CN118544360A - Robot vision detection method, system, terminal and medium based on laser compensation - Google Patents

Abstract

The application is suitable for the technical field of computer vision, and provides a robot vision detection method, a system, a terminal and a medium based on laser compensation, wherein the method comprises the steps of firstly acquiring terminal marker image set information, and then executing the processing on each terminal marker image information: acquiring first pixel coordinate information of a theoretical position of the tail end of the mechanical arm, carrying out distortion correction processing on the image information of the tail end marker according to the first pixel coordinate information and the distortion coefficient set information, determining second pixel coordinate information of an actual measurement position of the tail end of the mechanical arm, effectively determining coordinate error value information of the tail end of the mechanical arm, and finally carrying out compensation processing on the second pixel coordinate information to generate compensated pixel coordinate information. The application can compensate the global error condition, so that the accuracy of camera detection almost approaches to the motion error detected by the laser tracker, the accuracy of mechanical arm positioning is greatly improved, and monocular vision precision detection oriented to plane mechanical arm control is realized.

Description

Robot vision detection method, system, terminal and medium based on laser compensation

Technical Field

The present application relates to the technical field of computer vision, and in particular to a robot vision detection method, system, terminal and medium based on laser compensation.

Background Art

In the 1980s, computer vision technology was first applied to robot systems. Compared with traditional robots, visual robots are more outstanding in adaptability, control accuracy and robustness. With the increasing application of robots in industrial production, robot visualization has become an important research direction in the field of robotics, and is also widely recognized as the most important direction for the development of modern high-tech. The accuracy of robot visualization directly affects the motion accuracy of the operations guided by such robots. The accuracy of robot visualization is directly related to the calibration of robot vision. The accuracy of robot vision calibration directly affects the accuracy of the motion feedback of industrial robot arms. The complexity of calibration directly affects the rapidity of such robot calibration. Therefore, achieving high-precision positioning, identification and detection is crucial for the successful operation of robot arms.

At present, in the monocular vision precision inspection for planar robotic arm control, image deformation will occur when capturing the robotic arm movement in real time, resulting in serious deviations in the detection of the calibration object, affecting the accuracy of the robotic arm positioning, and there is a problem of low accuracy, which needs further improvement.

Summary of the invention

Based on this, the embodiments of the present application provide a robot vision detection method, system, terminal and medium based on laser compensation to solve the problem of low accuracy in the prior art.

In a first aspect, an embodiment of the present application provides a robot vision detection method based on laser compensation, the method comprising:

Based on a preset camera, continuously acquire end marker image set information, wherein the end marker image set information includes a plurality of continuous end marker image information, the shooting object of the end marker image information is a designated robotic arm end, and the robotic arm end is equipped with at least two light source markers;

For each of the end marker image information: obtaining first pixel coordinate information corresponding to the theoretical position of the end of the robotic arm, and performing distortion correction processing on the end marker image information according to the first pixel coordinate information and the preset distortion coefficient set information, to determine second pixel coordinate information corresponding to the measured position of the end of the robotic arm;

Determine the coordinate error value information of the end of the robotic arm according to the second pixel coordinate information and a preset end position motion error calculation function;

The second pixel coordinate information is compensated according to the coordinate error value information to generate compensated pixel coordinate information.

Compared with the prior art, the beneficial effect is as follows: in the robot vision detection method based on laser compensation provided in the embodiment of the present application, the terminal device can first use the camera to obtain the end marker image set information, and then perform the processing for each end marker image information: obtain the first pixel coordinate information corresponding to the theoretical position of the end of the robotic arm, and then perform distortion correction processing on the end marker image information based on the first pixel coordinate information and the distortion coefficient set information, accurately determine the second pixel coordinate information corresponding to the actual position of the end of the robotic arm, and then effectively determine the coordinate error value information of the end of the robotic arm based on the second pixel coordinate information and the end position motion error calculation function, and finally, based on the coordinate error value information, compensate the second pixel coordinate information to generate compensated pixel coordinate information, thereby achieving compensation for the global error situation, improving the accuracy of robotic arm positioning, and to a certain extent solving the current problem of low accuracy.

In a second aspect, an embodiment of the present application provides a robot vision detection system based on laser compensation, the system comprising:

The terminal marker image set information acquisition module is used to continuously acquire the terminal marker image set information based on a preset camera, wherein the terminal marker image set information includes a plurality of continuous terminal marker image information, and the shooting object of the terminal marker image information is a designated robotic arm terminal, and at least two light source markers are installed at the robotic arm terminal;

The first pixel coordinate information acquisition module is used to obtain the first pixel coordinate information corresponding to the theoretical position of the end of the robotic arm for each end marker image information, and perform distortion correction processing on the end marker image information according to the first pixel coordinate information and the preset distortion coefficient set information, so as to determine the second pixel coordinate information corresponding to the measured position of the end of the robotic arm;

A coordinate error value information determination module: used to determine the coordinate error value information of the end of the robotic arm according to the second pixel coordinate information and a preset end position motion error calculation function;

Compensated pixel coordinate information generating module: used to perform compensation processing on the second pixel coordinate information according to the coordinate error value information to generate compensated pixel coordinate information.

In a third aspect, an embodiment of the present application provides a terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method of the first aspect described above when executing the computer program.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method of the first aspect described above are implemented.

It can be understood that the beneficial effects of the second to fourth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art.

