Loose-leaf notebook inner page hole site machining self-adaptive positioning method and system and electronic equipment

Abstract

The invention relates to a loose-leaf notebook inner page hole site processing self-adaptive positioning method and system and electronic equipment, and the method comprises the steps: obtaining the morphological feature information of the edge of a to-be-processed paper pile, and determining a datum reference line and an actual edge contour curve according to the morphological feature information; and based on a deviation mapping model, determining a pose deviation vector according to the datum reference line and the actual edge contour curve, generating a positioning adjustment instruction according to the pose deviation vector by utilizing a self-adaptive compensation algorithm, and driving a positioning mechanism to carry out pose correction on the to-be-processed paper pile. And aligning the hole site machining center with the actual edge profile curve. Accurate sensing and self-adaptive compensation of the shape of the edge of the inner page of the loose-leaf notebook are achieved, the problem that traditional mechanical positioning is poor in adaptability is effectively solved, high precision and high quality of hole site machining are guaranteed, and the yield is increased.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of paper processing automation, and particularly relates to a method and system for adaptive positioning of hole positions in a notebook, and an electronic device. BACKGROUND

[0002] Notebooks are widely used in daily office work, study and professional recording due to their flexibility and replaceability. Hole position processing of notebook inner pages is a key link in production and manufacturing, and the core requirement is to ensure that the punched hole positions are accurately and consistently positioned with the edge of the inner pages, so that all the inner pages can be smoothly and neatly bound in the notebook binder.

[0003] At present, in order to ensure the page turning fluency and appearance neatness of the notebook inner pages, high-precision hole position processing is generally performed on the cut paper stack. The operation process is to place a stack of paper to be processed as inner pages on a processing platform, and to position the edge of the paper stack by using a fixed physical baffle or side push plate to limit the edge position. After positioning, a punching or drilling head processes along a preset path corresponding to a specific notebook specification (such as 26 holes, 6 holes, etc.).

[0004] When the paper material is soft or the cutting size has a tolerance, simple mechanical contact positioning can easily cause slight deformation or rebound of the edge of the paper stack, so that the actual punching center line is not parallel to the edge of the paper, and there is a deviation, which results in low accuracy and yield of hole position processing. SUMMARY

[0005] In view of this, in order to at least partially improve the above problems, the present application provides a method and system for adaptive positioning of hole positions in a notebook, and an electronic device.

[0006] The present application provides a method for adaptive positioning of hole positions in a notebook, comprising: obtaining morphological feature information of the edge of the paper stack to be processed, wherein the morphological feature information represents the edge flatness and relative position deviation of the paper stack in the stacked state; determining a reference line and an actual edge contour curve according to the morphological feature information; determining a pose deviation vector based on a deviation mapping model according to the reference line and the actual edge contour curve, wherein the deviation mapping model represents the geometric transformation relationship between the edge contour and the physical processing coordinate system, and the pose deviation vector includes a translation deviation component and a rotation deviation component; generating a positioning adjustment instruction according to the pose deviation vector by using an adaptive compensation algorithm, and driving a positioning mechanism to correct the pose of the paper stack to be processed, so that the hole position processing center is aligned with the actual edge contour curve.

[0007] The application further provides a hole position processing adaptive positioning system for inner pages of a notebook, which is used to implement the method according to any one of the above, comprising: An acquisition module is configured to acquire morphological feature information of an edge of a paper stack to be processed. A determination module is configured to determine a reference line and an actual edge contour curve according to the morphological feature information. A deviation calculation module is configured to determine a pose deviation vector according to the reference line and the actual edge contour curve based on a deviation mapping model. An execution control module is configured to generate a positioning adjustment instruction according to the pose deviation vector by using an adaptive compensation algorithm, and drive a positioning mechanism to correct the pose of the paper stack to be processed.

[0008] The application further provides an electronic device comprising a memory and a processor, wherein the memory stores a computer application program, and the processor is configured to execute the computer application program, and wherein the computer application program is configured to implement the adaptive positioning method for hole position processing of inner pages of a notebook according to any one of the above when executed by the processor.

