Data processing method, data processing system and full-automatic intelligent cutter production equipment

Abstract

The invention discloses a data processing method, a data processing system and full-automatic intelligent cutter production equipment. The data processing method is achieved based on full-automatic intelligent tool production equipment, the full-automatic intelligent tool production equipment comprises a camera shooting assembly, a mechanical arm and an edging grinder, and the full-automatic intelligent tool production equipment is used for edging of tools. The data processing method comprises the following steps: instructing the manipulator to grab a to-be-edged cutter; acquiring first image data of the camera component, wherein the image content of the first image data is a to-be-edged cutter grabbed by a manipulator; identifying a tool type in the first image data; querying preset track data matched with the tool type, wherein the preset track data comprises a first track and a second track; instructing the sharpening grinder to move to a target position according to a first track; and instructing the manipulator to grab the to-be-edged cutter to move to the initial position, and edging on the edging grinder according to a second track.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a data processing method, a data processing system and a fully automatic intelligent tool production device. Background Art

[0002] In the global kitchen knife and scissors market, competition is increasingly fierce, and companies are facing the dual challenges of improving production efficiency and reducing costs. Especially under the current situation, the difficulty in recruiting ordinary workers and the rising labor costs have put forward higher requirements for production efficiency and cost control. In the production process of kitchen knives and scissors, blade sharpening is a crucial link, which directly affects the product quality of the knives.

[0003] Traditionally, knife edge sharpening mainly relies on manual operation, which is inefficient and difficult to ensure the consistency and quality of the edge, and even more difficult to mass produce. With the development of industrial automation technology, especially the application of robotics in industrial production, some companies have begun to try to use robots and grinders to perform automated knife edge sharpening.

[0004] However, the existing robot sharpening method has obvious limitations: due to the large differences in shapes and sizes between different tools, the robot cannot identify them, resulting in the need to manually adjust the position of the grinder during the sharpening process to adapt to different tools, which cannot meet the current needs of enterprises for automated manufacturing of knives and scissors. Summary of the invention

[0005] In order to overcome the above technical defects, the present invention provides a data processing method, a data processing system and a fully automatic intelligent tool production equipment.

[0006] In order to solve the above problems, the present invention is implemented according to the following technical solutions:

[0007] In a first aspect, the present invention provides a data processing method, which is implemented based on a fully automatic intelligent tool production device, wherein the fully automatic intelligent tool production device includes a camera assembly, a manipulator, and a sharpening grinder, and the fully automatic intelligent tool production device is used for sharpening a tool; the data processing method includes:

[0008] Instructing the manipulator to grab the tool to be sharpened;

[0009] Acquire first image data of the camera assembly, wherein the image content of the first image data is a cutting tool to be sharpened grasped by the manipulator;

[0010] identifying a tool type in the first image data;

[0011] Querying preset trajectory data matching the tool type, the preset trajectory data including a first trajectory and a second trajectory;

[0012] Instructing the sharpening machine to move to a target position along a first trajectory;

[0013] The manipulator is instructed to grab the tool to be sharpened and move it to an initial position, and sharpen the tool on the sharpening machine according to a second trajectory.

[0014] In combination with the first aspect, the present invention further provides a first specific implementation manner of the first aspect. Specifically, before identifying the tool type in the first image data, the data processing method further includes:

[0015] Recognize the first image data, output a first target object and a second target object, and generate corresponding first target detection frames and second target detection frames respectively; wherein the first target object is a blade of a tool to be sharpened, and the second target object is a handle of the tool to be sharpened;

[0016] Determining whether the position of the first target object is correct;

[0017] When the position of the first target object is abnormal, instructing the manipulator to adjust the tool to be sharpened;

[0018] When the position of the first target object is correct, determining whether the blade of the first target object faces outwards;

[0019] When the blade of the first target object faces inward, instructing the manipulator to adjust the tool to be sharpened;

[0020] When the blade of the first target object faces outward, the type of the tool in the first image data is identified.

[0021] In combination with the first aspect, the present invention further provides a second specific implementation of the first aspect. Specifically, after the blade sharpening is completed, the data processing method further includes:

[0022] After the sharpening is completed, the manipulator is instructed to move the sharpened tool to a quality inspection position;

[0023] Acquire second image data of the camera assembly, wherein the image content of the second image data is a sharpened tool grasped by the manipulator;

[0024] According to the second image data, an edge detection step is performed and an edge detection result is output.

[0025] In combination with the first aspect, the present invention further provides a third specific implementation manner of the first aspect, specifically, the edge sharpening detection step comprises:

[0026] Preprocessing the second image data to obtain a blade image;

[0027] Using an edge detection algorithm to perform edge detection on the blade image to obtain blade contour information;

[0028] The standard profile information is compared with the blade profile information to obtain the edge sharpening detection result.

[0029] In combination with the first aspect, the present invention further provides a fourth specific implementation of the first aspect. Specifically, the data processing method further includes:

[0030] According to the edge detection result, the edge cutting tool is moved to the corresponding material cutting position;

[0031] Instruct the manipulator to grab the next tool to be sharpened.

[0032] In a second aspect, the present invention further provides a data processing system, which is applied to a fully automatic intelligent tool production device, wherein the fully automatic intelligent tool production device includes a camera assembly, a manipulator and a sharpening machine, and the fully automatic intelligent tool production device is used for sharpening a tool; the data processing system includes:

[0033] A first instruction module, which is used to instruct the manipulator to grab the tool to be sharpened;

[0034] A first acquisition module, which is used to acquire first image data of the camera assembly, wherein the image content of the first image data is a tool to be sharpened grasped by the manipulator;

[0035] A first recognition module, which is used to recognize the type of tool in the first image data;

[0036] A query module, which is used to query preset trajectory data matching the tool type, wherein the preset trajectory data includes a first trajectory and a second trajectory;

[0037] A second instruction module, which is used to instruct the sharpening machine to move to a target position along a first trajectory;

[0038] The third instruction module is used to instruct the manipulator to grab the tool to be sharpened and move it to an initial position, and sharpen it on a sharpening machine according to a second trajectory.