FIG1 is a schematic diagram of a flow chart of a robot vision detection method provided by an embodiment of the present application;

FIG2 is a first schematic diagram of a robotic arm provided by an embodiment of the present application;

FIG3 is a second schematic diagram of a robotic arm provided by an embodiment of the present application;

FIG4 is a schematic diagram of a flow chart before step S100 in a robot vision detection method provided in an embodiment of the present application;

FIG5 is a schematic diagram of camera imaging provided by an embodiment of the present application;

FIG6 is a schematic diagram of image information of a calibration plate provided in an embodiment of the present application;

FIG7 is a schematic diagram of corner point information of a calibration plate provided in an embodiment of the present application;

FIG8 is a schematic diagram of reprojection error information provided by an embodiment of the present application;

FIG9 is a flow chart of step S200 in the robot vision detection method provided in one embodiment of the present application;

FIG10 is a schematic diagram of edge fitting of a first circular marker provided in an embodiment of the present application;

FIG11 is a schematic diagram of edge fitting of a second circular marker provided in an embodiment of the present application;

FIG12 is a third schematic diagram of a robotic arm provided in an embodiment of the present application;

FIG13 is a schematic diagram of point position distribution provided by an embodiment of the present application;

FIG14 is a schematic diagram of a flow chart after step S240 in a robot vision detection method provided in an embodiment of the present application;

FIG15 is a schematic diagram of a flow chart before step S300 in a robot vision detection method provided in an embodiment of the present application;

FIG16 is a schematic diagram of a predicted position provided by an embodiment of the present application;

FIG. 17 is a flow chart of step S400 in the robot vision detection method provided in one embodiment of the present application;

FIG. 18 is a schematic diagram of a flow chart after step S400 in a robot vision detection method provided in an embodiment of the present application;

FIG19 is a block diagram of a robot vision detection system provided by an embodiment of the present application;

FIG. 20 is a schematic diagram of a terminal device provided in an embodiment of the present application.

DETAILED DESCRIPTION

In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

In the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and should not be understood as indicating or implying relative importance.

References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

In order to illustrate the technical solution described in this application, a specific embodiment is provided below for illustration.

Please refer to FIG. 1, which is a flow chart of a robot vision detection method based on laser compensation provided in an embodiment of the present application. In this embodiment, the execution subject of the robot vision detection method is a terminal device. It is understandable that the types of terminal devices include but are not limited to mobile phones, tablet computers, laptop computers, ultra-mobile personal computers (Ultra-Mobile Personal Computer, UMPC), netbooks, personal digital assistants (Personal Digital Assistant, PDA), etc., and the embodiments of the present application do not impose any restrictions on the specific types of terminal devices.

Please refer to FIG1 . The robot vision detection method provided in the embodiment of the present application includes but is not limited to the following steps:

In S100, based on a preset camera, terminal marker image set information is continuously acquired.

Exemplarily, referring to FIG. 2 , the robot vision detection method may be applicable to a robotic arm.

Specifically, the terminal device can photograph the end of the robotic arm based on a preset camera, and continuously obtain end marker image set information, wherein the end marker image set information includes multiple continuous end marker image information; the end marker image information is used to describe the image obtained by the camera photographing the end of the robotic arm; the end of the robotic arm is used to describe the end of the robotic arm that directly interacts with the work object or environment; the shooting object of the end marker image information is the specified end of the robotic arm; at least two light source markers are installed at the end of the robotic arm.

In one possible implementation, the light source marker may be a small white light lamp. When there are two light source markers, the two light source markers may be located at diagonal positions, respectively, and the two small white light lamps may be installed and fixed in the circular hole of the part, and then light is guided through acrylic rods, and then diffuse light is processed by sticking a white film on the hole surface.

Without loss of generality, referring to FIG. 3 , the terminal device can construct an inverse kinematics equation for the motion position of the end of the robotic arm and the joint angles of the two arms of the robotic arm, so as to facilitate determining the combination of joint angles through the operation posture, wherein the inverse kinematics equation can be:

,

In the formula, is the length corresponding to the first arm of the robot, is the length of the second arm of the robot, is the angle between the first arm and the positive direction of the X-axis, is the angle between the second arm and the positive direction of the X-axis, and P(A,B) is the end position of the robot arm.

In the formula, ;

.

In some possible implementations, in order to improve accuracy, please refer to FIG. 4 , before step S100, the method further includes but is not limited to the following steps:

In S101 , an imaging model of a camera is constructed.

Specifically, the terminal device may first construct an imaging model of the camera. For example, the camera may be an industrial camera of the specific model ME2P-2621-15U3M.

Without loss of generality, please refer to Figure 5. A camera can be composed of a sensor and a lens. The function of the lens is to converge the light emitted from a point in the external environment to a point on the photosensitive sensor to achieve clear imaging. According to different projection methods, it can be divided into ordinary lenses and telecentric lenses. The former is perspective projection and the latter is parallel projection. The imaging model of the camera can be simplified to a pinhole imaging model, which is characterized in that all light from the scene passes through a projection center, that is, the center of the lens.

Exemplarily, since the pinhole projection converts the entire three-dimensional coordinates into two-dimensional pixel coordinates, it can be regarded as a process of converting a three-dimensional world coordinate system coordinate point P (Xw, Yw, Zw) in the vision into a coordinate point P (Xc, Yc, Zc) in the camera coordinate system through the relationship with the camera coordinate system, and then converting the coordinate point P (Xc, Yc, Zc) in the camera coordinate system into a coordinate point P (u, v) in the pixel coordinate, and finally converting from the pixel coordinate to the image coordinate point P (x, y). Therefore, without considering the distortion, the imaging model can be:

,

Where; represents the imaging model; Indicates the depth information of the photographed object; The actual horizontal coordinate of the image center representing the terminal marker image information in a preset image coordinate system; The actual vertical coordinate of the image center representing the terminal marker image information in a preset image coordinate system; Indicates the internal parameter information of the camera; the internal parameter information may include 5 unknown numbers; , , Indicates the focal length information of the camera, in millimeters; Indicates the first pixel size with respect to the horizontal axis; Indicates the second pixel size with respect to the vertical axis; Indicates the intrinsic and extrinsic parameter information of the camera; The theoretical horizontal coordinate of the image center representing the terminal marker image information in the preset image coordinate system; The theoretical ordinate of the image center representing the terminal marker image information in the preset image coordinate system.

In the formula, Represents the rotation matrix of the camera coordinates relative to the world coordinates, ; Represents the translation vector of the camera coordinates relative to the world coordinates, .

It should be noted that although the calibration of a monocular camera can only obtain internal and external parameters and cannot determine the image depth of an object, the image depth can be eliminated when the three-dimensional world coordinate points are converted into pixel coordinate points. Therefore, when the camera and the robotic arm are coplanar, the terminal device can determine the specific number of pixels occupied by a specific unit distance of the three-dimensional world coordinates based on the first function, the second function, and the third function, thereby using the number of pixels between two points between the planes to achieve coplanar ranging of the monocular camera.

