Intelligent steel bar binding system and operation method thereof

Abstract

The invention discloses an intelligent steel bar binding system and an operation method thereof, and belongs to the technical field of steel bar binding, the intelligent steel bar binding system comprises an image acquisition module, an image identification and positioning module, an image preprocessing module, a result output module, a control operation module, a threshold setting module and an operation compensation module, according to the system, through mutual cooperative operation of all the modules, reinforcing steel bar binding operation of automatic recognition and autonomous operation is achieved, automatic recognition and positioning of reinforcing steel bar binding points are achieved through the image recognition technology, grabbing and binding operation is automatically completed through the operation and control operation module, and therefore complexity of traditional manual operation is reduced, and efficiency is improved. The whole operation efficiency is effectively improved, manual participation can be greatly reduced during system operation, therefore, the labor intensity can be greatly reduced, the stability of the binding quality can be guaranteed through automatic binding of the system, and the automatic binding system has wide application prospects.

Description

Technical Field

[0001] The present invention relates to the technical field of steel bar binding, and in particular to an intelligent steel bar binding system and an operating method thereof. Background Art

[0002] Steel bars are often used as skeleton supports in buildings. When in use, the steel bars need to be tied and fixed with steel wires to prevent them from being dispersed by concrete. Steel bar binding refers to the process of weaving steel bar structures to ensure the safety and quality of construction. At present, steel bar binding is traditionally done manually. However, the traditional manual binding method has a large workload, high labor cost, and a long time consumption. There are also certain problems of labor intensity and unstable binding quality. Therefore, an intelligent steel bar binding system and an operation method thereof are proposed. Summary of the invention

[0003] The object of the present invention is to provide an intelligent steel bar binding system and an operation method thereof to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent steel bar binding system, comprising an image acquisition module, an image recognition and positioning module, an image preprocessing module, a result output module, a control operation module, a threshold setting module and an operation compensation module, wherein the image acquisition module, the image recognition and positioning module, the image preprocessing module, the result output module and the control operation module are wirelessly connected, the image preprocessing module, the threshold setting module and the operation compensation module are wirelessly connected, and the image recognition and positioning module, the threshold setting module and the operation compensation module are wirelessly connected;

[0005] The image acquisition module captures the working area through an acquisition device to obtain image information data, wherein the acquisition device may be a camera or a scanner, etc.;

[0006] The image recognition and positioning module is used to recognize the acquired image information data, so as to obtain and determine the position and number of the lashing points;

[0007] The image preprocessing module is used to preprocess the collected image information data, so as to improve the accuracy of subsequent image recognition, and the image preprocessing module includes a steel bar recognition unit and a binding point confirmation unit, and the steel bar recognition unit is wirelessly connected to the binding point confirmation unit;

[0008] The steel bar identification unit is used to identify and segment the steel bars in the image through an image processing algorithm, thereby extracting the position and shape information of the steel bars;

[0009] The binding point confirmation unit determines the position and number of binding points according to the position and shape information of the steel bars in combination with preset binding rules.

[0010] As a further preferred embodiment of the present technical solution: the result output module is used to output the result information after the image preprocessing module is run to the control operation module, wherein the result information includes the image recognition result and the target position, etc.;

[0011] The control operation module is used to receive the result information and perform subsequent operations according to the result information, and the control operation module includes a transport and grabbing unit and a lashing operation unit, and the transport and grabbing unit is wirelessly connected to the lashing operation unit;

[0012] The transport and grabbing unit is used to grab the corresponding number and position of steel bars through the grabbing device according to the result information, and place them at the position to be tied;

[0013] The binding operation unit performs binding work on the steel bars placed on the confirmed binding points through binding equipment.