[0039] In combination with the first aspect, the present invention further provides a first specific implementation manner of the first aspect, specifically, the data processing system further includes:

[0040] A second recognition module, which is used to recognize the first image data, output a first target object and a second target object, and generate corresponding first target detection frames and second target detection frames respectively; wherein the first target object is a blade of a tool to be sharpened, and the second target object is a handle of the tool to be sharpened;

[0041] A judging module, used to judge whether the position of the first target object is correct;

[0042] When the position of the first target object is abnormal, instructing the manipulator to adjust the tool to be sharpened;

[0043] When the position of the first target object is correct, determining whether the blade of the first target object faces outwards;

[0044] When the blade of the first target object faces inward, instructing the manipulator to adjust the tool to be sharpened;

[0045] When the blade of the first target object faces outward, the type of the tool in the first image data is identified.

[0046] In combination with the first aspect, the present invention further provides a second specific implementation manner of the first aspect, specifically, the data processing system further includes:

[0047] A fourth instruction module, which is used to instruct the manipulator to grab the sharpened tool and move it to a quality inspection position after the sharpening is completed;

[0048] A second acquisition module, which is used to acquire second image data of the camera assembly, the image content of the second image data is a sharpened tool grasped by the manipulator;

[0049] A detection module is used to perform a sharp edge detection step according to the second image data and output a sharp edge detection result.

[0050] In combination with the first aspect, the present invention further provides a third specific implementation manner of the first aspect, specifically, the edge sharpening detection step comprises:

[0051] Preprocessing the second image data to obtain a blade image;

[0052] Using an edge detection algorithm to perform edge detection on the blade image to obtain blade contour information;

[0053] The standard profile information is compared with the blade profile information to obtain the edge sharpening detection result.

[0054] In a third aspect, the present invention further provides a fully automatic intelligent tool production device, the fully automatic intelligent tool production device comprises a camera assembly, a manipulator and a sharpening machine, and the fully automatic intelligent tool production device is used for sharpening the tool;

[0055] The fully automatic intelligent tool production equipment comprises a controller connected to the camera assembly, the manipulator and the sharpening machine respectively, the controller is provided with a data processing system according to the second aspect, and the controller is used to implement the following data processing method, which comprises:

[0056] Instructing the manipulator to grab the tool to be sharpened;

[0057] Acquire first image data of the camera assembly, wherein the image content of the first image data is a cutting tool to be sharpened grasped by the manipulator;

[0058] identifying a tool type in the first image data;

[0059] Querying preset trajectory data matching the tool type, the preset trajectory data including a first trajectory and a second trajectory;

[0060] Instructing the sharpening machine to move to a target position along a first trajectory;

[0061] The manipulator is instructed to grab the tool to be sharpened and move it to an initial position, and sharpen the tool on the sharpening machine according to a second trajectory.

[0062] Compared with the prior art, the present invention has the following beneficial effects:

[0063] The present invention provides a data processing method, which is implemented based on a fully automatic intelligent tool production device, wherein the fully automatic intelligent tool production device includes a camera assembly, a manipulator and a sharpening machine, and the fully automatic intelligent tool production device is used for sharpening the tool. The data processing method includes: instructing the manipulator to grab the tool to be sharpened; acquiring the first image data of the camera assembly, wherein the image content of the first image data is the tool to be sharpened grabbed by the manipulator; identifying the tool type in the first image data; querying the preset trajectory data matching the tool type, wherein the preset trajectory data includes a first trajectory and a second trajectory; instructing the sharpening machine to move to a target position according to the first trajectory; instructing the manipulator to grab the tool to be sharpened and move it to an initial position, and sharpen it on the sharpening machine according to the second trajectory.

[0064] This data processing method is based on the collaborative work of the camera assembly, the manipulator and the sharpening machine, and is compatible with and matches different types of tools to perform automated sharpening operations of the tools. This data processing method accurately identifies the tool type through image recognition technology, automatically queries and applies the matching trajectory data, and ensures the accuracy of the sharpening process. This method not only reduces manual intervention and operational risks, reduces labor costs, has good adaptability, can handle different types of tools, meets the current needs of enterprise intelligent manufacturing, and is more practical. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] The specific embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings, wherein:

[0066] Figure 1 It is a flow chart of a data processing method of the present invention;

[0067] Figure 2 It is a three-dimensional schematic diagram of the fully automatic intelligent tool production equipment of the present invention;

[0068] Figure 3 It is a schematic diagram of the assembly of a sharpening machine of the fully automatic intelligent tool production equipment of the present invention;

[0069] In the figure:

[0070] 10-opening sharpener, 11-frame, 12-moving mechanism, 13-grinding machine;

[0071] 20-manipulator, 21-fixture. DETAILED DESCRIPTION

[0072] The following specifically illustrates the implementation mode of the present invention in conjunction with the accompanying drawings. The embodiments are provided for illustrative purposes only and cannot be understood as limiting the present invention. The accompanying drawings are provided for reference and illustration only and do not constitute a limitation on the scope of patent protection of the present invention, because many changes can be made to the present invention without departing from the spirit and scope of the present invention.