[0009] The application has the following beneficial effects: By acquiring morphological feature information of an edge of a paper stack to be processed, determining a reference line and an actual edge contour curve according to the morphological feature information, determining a pose deviation vector according to the reference line and the actual edge contour curve based on a deviation mapping model, generating a positioning adjustment instruction according to the pose deviation vector by using an adaptive compensation algorithm, and driving a positioning mechanism to correct the pose of the paper stack to be processed, the hole position processing center is aligned with the actual edge contour curve. The application realizes accurate perception and adaptive compensation of the edge morphology of inner pages of a notebook, effectively solves the problem of poor adaptability of traditional mechanical positioning, ensures high precision and high quality of hole position processing, and improves the yield rate. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the application, and therefore should not be considered as a limitation to the scope. For those skilled in the art, other related drawings can also be obtained without creative labor.

[0011] Figure 1 is a flowchart of an adaptive positioning method for hole position processing of inner pages of a notebook according to an embodiment of the application.

[0012] Figure 2 is a flowchart of one step of an adaptive positioning method for hole position processing of inner pages of a notebook according to an embodiment of the application.

[0013] Figure 3 is a structure diagram of a differential drive side rule structure in one step of a method for adaptive positioning of a hole position in a notebook.

[0014] Figure 4 is a structure diagram of a system for adaptive positioning of a hole position in a notebook according to another embodiment of the present application.

[0015] Figure 5 is a block diagram of an electronic device according to an embodiment of the present application.

[0016] Figure 6 is a block diagram of a readable storage medium according to an embodiment of the present application. DETAILED DESCRIPTION

[0017] The terms “first”, “second”, and “third” and the like in the description and in the claims of the present application and the above drawings are used for distinguishing between similar objects, not necessarily for describing a specific sequential or chronological order. Moreover, the terms “comprises”, “comprising”, “includes”, “including” and the like are meant to encompass not exclusively a situation wherein for example a process, method, object, or device comprises, includes or consists of steps or modules listed in the claims, but can also consist of additional steps or modules, not listed, or consist only of inherently the steps or modules for the process, method, object or device. The terms “comprises”, “comprising”, “includes”, “including” and the like mean, when used in the description and in the claims of the present application, that an element preceded by the word “comprises”, “comprising”, “includes”, “including” or “has” is not limited to a list of elements recited, but can optionally include other elements not expressly listed or inherent to such process, method, object or apparatus. The terms “plurality” and “a plurality” mean “two or more” of the specified feature unless otherwise indicated.

[0018] In order to make the technical personnel in the art better understand the scheme of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0019] First, the terms used in the following various embodiments are explained.

[0020] Morphological feature information means a set of data reflecting the geometric distribution of the edges of the paper stack in space, usually represented as a series of discrete two-dimensional coordinate points.

[0021] Pose deviation vector refers to a vector representing the difference between the actual position of an object and the ideal position, including a translation component and a rotation component.

[0022] Homography matrix is a matrix describing the projection mapping relationship between two planes, used to convert image pixel coordinates to physical world coordinates.

[0023] Random Sample Consensus (RANSAC) is an algorithm for estimating the parameters of a mathematical model from a data set containing outliers by an iterative method.

[0024] The method can be applied to an automatic punching device equipped with a visual guidance system. The method accurately perceives the edge shape of a paper stack to be processed through machine vision, and drives a high-precision positioning mechanism to perform adaptive pose correction, effectively solving the hole position deviation problem. Please refer to Figures 1-3 The method can include: S1, acquiring shape feature information of an edge of a paper stack to be processed, wherein the shape feature information represents the edge flatness and relative position deviation of the paper stack in a stacked state.

[0025] Illustratively, the paper stack to be processed refers to a collection of loose-leaf inner page paper placed on a processing table, waiting for punching operation. Due to the possible displacement, warping or uneven edge of the paper during stacking and conveying, the overall edge contour will deviate from the ideal state. The shape feature information is used to represent this deviation. Specifically, the data set of the target edge geometry of the paper stack can be collected by a visual system in the image pixel coordinate system, which includes the accurate coordinates of multiple edge feature (pixel) points.

