Object edge polishing method, device, robot, and storage medium
By using image recognition and calculation to determine the edge to be polished, the automated polishing equipment can accurately polish different workpieces, solving the problem of poor adaptability of traditional equipment and improving polishing efficiency and accuracy.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUANGDONG XIRUI INTELLIGENT TECH CO LTD
- Filing Date
- 2025-10-28
- Publication Date
- 2026-07-31
AI Technical Summary
Existing automated grinding equipment cannot adapt to the grinding areas of different workpieces, resulting in inaccurate grinding, accelerated wear of parts, and low efficiency.
By acquiring an image of the target object, identifying the object type and extracting the outer surface edge, using a computing component to determine the edge to be polished, and controlling the polishing component to perform precise polishing.
It improves the accuracy and efficiency of polishing and reduces wear on parts caused by unnecessary movement.
Smart Images
Figure CN121374294B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of robotics technology, specifically to methods, apparatus, robots, and storage media for grinding the edges of objects. Background Technology
[0002] In industrial settings, grinding is a common step in workpiece processing to deburr and polish the edges. Different workpieces have different edge shapes, requiring different travel paths for the grinding machine and other grinding components. Traditional manual grinding methods require operators to manually move the grinding components, resulting in low efficiency. While some automated grinding equipment can grind automatically along predetermined paths, the grinding area varies depending on the size and shape of the workpiece. Predetermined travel paths cannot adapt to different workpieces, leading to inaccurate grinding after changing workpieces and accelerated wear of the grinding components due to unnecessary movement. Summary of the Invention
[0003] This application discloses a method, apparatus, robot, and storage medium for grinding the edges of objects, which can grind specific edges to be ground in the object's edge contour according to the type of the object, thereby improving the accuracy and efficiency of grinding.
[0004] This application discloses a method for polishing the edge of an object, the method comprising:
[0005] Acquire an image of the target object to be polished;
[0006] Identifying the type of the target object based on the image includes:
[0007] Extract the target corner points of the target object from the image.
[0008] The target corner points in the image are matched with template corner points in a template library. The template library includes object information for at least two different types of objects. The object information includes a type identifier representing the object type and a template corner point corresponding to each type identifier.
[0009] The type identifier corresponding to the template corner point that matches the target corner point is determined as the type of the target object;
[0010] Extract the outer surface edge of the target object from the image;
[0011] Based on the type of the target object and the curvature of the outer surface edge of the target object, determine the edge to be polished in the outer surface edge;
[0012] The grinding component is controlled to grind the edge to be ground.
[0013] As an optional implementation, extracting the target corner points of the target object from the image includes:
[0014] The target corner points of the target object are extracted from the image using the Harris operator.
[0015] As an optional implementation, determining the edge to be polished in the outer surface edge according to the type of the target object includes:
[0016] Determine the curvature sensitivity value of the target object;
[0017] The curvature of several edge data points on the outer surface edge is calculated based on the curvature sensitivity value of the target object.
[0018] The linear shape corresponding to each edge data point among the plurality of edge data points is determined based on the curvature;
[0019] Edge data points whose linear shape corresponds to the type of the target object are identified from the plurality of edge data points and are used as the edges to be polished.
[0020] As an alternative implementation, the line shape corresponding to the edge data points includes a straight line or an arc.
[0021] As an optional implementation, determining the linear shape corresponding to each of the plurality of edge data points based on curvature includes:
[0022] Among the given number of edge data points, the line shape corresponding to the edge data points whose curvature satisfies the first curvature range is determined as a straight line; or,
[0023] Among the plurality of edge data points, the line shape corresponding to the edge data point whose curvature satisfies the second curvature range is determined as an arc.
[0024] As an optional implementation, determining the curvature sensitivity value of the target object includes:
[0025] The curvature sensitivity value of the target object is determined based on the type of the target object.
[0026] This application discloses an object edge polishing device, the device comprising:
[0027] The acquisition module is used to acquire images of the target object to be polished.