Exemplarily, the first functional formula may be:

In S102, the calibration plate image information is acquired based on the camera.

Without loss of generality, it is essential to obtain three-dimensional geometric information from two-dimensional images, and the accuracy of camera calibration is also very important for visual measurement. The main content of camera calibration is to determine the internal parameters and external parameters of the camera and obtain the distortion coefficient of the camera in order to accurately map the pixel coordinates in the image to the physical coordinates in the real world.

For example, please refer to FIG6. Since radial distortion and tangential distortion are introduced during the manufacture and use of the camera lens, these distortions will cause the straight lines in the image to be deformed or the corners to be inaccurate. The terminal device can calculate the distortion coefficient through camera calibration, and then correct the image during the image processing process to ensure that the object in the image maintains an accurate shape and position, thereby obtaining pixel information of the object's real position. Therefore, after the terminal device builds the imaging model of the camera, the terminal device can obtain the calibration plate image information based on the camera, wherein the calibration plate image information is used to describe the image obtained by shooting the calibration plate with the camera; the calibration plate can be a checkerboard or a dot matrix, for example: a checkerboard calibration plate with a specific specification of GP290-20-12*9, and the calibration plate can be pre-placed on a plane at the same height as the plane to be detected, so as to facilitate the subsequent detection of the characteristic pixels of the light source moving on the fixed height plane.

In S103, grayscale processing is performed on the calibration plate image information to generate grayscale image information.

Specifically, after the terminal device acquires the calibration plate image information, the terminal device may perform grayscale processing on the calibration plate image information to generate grayscale image information, wherein the grayscale image information is used to describe the calibration plate image information after the grayscale processing.

In S104, based on a preset corner point detection algorithm and the grayscale image information, multiple calibration plate corner point information of the grayscale image information is determined.

Exemplarily, please refer to Figure 7. After the terminal device generates grayscale image information, the terminal device can determine multiple calibration plate corner point information of the grayscale image information based on a preset corner point detection algorithm and the grayscale image information, wherein the calibration plate corner point information is used to describe the corner points on the calibration plate, and these corner points are used for subsequent calculations. The perfect circle in Figure 7 represents the calibration plate corner point information.

In S105, for each calibration plate corner point information: obtaining the actual physical coordinate information of the corner point of the calibration plate corner point information;

Exemplarily, please refer to Figure 7. After the terminal device determines multiple calibration plate corner point information, the terminal device can perform this processing for each calibration plate corner point information: obtain the actual physical coordinate information of the corner point of the calibration plate corner point information, so as to facilitate the association of image coordinates and physical coordinates, wherein the actual physical coordinate information of the corner point is used to describe the actual physical coordinates of the calibration plate corner point information. Exemplarily, the actual physical coordinates can be the grid point coordinates of the calibration plate corner point information on the calibration plate.

In S106, reprojection error information is generated according to a preset Euclidean distance calculation function, the calibration plate corner point information, and the actual physical coordinate information of the corner points.

Specifically, after the terminal device obtains the actual physical coordinate information of the corner point, the terminal device can input the calibration plate corner point information and the actual physical coordinate information of the corner point into a preset Euclidean distance calculation function to generate reprojection error information, thereby evaluating the calibration result to facilitate determining the accuracy of the calibration, wherein the reprojection error information is used to describe the projection error value between the calibration plate corner point information and the actual physical coordinate information of the corner point.

Exemplarily, please refer to FIG. 8 , which shows the reprojection error information calculated for twenty-five calibration images. As can be seen from FIG. 8 , the maximum reprojection error information is 0.1 pixels, and the average reprojection error information is 0.07 pixels.

It should be noted that the camera resolution can be 5120 5120pixel; focal length can be 535.460 mm; internal reference can be ; The representative distance of a single pixel on the plane to be measured can be 0.2235 mm.

In S107, calibration accuracy result information is generated according to the re-projection error information and the preset error threshold information.

Specifically, after the terminal device generates the reprojection error information, the terminal device can compare the reprojection error information with the preset error threshold information to generate calibration accuracy result information, wherein the calibration accuracy result information is calibration pass information or calibration fail information, the calibration pass information is used to describe the calibration pass, and the calibration fail information is used to describe the calibration fail; the error threshold information can be pre-defined. Exemplarily, if the reprojection error information is less than the error threshold information, the terminal device can generate calibration pass information, otherwise the terminal device can generate calibration fail information.

Accordingly, the above step S100 includes but is not limited to the following steps:

In S110 , if the calibration accuracy result information is calibration qualified information, the terminal marker image set information is continuously acquired based on a preset camera.

Specifically, if the calibration accuracy result information is calibration qualified information, the terminal device can continuously acquire the terminal marker image set information based on a preset camera.

In S200, for each end marker image information: the first pixel coordinate information corresponding to the theoretical position of the end of the robotic arm is obtained, and based on the first pixel coordinate information and the preset distortion coefficient set information, the end marker image information is subjected to distortion correction processing to determine the second pixel coordinate information corresponding to the actual measured position of the end of the robotic arm.

Specifically, after the terminal device continuously obtains the end marker image set information, the terminal device can perform the processing for each end marker image information: obtain the first pixel coordinate information corresponding to the theoretical position of the end of the robotic arm, and then perform distortion correction processing on the end marker image information according to the first pixel coordinate information and the preset distortion coefficient set information, and effectively determine the second pixel coordinate information corresponding to the measured position of the end of the robotic arm. In a possible implementation, the distortion coefficient set information includes the first distortion coefficient information and the second distortion coefficient information. The first distortion coefficient information and the second distortion coefficient information can effectively remove the distortion of the image. It should be noted that the first distortion coefficient information and the second distortion coefficient information can be obtained through camera calibration.

In some possible implementations, in order to implement the dedistortion processing on the image, please refer to FIG. 9 , step S200 includes but is not limited to the following steps:

In S210 , for each end marker image information: first pixel coordinate information corresponding to the theoretical position of the end of the robot arm is obtained.