[0014] As a further preferred embodiment of the present technical solution: the specific operation mode of the transport and grabbing unit includes the following steps:

[0015] A1. Image recognition and positioning: Use machine vision technology to perform image recognition, positioning and posture estimation on the steel bars to be grasped, so as to obtain the position and posture information of the steel bars;

[0016] A2. Grasping planning and path planning: Based on the result information, the grasping planning algorithm is used to calculate the grasping posture and path planning of the grasping device, thereby ensuring that the grasping device can accurately grasp the steel bars;

[0017] A3. Visual servo control: By real-time monitoring of the relative position and posture of the end effector of the grasping device and the target steel bar, visual feedback is used to control the grasping action of the grasping device, and the result information is integrated with the control system of the grasping device to calculate the motion trajectory of the grasping device, so that the end effector can accurately grasp the steel bar according to the target position and posture;

[0018] A4. Gripping force control: Install sensors on the end effector of the gripping device, detect the contact force between the gripping device and the steel bar, use the force control algorithm to achieve safe gripping and placement of the gripping device, and adjust the motion trajectory and force of the gripping device by real-time monitoring of the contact force, thereby achieving stable gripping and placement operations;

[0019] A5. Safety protection mechanism: preset safety protection mechanism to ensure the safety and stability of the grabbing equipment operation;

[0020] A6. Real-time monitoring and feedback: During the grabbing process, the grabbing action and the position of the steel bars are monitored in real time, and feedback information is provided for adjustment and correction.

[0021] As a further preferred embodiment of the present technical solution: in A4, the installed sensor is a torque sensor, and in A5, the preset safety protection mechanism includes collision detection and force control.

[0022] As a further preferred embodiment of the present technical solution: the specific operation mode of the lashing operation unit includes the following steps:

[0023] B1. Automated control binding system: Design and implement an automated control system to automate the steel bar binding process;

[0024] B2. Based on the result information, coordinate the coordinated operation of the grabbing equipment, tying equipment and cutting equipment to achieve the automated operation of steel bar tying;

[0025] B3. Sensing and feedback control: sensors are integrated into the grabbing equipment, lashing equipment and cutting equipment to monitor the parameters of the lashing process in real time and perform feedback control;

[0026] B4. Security protection mechanism: Design a security protection mechanism to improve the security of the overall operation of the system;

[0027] B5. Real-time monitoring and recording: During the binding process, the binding status and quality are monitored in real time, and relevant data are recorded to facilitate subsequent analysis and tracing.

[0028] As a further preferred embodiment of the present technical solution: in B1, the automated control system includes integrated machine vision positioning, grabbing and cutting mechanisms, wire roll rotation mechanisms, etc.; in B3, real-time monitoring parameters include force, position and posture, etc.; in B4, the designed safety protection mechanisms include collision detection, emergency stop, etc., so as to ensure the safe operation of the grabbing equipment and the binding equipment; in B5, the recorded relevant data include binding time and force, etc.

[0029] As a further preferred embodiment of the present technical solution: through the feedback information of the sensor, the system as a whole can make real-time adjustments to the actions of the grabbing equipment and the binding equipment, so as to ensure the accuracy and stability of the binding, and the system as a whole can monitor and respond to any unexpected situations in real time to ensure the safety of the operators and equipment. At the same time, the system as a whole can optimize the binding process and improve efficiency and quality through data recording and analysis.

[0030] As a further preferred embodiment of the present technical solution: the threshold setting unit is used to monitor the number of operations of the image preprocessing module, and when the number of operations of the image preprocessing module reaches a set threshold, the operation information of the image preprocessing module is fed back to the image recognition and positioning module, and the operation compensation module is triggered;

[0031] The running compensation module is used to accurately obtain the result information processed by the image preprocessing module.

[0032] An operation method of an intelligent steel bar binding system comprises the following steps:

[0033] S1. Using an image acquisition module to capture the work area with an acquisition device to obtain image information data;

[0034] S2, after S1 is executed, the acquired image information data is identified through the image recognition and positioning module, so as to obtain and determine the position and number of the lashing points;

[0035] S3, after S2 is executed, the collected image information data is preprocessed by an image preprocessing module, so as to improve the accuracy of subsequent image recognition;

[0036] S4, after S3 is executed, the result information after the image preprocessing module is executed is output to the control operation module through the result output module;

[0037] S5, after S4 is run, the control operation module receives the result information and performs subsequent operations according to the result information;

[0038] S6, after S3 is executed, the operation times of the image preprocessing module are monitored by the threshold setting unit, and when the operation times reach the set threshold, the monitoring information is fed back to the image recognition and positioning module, and the operation compensation module is triggered;

[0039] S7, after S6 is executed, the compensation module is executed to accurately obtain the result information processed by the image preprocessing module.