[0026] S2, determining a reference line and an actual edge contour curve according to the shape feature information.

[0027] Illustratively, a straight line that best represents the overall edge trend of the paper stack can be robustly fitted from the edge pixel points that may contain noise such as paper burrs and paper dust according to the shape feature information, i.e. the actual edge contour curve, and associated with the inherent reference line of the automatic punching device. Considering that the least squares fitting of the shape feature information will be disturbed by noise, the improved random sample consensus algorithm is used for robust fitting in this embodiment.

[0028] S3, determining a pose deviation vector based on a deviation mapping model according to the reference line and the actual edge contour curve, wherein the deviation mapping model represents the geometric transformation relationship between the edge contour and the physical processing coordinate system, and the pose deviation vector includes a translation deviation component and a rotation deviation component.

[0029] Exemplarily, the deviation mapping model refers to a mathematical relationship or algorithm model for accurately converting the image deviation recognized by the vision system, i.e., pixel coordinates, into motion compensation instructions (e.g., millimeters, degrees, etc.) that can be executed by the mechanical system in the actual physical world. The deviation mapping models in the prior art include an affine transformation model, a projection geometry model, an error Jacobian matrix model, etc. These models can generally only convert coordinates, but cannot decompose and control the action of the actual mechanical device according to the deviation between the actual mechanical motion and the ideal mechanical motion. Based on this, in the present embodiment, the deviation mapping model is divided into a front-end vision geometry model, which can complete the mapping of the image pixel domain corresponding to the actual edge contour curve to the physical coordinate domain of the physical machining coordinate system, so as to obtain a position deviation vector by comparing with the reference baseline, and a back-end analytical error Jacobian matrix model, which can decouple the unified pose deviation vector into two independent control quantities of the motor, i.e., displacement control quantity and angle control quantity that can be directly used for motion control. In addition, the back-end can also use a kinematics inverse solution model, which is not limited herein.

[0030] S4, generating a positioning adjustment instruction according to the pose deviation vector by using an adaptive compensation algorithm, and driving the positioning mechanism to correct the pose of the paper stack so that the hole position machining center is aligned with the actual edge contour curve.

[0031] Exemplarily, the back-end can use a kinematics inverse solution model based on the adaptive compensation algorithm to dynamically calculate the motion amount of the positioning mechanism according to the size and change trend of the pose deviation vector.

[0032] Exemplarily, the adaptive compensation algorithm applies a kinematics inverse solution model of the differential side gauge when calculating the control quantity, which can dynamically calculate the independent motion amounts of the two motors in the positioning mechanism and generate corresponding adjustment instructions according to the size and change trend of the pose deviation vector.

[0033] To realize adaptive deviation correction of the paper stack translation and rotation, the positioning mechanism of the present embodiment adopts a differential drive side gauge structure, please refer to the attached drawings. The structure mainly includes a side gauge push plate 4, a first servo motor 1 (motor A) and a second servo motor 2 (motor B). The drive rods of the first servo motor 1 and the second servo motor 2 are respectively hinged or fixed to the front and rear ends of the side gauge push plate 4 along the length direction, for driving the side gauge push plate 4 to move along the X-axis direction of the physical coordinate system.

[0034] The differential working principle is as follows: When the same motion command is sent to both motors, they move synchronously, driving the side guide push plate 4 to translate as a whole, pushing the paper stack 3 for lateral positioning; when the controller sends different motion commands to the two motors, they generate differential motion, causing the side guide push plate 4 to deflect while translating, thereby applying a rotational torque to the paper stack 3 to achieve its angle correction. Through the above-mentioned combined translation and rotation motion, the actual edge contour of the paper stack 3 is finally aligned with the hole processing center. It should be noted that the differential drive side guide structure in this embodiment is only an example, and other differential drive side guide structures that can achieve the functions of this application can also be used, and are not limited to this one.