[0073] In the global kitchen knife and scissors market, as competition becomes increasingly fierce, companies are facing the dual pressure of improving production efficiency and reducing costs. Especially in the current economic environment, it is difficult to recruit ordinary workers and labor costs continue to rise, making it an urgent need for companies to improve production efficiency and control costs. In the production process of kitchen knives and scissors, blade sharpening is a crucial link, which directly affects the quality and performance of the knife.

[0074] Traditional knife edge sharpening mainly relies on manual operation, but this method is not only inefficient, but also difficult to ensure the consistency and quality of the edge, especially in large-scale production, the problem is more obvious. With the continuous advancement of industrial automation technology, especially the widespread application of robotics in the production field, many companies have begun to explore the use of robots and grinders to automate knife edge sharpening.

[0075] However, the existing robot sharpening method has significant limitations: due to the large differences in shapes and sizes of different tools, the robot cannot automatically identify the tool type, which requires manual adjustment of the grinder position during the sharpening process. The purpose is to reduce the complex programming and path adjustment of the robot, and to be able to flexibly respond to the processing needs of various tools to better adapt to the processing needs of different tools.

[0076] However, this reliance on manual intervention makes it difficult for existing automation solutions to meet the requirements of efficient and flexible production. Therefore, the existing blade sharpening technology urgently needs further innovation to break through the existing limitations in order to achieve a more accurate, efficient and intelligent automation solution.

[0077] For this purpose, refer to Figure 1 , an embodiment of the present invention provides a flow chart of a data processing method. The present invention mainly obtains image data of a tool to be sharpened through a camera assembly, identifies the tool type, and then queries the preset trajectory data matching it. Then, through the coordinated work of the manipulator and the sharpening grinder, the tool grabbing, positioning and sharpening process are controlled according to the preset trajectory, thereby realizing the automatic sharpening of the tool.

[0078] Embodiment 1

[0079] like Figure 2-3 As shown, a preferred structure of a fully automatic intelligent tool production equipment shown in the present invention.

[0080] In the present invention, the fully automatic intelligent tool production equipment is used for sharpening the blade of a tool. The fully automatic intelligent tool production equipment includes a camera assembly, a manipulator, a sharpening machine and a controller, and the controller is connected to the camera assembly, the manipulator and the sharpening machine respectively. The controller is deployed with the data processing system described below to execute the data processing method described below.

[0081] like Figure 2 As shown, the manipulator is arranged on the side of the sharpening machine, and the manipulator has a tool fixture, and the tool fixture uses a clamp to grab the tool to be sharpened. Specifically, the manipulator grabs the back of the tool to be sharpened, and the back of the tool is the position other than the blade and the handle.

[0082] Specifically, the manipulator moves the tool to be sharpened from one position to another through precisely controlled movements, such as from a loading position to a sharpening machine, or moves a tool that has been sharpened to a corresponding designated position, such as a quality inspection position or a unloading position. The controller generates control instructions based on the data processing results and transmits them to the manipulator. For example, the controller instructs the manipulator to grasp and position the tool based on the tool type and target trajectory, and instructs the sharpening machine to perform the sharpening operation of the tool according to the preset trajectory. Specifically, the manipulator is a commonly used device in this field and will not be described in detail here.

[0083] In a preferred embodiment, the fully automatic intelligent tool production equipment further includes a loading conveyor belt and a unloading conveyor belt, and the loading conveyor belt and the unloading conveyor belt are both provided with a plurality of placement racks for placing the tools. The loading conveyor belt conveys a plurality of tools to be sharpened to the side of the manipulator through the plurality of placement racks, and the manipulator grabs the tools to be sharpened from the placement rack on the loading conveyor belt. After the sharpening is completed, the manipulator places the sharpened tool on the placement rack of the unloading conveyor belt.

[0084] like Figure 2As shown, the sharpening machine is a grinding wheel grinder, which mainly grinds the tool through a high-speed rotating grinding wheel to complete the sharpening of the blade. In the automation system, the grinding wheel grinder works in coordination with the manipulator and the controller to ensure the accuracy and efficiency of the tool sharpening.

[0085] In a specific implementation, the sharpening machine includes a frame, a moving mechanism and a grinding wheel machine, the moving mechanism is fixedly connected to the upper surface of the frame, the grinding wheel machine is bolted to the upper surface of the moving mechanism, and the moving mechanism is used to drive the grinding wheel machine to move.

[0086] Specifically, the grinding machine is provided with a grinding wheel for sharpening, and the grinding wheel can be a flat grinding wheel or a cylindrical grinding wheel, or can be other types of grinding wheels. Specifically, the grinding machine has a control module, and the control module is used to control the rotation speed of the grinding machine and control the movement of the moving mechanism.

[0087] Specifically, the moving mechanism includes a transverse screw assembly and a longitudinal screw assembly, wherein the transverse screw assembly is fixedly connected to the top of the frame. The longitudinal screw assembly is connected to the transverse screw assembly, and the transverse screw assembly drives the longitudinal screw assembly to move on the X axis. The grinding machine is connected to the longitudinal screw assembly, and the longitudinal screw assembly drives the grinding machine to move on the Y axis. Through the coordination of the transverse screw assembly and the longitudinal screw assembly, the movement of the grinding machine on the X axis and the Y axis is realized.

[0088] The transverse screw assembly and the longitudinal screw assembly both include a servo motor, a screw, a guide rail, a slider, a mounting seat, and other components. The screw assembly is a well-known technology in the art and will not be described in detail here.