[0040] As a further preferred embodiment of the present technical solution: in S3, the image preprocessing module adopts a machine vision method, and the specific operation mode of the machine vision method includes the following steps:

[0041] C1. Image preprocessing: preprocess the acquired image information to improve the accuracy of subsequent processing;

[0042] C2. Feature extraction: Use feature extraction algorithm to extract key features in the steel bar image;

[0043] C3. Target detection: Based on feature extraction, target detection algorithms can be used to accurately locate and identify steel bars;

[0044] C4. Positioning and identification: The position and direction information of the steel bars are obtained through the output results of the target detection algorithm.

[0045] As a further preferred embodiment of the present technical solution: in C1, the preprocessing operation includes image denoising, image enhancement and image smoothing;

[0046] Among them, the denoising operation uses a filter (median filter or Gaussian filter) to eliminate noise in the image, and the enhancement operation uses methods such as histogram equalization or contrast stretching to enhance the characteristics of the steel bar image.

[0047] As a further preferred embodiment of the present technical solution: in C2, feature extraction algorithms include SIFT (Scale Invariant Feature Transform), SURF (Speeded Up Robust Features) and ORB (Oriented FAST and Rotated BRIEF), etc., through which feature points or feature descriptors with robustness and uniqueness can be effectively extracted for subsequent positioning and recognition.

[0048] As a further preferred embodiment of the present technical solution: in C3, the target detection algorithm can be divided into a traditional classifier method (Haar feature detector, HOG feature + SVM classifier) ​​and a cascade classifier-based method (Viola-Jones algorithm).

[0049] As a further preferred embodiment of the present technical solution: according to the requirements, the steel bars can be further positioned and identified;

[0050] Among them, positioning can be achieved by calculating parameters such as the center point coordinates and bounding box coordinates of the steel bar, and identification can be achieved by matching the steel bar with a pre-established steel bar library to determine the type or other properties of the steel bar.

[0051] As a further preferred embodiment of the present technical solution: in S7, the operation compensation module adopts a deep learning method, and the specific operation mode of the deep learning method includes the following steps:

[0052] D1. Collect a large amount of steel bar image data and mark them;

[0053] D2. Use the location and shape information of the steel bars as labels;

[0054] D3. Use these labeled data to train the CNN model. During the training process, the CNN model autonomously learns the feature representation and location method of the steel bars.

[0055] D4. Apply the trained CNN model to the new steel bar image and determine the position and orientation of the steel bars through the output of the model.

[0056] As a further preferred embodiment of the present technical solution: a convolutional neural network (CNN) is used as the deep learning model. In D3, a labeled data set is used to divide it into a training set and a test set. After that, the training set is input into the deep learning model for training. The goal of the training is to adjust the weights and parameters of the model so that it can accurately predict the position and shape of the steel bars. During the training process, optimization techniques are used to improve the performance of the model. Methods such as learning rate adjustment, regularization, and batch normalization are used to optimize the training process and results of the model.

[0057] In this embodiment, specifically: after the training is completed, the model is evaluated using a test set. The performance of the model is evaluated by calculating indicators such as the accuracy, recall, and precision of the model on the test set. If the performance of the model is not ideal, timely adjustments and improvements can be made, such as increasing training data, adjusting the model structure, etc.

[0058] In this embodiment, specifically: the trained deep learning model can be used for the prediction and application of new steel bar images. By inputting the new image into the model, the position and shape information of the steel bar can be obtained, thereby achieving accurate positioning and identification of the steel bar.