Specifically, the terminal device can perform this processing on each end marker image information: obtain the first pixel coordinate information corresponding to the theoretical position of the end of the robotic arm.

Exemplarily, the first pixel coordinate information may be:

,

In the formula, The horizontal coordinate representing the first pixel coordinate information, The actual horizontal coordinate of the image center representing the terminal marker image information in the preset image coordinate system, The theoretical horizontal coordinate of the image center representing the terminal marker image information in the preset image coordinate system, , Indicates the focal length information of the camera. mm, Represents the first pixel size with respect to the horizontal axis, The ordinate representing the first pixel coordinate information, The actual vertical coordinate of the image center representing the terminal marker image information in the preset image coordinate system, The theoretical ordinate of the image center representing the terminal marker image information in the preset image coordinate system, , Indicates the second pixel size with respect to the vertical axis.

In S220, distortion correction processing is performed on the end marker image information according to the first pixel coordinate information, the preset first distortion coefficient information and the second distortion coefficient information to determine the distortion corrected pixel coordinate information corresponding to the measured position of the end of the robot arm.

Specifically, due to the non-ideal characteristics of the camera's own lens and the non-parallel relationship between the imager and the lens during the image acquisition process, radial distortion and tangential distortion will be formed, resulting in the image being unable to reflect the actual state. When a camera is used to take a photo to obtain the pixels of the marked point position, the image distortion will cause a deviation between the actual pixels and the ideal pixels. In order to obtain the ideal pixel coordinates, the image distortion problem must be dealt with. The above-mentioned calibration process based on the calibration plate image information only deals with larger radial distortions. Therefore, after the first pixel coordinate information of the terminal device, the terminal device can perform distortion correction processing on the end marker image information according to the first pixel coordinate information, the preset first distortion coefficient information and the preset second distortion coefficient information, and determine the distortion corrected pixel coordinate information corresponding to the actual measured position of the end of the robotic arm, wherein the distortion corrected pixel coordinate information is used to describe the first pixel coordinate information after the initial removal of the distortion.

For example, assuming that the ideal undistorted image coordinates are (x, y), the image coordinates actually captured by the camera can be expressed as ( , ), that is, the above-mentioned distortion-corrected pixel coordinate information can be:

,

In the formula, The horizontal coordinate representing the distortion-corrected pixel coordinate information, The horizontal coordinate representing the first pixel coordinate information, Represents the first distortion coefficient information, which is used to describe the first-order radial distortion coefficient. , , Represents the second distortion coefficient information, which is used to describe the second-order radial distortion coefficient. , Represents the preset third-order radial distortion coefficient, represents the preset first-order tangential distortion coefficient, represents the preset second-order tangential distortion coefficient, The ordinate represents the distortion-corrected pixel coordinate information, The vertical coordinate representing the first pixel coordinate information.

In S230, for each distortion-corrected pixel coordinate information of the first marker and each distortion-corrected pixel coordinate information of the second marker: least square fitting processing is performed on the distortion-corrected pixel coordinate information to generate ideal circle feature set information of the target ideal circle.

Specifically, after the terminal device determines the distortion-corrected pixel coordinate information, the terminal device can perform this processing on each distortion-corrected pixel coordinate information of the first marker and each distortion-corrected pixel coordinate information of the second marker: perform least squares fitting processing on each distortion-corrected pixel coordinate information to generate ideal circle feature set information of the target ideal circle, thereby performing least squares fitting of circles on the edges of the two markers respectively to obtain the centers of the two markers respectively, wherein the first marker is used to describe any one light source marker, and the second marker is used to describe any other light source marker, and the ideal circle feature set information includes the center pixel coordinate information and radius information of the target ideal circle.

In a possible implementation, please refer to Figures 10 and 11, Figure 10 shows an edge fitting diagram of the first circular marker, and Figure 11 shows an edge fitting diagram of the second circular marker; after the terminal device grayscales the captured image to retain the light source marker pixels of a specific grayscale threshold, the terminal device can fit the edge of the retained circular light source using the moment estimate of the grayscale distribution of the edge point neighborhood determined by the Canny operator, wherein the use of the Canny operator can improve the image noise resistance and improve the edge connection effect.

Without loss of generality, after obtaining the pixel information corresponding to the end position of the robotic arm detected by the camera, the terminal device can install the laser tracking ball to a position aligned with the end position, and then use the laser tracker to collect the end position of the robotic arm, and then use the information from the laser tracker to judge the accuracy of the camera's detection of the end position of the robotic arm.

In S240, second pixel coordinate information is determined according to a plurality of circle center pixel coordinate information.

Specifically, after the terminal device generates the ideal circle feature set information, the terminal device can effectively determine the second pixel coordinate information based on the multiple circle center pixel coordinate information, so as to achieve the pixel coordinates of the end position of the robot arm under camera detection by summing the two circle center pixel coordinates after fitting the circle center pixel coordinates of the two circular markers by extracting the edge pixel information of the two circular markers, wherein the second pixel coordinate information can be:

,

In the formula, a horizontal coordinate representing the second pixel coordinate information, The vertical coordinate representing the second pixel coordinate information.

For example, referring to FIG. 12 , for a discrete measurement point set ( , ), we can assume in advance that the center of the target ideal circle is , the radius of the target ideal circle is R, where i = 1, 2…m; since the requirement of the least squares fitting method is usually to minimize the sum of the squares of the distances, , and then the terminal device is further simplified to obtain ,and At the same time, since the least squares method should satisfy ,but , assuming , it can be determined , further determined that: ,and ,as well as .

Without loss of generality, by establishing the relationship between the robot body coordinate system and the pixel coordinate system for the robot origin, the relationship between the pixel information of the robot end and the position of the robot end can be realized. Please refer to Figure 13, which is a point position distribution diagram of the robot end position obtained by detecting 25 points corresponding to the circular trajectories of 100mm, 200mm, 300mm and 400mm respectively by the camera.