[0089] In a specific implementation, the grinding wheel of the sharpening grinder of the present invention is of mobile design. Since the length, width, thickness and other dimensions of each tool are different, the shape and cutting angle of the blade are also different. Adjusting the position of the grinder can ensure that the contact position and angle between the grinder and the tool are accurate, thereby obtaining an ideal blade. Different types of tools may require different sharpening processes, and the automatic adjustment grinder can flexibly cope with these differences, and for tools of the same type with small differences in specifications, there is no need to change / slightly adjust the processing path of the manipulator to meet the processing requirements of different tools.

[0090] It can be seen that the automatic adjustment of the mill position can be accurately adjusted according to the specific shape, size, angle, etc. of each tool to ensure that each tool can obtain the appropriate processing method. This method is particularly effective for tools with small differences in shape and size, and can flexibly respond to the processing needs of various tools without the need for complex programming and path adjustment of the robot. In production, especially for small batch tool production, this method is more flexible and economical. Because in comparison, changing the position of the mill is simple and easier to achieve. And for some tools, adjusting the position of the mill may be faster than adjusting the robot path, and adjustments can be made quickly without having to wait for the complex path of the robot to be designed.

[0091] In a specific implementation, the camera assembly includes a bracket and a camera device, the bracket can be fixedly mounted on one side of the manipulator, and the camera device is mounted on the bracket. The camera device can be an industrial camera or a CCD camera. In a preferred implementation, the camera assembly uses a high-resolution industrial camera, combined with an appropriate light source (such as an LED light source), to capture the image of the tool grasped by the manipulator in real time. The camera is calibrated by a lens to ensure that the image is free of distortion.

[0092] Embodiment 2

[0093] refer to Figure 1 , Embodiment 2 of the present invention provides a flow chart of a data processing method. The present invention provides a data processing technology, which is based on the collaborative work of a camera assembly, a manipulator, and a sharpening grinder, and is compatible with and matches different types of tools to perform automated sharpening operations of tools. This data processing method accurately identifies the type of tool through image recognition technology, automatically queries and applies the trajectory data that matches it, and ensures the accuracy of the sharpening process. This method not only reduces manual intervention and operational risks, reduces labor costs, but also has good adaptability, can handle different types of tools, meets the current needs of enterprise intelligent manufacturing, and is more practical.

[0094] The data processing method of the present invention can be executed by a data processing system, which can be implemented in the form of hardware and / or software. The system can be configured in an electronic device of the fully automatic intelligent tool production equipment, such as a controller. The data processing method of the present invention is implemented based on the fully automatic intelligent tool production equipment. Specifically, the fully automatic intelligent tool production equipment is exactly the same as that described in Example 1. Figure 1 As shown, the data processing method includes:

[0095] S100: Instructing the robot arm to grab the cutting tool to be sharpened.

[0096] In a specific implementation, the controller sends a command to the robot to grab the tool to be sharpened. For example, the robot control system, such as the SPEL+ language instruction set of the EPSON robot, is used to write a control program so that the robot can automatically identify and grab the tool to be sharpened in the set position of the feeding conveyor. After the grabbing is completed, the robot controller sends a confirmation signal, indicating that the grabbing action is completed.

[0097] In the specific implementation, in the manipulator control system, the joint axis position of the manipulator needs to be set. This includes determining the original position, target position and path point of the manipulator. These position information will be used to guide the manipulator to accurately move to the position of the tool to be sharpened and then grab it. The manipulator is a well-known technology in the field and will not be explained in detail.

[0098] S200: Acquire first image data of the camera assembly, where the image content of the first image data is a cutting tool to be sharpened grasped by a robot arm.

[0099] In a specific implementation, after the robot grabs the tool to be sharpened, it will automatically move to the shooting position of the camera assembly, which has been preset, preferably in front of the camera lens. After the robot reaches the shooting position, the robot feeds back a shooting signal to the controller to instruct the camera shutter to work and shoot.

[0100] S300: Identify the tool type in the first image data.

[0101] In the present invention, the controller database has pre-stored the tool data of the enterprise's own products, and the tool data is the tool type suitable for processing by the fully automatic intelligent tool production equipment. The purpose of identifying the tool type is to determine the type of tool grasped by the manipulator so as to call the corresponding tool processing data.

[0102] It is understandable that the type of tool grasped by the robot is automatically determined by image recognition technology. After the tool type is identified, the controller will query the processing data matching the tool type from the database, including its processing trajectory, grinding parameters, etc., and use it for subsequent processing.

[0103] In a preferred implementation, an image processing algorithm, such as a convolutional neural network (CNN), can be used to extract features from the image. The types of knives (such as slicing knives, boning knives, boning knives, etc.) are classified and identified through a trained model. The model makes a judgment based on the shape, size, angle and other features of the tool. The tool data of the company's own products has been pre-stored in the controller's database. The tool data is the type of tool suitable for processing by the fully automatic intelligent tool production equipment.

[0104] In a specific implementation, the S300 includes:

[0105] S310: performing image preprocessing on the first image data, including:

[0106] Denoising: Use filtering techniques to remove background noise and improve image quality.

[0107] Cropping and normalization: Crop the image according to the grasping position of the tool, keep only the tool area, unify the image size and brightness, and make the image more standardized.

[0108] Grayscale or color adjustment: Convert the image to a grayscale image or enhance the image contrast to facilitate subsequent feature extraction.

[0109] S320: Feature Extraction

[0110] Using deep learning algorithms such as convolutional neural networks (CNN), a model specifically for tool type classification is trained. The model learns the shape, edge, angle and other features of the tool, and can identify different types of tools by learning from large-scale labeled data sets. Each tool type has its own unique features in the image, and CNN can automatically extract these features through multiple convolutional layers, classify through fully connected layers, and output recognition results.