[0059] Compared with the prior art, the present invention has the following beneficial effects:

[0060] 1. The present invention realizes automatic identification and autonomous operation of steel bar binding through the mutual cooperation between modules, realizes automatic identification and positioning of steel bar binding points through image recognition technology, and then uses the control operation module to automatically complete the grabbing and binding operations, thereby reducing the complexity of traditional manual operations and effectively improving the overall operation efficiency. The system operation can greatly reduce manual participation, so the labor intensity can be greatly reduced. Through the automatic binding of the system, the stability of the binding quality can be guaranteed, and it has a wide range of application prospects;

[0061] 2. In the present invention, the results of the machine vision method and the deep learning method are integrated, and a weighted average mechanism is tried to comprehensively utilize the advantages of the two methods to improve the accuracy and robustness of positioning. In addition, according to the situation of the machine vision method, the model is adjusted and optimized, the model parameters are adjusted, the training data is increased, and regularization is introduced to further improve the performance of steel bar binding point positioning;

[0062] 3. In the present invention, the cross-validation of the two recognition methods of machine vision and deep learning can improve the accuracy of binding point recognition. At the same time, in actual production, the accuracy, precision and other parameters of the model are continuously learned to achieve automatic iterative evolution. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a schematic diagram of the architecture of an intelligent steel bar binding system of the present invention;

[0064] Figure 2 A schematic diagram of a flow chart of an operation method of an intelligent steel bar binding system of the present invention;

[0065] Figure 3 It is a schematic diagram of the process of a handling and grabbing unit in an intelligent steel bar binding system of the present invention;

[0066] Figure 4 A schematic diagram of a process flow of a binding operation unit in an intelligent steel bar binding system of the present invention;

[0067] Figure 5 A schematic diagram of a flow chart of an image preprocessing module in an operation method of an intelligent steel bar binding system of the present invention;

[0068] Figure 6 The present invention is a schematic diagram of the process of operating a compensation module in an operating method of an intelligent steel bar binding system. DETAILED DESCRIPTION

[0069] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0070] Example

[0071] See also Figure 1-Figure 6 The present invention provides a technical solution: an intelligent steel bar binding system, comprising an image acquisition module, an image recognition and positioning module, an image preprocessing module, a result output module, a control operation module, a threshold setting module and an operation compensation module, wherein the image acquisition module, the image recognition and positioning module, the image preprocessing module, the result output module and the control operation module are wirelessly connected, the image preprocessing module, the threshold setting module and the operation compensation module are wirelessly connected, and the image recognition and positioning module, the threshold setting module and the operation compensation module are wirelessly connected;

[0072] The image acquisition module captures the working area through an acquisition device to obtain image information data, wherein the acquisition device may be a camera or a scanner, etc.;

[0073] The image recognition and positioning module is used to recognize the acquired image information data, so as to obtain and determine the position and number of the lashing points;

[0074] The image preprocessing module is used to preprocess the collected image information data, so as to improve the accuracy of subsequent image recognition, and the image preprocessing module includes a steel bar recognition unit and a binding point confirmation unit, and the steel bar recognition unit is wirelessly connected to the binding point confirmation unit;

[0075] The steel bar identification unit is used to identify and segment the steel bars in the image through an image processing algorithm, thereby extracting the position and shape information of the steel bars;

[0076] The binding point confirmation unit determines the position and number of binding points according to the position and shape information of the steel bars in combination with preset binding rules.

[0077] In this embodiment, specifically: the result output module is used to output the result information after the image preprocessing module is run to the control operation module;

[0078] The result information includes image recognition results and target location, etc.;

[0079] The control operation module is used to receive the result information and perform subsequent operations according to the result information, and the control operation module includes a transport and grabbing unit and a lashing operation unit, and the transport and grabbing unit is wirelessly connected to the lashing operation unit;

[0080] The transport and grabbing unit is used to grab the corresponding number and position of steel bars through the grabbing device according to the result information, and place them at the position to be tied;

[0081] The binding operation unit performs binding work on the steel bars placed on the confirmed binding points through binding equipment.