[0035] As described above, this application achieves accurate perception and adaptive compensation of the edge shape of the inner pages of the loose-leaf notebook, effectively solving the problem of poor adaptability of traditional mechanical positioning, ensuring high precision and high quality of hole processing, and improving yield.

[0036] In one possible implementation, step S1 may include: S11. Obtain the raw image data; For example, when the conveying mechanism transports the paper stack to a preset position on the processing table of the automated punching equipment, the photoelectric sensor installed above the processing table sends an arrival signal, which triggers a fixed high-resolution industrial camera to take a picture and acquire a grayscale image containing the entire side of the paper stack target.

[0037] S12. Based on the grayscale gradient threshold, extract the region of interest from the original image data; For example, in order to reduce computation and eliminate irrelevant background noise interference, a long strip area can be set in the grayscale image according to the standard size of the paper stack to be processed, such as A4 or A5. This area should completely cover the target edge of the paper stack. By calculating the grayscale gradient of this area and setting a grayscale gradient threshold according to the significant contrast between the paper stack and the background, such as a dark processing table, the precise region of interest can be cropped according to the grayscale gradient threshold.

[0038] S13. Perform adaptive binarization and edge detection on the region of interest to obtain a set of edge feature point coordinates, and use the set of edge feature point coordinates as the morphological feature information.

[0039] Edge detection can employ either the Canny or Sobel operator. As an example, Gaussian filtering can be applied to the region of interest to smooth noise. Subsequently, the Canny edge detection operator is used for preliminary edge extraction. To further improve positioning accuracy, a moment-based sub-pixel edge localization algorithm is applied to the initially extracted edge points. This algorithm analyzes the grayscale distribution of the edge point's neighborhood, improving edge localization accuracy to the sub-pixel level, such as 0.1 pixels. The final output is an ordered, high-precision set of edge feature points: P edge ={(u1,v1),(u2,v2),...,(u n ,v n )}, where, (u n ,v n P represents the coordinates of the nth edge feature point in the image pixel coordinate system, where n = 1, 2, ..., n; edge This refers to morphological feature information.

[0040] Considering the potential interference from burrs or paper scraps at the edge of the paper stack, i.e., noise, to improve the accuracy of straight line fitting, in one possible implementation, step S2, which determines the reference line and the actual edge contour curve based on the morphological feature information, may include: Step S21: Iteratively sample the morphological feature information based on the random sampling consensus algorithm, remove out-of-local noise, and retain the in-local point set; For example, the morphological feature information is iteratively sampled based on a random sampling consensus algorithm, such as RANSAC. Specifically, the morphological feature information P can be obtained from... edge Two edge feature points are randomly selected from the data to determine a straight line, and the morphological feature information P is calculated. edge Calculate the Euclidean distance from all other edge feature points to the line. Set a distance threshold; if the distance from a point to the line is less than the threshold, mark it as an inlier; otherwise, mark it as an outlier. Multiple inliers form an inlier set. Remove outliers, i.e., remove burrs or paper scraps. Repeat the above process m times. After each iteration, count the number of inliers obtained in each sampling. Select the line and inlier set corresponding to the sampling with the most inliers as the optimal result and save it.

[0041] Step S22: Obtain the initial pixel contour in the image pixel coordinate system based on the local point set; For example, the least squares method is used to fit a straight line to the selected optimal result and the set of interior points to obtain the initial pixel contour line equation in the image pixel coordinate system: v=k ing *u+b ing ; Wherein, the king b represents the slope of a straight line. ing denoted by , u represents the horizontal coordinate of the pixel in the image pixel coordinate system, and v represents the vertical coordinate of the pixel in the image pixel coordinate system.

[0042] Step S23: Obtain the calibration parameters of the vision system; For example, to achieve accurate mapping from image pixel space to physical processing space, the vision system needs to be calibrated beforehand, and calibration parameters need to be obtained. In this embodiment, the calibration parameters mainly include a homography matrix, which is a mathematical model describing the projection transformation relationship between two planes, namely the image plane of the vision system and the physical plane of the processing table. This matrix includes the camera's internal parameters, such as focal length and principal point, as well as the camera's external parameters relative to the processing table, such as rotation angle and translation distance. It can directly and linearly map the pixel coordinates in the image to physical coordinates in the physical processing coordinate system, thereby avoiding the complex step-by-step coordinate system transformation process and improving computational efficiency and positioning accuracy.