[0028] The recognition module is used to identify the type of the target object based on the image;
[0029] The extraction module is used to extract the outer surface edge of the target object from the image;
[0030] The determination module is used to determine the edge to be polished in the outer surface edge based on the type of the target object and the curvature of the outer surface edge of the target object;
[0031] The control module is used to control the grinding component to grind the edge to be ground.
[0032] This application discloses a robot. The electronic device includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor enables the processor to implement any of the object edge polishing methods disclosed in this application.
[0033] This application discloses a computer-readable storage medium storing a computer program thereon, characterized in that the computer program, when executed by a processor, implements any of the object edge polishing methods disclosed in this application.
[0034] Compared with related technologies, the embodiments of this application have the following beneficial effects:
[0035] In this embodiment, by acquiring an image of the target object to be polished, the type of the target object is identified from the image based on visual recognition, and the outer surface edge of the target object is extracted. Therefore, the edge to be polished can be determined based on the type of the target object and the curvature of the outer surface edge, and the polishing component can be controlled to polish this edge. In other words, it is possible to polish specific edges within the object's contour according to the object's type, improving the accuracy and efficiency of polishing. Attached Figure Description
[0036] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0037] Figure 1 This is a schematic diagram of the architecture of a robot disclosed in an embodiment of this application.
[0038] Figure 2 This is a schematic flowchart of an object edge polishing method disclosed in an embodiment of this application.
[0039] Figure 3 This is an example diagram of an edge to be polished, as disclosed in an embodiment of this application.
[0040] Figure 4 This is a schematic flowchart of a method for identifying target object types based on images, as disclosed in an embodiment of this application.
[0041] Figure 5 This is an example diagram illustrating the rotation and scaling of the interior angles of a triangle as disclosed in an embodiment of this application.
[0042] Figure 6 This is a schematic flowchart of a method for determining the edge to be polished in the outer surface edge, as disclosed in the application embodiment.
[0043] Figure 7 This is a schematic diagram of the structure of an object edge grinding device disclosed in an embodiment of this application.
[0044] Figure 8 This is a schematic diagram of the structure of a robot disclosed in an embodiment of this application. Detailed Implementation
[0045] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0046] It should be noted that the terms "comprising" and "having," and any variations thereof, in the embodiments and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.
[0047] This application discloses a method, apparatus, robot, and storage medium for grinding the edges of objects, which can grind specific edges to be ground in the object's edge contour according to the type of the object, thereby improving the accuracy and efficiency of grinding.
[0048] Please see Figure 1 , Figure 1 This is a schematic diagram of the architecture of a robot disclosed in an embodiment of this application. Figure 1As shown, the robot includes at least a camera component 110, a computing component 120, and a polishing component 130. The camera component 110 may include a camera or other imaging device capable of capturing images of the target object to be polished. The field of view of the imaging device varies depending on the working distance. Optionally, the camera component 110 can adjust its field of view to capture only the edge contour of a single outer surface of the target object. For example, if the top surface of the target object is to be polished, the camera component 110 can capture only the outer surface of the top surface of the target object without needing to magnify the field of view to capture a three-dimensional view of the entire target object. The adjustment of the field of view can be achieved manually by the operator moving the distance between the camera component 110 and the target object, or by the robot moving the distance between the camera component 110 and the target object using a telescopic mechanism or other moving structure, or by manually or automatically adjusting the focal length of the camera component 110; the specific method is not limited.
[0049] The computing component 120 includes components with computing capabilities, such as a central processing unit, a microprocessor, etc., without being specifically limited, and is used to perform various data transmission or processing operations disclosed in the embodiments of this application.
[0050] The polishing component 130 is a component used to perform polishing actions, and may include one or more polishing tools such as sandpaper, file, and flap wheel, without being specifically limited.
[0051] In this embodiment, the imaging component 110, the calculation component 120, and the polishing component 120 can be separate units or integrated into the same mechanical device, with communication connections between the components for data transmission.