[0082] In this embodiment, specifically: the specific operation mode of the transport and grabbing unit includes the following steps:

[0083] A1. Image recognition and positioning: Use machine vision technology to perform image recognition, positioning and posture estimation on the steel bars to be grasped, so as to obtain the position and posture information of the steel bars;

[0084] A2. Grasping planning and path planning: Based on the result information, the grasping planning algorithm is used to calculate the grasping posture and path planning of the grasping device, thereby ensuring that the grasping device can accurately grasp the steel bars;

[0085] A3. Visual servo control: By real-time monitoring of the relative position and posture of the end effector of the grasping device and the target steel bar, visual feedback is used to control the grasping action of the grasping device, and the result information is integrated with the control system of the grasping device to calculate the motion trajectory of the grasping device, so that the end effector can accurately grasp the steel bar according to the target position and posture;

[0086] A4. Gripping force control: Install sensors on the end effector of the gripping device, detect the contact force between the gripping device and the steel bar, use the force control algorithm to achieve safe gripping and placement of the gripping device, and adjust the motion trajectory and force of the gripping device by real-time monitoring of the contact force, thereby achieving stable gripping and placement operations;

[0087] A5. Safety protection mechanism: preset safety protection mechanism to ensure the safety and stability of the grabbing equipment operation;

[0088] A6. Real-time monitoring and feedback: During the grabbing process, the grabbing action and the position of the steel bars are monitored in real time, and feedback information is provided for adjustment and correction.

[0089] In this embodiment, specifically: in A4, the installed sensor is a torque sensor, and in A5, the preset safety protection mechanism includes collision detection and force control.

[0090] In this embodiment, specifically: the specific operation mode of the binding operation unit includes the following steps:

[0091] B1. Automated control binding system: Design and implement an automated control system to automate the steel bar binding process;

[0092] B2. Based on the result information, coordinate the coordinated operation of the grabbing equipment, tying equipment and cutting equipment to achieve the automated operation of steel bar tying;

[0093] B3. Sensing and feedback control: sensors are integrated into the grabbing equipment, lashing equipment and cutting equipment to monitor the parameters of the lashing process in real time and perform feedback control;

[0094] B4. Security protection mechanism: Design a security protection mechanism to improve the security of the overall operation of the system;

[0095] B5. Real-time monitoring and recording: During the binding process, the binding status and quality are monitored in real time, and relevant data are recorded to facilitate subsequent analysis and tracing.

[0096] In this embodiment, specifically: in B1, the automated control system includes integrated machine vision positioning, grabbing and cutting mechanisms, wire roll rotation mechanisms, etc.; in B3, real-time monitoring parameters include force, position and posture, etc.; in B4, the designed safety protection mechanisms include collision detection, emergency stop, etc., so as to ensure the safe operation of the grabbing equipment and the binding equipment; in B5, the recorded relevant data include binding time and strength, etc.

[0097] In this embodiment, specifically: through the feedback information of the sensor, the system as a whole can make real-time adjustments to the actions of the grabbing equipment and the binding equipment to ensure the accuracy and stability of the binding, and the system as a whole can monitor and respond to any unexpected situations in real time to ensure the safety of operators and equipment. At the same time, the system as a whole can optimize the binding process through data recording and analysis to improve efficiency and quality.

[0098] In this embodiment, specifically: the threshold setting unit is used to monitor the number of times the image preprocessing module is run, and when the number of times the image preprocessing module is run reaches a set threshold, the running information of the image preprocessing module is fed back to the image recognition and positioning module, and the running compensation module is triggered;

[0099] The running compensation module is used to accurately obtain the result information processed by the image preprocessing module.