[0043] Step S24: Based on the calibration parameters, map the initial pixel contour to the physical processing coordinate system and fit the actual edge contour curve; For example, two key points on the initial pixel contour can be selected, such as the endpoints of the fitted straight line. Based on the inverse transformation of the homography matrix, the coordinates (u,v) of its edge feature points are mapped to coordinates (x,y) in the physical processing coordinate system. Based on the set of coordinates in the physical processing coordinate system obtained after mapping, a straight line is fitted again using the least squares method. This straight line is in millimeters in the physical processing coordinate system, which is the actual edge contour curve, and its equation can be expressed as: y=k phy *x+b phy ; Wherein, the k phy b represents the slope of the straight line in the physical machining coordinate system. phy y represents the intercept in the physical machining coordinate system, x represents the horizontal coordinate of the edge feature point in the physical machining coordinate system, and y represents the vertical coordinate of the edge feature point in the physical machining coordinate system.

[0044] Step S25: Obtain the mechanical zero-point parameters of the automated drilling equipment, and generate the reference line in the physical machining coordinates based on the mechanical zero-point parameters.

[0045] For example, the reference line is the absolute reference for hole machining, derived from the mechanical coordinate system of the automated drilling equipment. Specifically, the coordinate position of the preset mechanical zero point in the physical machining coordinate system and the direction of each axis of the machine tool are obtained from the automated drilling equipment. This mechanical zero point is the physical reference origin determined after the automated drilling equipment is assembled and debugged, and cannot be changed. Machining process requirements are obtained, such as all holes need to be parallel to the Y-axis of the equipment and x0 mm away from the mechanical zero point. The equation of the reference line can then be calculated using the mechanical zero point parameters. For example, the reference line can be defined as a straight line parallel to the Y-axis and passing through the point (x0, 0), then its equation is x = x0.

[0046] To accurately quantify camera imaging and its geometric relationship with the processing table, as an example, step S23 of obtaining the calibration parameters of the vision system may include: Step S231: Control the vision system to acquire calibration board images, wherein the calibration board images include a calibration board placed on the processing table, and the surface of the calibration board is drawn with specific geometric patterns arranged according to a preset rule; For example, a high-precision calibration plate is first selected, such as one made of optical glass or ceramic with a low coefficient of thermal expansion, whose surface is decorated with specific geometric patterns arranged according to a preset rule. This embodiment uses the internationally widely used black and white checkerboard pattern as an example, which has the advantages of clear corner points and ease of automatic detection. During calibration data acquisition, a mechanical limiting device, such as a high-precision L-shaped positioning ruler or positioning pin, is used to fix the calibration plate at a specific preset position on the processing table. The position of this mechanical limiting device is precisely adjusted so that the origin of the specific geometric pattern on the calibration plate, such as the first corner point of the upper left corner of the checkerboard, coincides with the mechanical zero point of the hole-making equipment, such as the intersection of the center line of the drilling mold and the back gauge. A vision system, such as an industrial camera, mounted above the processing table is controlled to focus and capture one or more images containing the complete calibration plate, i.e., calibration plate images. It should be noted that to ensure calibration accuracy, uniform lighting must be ensured so that the pattern has clear contrast and no strong reflections or shadows.

[0047] Step S232: Obtain the unit physical dimensions of the specific geometric pattern to determine multiple key feature points, and establish the physical coordinates of several key feature points on the specific geometric pattern in the physical processing coordinate system according to the arrangement rule; For example, for a checkerboard pattern, the physical dimensions of the unit, i.e., the side length of each black and white square, are known. A two-dimensional physical machining coordinate system can be defined on the calibration board plane, for example, with the first interior corner point (the intersection of the black and white squares) at the top left corner of the checkerboard as the origin O. w Let X be the horizontal direction to the right. w The axis, with the vertical downward direction as Y. wBased on the arrangement of the chessboard grid and the known physical dimensions of the units, each inner corner point, i.e., the coordinates of the key feature points in this physical coordinate system, can be uniquely determined. For example, the corner point of the i-th row and j-th column. Thus, the physical coordinate set of all key feature points can be established.