[0111] When performing tool type recognition, the model's training data should cover the tool types of all the company's tool products. During the training process, the model will learn details such as the size, shape, and angle of the tool, and improve recognition accuracy through continuous optimization.

[0112] S330: Input the collected first image into the trained CNN model, determine the type of tool according to the features of the first image, and output the category of the tool.

[0113] In the present invention, the database is mainly used to store various data related to tool processing. Through the database management system (DBMS), the controller can quickly find and obtain corresponding tool processing data according to the identified tool type, and provide accurate processing instructions for the fully automatic intelligent tool production equipment.

[0114] S400: Query preset trajectory data matching the tool type, wherein the preset trajectory data includes a first trajectory and a second trajectory.

[0115] In a specific implementation, the database is queried according to the identified tool type, and preset trajectory data matching the tool type is returned to control the motion path of the sharpening machine and the manipulator, thereby achieving high-precision tool sharpening operations. The data table in the database should store the mapping relationship between the tool type and the processing trajectory data, so as to dynamically query and call the relevant processing trajectory according to the tool type.

[0116] It is understandable that in step S300, the image identifies the tool type, which will be used as a key field to query the database to match the corresponding preset trajectory data. Then in step S400, the controller obtains the processing trajectory data and processing parameters matching the tool type by querying the database.

[0117] In one example, the controller further queries the tool processing trajectory table (ToolProcessingTrajectory) based on the query results, finds the processing trajectory ID (such as "G001") that matches the "kitchen knife", and extracts the corresponding trajectory path data of the tool (such as the robot from "point A→point B→point C→point D").

[0118] It is understood that the first trajectory refers to the path of the sharpening machine moving from the initial position to the target position. The first trajectory includes the coordinates of the target position, which is achieved by moving the transverse screw assembly and the longitudinal screw assembly of the sharpening machine.

[0119] It is understood that the second trajectory refers to the path that the robot takes to perform the sharpening operation on the sharpening machine after grabbing the tool to be sharpened. This trajectory needs to be precisely controlled to ensure that the tool is sharpened at the correct position and angle on the grinder. During the sharpening process, the robot needs to move the tool along the second trajectory. This path usually includes multiple stages, such as approach, contact, sharpening, and withdrawal.

[0120] Specifically, both the first trajectory and the second trajectory can be implemented by those skilled in the art and will not be described in detail.

[0121] S500: Instruct the sharpening machine to move to a target position along a first trajectory.

[0122] S600: Instruct the robot arm to grab the tool to be sharpened and move it to an initial position, and sharpen it on a sharpening machine according to a second trajectory.

[0123] Specifically, the present invention describes each step of the data processing method in detail.

[0124] S100: Instructing the robot arm to grab the cutting tool to be sharpened.

[0125] S200: Acquire first image data of the camera assembly, where the image content of the first image data is a cutting tool to be sharpened grasped by a robot arm.

[0126] S300: Identify the tool type in the first image data.

[0127] S400: Query preset trajectory data matching the tool type, wherein the preset trajectory data includes a first trajectory and a second trajectory.

[0128] S500: Instruct the sharpening machine to move to a target position along a first trajectory.

[0129] According to the queried first trajectory data, the controller issues a command to direct the open-blade grinder to move to the target position.

[0130] S600: Instruct the robot arm to grab the tool to be sharpened and move it to an initial position, and sharpen it on a sharpening machine according to a second trajectory.

[0131] According to the second trajectory data, the robot is instructed to move the substitute sharpening tool to the sharpening grinder and perform the sharpening operation according to the second trajectory.

[0132] This data processing method is based on the collaborative work of the camera assembly, the manipulator and the sharpening machine, and is compatible with and matches different types of tools to perform automated sharpening operations of the tools. This data processing method accurately identifies the tool type through image recognition technology, automatically queries and applies the matching trajectory data, and ensures the accuracy of the sharpening process. This method not only reduces manual intervention and operational risks, reduces labor costs, has good adaptability, can handle different types of tools, meets the current needs of enterprise intelligent manufacturing, and is more practical.

[0133] In a preferred implementation, before identifying the tool type in the first image data, the data processing method further includes:

[0134] S10: Identify the first image data, output a first target object and a second target object, and generate corresponding first target detection frames and second target detection frames respectively; wherein the first target object is the blade of the tool to be sharpened, and the second target object is the handle of the tool to be sharpened.

[0135] In a specific implementation, a deep learning algorithm, such as YOLO (You Only Look Once) or SSD (Single Shot MultiBox Detector), can be used to perform real-time analysis on the captured second image data to identify the blade and handle. After detecting the target object, the algorithm will mark the location of the target object in the video frame and generate a bounding box to identify these objects. The system will classify the detected target objects into the blade (first target object) and the handle (second target object) according to the preset classification criteria.

[0136] In a specific implementation, the target detection model is a Yolov5s model, which has identified the first target object and the second target object.

[0137] S20: Determine whether the position of the first target object is correct.

[0138] It is understandable that the blade of the tool is sharpened at the blade, and most blades are irregular shapes, with a certain arc shape. Therefore, the left and right distribution of the handle and the blade is very critical for the sharpening of the tool. If the handle and the blade are wrongly swapped, the tool will break or the grinder will be damaged during the sharpening process.

[0139] In a specific implementation, the preset trajectory data matching the tool type is queried, and the preset trajectory data also includes other tool processing parameters, the tool handle is located on the left side of the tool, and the blade is located on the right side of the tool. By judging whether the position of the first target object is located on the right side of the tool, it is determined whether the position of the first target object is correct, abnormal, etc.

[0140] S21: When the position of the first target object is abnormal, instruct the manipulator to adjust the tool to be sharpened.

