviation.

[0014] S6. Predict the material edge position using the trained physical model, compare the predicted material edge position with the real-time detected material edge position, and calculate the deviation; generate specific control commands based on the deviation to correct the deviation.

[0015] Preferably, step S3 preprocesses the acquired material edge image, specifically including the following operations: noise reduction, contrast enhancement, and edge sharpening.

[0016] Preferably, the specific process for edge detection of the preprocessed image is as follows:

[0017] A multi-scale space for the image is constructed using a Gaussian pyramid, with each scale corresponding to a resolution version of the image.

[0018] Edge detection is performed using an edge detection algorithm at each scale;

[0019] A confidence-based fusion method is used to fuse the edge detection results at different scales into an original edge map;

[0020] Continuous contour lines are extracted from the original edge map. The extracted contour lines are evaluated to check whether they conform to the expected material shape and size. If they do, the original edge map is output as the final edge detection result. If they do not conform, a graph theory-based method is used to optimize the contour. The optimized contour is then fused with the original edge map to generate the final material edge contour map for output.

[0021] Preferably, at each scale, the specific process of edge detection using the edge detection algorithm is as follows:

[0022] Calculate the gradient of each pixel in the image;

[0023] Determine if the gradient of the current pixel is greater than the set threshold; if so, identify it as an edge point; thus obtaining the preliminary edge detection result.

[0024] The preliminary edge detection results are processed using non-maximum suppression, and then isolated edge points are connected by an edge connection algorithm to form complete edge lines.

[0025] Preferably, the specific process of fusing edge detection results at various scales into an original edge map using a confidence-based fusion method is as follows:

[0026] For the edge detection results at each scale, calculate the confidence level of each edge point;

[0027] The confidence scores calculated at different scales are normalized.

[0028] Based on the confidence level of the edges at each scale, different weights are assigned to the edge detection results at each scale;

[0029] The edge detection results at different scales are fused by weighted averaging of edge point locations.

[0030] Preferably, the edge detection results are precisely aligned and fused with the sensor data in time and space, as follows:

[0031] Time alignment: Use timestamps or external synchronization signals to precisely align edge detection results with sensor data in time;

[0032] Spatial alignment: First, camera calibration is performed to obtain the camera's intrinsic and extrinsic parameters; based on the sensor's installation position and orientation, the mapping relationship between sensor data and image pixels is calculated; the sensor data is then mapped onto the image plane to achieve pixel-level alignment.

[0033] Fusion: Weighted averaging of data from different sources to obtain a comprehensive result.

[0034] Preferably, in step S5, the parameters of the physical model are dynamically adjusted using the gradient descent method.

[0035] Preferably, the specific process of dynamically adjusting the parameters of the physical model using the gradient descent method is as follows:

[0036] S11.

[0037] Define a loss function:

[0038] Where N is the number of samples, x i It is the input of the i-th sample, y i This corresponds to the detected edge position of the material, f(θ; x).i () is the physical model with parameters θ and input x i The predicted value below;

[0039] S12. Calculate the gradient of the loss function L(θ) with respect to the model parameters θ.

[0040] S13. Update the model parameter θ according to the following formula:

[0041]

[0042] Where α is the learning rate;

[0043] S14. Repeat S12 and S13 until L(θ) is less than the set threshold.

[0044] Meanwhile, the present invention also provides a correction controller that executes the high-precision material edge correction method described above during operation.

[0045] Compared with the prior art, the beneficial effects of the present invention are:

[0046] Improving correction accuracy: By introducing high-resolution industrial cameras and advanced image processing algorithms, combined with deep learning models, high-precision identification of material edges is achieved. Simultaneously, by incorporating sensor data, a multi-modal information fusion system is constructed to further enhance the accuracy and reliability of correction.

[0047] Enhance system adaptability: By using a real-time physical model to predict material movement and combining it with a dynamic feedback control mechanism, the system can adjust its correction strategy in real time to adapt to changes in different material types, transmission speeds, and environmental conditions.

[0048] Improve production efficiency: By achieving real-time, high-precision edge correction of materials, reduce production interruptions and scrap rates caused by inaccurate edge correction, and improve the overall efficiency and capacity of the production line. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart illustrating a high-precision material edge correction method.

[0051] Figure 2 This is a flowchart illustrating the edge detection algorithm.

[0052] Figure 3 This diagram illustrates the specific process of fusing edge detection results at various scales into an original edge map using a confidence-based fusion method. Detailed Implementation

[0053] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0054] Example 1

[0055] like Figure 1-3 As shown, the specific process of the high-precision material edge correction method is as follows:

[0056] Implement environmental preparation

[0057] 1. Deployment of machine vision system:

[0058] Install high-resolution industrial cameras alongside the material conveyor line to ensure clear images of the material edges.

[0059] Configure the image processing unit, including the necessary software and algorithm library, for subsequent image preprocessing and edge detection.

[0060] 2. Sensor Network Deployment:

[0061] Position sensors, speed sensors, and acceleration sensors are installed at key locations along the material transport path to ensure that the physical state data of the materials can be acquired in real time.