In some possible implementations, in order to implement repeated detection of random errors using a camera, please refer to FIG. 14 . After step S240, the method further includes but is not limited to the following steps:

In S241, based on the preset number of repeated detection times information, multiple repeated measurement sample data information at the same position of the end of the robot arm is obtained.

Specifically, when the camera acquires the pixel of the center of the light source marker, there will be very slight changes in the light source of the light source marker and the ambient light intensity, resulting in a slight impact on the extraction of the edges of the two markers. In order to reduce the adverse interference caused by this slight impact, the terminal device can obtain multiple repeated measurement sample data information at the same position of the end of the robotic arm based on the preset number of repeated detections, where the number of repeated detections can be a preset value.

In S242, repeated measurement sample data average value information is generated based on a plurality of repeated measurement sample data information.

Specifically, after the terminal device obtains a plurality of repeated measurement sample data information, the terminal device may generate repeated measurement sample data average value information according to the plurality of repeated measurement sample data information, wherein the repeated measurement sample data average value information may be:

,

In the formula, Indicates the mean value information of the repeated measurement sample data. Represents repeated measures sample data information.

In S243, the standard deviation value information of the end of the robot arm at the same position is determined based on the average value information of the repeated measurement sample data.

Specifically, after the terminal device generates the average value information of the repeated measurement sample data, the terminal device can determine the standard deviation value information of the end of the robot arm at the same position according to the average value information of the repeated measurement sample data, wherein the standard deviation value information can be:

,

In the formula, Indicates standard deviation value information.

Without loss of generality, the terminal device can perform repeated detection five times at intervals of thirty degrees on the thirteen points corresponding to the circular trajectories of the robotic arm of 100 mm, 200 mm, 300 mm and 400 mm respectively, and obtain the standard deviation value information of 0.1 mm, that is, when the camera captures two markers and performs least squares fitting, the repeated detection pixel error is in the range of 0 to 0.05 pixels.

In S300, coordinate error value information of the end of the robot arm is determined according to the second pixel coordinate information and a preset end position motion error calculation function.

Specifically, after the terminal device determines the second pixel coordinate information, the terminal device can effectively determine the coordinate error value information of the end of the robotic arm according to the second pixel coordinate information and a preset end position motion error calculation function.

Without loss of generality, since the motion robot belongs to an open-loop motion mode, there will be a certain motion error with the ideal circular trajectory when it moves point-to-point. The terminal device can use the camera to detect the motion trajectory of thirteen points on the circular trajectory of 100mm, 200mm, 300mm and 400mm respectively, which are separated by thirty degrees. The motion error of the robot under camera detection is obtained by comparing the difference between the detection data and the theoretical position, which can be combined with the function:

,

In the formula, represents the motion error, ( , ) represents the value of the end position of the robot arm detected by the camera, ( , ) represents the value of the theoretical position.

Exemplarily, when a camera is used to detect the motion error of the robot's motion trajectory, the larger the working range of the robot's motion, the larger the detected motion error will be, and the farther away from the optical center, the larger the distortion error corresponding to the detected field of view will be. After the image is distorted, the camera will still have a certain system error, that is, the larger the planar motion range of the robot, the larger the camera detection error will be. At the same time, due to the motion deviation that will exist when the robot's body moves in the same trajectory plane, the terminal device can repeat the same trajectory of the robot's end position plane several times, detect and take the average value, and then compare the average value with the value of each motion on the same trajectory to obtain the motion deviation of the robot under the influence of the system error. Exemplarily, the terminal device can use a laser tracker to detect the forward and reverse motion of the robotic arm five times in sequence along the same trajectory. For example, the laser tracker can be used to detect the 25 points corresponding to the circular trajectories of 100mm, 200mm, 300mm and 400mm respectively, to detect the motion error of the robotic arm. This is conducive to comparing the accuracy of the camera's detection of the end position of the robotic arm with the data of the laser tracker to judge the accuracy of the camera's detection effect, and comparing the error of the camera detection with the error of the laser tracker detection to evaluate the camera's detection effect.

In some possible implementations, in order to realize the generation of multiple predicted pixel coordinate information, please refer to FIG. 15 . Before step S300, the method further includes but is not limited to the following steps:

In S301, spatial interpolation distance information is generated based on the second pixel coordinate information and a preset distance inverse weight interpolation function.

In S302, a plurality of predicted pixel coordinate information is generated in an equidistant array according to the spatial interpolation distance information and the second pixel coordinate information.

Specifically, after the terminal device generates spatial interpolation distance information, the terminal device can, wherein the predicted pixel coordinate information is used to describe the second pixel coordinate information generated by an equally spaced array, the distance between multiple predicted pixel coordinate information is the spatial interpolation distance information, and the distance between the predicted pixel coordinate information and the second pixel coordinate information is the spatial interpolation distance information.

Accordingly, the above step S300 includes but is not limited to the following steps:

In S310, coordinate error value information of the end of the robot arm is determined according to the horizontal coordinate of the second pixel coordinate information and a preset end position motion error calculation function.

Specifically, the terminal device can determine the coordinate error value information of the end of the robotic arm according to the horizontal coordinate of the second pixel coordinate information and a preset end position motion error calculation function, wherein the end position motion error calculation function can be:

,

In the formula, Indicates coordinate error value information, Indicates the total amount of second pixel coordinate information, Indicates the order of the second pixel coordinate information, Indicates The difference between the horizontal coordinate corresponding to the second pixel coordinate information and the horizontal coordinate corresponding to the previous second pixel coordinate information.

Exemplarily, since it is difficult for the terminal device to perform camera measurement on the motion of each point at the end position of the robot arm, in order to reduce the limitations of experimental measurement and improve time utilization, the terminal device can first measure the entire plane part of the points when the end position of the robot arm moves in the plane, and then use the known data to perform the estimation method to obtain the motion of the desired position. For example, the terminal device can perform estimation prediction on the positions of 150mm, 250mm and 350mm motion trajectory parts for the information of the position points detected for the motion trajectory parts of 100mm, 200mm, 300mm and 400mm, respectively, to obtain error compensation, and then compare the prediction results with the actual detection data to evaluate the effect. Among them, the inverse distance weighted interpolation method is one of the spatial interpolation methods, that is, the closer the sample point is to the point to be interpolated, the greater the weight is assigned, and its weight contribution is inversely proportional to the distance. The inverse distance weighted interpolation method can be combined with the following function:

,and ,as well as .