[0141] In one example, the manipulator can rotate the fixture to make the sharpening tool normal. On the other hand, the adjustment can be to replace a new tool to be sharpened.

[0142] S22: When the position of the first target object is correct, determine whether the blade of the first target object is facing outwards.

[0143] It is understandable that after confirming that the blade is in the correct position, the direction of the blade is further checked to ensure that the blade is facing outwards, which is a necessary condition for the sharpening operation. The blade of the tool is sharpened at the blade of the blade, not the handle. The robot grips the blade from the back of the blade, so it is necessary to determine whether the blade is facing outwards, otherwise the sharpening process is invalid.

[0144] S31: When the blade of the first target object faces inward, instruct the manipulator to adjust the tool to be sharpened.

[0145] S32: When the blade of the first target object faces outward, execute S200 to identify the type of the tool in the first image data.

[0146] In actual production scenarios, due to the feeding conveyor belt and the placement rack, the possibility of tool position failure is low. The above steps are set as a bottom-up detection to avoid abnormal situations and minimize the damage to the equipment and the impact on the processing progress caused by abnormal events.

[0147] In the present invention, after the sharpening is completed, the data processing method further includes:

[0148] S700: After the sharpening is completed, the robot is instructed to move the sharpening tool to a quality inspection position.

[0149] In the specific implementation, after the tool is sharpened, the control system generates a movement command to instruct the manipulator to move the tool from the sharpening position to the set quality inspection position. After receiving the movement command, the manipulator moves the tool to the quality inspection position according to the planned path.

[0150] Specifically, in the quality inspection position, the tool is located in front of the camera lens to reduce image distortion. After the robot moves to the quality inspection position, it feeds back information to the controller.

[0151] S800: Acquire second image data of the camera assembly, where the image content of the second image data is a sharpened tool grasped by the robot arm.

[0152] S900: Execute a sharp edge detection step according to the second image data, and output a sharp edge detection result.

[0153] In a specific implementation, before executing the edge detection step, S900 further includes a step of preprocessing the second image data at S910, including:

[0154] S911: Convert the second image data into a grayscale image and output it, in order to reduce computational complexity.

[0155] The converted grayscale image is output for use in subsequent steps. Grayscale images reduce the amount of data, thereby reducing computational complexity.

[0156] S912: Use a guided filter to enhance edge information in the grayscale image in order to maintain edge sharpness.

[0157] In the specific implementation, a guided filter is applied to process the grayscale image. The filter can adjust the filter strength according to the local statistical characteristics of the image and enhance the edge information. The input of the guided filter includes the grayscale image and the edge weight image, and the output is the enhanced image. The guided filter can maintain the sharpness of the edge while reducing the noise in the non-edge area, which is very important for subsequent edge detection and contour segmentation.

[0158] S913: Apply the Canny algorithm to detect the tool edge in the grayscale image and output the edge image.

[0159] In the specific implementation, the Canny edge detection algorithm performs preliminary edge positioning on the tool image to be detected and obtains its edge point set. Specifically, the specific steps of the Canny operator are as follows:

[0160] 1. Gaussian smoothing of the image to obtain the smoothed image I(x, y);

[0161] 2. Use the first-order partial derivative to obtain the grayscale gradient gx and gy in the horizontal and vertical directions of the image, where the gradient amplitude M(x, y) and the gradient direction θ(x, y) can be expressed as:

[0162]

[0163] 3. Perform non-maximum suppression on the obtained gradient amplitude;

[0164] 4. Use the hysteresis threshold method to screen all possible edge points. Define two thresholds, Max and Min, for the gradient amplitude. If it is less than Min, the edge point is eliminated; if it is greater than Max, the point is a definite edge and is retained; if it is between max and min, if the point is connected to the definite edge, the point is retained, otherwise the point is discarded.

[0165] Therefore, the Canny operator obtains the tool edge contour with continuous edges, excludes discontinuous edge points, and also discards false edge points.

[0166] S914: Perform contour tracing on the edge image, and output a final blade image including the blade edge contour.

[0167] In the present invention, the edge obtained by using pixel-level operators such as Canny is at the pixel level, which is a continuous edge composed of several single pixels connected. Subsequent processing needs to convert discrete edges into ordered edge storage, so it is necessary to track the contour of the edge image.

[0168] In the specific implementation, the tool contour tracking algorithm is designed based on the eight-neighborhood method. The specific steps of contour tracking are as follows:

[0169] 1) Based on pixel-level operators, discrete edge points are obtained.

[0170] 2) According to the characteristics of the tool blade image, the lower boundary edge point of the image is set as the starting point of the contour.

[0171] 3) Search for the next edge point in a clockwise direction using the eight neighborhoods around the starting point.

[0172] 4) The termination condition is that there are no edge points in the eight neighborhoods.

[0173] In a specific implementation, the edge detection step includes:

[0174] S920: Perform edge detection on the blade image using an edge detection algorithm to obtain blade contour information.

[0175] In a specific implementation, a sub-pixel edge detection algorithm is used based on Sigmoid function fitting. Sigmoid function has the characteristics of continuity, smoothness, monotony and is widely used. In order to improve the positioning accuracy of the edge of the tool contour, floating point numbers can be used to represent the position coordinates of the edge points. Compared with using integers to represent the edge at the pixel level, the positioning accuracy of the edge can be significantly improved.

[0176] In the specific implementation, the edge model of the Sigmoid function is established as:

[0177]

[0178] Where a is the difference between the maximum and minimum grayscale values; b is the translation distance of the S function in the x-axis direction; c represents the tilt of the edge, which is generally 0.5; d is the translation distance of the S function in the y-axis direction, that is, the minimum grayscale value of the edge.