[0100] An operation method of an intelligent steel bar binding system comprises the following steps:

[0101] S1. Using an image acquisition module to capture the work area with an acquisition device to obtain image information data;

[0102] S2, after S1 is executed, the acquired image information data is identified through the image recognition and positioning module, so as to obtain and determine the position and number of the lashing points;

[0103] S3, after S2 is executed, the collected image information data is preprocessed by an image preprocessing module, so as to improve the accuracy of subsequent image recognition;

[0104] S4, after S3 is executed, the result information after the image preprocessing module is executed is output to the control operation module through the result output module;

[0105] S5, after S4 is run, the control operation module receives the result information and performs subsequent operations according to the result information;

[0106] S6, after S3 is executed, the operation times of the image preprocessing module are monitored by the threshold setting unit, and when the operation times reach the set threshold, the monitoring information is fed back to the image recognition and positioning module, and the operation compensation module is triggered;

[0107] S7, after S6 is executed, the compensation module is executed to accurately obtain the result information processed by the image preprocessing module.

[0108] In this embodiment, specifically: in S3, the image preprocessing module adopts a machine vision method, and the specific operation mode of the machine vision method includes the following steps:

[0109] C1. Image preprocessing: preprocess the acquired image information to improve the accuracy of subsequent processing;

[0110] C2. Feature extraction: Use feature extraction algorithm to extract key features in the steel bar image;

[0111] C3. Target detection: Based on feature extraction, target detection algorithms can be used to accurately locate and identify steel bars;

[0112] C4. Positioning and identification: The position and direction information of the steel bars are obtained through the output results of the target detection algorithm.

[0113] In this embodiment, specifically: in C1, the preprocessing operation includes image denoising, image enhancement and image smoothing;

[0114] Among them, the denoising operation uses a filter (median filter or Gaussian filter) to eliminate noise in the image, and the enhancement operation uses methods such as histogram equalization or contrast stretching to enhance the characteristics of the steel bar image.

[0115] In this embodiment, specifically: in C2, feature extraction algorithms include SIFT (Scale Invariant Feature Transform), SURF (Speeded Up Robust Features) and ORB (Oriented FAST and Rotated BRIEF), etc., through which feature points or feature descriptors with robustness and uniqueness can be effectively extracted, so as to be used for subsequent positioning and recognition;

[0116] In this embodiment, specifically: in C3, the target detection algorithm can be divided into a traditional classifier method (Haar feature detector, HOG feature + SVM classifier) ​​and a cascade classifier-based method (Viola-Jones algorithm).

[0117] In this embodiment, specifically: according to the needs, the steel bars can be further positioned and identified;

[0118] Among them, positioning can be achieved by calculating parameters such as the center point coordinates and bounding box coordinates of the steel bar, and identification can be achieved by matching the steel bar with a pre-established steel bar library to determine the type or other properties of the steel bar.

[0119] In this embodiment, specifically: in S7, the compensation module is operated using a deep learning method, and the specific operation mode of the deep learning method includes the following steps:

[0120] D1. Collect a large amount of steel bar image data and mark them;

[0121] D2. Use the location and shape information of the steel bars as labels;

[0122] D3. Use these labeled data to train the CNN model. During the training process, the CNN model autonomously learns the feature representation and location method of the steel bars.

[0123] D4. Apply the trained CNN model to the new steel bar image and determine the position and orientation of the steel bars through the output of the model.

[0124] In this embodiment, specifically: the deep learning model adopts a convolutional neural network (CNN). In D3, a labeled data set is used to divide it into a training set and a test set. Then, the training set is input into the deep learning model for training. The goal of the training is to adjust the weights and parameters of the model so that it can accurately predict the position and shape of the steel bars. During the training process, optimization techniques are used to improve the performance of the model. Methods such as learning rate adjustment, regularization, and batch normalization are used to optimize the training process and results of the model.

[0125] In this embodiment, specifically: after the training is completed, the model is evaluated using a test set. The performance of the model is evaluated by calculating indicators such as the accuracy, recall, and precision of the model on the test set. If the performance of the model is not ideal, timely adjustments and improvements can be made, such as increasing training data, adjusting the model structure, etc.