[0062] Calibrate the sensors to ensure the accuracy and consistency of the data.

[0063] Implementation steps

[0064] S1. Machine vision system acquires images of material edges.

[0065] Start the machine vision system to capture images of the material edges in real time.

[0066] S2. Sensor network acquires physical data

[0067] The sensor network is activated synchronously to collect data on the physical position, velocity, and acceleration of the material in real time.

[0068] S3. Image Preprocessing and Edge Detection

[0069] Preprocessing:

[0070] The acquired images are denoised to reduce noise interference.

[0071] Enhance image contrast to make material edges clearer.

[0072] The image is sharpened at the edges to highlight the details of the material edges.

[0073] Edge detection:

[0074] Constructing a multi-scale space for images using Gaussian pyramids.

[0075] At each scale, an edge detection algorithm (such as the Canny algorithm) is used to calculate the gradient of each pixel in the image, and edge points are determined based on the gradient magnitude and a set threshold.

[0076] By applying nonmaximum suppression and edge connection algorithms, isolated edge points are connected into complete edge lines.

[0077] By using a confidence-based fusion method, the edge detection results at various scales are fused into the final material edge contour map.

[0078] Continuous contour lines are extracted from the original edge map. The extracted contour lines are evaluated to check whether they conform to the expected material shape and size. If they do, the original edge map is output as the final edge detection result. If they do not conform, a graph theory-based method is used to optimize the contour. The optimized contour is then fused with the original edge map to generate the final material edge contour map for output.

[0079] S4. Data Alignment and Fusion

[0080] Time alignment: Using timestamps or external synchronization signals, edge detection results are precisely aligned with sensor data in time.

[0081] Spatial alignment:

[0082] Perform camera calibration to obtain the camera's intrinsic and extrinsic parameters.

[0083] Based on the sensor's installation location and orientation, the mapping relationship between sensor data and image pixels is calculated.

[0084] Sensor data is mapped onto the image plane to achieve pixel-level alignment.

[0085] The aligned data are weighted and averaged to obtain a comprehensive multidimensional dataset.

[0086] S5. Physics Model Training

[0087] A physical model of material movement is established based on material properties and the dynamics of the transport system.

[0088] The physical model is trained by using a multidimensional dataset as input.

[0089] Define a loss function to measure the deviation between the model's predicted values ​​and the actual detected values.

[0090] The gradient descent method is used to dynamically adjust the parameters of the physical model and minimize the loss function value.

[0091] Repeat the training process until the loss function value is less than the set threshold or the predetermined number of training rounds is reached.

[0092] S6. Edge Location Prediction and Correction

[0093] The edge position of the material is predicted using a trained physical model.

[0094] The predicted material edge position is compared with the real-time detected material edge position, and the deviation is calculated.

[0095] Specific control commands are generated based on the deviation, such as adjusting the speed, direction, or position of the conveyor belt.

[0096] Control commands are sent to the actuators (such as motors, cylinders, etc.) to adjust the material position in real time.

[0097] S7. Monitoring and Optimization

[0098] During system operation, the correction effect at the material edge is continuously monitored.

[0099] Based on actual feedback, the physical model and control algorithm are optimized and adjusted as necessary to improve the stability and accuracy of the system.

[0100] Example 2

[0101] On an automated packaging line, materials of various shapes and sizes (such as food boxes and medicine bottles) need to be precisely aligned and packaged. During transport, materials may experience edge misalignment due to vibration, speed variations, or mechanical errors, affecting subsequent packaging quality. Therefore, a high-precision material edge correction method is introduced to ensure accurate alignment of materials before packaging.

[0102] Implementation steps

[0103] 1. System Integration and Deployment

[0104] Machine vision system: A high-resolution industrial camera is installed at the material inlet of the packaging line, and an image processing unit is configured to capture images of the material edges in real time.

[0105] Sensor network: Deploy position sensors, speed sensors, and acceleration sensors on the material conveyor belt to obtain real-time physical state data of the material.

[0106] Control system: A control system that integrates machine vision system, sensor network and actuators (such as pneumatic actuators, servo motors, etc.) to receive processed data and generate control commands.

[0107] 2. Material Edge Image Acquisition and Processing

[0108] When material enters the field of view of the machine vision system, the camera begins to capture images.

[0109] The acquired images are preprocessed, including noise reduction, contrast enhancement, and edge sharpening, to improve the accuracy of edge detection.

[0110] By applying a multi-scale edge detection algorithm, combined with Gaussian pyramid and Canny edge detector, the precise edge contour of the material is extracted.

[0111] 3. Data Alignment and Fusion

[0112] Timestamps are used to precisely align image data acquired by machine vision systems with physical data acquired by sensor networks in time.

[0113] By calibrating the camera and adjusting the sensor position, spatial alignment between image data and physical data is achieved.

[0114] The aligned data is then fused to form a dataset containing multi-dimensional information such as material edge position, velocity, and acceleration.

[0115] 4. Physical model prediction and parameter adjustment

[0116] A physical model of material movement is established based on material properties and the dynamics of the transport system.