[0179] Obtain the grayscale value and x-coordinate value within the edge range, and use the nonlinear least squares method to solve the acquired data. The parameters a, b, c, and d in the edge model can be obtained in this way. To locate the sub-pixel edge position of the contour image, Taking the first-order derivative, we get:

[0180]

[0181] General Taking the derivative of y and setting it equal to 0, we get:

[0182] Depend on We can get: The y value obtained at this time is the coordinate of the point where the slope is the largest in the edge. According to the definition of the gradient edge concept, the position of this point is the sub-pixel edge position.

[0183] S930: Compare the standard profile information with the blade profile information to obtain a sharpening detection result.

[0184] In a specific implementation, the standard profile information is usually an ideal blade profile generated based on historical data, product design requirements or high-precision measuring equipment. It represents the ideal shape, curvature, angle and other geometric features that the tool should have. The average blade standard profile can be generated by collecting images of qualified tools produced in multiple batches, or the standard profile can be directly obtained by precision measuring instruments.

[0185] In a specific implementation, an image comparison algorithm, such as template matching, feature point matching or geometric comparison method, is used to compare the detected blade profile with the standard profile, and the difference between the blade profile and the standard profile is analyzed, including deviations in shape, size and position.

[0186] In another specific implementation, the contour matching algorithm uses dynamic time warping (DTW) or Hausdorff distance to measure the similarity between two contours. This method can determine the degree of contour matching by calculating the minimum distance between the two.

[0187] Hausdorff distance: Calculate the maximum and minimum distance between the standard contour and the blade contour. If the distance is less than the preset threshold, the blade is considered to meet the standard.

[0188] Specifically, standard contour data loading: load the standard blade contour data stored in the database into the memory. Contour comparison: use the selected contour comparison algorithm to compare the standard contour and the detected blade contour point by point. Error calculation and result judgment: calculate the error value after comparison. If the error is within the allowable range, it is considered that the blade meets the blade sharpening standard. Otherwise, it means that the tool sharpening is unqualified and may need to be reprocessed or trimmed.

[0189] In a preferred implementation, the data processing method further includes:

[0190] S1000: According to the sharpening detection result, the sharpening tool is moved to the corresponding material unloading position.

[0191] In the specific implementation, according to the edge sharpening detection result output in step S900, the blade of each tool is analyzed to see whether it meets the quality standard. According to the analysis result, a classification instruction is generated, and qualified tools and unqualified tools are moved to different unloading positions respectively. According to the classification instruction and path, the robot moves the tool to the specified unloading position.

[0192] S2000: Instruct the robot to grab the next tool to be sharpened.

[0193] In the present invention, in order to realize continuous operation, the above steps are cycled until the tool processing is completed.

[0194] The present invention also provides a data processing system, which is applied to a fully automatic intelligent tool production device, wherein the fully automatic intelligent tool production device includes a camera assembly, a manipulator and a sharpening machine, and the fully automatic intelligent tool production device is used for sharpening the tool. The data processing system is used to implement the above-mentioned data processing method. Specifically, the data processing system includes:

[0195] A first instruction module, which is used to instruct the manipulator to grab the tool to be sharpened;

[0196] A first acquisition module, which is used to acquire first image data of the camera assembly, wherein the image content of the first image data is a tool to be sharpened grasped by the manipulator;

[0197] A first recognition module, which is used to recognize the type of tool in the first image data;

[0198] A query module, which is used to query preset trajectory data matching the tool type, wherein the preset trajectory data includes a first trajectory and a second trajectory;

[0199] A second instruction module, which is used to instruct the sharpening machine to move to a target position along a first trajectory;

[0200] The third instruction module is used to instruct the manipulator to grab the tool to be sharpened and move it to an initial position, and sharpen it on a sharpening machine according to a second trajectory.

[0201] Preferably, the data processing system further includes:

[0202] A second recognition module, which is used to recognize the first image data, output a first target object and a second target object, and generate corresponding first target detection frames and second target detection frames respectively; wherein the first target object is a blade of a tool to be sharpened, and the second target object is a handle of the tool to be sharpened;

[0203] A judging module, used to judge whether the position of the first target object is correct;

[0204] When the position of the first target object is abnormal, instructing the manipulator to adjust the tool to be sharpened;

[0205] When the position of the first target object is correct, determining whether the blade of the first target object faces outwards;

[0206] When the blade of the first target object faces inward, instructing the manipulator to adjust the tool to be sharpened;

[0207] When the blade of the first target object faces outward, the type of the tool in the first image data is identified.

[0208] Preferably, the data processing system further includes:

[0209] A fourth instruction module, which is used to instruct the manipulator to grab the sharpened tool and move it to a quality inspection position after the sharpening is completed;

[0210] A second acquisition module, which is used to acquire second image data of the camera assembly, the image content of the second image data is a sharpened tool grasped by the manipulator;

[0211] A detection module is used to perform a sharp edge detection step according to the second image data and output a sharp edge detection result.

[0212] Preferably, the edge sharpening detection step comprises:

[0213] Preprocessing the second image data to obtain a blade image;

[0214] Using an edge detection algorithm to perform edge detection on the blade image to obtain blade contour information;

[0215] The standard profile information is compared with the blade profile information to obtain the edge sharpening detection result.

[0216] For other structures of a data processing method, a data processing system and a fully automatic intelligent tool production equipment described in this embodiment, refer to the prior art.

[0217] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Therefore, any modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the technical solution of the present invention shall still fall within the scope of the technical solution of the present invention.