[0126] In this embodiment, specifically: the trained deep learning model can be used for the prediction and application of new steel bar images. By inputting the new image into the model, the position and shape information of the steel bar can be obtained, thereby achieving accurate positioning and identification of the steel bar.

[0127] In this embodiment, specifically: by fusing the results of the machine vision method and the deep learning method, a weighted average mechanism is tried to comprehensively utilize the advantages of the two methods to improve the accuracy and robustness of positioning, and according to the situation of the machine vision method, the model is adjusted and optimized, the model parameters are adjusted, the training data is increased, regularization is introduced, etc., to further improve the performance of steel bar binding point positioning.

[0128] In this embodiment, specifically: the accuracy of binding point recognition can be improved through cross-validation of the two recognition methods of machine vision and deep learning. At the same time, in actual production, the accuracy, precision and other parameters of the model are continuously learned to achieve automatic iterative evolution.

[0129] Working principle: When in use, first use the image acquisition module to capture the working area with the acquisition equipment to obtain image information data, and then use the image recognition and positioning module to identify its internal data of the acquired image information, so as to obtain and determine the position and number of the binding points, and then use the image preprocessing module to preprocess the collected image information data, so as to improve the accuracy of subsequent image recognition, and at the same time use the threshold setting unit to monitor the number of operations of the image preprocessing module. When the number of operations reaches the set threshold, the monitoring information is fed back to the image recognition and positioning module, and the operation compensation module is triggered. The operation compensation module runs the result information processed by the accurate image preprocessing module, and finally outputs the result information after the image preprocessing module is run to the control operation module through the result output module. After receiving the result information, the control operation module performs subsequent operation work according to the result information.

[0130] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent steel bar binding system, comprising an image acquisition module, an image recognition and positioning module, an image preprocessing module, a result output module, a control operation module, a threshold setting module and an operation compensation module, characterized in that: The image acquisition module, image recognition and positioning module, image preprocessing module, result output module and control operation module are wirelessly connected, the image preprocessing module, threshold setting module and operation compensation module are wirelessly connected, and the image recognition and positioning module, threshold setting module and operation compensation module are wirelessly connected; The image acquisition module captures the working area through an acquisition device to obtain image information data; The image recognition and positioning module is used to recognize the acquired image information data, so as to obtain and determine the position and number of the lashing points; The image preprocessing module is used to preprocess the collected image information data, so as to improve the accuracy of subsequent image recognition, and the image preprocessing module includes a steel bar recognition unit and a binding point confirmation unit, and the steel bar recognition unit is wirelessly connected to the binding point confirmation unit; The steel bar identification unit is used to identify and segment the steel bars in the image through an image processing algorithm, thereby extracting the position and shape information of the steel bars; The binding point confirmation unit determines the position and number of binding points according to the position and shape information of the steel bars in combination with preset binding rules.

2. The intelligent steel bar tying system according to claim 1, characterized in that: The result output module is used to output the result information after the image preprocessing module is run to the control operation module; The control operation module is used to receive the result information and perform subsequent operations according to the result information, and the control operation module includes a transport and grabbing unit and a lashing operation unit, and the transport and grabbing unit is wirelessly connected to the lashing operation unit; The transport and grabbing unit is used to grab the corresponding number and position of steel bars through the grabbing device according to the result information, and place them at the position to be tied; The binding operation unit performs binding work on the steel bars placed on the confirmed binding points through binding equipment.