[0052] Please see Figure 2 , Figure 2 This is a schematic flowchart of an object edge polishing method disclosed in an embodiment of this application. Figure 2 The method shown can be executed by the aforementioned computing unit, and there are no specific limitations. For example... Figure 2 As shown, the method may include the following steps:
[0053] 210. Obtain an image of the target object to be polished.
[0054] In this embodiment, the image includes the target object to be polished. The computing component can acquire the image captured by the polishing component through any one or more existing methods, some examples of which are given below.
[0055] The computing unit can acquire images directly from the sensor interface of the polishing component, such as a USB interface, GigE interface, or Camera Link interface. Alternatively, the computing unit can acquire images through a file system, such as local storage devices (SD cards, silver disks, etc.) or network shared storage.
[0056] 220. Identify the type of target object based on image.
[0057] In this embodiment, the type of the target object may include product types, such as screws, nuts, bows and arrows, etc., and is not specifically limited. The image includes features such as the shape, outline, and color of the target object, which can be used to identify the type of the target object.
[0058] As an optional implementation, the computing component can identify the type of the target object based on feature extraction. For example, it can extract one or more features of the target object from the image, such as edge features, corner features, and scale-invariant features, and then match the extracted features with template features in a database, thereby determining the type of the target object based on the matched template features. Template-matching-based type identification has relatively low computational cost and speed.
[0059] As an alternative implementation, the computing unit can also utilize a pre-trained neural network model to perform type recognition on images including the target object. This neural network model can be stored locally on the computing unit or stored on a cloud server and accessed by the computing unit; the specific method is not limited. The neural network can be trained using supervised, unsupervised, or model distillation methods to acquire the ability to distinguish different types of objects. After acquiring an image including the target object, the computing unit inputs the image into the trained neural network and then obtains the type of the target object from the neural network's output. The method based on neural networks offers better generalization capabilities and can flexibly adapt to different types of objects to be polished in different polishing scenarios, independent of feature rule design.
[0060] 230. Extract the outer surface edge of the target object from the image.
[0061] In this embodiment, the outer surface edge of the target object may include all or part of the edge of any one outer surface of the target object. The image captured by the imaging component may include one or more outer surfaces of the target object. When the image includes one outer surface, the calculation component may extract all or part of the edge from one outer surface; when the image includes at least two outer surfaces, the calculation component may extract the edge of the outer surface with the largest area from the two or more outer surfaces; or extract the outer surface edge specified by the user based on user input; or extract all outer surface edges.
[0062] In the embodiments of this application, the computing device may also extract the outer surface edge of the target object based on one or more feature matching methods such as gradient operator, Laplacian operator, Canny edge detection; or, it may use another trained neural network model to extract the outer surface edge of the target object based on machine learning, without any specific limitation.
[0063] 240. Based on the type of the target object and the curvature of the outer surface edge of the target object, determine the edge to be polished in the outer surface edge.
[0064] In this embodiment, the computing component may extract one or more outer surface edges during the aforementioned step 230. The computing component can determine the edge to be ground based on the type of the target object and the curvature of each extracted outer surface edge. Different types of objects correspond to different types of edges to be ground. For example, for objects such as plates and hexagonal nuts whose edges are composed of rectangles, triangles, or other straight lines, the edge to be ground should be a straight line rather than a curve. For objects such as round handles and spherical cone plugs whose edges are composed of curves, the edge to be ground should be a curve rather than a straight line. For irregularly shaped workpieces with a mixture of straight and curved outer surface edges, the shape of the edge to be ground can be customized by the user. For example, the robot disclosed in this embodiment can first automatically grind the straight segments of the irregular workpiece, and then the curved segments of the irregular workpiece can be manually ground; or conversely, the robot can first automatically grind the curved segments of the irregular workpiece, and then the straight segments of the irregular workpiece can be manually ground; or the robot can automatically grind both curved and straight segments, with the grinding order of the straight and curved segments customized by the user, grinding one type of edge at a time.
[0065] Therefore, in this embodiment, the curvature of the outer surface edge of the target object can be used to determine the line shape of each segment of the outer surface edge, such as a curve or a straight line. Then, based on the type of the target object, it is determined which line shape in the outer surface edge corresponds to the edge to be polished.