[0117] The fused dataset is used as input to train a physical model to predict the future location of material edges.

[0118] The gradient descent method is used to dynamically adjust the parameters of the physical model in order to minimize the deviation between the predicted values ​​and the actual detected values.

[0119] 5. Real-time correction control

[0120] Before the material is about to reach the packaging area, its edge position is predicted using a trained physical model.

[0121] The predicted location is compared with the real-time detected location, and the deviation is calculated.

[0122] Based on the deviation, specific control commands are generated, such as adjusting the speed and direction of the conveyor belt or triggering pneumatic actuators for fine-tuning.

[0123] The actuator adjusts the material position in real time according to control commands to ensure that the material is accurately aligned before packaging.

[0124] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0125] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0126] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0127] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0128] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A high-precision material edge correction method, characterized in that: Includes the following steps: S1. Deploy a machine vision system to acquire images of material edges; S2. Deploy a sensor network along the material transport path to collect data on the physical position, velocity, and acceleration of the material; the data collected by the sensor network is collectively referred to as sensor data. S3. Preprocess the acquired material edge image, perform edge detection on the preprocessed image, and extract the precise contour of the material edge; precisely align and fuse the edge detection result with the sensor data in time and space to obtain a multidimensional dataset; S4. Based on the material properties and the dynamics of the transport system, establish a physical model of the material movement to predict the future position of the material edge; S5. Use the multidimensional dataset as input to train the physical model. The training process is as follows: The physical model outputs the predicted material edge position; the predicted material edge position is compared with the detected material edge position, the deviation is calculated, and the parameters of the physical model are dynamically adjusted based on the deviation. S6. Predict the material edge position using the trained physical model, compare the predicted material edge position with the real-time detected material edge position, and calculate the deviation; generate specific control commands based on the deviation to correct the deviation.

2. The high-precision material edge correction method according to claim 1, characterized in that: Step S3 preprocesses the acquired material edge image, specifically including the following operations: noise reduction, contrast enhancement, and edge sharpening.

3. The high-precision material edge correction method according to claim 1, characterized in that: The specific process of edge detection for the preprocessed image is as follows: A multi-scale space for the image is constructed using a Gaussian pyramid, with each scale corresponding to a resolution version of the image. Edge detection is performed using an edge detection algorithm at each scale; A confidence-based fusion method is used to fuse the edge detection results at different scales into an original edge map; Continuous contour lines are extracted from the original edge map. The extracted contour lines are evaluated to check whether they conform to the expected material shape and size. If they do, the original edge map is output as the final edge detection result. If they do not conform, a graph theory-based method is used to optimize the contour. The optimized contour is then fused with the original edge map to generate the final material edge contour map for output.

4. The high-precision material edge correction method according to claim 3, characterized in that: The specific process of edge detection using edge detection algorithms at each scale is as follows: Calculate the gradient of each pixel in the image; Determine if the gradient of the current pixel is greater than the set threshold; if so, identify it as an edge point; thus obtaining the preliminary edge detection result. The preliminary edge detection results are processed using non-maximum suppression, and then isolated edge points are connected by an edge connection algorithm to form complete edge lines.

5. The high-precision material edge correction method according to claim 3, characterized in that: The specific process of fusing edge detection results at various scales into an original edge map using a confidence-based fusion method is as follows: For the edge detection results at each scale, calculate the confidence level of each edge point; The confidence scores calculated at different scales are normalized. Based on the confidence level of the edges at each scale, different weights are assigned to the edge detection results at each scale; The edge detection results at different scales are fused by weighted averaging of edge point locations.

6. The high-precision material edge correction method according to claim 1, characterized in that: The edge detection results are precisely aligned and fused with sensor data in time and space, as follows: Time alignment: Use timestamps or external synchronization signals to precisely align edge detection results with sensor data in time; Spatial alignment: First, camera calibration is performed to obtain the camera's intrinsic and extrinsic parameters; based on the sensor's installation position and orientation, the mapping relationship between sensor data and image pixels is calculated. By mapping sensor data onto the image plane, pixel-level alignment is achieved; Fusion: Weighted averaging of data from different sources to obtain a comprehensive result.

7. The high-precision material edge correction method according to claim 1, characterized in that: In step S5, the parameters of the physical model are dynamically adjusted using the gradient descent method.

8. The high-precision material edge correction method according to claim 7, characterized in that: The specific process of dynamically adjusting the parameters of the physical model using gradient descent is as follows: S11. Define a loss function: Where N is the number of samples, x i It is the input of the i-th sample, y i This corresponds to the detected edge position of the material, f(θ; x). i () is the physical model with parameters θ and input x i The predicted value below; S12. Calculate the gradient of the loss function L(θ) with respect to the model parameters θ. S13. Update the model parameter θ according to the following formula: Where α is the learning rate; S14. Repeat S12 and S13 until L(θ) is less than the set threshold.

9. A web correction controller, characterized in that: When it is in operation, it performs the high-precision material edge correction method according to any one of claims 1-8.

Images