[0048] Step S233: Perform image processing on the calibration board image to extract the pixel coordinates that correspond one-to-one with the several key feature points; For example, the checkerboard pattern in the calibration board image can be automatically identified based on the findChessboardCorners function, and the preliminary pixel coordinates of all key feature points can be initially located to obtain an initial set of pixel coordinates A. To further improve the positioning accuracy, a small neighborhood, such as a 5x5 pixel window, can be defined centered on the initial set of pixel coordinates A, and the cornerSubPix function can be used for sub-pixel refinement. This algorithm can improve the corner point positioning accuracy to 0.1 pixels or even higher by fitting the gray-level gradient within the neighborhood. The refined coordinate set is the final extracted pixel coordinates that correspond one-to-one with the physical coordinate set.

[0049] Step S234: Based on the mapping relationship between the corresponding pixel coordinates and the physical coordinates, the homography matrix is ​​obtained by using the least squares method or the direct linear transformation algorithm, and the homography matrix is ​​used as the calibration parameter.

[0050] For example, calibration parameters can be solved using a mathematical model based on multiple sets of corresponding physical-pixel coordinate pairs. Specifically, since camera imaging follows the pinhole imaging principle, i.e., a physical spatial point [X... w ,Y w Z w ] T (Since the calibration board is planar, the Z-coordinate is 0) and its projected pixel [u,v] T The relationship can be expressed by the homography transformation formula. To use a matrix to uniformly represent the transformation, homogeneous coordinates are adopted, and the coordinates of the physical space point are set as [X...]. w ,Y w ,1] T The homogeneous coordinates of the corresponding pixel in the image are [u, v, 1]. T Then the two satisfy the following relationship: [u,v,1] T =s*H*[X w ,Y w ,1] T ; Where s represents the scale factor, and H is the homography matrix to be determined, which is a 3*3 matrix; In this embodiment, the homography matrix H comprehensively includes the camera's intrinsic parameters, such as focal length and principal point, as well as the camera's extrinsic parameters relative to the processing table, such as rotation and translation. For the specific application of "processing the inner page holes of loose-leaf notebooks" described in this application, since the target to be measured, i.e., the edge of the paper stack, is always located on a fixed physical processing table plane, directly calibrating and storing the homography matrix H is the most efficient and direct mapping method. It can transform the image pixel coordinates (u, v) into physical processing coordinates (x, y) in one step without explicitly separating the intrinsic parameter matrix, rotation matrix, and translation vector. Finally, the homography matrix H is solidified and can be called upon during subsequent online positioning to achieve accurate mapping from pixel space to physical space.

[0051] In one possible implementation, step S3 may include: S31. In the physical machining coordinate system, calculate the angle between the normal vector of the actual edge contour curve and the normal vector of the reference line, and use the angle as a rotational deviation component. For example, the rotational deviation component characterizes the overall tilt angle of the actual edge of the paper stack, which can be obtained by calculating the angle between the normal vectors of two straight lines; specifically, let the equation of the reference line be x=x0, this line is perpendicular to the x-axis, and its unit normal vector n ref It can be defined as (1, 0), with the direction pointing towards the positive X-axis. Let the equation of the actual edge contour curve in the physical machining coordinate system be y=k. phy *x+b phy Its direction vector is (1, k) phy If ), then the unit normal vector n perpendicular to it is... act It can be obtained through calculation, for example: n act = (-k phy , 1) Then normalize; the rotational deviation component Δθ is the same as the two normal vectors n mentioned above. ref With n act The angle between them can be calculated using the dot product formula: Δθ = arccos(n act ·n ref ), where · represents the dot product operation, and Δθ is in radians, which can be converted to degrees as needed.