[0066] For example, please refer to Figure 3 , Figure 3 This is an example diagram of an edge to be polished, as disclosed in an embodiment of this application. Figure 3 (a) is an example diagram of a target object 310 disclosed in an embodiment of this application. The target object 310 is a rectangular cube. Figure 3 (a) is a front view of the target object 310. For example... Figure 3 As shown in (b), the image captured by the imaging component may include at least the top surface of the target object 310, wherein the outer surface edges of the top surface may include edges 310a, 310b, 310c, and 310d. The two long edges of the top surface of the target object may be polished first; therefore, the edges to be polished identified by the calculation component in step 240 may include edges 310a and 310c.
[0067] 250. Control the grinding components to grind the edges to be ground.
[0068] In this embodiment, after determining the edge to be polished, the calculation component can further calculate the movement trajectory of the polishing component based on the edge to be polished. Then, the polishing component is controlled to move along the aforementioned trajectory to polish the edge to be polished.
[0069] By implementing the embodiments of this application, the robot can automatically identify the type and outer contour of an object from an image, and then further polish the specific edges to be polished in the object's edge contour according to the type of the object, thereby improving the accuracy and efficiency of polishing.
[0070] To better illustrate the methods disclosed in the embodiments of this application, please refer to... Figure 4 , Figure 4 This is a schematic flowchart of a method for identifying target object types based on images, as disclosed in an embodiment of this application. Figure 4 The illustrated flow is a method for identifying the type of a target object based on feature extraction, specifically one of the aforementioned steps 220 disclosed in the embodiments of this application. Figure 4 As shown, the following steps may be included:
[0071] 2210. Extract the target corner points of the target object from the image.
[0072] In this embodiment, corner features are selected as matching features. Based on the principle of triangle similarity and its resistance to rotation and scaling, the interior angles of a triangle are invariant to translation and rotation, and also invariant to scaling. Please refer to... Figure 5 , Figure 5 This is an example diagram illustrating the rotation and scaling changes of the interior angles of a triangle as disclosed in an embodiment of this application. Figure 5 As shown, the original triangle is as follows Figure 5 As shown in (a), after rotating the original triangle, it becomes... Figure 5 As shown in (b), the original triangle is enlarged as follows: Figure 5 As shown in (c). Comparison Figure 5 (a) and Figure 5 (b) It can be seen that the interior angles α and β did not change before and after the rotation; in contrast... Figure 5 (a) and Figure 5 (c) As can be seen, the interior angles α and β remain unchanged before and after magnification. Therefore, corner features have good translation, rotation, and scaling invariance, and feature matching accuracy can be improved based on corner features. Corner features may include Harris corners or Shi-Tomasi corners, and there is no specific limitation.
[0073] Optionally, when performing step 2210, the computing unit can use the Harris operator to extract Harris corner points as target corner points. Harris corner points have the advantages of low computational complexity and good real-time performance, which is suitable for the timeliness requirements of industrial grinding scenarios. For example, the Harris corner point detection formula can be expressed as follows:
[0074]
[0075] in, For window functions, The gradient value of the image grayscale, for each small displacement. Equation (1) can be bilinearly approximated as follows:
[0076]
[0077] in
[0078] E can be approximated as a local cross-correlation function, describing the shape at that point. Let... If are two eigenvalues of matrix M, then they can represent the curvature of the local autocorrelation function. Due to isotropy, M retains rotation invariance. By analyzing the two eigenvalues of matrix M, the following three cases can be derived:
[0079] 1) If both feature values are very small, it means that the gray level of the area where the window is located is approximately constant, and the function E changes very little when moved in any direction.
[0080] 2) If one eigenvalue is large and another is small, it indicates a roof-like shape, such as an edge. Moving along the edge direction results in a small change in the function E, while moving perpendicular to the edge results in a larger change.
[0081] 3) If both eigenvalues are very large, it indicates that the peaks are sharp and any movement in any direction will cause E to increase sharply.