In the formula, Indicates the value of the point to be estimated; x represents the horizontal coordinate of the position of the point to be estimated, and y represents the vertical coordinate of the position of the point to be estimated. represents the value of a known point, Represents the distance between the point to be estimated and the known point.

Exemplarily, referring to FIG. 16 , the terminal device may perform compensation processing on the second pixel coordinate information according to the coordinate error value information to generate compensated pixel coordinate information. In FIG. 16 , a plurality of asterisk marks located in the same circle as the asterisk mark indicated by point A represent the measured position, and a plurality of asterisk marks located in the same circle as the asterisk mark indicated by point B represent the predicted position.

Without loss of generality, the inventors found through a large number of experiments that the error estimate of the position to be estimated is very close to the overall motion error trend of the robotic arm, indicating that it is feasible to use the inverse distance interpolation method for the selected data points to estimate the position error.

Exemplarily, the terminal device can detect the error of the estimated position by interpolating the motion trajectory of the end position of the robot arm at 150mm, 250mm and 350mm through a camera, and then perform error detection comparison with the laser tracker at the same position to determine the actual detection error of the camera at that position. For the motion trajectory of 150mm, the average value is 0.534mm, for the motion trajectory of 250mm, the average value is 0.931mm, and for the motion trajectory of 350mm, the average value is 1.405mm, with a total average value of 0.967mm. At the same time, after obtaining the planar motion error of the position to be estimated by interpolating the inverse distance method, the terminal device can compensate the error of the points corresponding to the camera detection motion trajectory one by one using the error value estimated by the inverse distance interpolation. After a large number of experiments, the inventor found that after interpolation compensation, the overall error value of the 150mm trajectory detection position was reduced from 0.534mm to 0.136mm, the overall error value of the 250mm trajectory detection position was reduced from 0.931mm to 0.229mm, and the overall error value of the 350mm trajectory detection position was reduced from 1.405mm to 0.262mm. The overall error average value of the three detection trajectories was reduced from 0.967mm to 0.209mm.

In a possible implementation, the terminal device can also use an RBF neural network to predict the error of the position to be estimated, where the RBF neural network is a radial basis function network, which is an artificial neural network that uses a radial basis function as an activation function. The RBF neural network can approximate any nonlinear function, can handle the regularity that is difficult to analyze within the system, has good generalization ability, and has a fast learning convergence speed, and can be successfully applied to nonlinear function approximation.

Without loss of generality, the RBF neural network can be composed of three layers: input layer, hidden layer, and output layer. The first layer is the input layer, which is composed of input data; the second layer is the hidden layer, which is the transformation function of each unit, mapping the low-dimensional input to a high-dimensional space through a nonlinear Gaussian function; the third layer is the output layer, which responds to the input signal. The output from the hidden layer to the output is obtained by linear weighted evaluation to obtain the output value of the output layer, where the weight is The activation function used in the RBF neural network can be the Gauss radial basis function, and its expression can be:

,

In the formula, r= , represents the jth input data, represents the center point of the i-th sample data, is the average deviation of the core function in the hidden layer;

Therefore, the output function expression of RBF neural network training can be:

;

Without loss of generality, some other parameters of the RBF neural network can be as follows: the number of training samples is 125, the number of test samples is 20, the radial basis velocity is 2.0, and the number of hidden layer neurons is 125. After the planar motion error of the position to be estimated is predicted by the RBF neural network, the terminal device can compensate the error of the points corresponding to the camera detection motion trajectory one by one with its predicted error value. After a large number of experiments, the inventor found that after compensation by the RBF neural network, the overall error value of the 150mm trajectory detection position was reduced from 0.534mm to 0.0768mm, the overall error value of the 250mm trajectory detection position was reduced from 0.931mm to 0.147mm, and the overall error value of the 350mm trajectory detection position was reduced from 1.405mm to 0.221mm. The overall error average of the three detection trajectories was reduced from 0.967mm to 0.148mm. Therefore, the effect of the RBF neural network on the error prediction of the position to be estimated and the compensation effect of the inverse distance interpolation method are both in line with expectations.

In S400, the second pixel coordinate information is compensated according to the coordinate error value information to generate compensated pixel coordinate information.

Specifically, after the terminal device determines the coordinate error value information, the terminal device may perform compensation processing on the second pixel coordinate information according to the coordinate error value information to generate compensated pixel coordinate information, wherein the compensated pixel coordinate information is used to describe the compensated second pixel coordinate information.

In some possible implementations, in order to realize the generation of compensated pixel coordinate information, please refer to FIG. 17 , step S400 includes but is not limited to the following steps:

In S410, compensation direction information is determined according to the first pixel coordinate information and the second pixel coordinate information.

Specifically, the terminal device can determine the compensation direction information based on the first pixel coordinate information and the second pixel coordinate information, with the first pixel coordinate information as the starting point and the second pixel coordinate information as the end point, wherein the compensation direction information is used to describe the direction of compensation.

In S420, based on the compensation direction information and the coordinate error value information, the second pixel coordinate information is compensated to generate compensated pixel coordinate information.

Specifically, after the terminal device determines the compensation direction information, the terminal device can perform compensation processing on the second pixel coordinate information based on the compensation direction information and the coordinate error value information to generate compensated pixel coordinate information, thereby improving the accuracy of robot arm positioning, wherein the compensated pixel coordinate information is used to describe the second pixel coordinate information after the coordinate error value information is compensated for the compensation direction information.

In some possible implementations, in order to achieve precise operation of the robot arm, please refer to FIG. 18 . After step S400, the method further includes but is not limited to the following steps:

In S500, compensated motion trajectory information is generated according to a plurality of compensated pixel coordinate information.

Specifically, the terminal device may sequentially connect multiple compensation pixel coordinate information to generate compensation motion trajectory information, wherein the compensation motion trajectory information is used to describe the motion trajectory composed of multiple compensation pixel coordinate information.