Claims

1. A data processing method, characterized in that: The data processing method is implemented based on a fully automatic intelligent tool production device, the fully automatic intelligent tool production device includes a camera assembly, a manipulator and a sharpening machine, and the fully automatic intelligent tool production device is used for sharpening the tool; the data processing method includes: Instructing the manipulator to grab the tool to be sharpened; Acquire first image data of the camera assembly, wherein the image content of the first image data is a cutting tool to be sharpened grasped by the manipulator; identifying a tool type in the first image data; Querying preset trajectory data matching the tool type, the preset trajectory data including a first trajectory and a second trajectory; Instructing the sharpening machine to move to a target position along a first trajectory; The manipulator is instructed to grab the tool to be sharpened and move it to an initial position, and sharpen the tool on the sharpening machine according to a second trajectory.

2. A data processing method according to claim 1, characterized in that: Before identifying the tool type in the first image data, the data processing method further includes: Recognize the first image data, output a first target object and a second target object, and generate corresponding first target detection frames and second target detection frames respectively; wherein the first target object is a blade of a tool to be sharpened, and the second target object is a handle of the tool to be sharpened; Determining whether the position of the first target object is correct; When the position of the first target object is abnormal, instructing the manipulator to adjust the tool to be sharpened; When the position of the first target object is correct, determining whether the blade of the first target object faces outwards; When the blade of the first target object faces inward, instructing the manipulator to adjust the tool to be sharpened; When the blade edge of the first target object faces outward, the type of the tool in the first image data is identified.

3. A data processing method according to claim 1, characterized in that: After the sharpening is completed, the data processing method further includes: After the sharpening is completed, the manipulator is instructed to move the sharpened tool to a quality inspection position; Acquire second image data of the camera assembly, wherein the image content of the second image data is a sharpened tool grasped by the manipulator; According to the second image data, an edge detection step is performed and an edge detection result is output.

4. A data processing method according to claim 3, characterized in that: The edge detection step comprises: Preprocessing the second image data to obtain a blade image; Using an edge detection algorithm to perform edge detection on the blade image to obtain blade contour information; The standard profile information is compared with the blade profile information to obtain the edge sharpening detection result.

5. A data processing method according to claim 4, characterized in that: The data processing method further includes: According to the edge detection result, the edge cutting tool is moved to the corresponding material cutting position; Instruct the manipulator to grab the next tool to be sharpened.

6. A data processing system, characterized in that: The data processing system is applied to a fully automatic intelligent tool production device, wherein the fully automatic intelligent tool production device includes a camera assembly, a manipulator and a sharpening machine, and the fully automatic intelligent tool production device is used for sharpening the tool; the data processing system includes: A first instruction module, which is used to instruct the manipulator to grab the tool to be sharpened; A first acquisition module, which is used to acquire first image data of the camera assembly, wherein the image content of the first image data is a tool to be sharpened grasped by the manipulator; A first recognition module, which is used to recognize the type of tool in the first image data; A query module, which is used to query preset trajectory data matching the tool type, wherein the preset trajectory data includes a first trajectory and a second trajectory; A second instruction module, which is used to instruct the sharpening machine to move to a target position along a first trajectory; The third instruction module is used to instruct the manipulator to grab the tool to be sharpened and move it to an initial position, and sharpen it on a sharpening machine according to a second trajectory.

7. A data processing system according to claim 6, characterized in that: The data processing system further comprises: A second recognition module, which is used to recognize the first image data, output a first target object and a second target object, and generate corresponding first target detection frames and second target detection frames respectively; wherein the first target object is a blade of a tool to be sharpened, and the second target object is a handle of the tool to be sharpened; A judging module, used to judge whether the position of the first target object is correct; When the position of the first target object is abnormal, instructing the manipulator to adjust the tool to be sharpened; When the position of the first target object is correct, determining whether the blade of the first target object faces outwards; When the blade of the first target object faces inward, instructing the manipulator to adjust the tool to be sharpened; When the blade edge of the first target object faces outward, the type of the tool in the first image data is identified.

8. A data processing system according to claim 7, characterized in that: The data processing system further comprises: A fourth instruction module, which is used to instruct the manipulator to grab the sharpened tool and move it to a quality inspection position after the sharpening is completed; A second acquisition module, which is used to acquire second image data of the camera assembly, the image content of the second image data is a sharpened tool grasped by the manipulator; A detection module is used to perform a sharp edge detection step according to the second image data and output a sharp edge detection result.

9. A data processing system according to claim 8, characterized in that: The edge detection step comprises: Preprocessing the second image data to obtain a blade image; Using an edge detection algorithm to perform edge detection on the blade image to obtain blade contour information; The standard profile information is compared with the blade profile information to obtain the edge sharpening detection result.

10. A fully automatic intelligent tool production equipment, characterized in that: The fully automatic intelligent tool production equipment includes a camera assembly, a manipulator and a sharpening machine, and the fully automatic intelligent tool production equipment is used for sharpening the tool; The fully automatic intelligent tool production equipment comprises a controller connected to the camera assembly, the manipulator and the sharpening machine respectively, and the controller is provided with a data processing system according to claim 9, and the controller is used to implement the following data processing method, which comprises: Instructing the manipulator to grab the tool to be sharpened; Acquire first image data of the camera assembly, wherein the image content of the first image data is a cutting tool to be sharpened grasped by the manipulator; identifying a tool type in the first image data; Querying preset trajectory data matching the tool type, the preset trajectory data including a first trajectory and a second trajectory; Instructing the sharpening machine to move to a target position along a first trajectory; The manipulator is instructed to grab the tool to be sharpened and move it to an initial position, and sharpen the tool on the sharpening machine according to a second trajectory.

Images