[0082] The analysis of the above three cases can be used to detect Harris corners.
[0083] In this embodiment, the target corner point can be all or part of the corner point features identified based on the aforementioned corner point features. For example, after identifying corner point features from an image using the Harris algorithm, the computing unit can determine all identified corner point features as the target corner points of the target object, or it can select part of the identified corner point features as the target corner points of the target object. The selection rules can be set empirically and are not specifically limited. For example, corner points with a distance below a threshold can be removed from all corner point features to reduce unnecessary computation caused by duplicate features.
[0084] 2220. Match the target corner points in the image with the template corner points in the template library.
[0085] In this embodiment, the template library may include object information for at least two different types of objects. The object information includes a type identifier representing the object type and the template corner points corresponding to each type identifier. The data in the template library can be pre-verified manually before storage, ensuring high accuracy.
[0086] 2230. Determine the type identifier corresponding to the template corner point that matches the target corner point as the type of the target object.
[0087] In this embodiment, the type identifier corresponding to the template corner point is the type of the target object. For example, if the type identifier corresponding to the template corner point is a nut, then the type of the target object is a nut; if the type identifier corresponding to the template corner point is a sheet material, then the type of the target object is a sheet material.
[0088] Based on such Figure 4 The method shown allows the computational component to balance computational load and accuracy, identifying the type of target object from an image relatively accurately with less computational load.
[0089] Please see Figure 6 , Figure 6 This is a schematic flowchart of a method for determining the edge to be polished in the outer surface edge, as disclosed in the application embodiment. Figure 6 As shown, the following steps may be included:
[0090] 2410. Determine the curvature sensitivity value of the target object.
[0091] 2420. Calculate the curvature of several edge data points on the outer surface edge based on the curvature sensitivity value of the target object.
[0092] In this embodiment, the curvature sensitivity value is an indicator used to distinguish linear shapes, and can at least be used to divide straight lines or curves. The curvature sensitivity value can be a pre-set fixed value based on experience, or it can be a variable value that varies according to the type of object; that is, different types of objects may have different curvature sensitivity values. Optionally, if the curvature sensitivity value is a variable value, the curvature sensitivity value of the target object can be determined according to the type of the target object.
[0093] The outer surface edges of the target object are discrete signals in the image, comprising multiple edge data points. For example, the curvature can be calculated using the following formula:
[0094]
[0095] In the formula, x(i) and y(i) are the row coordinates and column coordinates of the edge position sequence, respectively, and i is the position sequence number. and Let x(i) and y(i) be the approximate first derivatives of x(i) and y(i) with respect to i, respectively. and Let x(i) and y(i) be the approximate second reciprocals of i, respectively, where:
[0096]
[0097]
[0098] In the formula It is the increment of variable i, i.e., the curvature sensitivity value.
[0099] In other words, the curvature of several edge data points in the outer surface edge can be calculated. The larger the curvature sensitivity value, that is, the larger the increment of variable i, the more edge data points are added to the curvature calculation.
[0100] 2430. Determine the linear shape corresponding to each edge data point among several edge data points based on the curvature.
[0101] In the embodiments of this application, curvature can be used to characterize the arc of the line segment corresponding to the edge data point. Curvature can be used to determine the line shape corresponding to the edge data point, and the line shape may include a straight line or an arc. It should be noted that in some other possible embodiments, the line shape may also be subdivided by different functions. For example, the line shape may include a parabola, a hyperbola, etc., and there is no specific limitation.
[0102] Optionally, if the line shape includes straight lines or arcs, the line shape corresponding to each edge data point can be determined directly based on the curvature of each edge data point using a threshold judgment method. For example, a first threshold can be set to characterize the curvature of a straight line. and a second threshold used to characterize the curvature of the circular arc. According to the first threshold Set the first curvature range, based on the second threshold. Set a second curvature range. The first curvature range can be a first threshold. It can also include the first threshold. and the first threshold The range of preset values that float upwards and / or downwards from the baseline can be set empirically, and there is no specific limitation. The second curvature range is similar, and will not be elaborated further below.