[0052] S32. Under the physical processing coordinate system, calculate the vertical Euclidean distance from the midpoint of the actual edge contour curve to the reference line, and use the vertical Euclidean distance as the translational deviation component. For example, the translational deviation component Δx characterizes the overall positional offset of the paper stack edge in the direction perpendicular to the reference line; a representative point on the actual edge profile curve can be selected, and in this embodiment, the midpoint C of the actual edge profile curve is used. actAs a feature point, this midpoint can be calculated by averaging the X coordinates of all valid edge points mapped to the physical machining coordinate system. mean， Substitute it into the equation of the straight line y=k phy *x+b phy Find the corresponding Y coordinate. mean Thus, feature points (X) are obtained. mean Y mean Then, the vertical distance from the feature point to the reference line is calculated, which is the translational deviation component Δx.

[0053] S33. Construct the pose deviation vector based on the rotation deviation component and the translation deviation component.

[0054] For example, the pose deviation vector D = [Δx, Δθ] T , where D represents the pose deviation vector.

[0055] It should be understood that, in the embodiments of this application, by determining the pose deviation vector based on the deviation mapping model, the complex edge morphology is transformed into quantifiable control parameters, which reduces the reliance on human experience and enhances the robustness of the positioning method.

[0056] In order to convert the deviation into a mechanical action that can eliminate the deviation in real time and accurately, in one possible implementation, step S4 may include: S41. Determine the predicted deviation D based on the historical deviation vector and historical compensation gain of the previous processing cycle. pred ; For example, the historical sequence of pose deviation vectors from the previous processing cycle or the previous N cycles and their corresponding, actually effective compensation gains can be obtained. By using linear extrapolation or exponential smoothing, the prediction deviation D for the current cycle can be determined. pred ; S42, Combine the pose deviation vector with the prediction deviation D pred The input is fed into the position closed-loop control model to determine the target compensation amount. The position closed-loop control model is constructed based on the fuzzy PID control algorithm. The controller of the fuzzy PID control algorithm dynamically adjusts the proportional coefficient, integral coefficient and derivative coefficient according to the magnitude of the pose deviation vector. For example, the current actual pose deviation vector is denoted as D. k =[Δx k , Δθ k ] T , and the prediction deviation D pred As inputs to the fuzzy PID controller, the translation deviation Δx is respectively... k and rotational deviation Δθ kThe absolute value is fuzzified and divided into linguistic variables such as large, medium, small, and zero. Based on preset fuzzy rules, for example, according to the magnitude of the pose deviation vector, two independent PID parameters are dynamically adjusted, including proportional coefficient, integral coefficient, and derivative coefficient. These two PID parameters correspond to the translation deviation Δx. k and rotational deviation Δθ k The channel; after defuzzification and PID calculation, outputs the target compensation amount in the physical pose space for the current processing cycle: (C x C θ ) T ; S43. The target compensation amount is converted into a multi-axis drive signal of the positioning mechanism as the positioning adjustment command.

[0057] For example, the target pose compensation amount is input to the back end of the deviation mapping model. In the inverse kinematics model, the inverse kinematics model can be based on the distance L between the driving points of the two motors of the positioning mechanism, that is, let the horizontal distance L between the action points of the first servo motor (motor A) and the second servo motor (motor B) on the side guide push plate be L. In order to achieve the target compensation amount (C x C θ ), calculate the target feed amount (S) required to drive the two motors. A S B ), that is, the target displacement, where, ; ; Wherein, the S A S is the target feed amount for motor A. B For the target feed rate of motor B, C θ This represents the total rotational compensation amount that needs to be compensated. If C θ If the value is not equal to 0, then motor A and motor B will generate differential displacement, driving the push plate to deflect to correct the rotational error. This represents the total translational compensation amount required, and P represents the gain coefficient set for the mechanical transmission. The controller will then calculate the target feed amount (S). A S B The signal is converted into a multi-axis drive signal for the positioning mechanism, such as a pulse signal or a direction signal, and sent to the servo driver as a positioning adjustment command.