In S510, the movement of the end of the robot arm is controlled according to the compensation motion trajectory information.

Specifically, after the terminal device generates the compensation motion trajectory information, the terminal device can control the end of the robot arm to move according to the compensation motion trajectory information.

The implementation principle of the robot vision detection method based on laser compensation in the embodiment of the present application is: the terminal device can first obtain the end marker image set information based on the camera, and then perform the processing for each end marker image information: obtain the first pixel coordinate information of the theoretical position of the end of the robot arm, and then perform distortion correction processing on the end marker image information based on this, determine the second pixel coordinate information corresponding to the actual measured position of the end of the robot arm, and then further determine the coordinate error value information of the end of the robot arm in combination with the end position motion error calculation function, and finally, according to the coordinate error value information, compensate the second pixel coordinate information to generate compensated pixel coordinate information, thereby achieving compensation for the global error situation and improving the accuracy of robot arm positioning.

It should be noted that the size of the serial numbers of the steps in the above embodiments does not mean 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 the present application.

The embodiment of the present application further provides a robot vision detection system based on laser compensation. For ease of description, only the parts related to the present application are shown. As shown in FIG. 19 , the system 190 includes:

The terminal marker image set information acquisition module 1911 is used to continuously acquire terminal marker image set information based on a preset camera, wherein the terminal marker image set information includes a plurality of continuous terminal marker image information, and the shooting object of the terminal marker image information is a specified robotic arm terminal, and at least two light source markers are installed at the robotic arm terminal;

The first pixel coordinate information acquisition module 192 is used to obtain the first pixel coordinate information corresponding to the theoretical position of the end of the robot arm for each end marker image information, and perform distortion correction processing on the end marker image information according to the first pixel coordinate information and the preset distortion coefficient set information, so as to determine the second pixel coordinate information corresponding to the actual measured position of the end of the robot arm;

Coordinate error value information determination module 193: used to determine the coordinate error value information of the end of the robot arm according to the second pixel coordinate information and a preset end position motion error calculation function;

The compensated pixel coordinate information generating module 194 is used to perform compensation processing on the second pixel coordinate information according to the coordinate error value information to generate compensated pixel coordinate information.

It should be noted that the information interaction, execution process and other contents between the above-mentioned modules are based on the same concept as the method embodiment of the present application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

The embodiment of the present application also provides a terminal device, as shown in FIG20, the terminal device 200 of the embodiment includes: a processor 201, a memory 202, and a computer program 203 stored in the memory 202 and executable on the processor 201. When the processor 201 executes the computer program 203, the steps in the above-mentioned robot vision detection method embodiment are implemented, such as steps S100 to S400 shown in FIG1; or, when the processor 201 executes the computer program 203, the functions of each module in the above-mentioned device are implemented, such as the functions of modules 191 to 194 shown in FIG19.

The terminal device 200 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server, and the terminal device 200 includes but is not limited to a processor 201 and a memory 202. Those skilled in the art will appreciate that FIG. 20 is merely an example of the terminal device 200 and does not limit the terminal device 200, and may include more or fewer components than shown in the figure, or may combine certain components, or different components, for example, the terminal device 200 may also include input and output devices, network access devices, buses, etc.

Among them, the processor 201 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.; the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

The memory 202 may be an internal storage unit of the terminal device 200, such as a hard disk or memory of the terminal device 200, or the memory 202 may be an external storage device of the terminal device 200, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card (FlashCard), etc. equipped on the terminal device 200; further, the memory 202 may also include both an internal storage unit and an external storage device of the terminal device 200, and the memory 202 may also store a computer program 203 and other programs and data required by the terminal device 200, and the memory 202 may also be used to temporarily store data that has been output or is to be output.

An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the steps of each of the above method embodiments can be implemented. Among them, the computer program includes computer program code, which can be in source code form, object code form, executable file or some intermediate form, etc.; the computer-readable medium may include: any entity or device that can carry computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, all equivalent changes made according to the methods, principles, and structures of the present application should be included in the protection scope of the present application.

Claims (10)