[0103] Based on this, the line shape corresponding to the edge data points whose curvature satisfies the first curvature range among a number of edge data points can be determined as a straight line; and / or, the line shape corresponding to the edge data points whose curvature satisfies the second curvature range among a number of edge data points can be determined as an arc.
[0104] 2440. Identify the edge data points whose line shape corresponds to the type of the target object from a number of edge data points as the edges to be polished.
[0105] In this embodiment, the contour edge represented by the edge data points corresponding to the type of the target object is the edge of the target object to be polished. For example, if the type of the target object is a board, the line corresponding to the board is a straight line. Therefore, the edge data points with a straight line are identified from a number of edge data points as the edge to be polished. This can remove some arc parts that are mistakenly identified as edges from the edge to be polished. It can also perform two-stage polishing on irregular board materials that mix straight lines and arcs. For example, the straight edges are polished first, and then the curved edges are polished. The polishing precision used in different stages can be different, thereby improving the overall polishing precision.
[0106] As can be seen, by implementing the embodiments of this application, a robot including the aforementioned computing components can automatically identify the line shape of the outer surface contour, thereby determining the edge to be polished based on the line shape and the type of the target object, achieving a precise polishing effect.
[0107] In summary, by implementing the object edge polishing method disclosed in the embodiments of this application, the robot can automatically identify the object type, automatically classify the edge line shape, and polish specific edges to be polished in the object edge contour according to the object type, thereby improving the accuracy and efficiency of polishing.
[0108] Please see Figure 7 , Figure 7 This is a schematic diagram of an object edge polishing device disclosed in an embodiment of this application. This object edge polishing device can be applied to the aforementioned computing unit, and further to the aforementioned robot. Figure 7 As shown, the device includes:
[0109] The acquisition module 710 is used to acquire an image of the target object to be polished.
[0110] Recognition module 720 is used to identify the type of the target object based on the image;
[0111] Extraction module 730 is used to extract the outer surface edge of the target object from the image;
[0112] The determining module 740 is used to determine the edge to be polished in the outer surface edge according to the type of the target object and the curvature of the outer surface edge of the target object;
[0113] The control module 750 is used to control the grinding component to grind the edge to be ground.
[0114] As an optional implementation, the recognition module 720 can also be used to extract target corner points of the target object from the image; and to match the target corner points in the image with template corner points in a template library; the template library includes object information of at least two different types of objects, the object information including a type identifier for characterizing the object type and a template corner point corresponding to each type identifier; and to determine the type identifier corresponding to the template corner point that matches the target corner point as the type of the target object.
[0115] Further optionally, the recognition module 720 can also be used to extract the target corner points of the target object from the image using the Harris operator.
[0116] As an optional implementation, the determining module 740 can also be used to determine the curvature sensitivity value of the target object; and to calculate the curvature of a plurality of edge data points in the outer surface edge based on the curvature sensitivity value of the target object; and to determine the line shape corresponding to each of the plurality of edge data points based on the curvature; and to identify edge data points whose line shape corresponds to the type of the target object from the plurality of edge data points as edges to be polished.
[0117] Optionally, the line shape corresponding to the edge data points can be a straight line or an arc.
[0118] Further optionally, the determining module 740 can also be used to determine the line shape corresponding to the edge data point whose curvature satisfies the first curvature range as a straight line among the plurality of edge data points; or, to determine the line shape corresponding to the edge data point whose curvature satisfies the second curvature range among the plurality of edge data points as an arc.
[0119] Optionally, the determining module 740 can also be used to determine the curvature sensitivity value of the target object based on the type of the target object.
[0120] The object edge grinding apparatus disclosed in this application can grind specific edges to be ground in the object's edge contour according to the object's type, improving grinding accuracy and efficiency.
[0121] Please see Figure 8 , Figure 8 This is a schematic diagram of the structure of a robot disclosed in an embodiment of this application. Figure 8 As shown, the robot may include:
[0122] Memory 810 containing computer programs;
[0123] Processor 820 coupled to memory 810;
[0124] When the computer program stored in the memory 810 is executed by the processor 820, the processor 820 implements any of the grinding trajectory generation methods disclosed in the embodiments of this application.