3. The intelligent steel bar tying system according to claim 2, characterized in that: The specific operation mode of the transport and grabbing unit includes the following steps: A1. Image recognition and positioning: Use machine vision technology to perform image recognition, positioning and posture estimation on the steel bars to be grasped, so as to obtain the position and posture information of the steel bars; A2. Grasping planning and path planning: Based on the result information, the grasping planning algorithm is used to calculate the grasping posture and path planning of the grasping device, thereby ensuring that the grasping device can accurately grasp the steel bars; A3. Visual servo control: By real-time monitoring of the relative position and posture of the end effector of the grasping device and the target steel bar, visual feedback is used to control the grasping action of the grasping device, and the result information is integrated with the control system of the grasping device to calculate the motion trajectory of the grasping device, so that the end effector can accurately grasp the steel bar according to the target position and posture; A4. Gripping force control: Install sensors on the end effector of the gripping device, detect the contact force between the gripping device and the steel bar, use the force control algorithm to achieve safe gripping and placement of the gripping device, and adjust the motion trajectory and force of the gripping device by real-time monitoring of the contact force, thereby achieving stable gripping and placement operations; A5. Safety protection mechanism: preset safety protection mechanism to ensure the safety and stability of the grabbing equipment operation; A6. Real-time monitoring and feedback: During the grabbing process, the grabbing action and the position of the steel bars are monitored in real time, and feedback information is provided for adjustment and correction.

4. The intelligent steel bar tying system according to claim 2, characterized in that: The specific operation mode of the lashing operation unit includes the following steps: B1. Automated control binding system: Design and implement an automated control system to automate the steel bar binding process; B2. Based on the result information, coordinate the coordinated operation of the grabbing equipment, tying equipment and cutting equipment to achieve the automated operation of steel bar tying; B3. Sensing and feedback control: sensors are integrated into the grabbing equipment, lashing equipment and cutting equipment to monitor the parameters of the lashing process in real time and perform feedback control; B4. Security protection mechanism: Design a security protection mechanism to improve the security of the overall operation of the system; B5. Real-time monitoring and recording: During the binding process, the binding status and quality are monitored in real time, and relevant data are recorded to facilitate subsequent analysis and tracing.

5. The intelligent steel bar tying system according to claim 1, characterized in that: The threshold setting unit is used to monitor the number of times the image preprocessing module is run, and when the number of times the image preprocessing module is run reaches a set threshold, the running information of the image preprocessing module is fed back to the image recognition and positioning module, and the running compensation module is triggered; The running compensation module is used to accurately obtain the result information processed by the image preprocessing module.

6. An operating method of an intelligent steel bar tying system, characterized in that: The following steps are involved: S1. Using an image acquisition module to capture the work area with an acquisition device to obtain image information data; S2, after S1 is executed, the acquired image information data is identified through the image recognition and positioning module, so as to obtain and determine the position and number of the lashing points; S3, after S2 is executed, the collected image information data is preprocessed by an image preprocessing module, so as to improve the accuracy of subsequent image recognition; S4, after S3 is executed, the result information after the image preprocessing module is executed is output to the control operation module through the result output module; S5, after S4 is run, the control operation module receives the result information and performs subsequent operations according to the result information; S6, after S3 is executed, the operation times of the image preprocessing module are monitored by the threshold setting unit, and when the operation times reach the set threshold, the monitoring information is fed back to the image recognition and positioning module, and the operation compensation module is triggered; S7, after S6 is executed, the compensation module is executed to accurately obtain the result information processed by the image preprocessing module.

7. The operating method of the intelligent steel bar tying system according to claim 6, characterized in that: In S3, the image preprocessing module adopts a machine vision method, and the specific operation of the machine vision method includes the following steps: C1. Image preprocessing: preprocess the acquired image information to improve the accuracy of subsequent processing; C2. Feature extraction: Use feature extraction algorithm to extract key features in the steel bar image; C3. Target detection: Based on feature extraction, target detection algorithms can be used to accurately locate and identify steel bars; C4. Positioning and identification: The position and direction information of the steel bars are obtained through the output results of the target detection algorithm.

8. The operating method of the intelligent steel bar tying system according to claim 6, characterized in that: In S7, the compensation module is operated by using a deep learning method, and the specific operation mode of the deep learning method includes the following steps: D1. Collect a large amount of steel bar image data and mark them; D2. Use the location and shape information of the steel bars as labels; D3. Use these labeled data to train the CNN model. During the training process, the CNN model autonomously learns the feature representation and location method of steel bars. D4. Apply the trained CNN model to the new steel bar image and determine the position and orientation of the steel bars through the output of the model.

Images