1. A robot vision detection method based on laser compensation, characterized in that the method comprises: Based on a preset camera, continuously acquire end marker image set information, wherein the end marker image set information includes a plurality of continuous end marker image information, the shooting object of the end marker image information is a designated robotic arm end, and the robotic arm end is equipped with at least two light source markers; For each of the end marker image information: obtaining first pixel coordinate information corresponding to the theoretical position of the end of the robotic arm, and performing distortion correction processing on the end marker image information according to the first pixel coordinate information and the preset distortion coefficient set information, to determine second pixel coordinate information corresponding to the measured position of the end of the robotic arm; Determine the coordinate error value information of the end of the robotic arm according to the second pixel coordinate information and a preset end position motion error calculation function; The second pixel coordinate information is compensated according to the coordinate error value information to generate compensated pixel coordinate information. 2. The method according to claim 1, characterized in that before continuously acquiring the terminal marker image set information based on the preset camera, the method further comprises: Constructing an imaging model of the camera; Based on the camera, obtaining calibration plate image information; Performing grayscale processing on the calibration plate image information to generate grayscale image information; Based on a preset corner point detection algorithm and the grayscale image information, determining a plurality of calibration plate corner point information of the grayscale image information; For each of the calibration plate corner point information: obtaining the actual physical coordinate information of the corner point of the calibration plate corner point information; Generate reprojection error information according to a preset Euclidean distance calculation function, the calibration plate corner point information and the actual physical coordinate information of the corner point; Generate calibration accuracy result information according to the reprojection error information and preset error threshold information, wherein the calibration accuracy result information is calibration qualified information or calibration unqualified information; Accordingly, the method of continuously acquiring the terminal marker image set information based on the preset camera includes: If the calibration accuracy result information is the calibration qualified information, the terminal marker image set information is continuously acquired based on a preset camera. 3. The method according to claim 2, characterized in that the distortion coefficient set information includes first distortion coefficient information and second distortion coefficient information; for each of the end marker image information: obtaining first pixel coordinate information corresponding to the theoretical position of the end of the robotic arm, and performing distortion correction processing on the end marker image information according to the first pixel coordinate information and the preset distortion coefficient set information, and determining second pixel coordinate information corresponding to the measured position of the end of the robotic arm, comprising: For each of the end marker image information: obtain the first pixel coordinate information corresponding to the theoretical position of the end of the robotic arm, wherein the first pixel coordinate information is: , In the formula, is the horizontal coordinate of the first pixel coordinate information, is the actual horizontal coordinate of the image center of the terminal marker image information in the preset image coordinate system, is the theoretical horizontal coordinate of the image center of the terminal marker image information in a preset image coordinate system, , is the focal length information of the camera, mm, is the first pixel size with respect to the horizontal axis, is the ordinate of the first pixel coordinate information, is the actual ordinate of the image center of the terminal marker image information in the preset image coordinate system, is the theoretical ordinate of the image center of the terminal marker image information in a preset image coordinate system, , is the second pixel size with respect to the ordinate; According to the first pixel coordinate information, the preset first distortion coefficient information and the second distortion coefficient information, the end marker image information is subjected to distortion correction processing to determine the distortion corrected pixel coordinate information corresponding to the measured position of the end of the robotic arm, wherein the distortion corrected pixel coordinate information is: , In the formula, is the horizontal coordinate of the distortion-corrected pixel coordinate information, is the horizontal coordinate of the first pixel coordinate information, is the first distortion coefficient information, and the first distortion coefficient information is used to describe the first-order radial distortion coefficient. , , is the second distortion coefficient information, where the second distortion coefficient information is used to describe the second-order radial distortion coefficient. , is the preset third-order radial distortion coefficient, is the preset first-order tangential distortion coefficient, is the preset second-order tangential distortion coefficient, is the ordinate of the distortion-corrected pixel coordinate information, is the ordinate of the first pixel coordinate information; For each of the distortion-corrected pixel coordinate information of the first marker and each of the distortion-corrected pixel coordinate information of the second marker: performing least squares fitting processing on the distortion-corrected pixel coordinate information to generate ideal circle feature set information of a target ideal circle, wherein the first marker is used to describe any one of the light source markers, the second marker is used to describe any other of the light source markers, and the ideal circle feature set information includes the center pixel coordinate information and radius information of the target ideal circle The second pixel coordinate information is determined according to the plurality of circle center pixel coordinate information. 4. The method according to claim 3, characterized in that after determining the second pixel coordinate information according to the plurality of circle center pixel coordinate information, the method further comprises: Based on the preset number of repeated detection times, obtaining multiple repeated measurement sample data information of the end of the robotic arm at the same position; Generate repeated measurement sample data average value information according to the plurality of repeated measurement sample data information; The standard deviation value information of the end of the robot arm at the same position is determined according to the average value information of the repeated measurement sample data. 5. The method according to claim 4, characterized in that before determining the coordinate error value information of the end of the robotic arm according to the second pixel coordinate information and a preset end position motion error calculation function, the method further comprises: Generate spatial interpolation distance information based on the second pixel coordinate information and a preset distance inverse weight interpolation function; According to the spatial interpolation distance information and the second pixel coordinate information, a plurality of predicted pixel coordinate information is generated in an equally spaced array, wherein the predicted pixel coordinate information is used to describe the second pixel coordinate information generated in an equally spaced array, the distance between the plurality of predicted pixel coordinate information is the spatial interpolation distance information, and the distance between the predicted pixel coordinate information and the second pixel coordinate information is the spatial interpolation distance information; Correspondingly, determining the coordinate error value information of the end of the robotic arm according to the second pixel coordinate information and a preset end position motion error calculation function includes: The coordinate error value information of the end of the robotic arm is determined according to the horizontal coordinate of the second pixel coordinate information and a preset end position motion error calculation function, wherein the end position motion error calculation function is: , In the formula, is the coordinate error value information, is the total amount of the second pixel coordinate information, is the order of the second pixel coordinate information, For the The difference between the horizontal coordinate corresponding to the second pixel coordinate information and the horizontal coordinate corresponding to the previous second pixel coordinate information. 6. The method according to claim 4, characterized in that the step of compensating the second pixel coordinate information according to the coordinate error value information to generate compensated pixel coordinate information comprises: determining compensation direction information according to the first pixel coordinate information and the second pixel coordinate information; Based on the compensation direction information and the coordinate error value information, the second pixel coordinate information is compensated to generate compensated pixel coordinate information, wherein the compensated pixel coordinate information is used to describe the second pixel coordinate information after the coordinate error value information is compensated toward the compensation direction information. 7. The method according to claim 6, characterized in that after compensating the first pixel coordinate information according to the coordinate error value information to generate compensated pixel coordinate information, the method further comprises: Generate compensation motion trajectory information according to the plurality of compensation pixel coordinate information; The movement of the end of the robotic arm is controlled according to the compensated motion trajectory information. 8. A robot vision detection system based on laser compensation, characterized in that the system comprises: The terminal marker image set information acquisition module is used to continuously acquire the terminal marker image set information based on a preset camera, wherein the terminal marker image set information includes a plurality of continuous terminal marker image information, and the shooting object of the terminal marker image information is a designated robotic arm terminal, and at least two light source markers are installed at the robotic arm terminal; The first pixel coordinate information acquisition module is used to obtain the first pixel coordinate information corresponding to the theoretical position of the end of the robotic arm for each end marker image information, and perform distortion correction processing on the end marker image information according to the first pixel coordinate information and the preset distortion coefficient set information, so as to determine the second pixel coordinate information corresponding to the measured position of the end of the robotic arm; A coordinate error value information determination module: used to determine the coordinate error value information of the end of the robotic arm according to the second pixel coordinate information and a preset end position motion error calculation function; Compensated pixel coordinate information generating module: used to perform compensation processing on the second pixel coordinate information according to the coordinate error value information to generate compensated pixel coordinate information. 9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method according to any one of claims 1 to 7 when executing the computer program. 10. A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the method according to any one of claims 1 to 7.