[0125] It should be noted that, Figure 8 The robot shown may also include components not shown, such as a power supply, input buttons, screen, RF circuit, Wi-Fi module, and Bluetooth module, which will not be described in detail in this embodiment.
[0126] This application discloses a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements any of the object edge polishing methods disclosed in this application.
[0127] This application discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute any of the grinding trajectory generation methods disclosed in this application.
[0128] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. Those skilled in the art should also recognize that the embodiments described in the specification are optional embodiments, and the actions and modules involved are not necessarily essential to this application.
[0129] In the various embodiments of this application, it should be understood that the sequence number of each process does not necessarily imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.
[0130] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they can be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0131] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0132] If the integrated units described above are implemented as software functional units and sold or used as independent products, they can be stored in a computer-accessible memory. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several requests to cause a computer device (which can be a personal computer, server, or network device, specifically a processor in the computer device) to execute some or all of the steps of the methods described in the various embodiments of this application.
[0133] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compactdisc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.
[0134] The foregoing has provided a detailed description of the object edge polishing method, apparatus, robot, and storage medium disclosed in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.
Claims
1. A method of edge polishing of an object, characterized by, The method includes: Acquire an image of the target object to be polished; Identify the type of the target object based on the image; Extract the outer surface edge of the target object from the image; Determine the curvature sensitivity value of the target object; The curvature of several edge data points on the outer surface edge is calculated based on the curvature sensitivity value of the target object. The curvature determines the line shape corresponding to each edge data point among the plurality of edge data points; the curvature sensitivity value is an index used to distinguish line shapes, and can at least be used to divide straight lines or curves; From the plurality of edge data points, edge data points whose linear shape corresponds to the type of the target object are identified as the edges to be polished; The grinding component is controlled to grind the edge to be ground.
2. The method of claim 1, wherein, The step of identifying the type of the target object based on the image includes: Extract the target corner points of the target object from the image; The target corner point in the image is matched with the template corner point in the template library; the template library includes object information of at least two different types of objects, and the object information includes a type identifier for representing the object type and a template corner point corresponding to each type identifier; The type identifier corresponding to the template corner point that matches the target corner point is determined as the type of the target object.
3. The method of claim 2, wherein, Extracting the target corner points of the target object from the image includes: The target corner points of the target object are extracted from the image using the Harris operator.
4. The method of claim 1, wherein, The line shape corresponding to the edge data points can be either a straight line or an arc.
5. The method according to claim 4, characterized in that, The step of determining the line shape corresponding to each edge data point among the plurality of edge data points based on curvature includes: Among the given number of edge data points, the line shape corresponding to the edge data points whose curvature satisfies the first curvature range is determined as a straight line; or, Among the plurality of edge data points, the line shape corresponding to the edge data point whose curvature satisfies the second curvature range is determined as an arc.
6. The method according to claim 1, characterized in that, Determining the curvature sensitivity value of the target object includes: The curvature sensitivity value of the target object is determined based on the type of the target object.
7. An object edge grinding device, characterized in that, The device includes: The acquisition module is used to acquire images of the target object to be polished. The recognition module is used to identify the type of the target object based on the image; The extraction module is used to extract the outer surface edge of the target object from the image; A determining module is used to determine the curvature sensitivity value of the target object; and to calculate the curvature of a plurality of edge data points in the outer surface edge based on the curvature sensitivity value of the target object; and to determine the line shape corresponding to each edge data point in the plurality of edge data points based on the curvature; and to identify edge data points whose line shape corresponds to the type of the target object from the plurality of edge data points as edges to be polished; the curvature sensitivity value is an index used to distinguish line shapes, and can at least be used to divide straight lines or curves; The control module is used to control the grinding component to grind the edge to be ground.
8. A robot, characterized in that, The robot includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to implement the method as described in any one of claims 1 to 6.
9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 6.