[0058] As an optional embodiment, to ensure the final machining accuracy, the method may further include an iterative correction step: After the positioning mechanism executes the positioning adjustment command, it calculates the residual deviation vector based on the secondary morphological feature information, that is, by comparing the newly obtained actual edge contour curve with the same reference line. If the residual deviation vector is greater than the preset accuracy allowable threshold, a secondary fine-tuning command is generated based on the residual deviation vector to indicate that there is a residual error that needs to be corrected. This continues until the residual deviation vector converges within the accuracy allowable threshold. In other words, the tertiary morphological feature information can be obtained and iterated until the residual deviation vector is less than the preset accuracy allowable threshold. At this point, the positioning can be determined to be successful, and the drilling operation can be started.

[0059] Furthermore, before executing S4, in order to prevent equipment damage caused by foreign objects or malfunctions, the method also includes: determining whether the magnitude of the pose deviation vector exceeds a preset safety boundary threshold; if it exceeds, generating an abnormal alarm signal and interrupting the process, wherein the abnormal alarm signal indicates that the stacking state of the paper stack to be processed is severely disordered or the vision system calibration fails.

[0060] It is understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0061] Based on the same inventive concept, this application also provides an adaptive positioning system for processing the inner page holes of the loose-leaf book as described above. The solution provided by this system is similar to the solution described in the above method; therefore, the specific limitations in one or more system embodiments provided below can be found in the limitations of the method described above, and will not be repeated here.

[0062] In one exemplary embodiment, such as Figure 4 As shown, an adaptive positioning system for machining the inner page holes of a loose-leaf notebook is provided, including: an acquisition module 1, a determination module 2, a deviation calculation module 3, and an execution control module 4. The above system corresponds to the aforementioned method embodiment and can implement the corresponding method steps. Its implementation principle and technical effect are similar, and will not be described again here.

[0063] Please see Figure 5This application also provides an electronic device. The electronic device 90 may include a processor 91, a memory 92, and a computer application program, wherein: The memory 92 is used to store the computer application, and the memory may also be flash memory. The computer application is, for example, an application that implements the various method embodiments described above.

[0064] Processor 91 is configured to execute the computer application stored in the memory to implement the steps in the various method embodiments described above. For details, please refer to the relevant descriptions in the preceding method embodiments.

[0065] Alternatively, the memory 92 can be either standalone or integrated with the processor 91.

[0066] When the memory 92 is a device independent of the processor 91, the electronic device 90 may further include: Bus 93 is used to connect the memory 92 and the processor 91.

[0067] Please see Figure 6 This application also provides a readable storage medium storing a computer application program, which, when executed by a processor, implements the methods of the above-described method embodiments.

[0068] The readable storage medium can be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of computer programs from one location to another. A computer storage medium can be any available medium accessible to a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located on an Application-Specific Integrated Circuit (ASIC). s In an ASIC (Integrated Circuit-Based ASIC), the processor and readable storage medium can reside within the user equipment. Alternatively, the processor and readable storage medium can also exist as discrete components within the communication device.

[0069] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this application can be directly manifested as being executed by a hardware processor, or executed by a combination of hardware and software modules within the processor.

[0070] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein, and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for adaptive positioning of hole locations in a notebook, characterized in that, The method comprises the following steps: Obtain the morphological feature information of the edge of the paper stack to be processed, wherein the morphological feature information represents the edge flatness and relative position deviation of the paper stack in the stacked state; Determine the reference line and the actual edge contour curve according to the morphological feature information; Determine the pose deviation vector according to the reference line and the actual edge contour curve based on a deviation mapping model, wherein the deviation mapping model represents the geometric transformation relationship between the edge contour and the physical processing coordinate system, and the pose deviation vector includes a translation deviation component and a rotation deviation component; Generate a positioning adjustment instruction according to the pose deviation vector by using an adaptive compensation algorithm, and drive the positioning mechanism to correct the pose of the paper stack to be processed, so that the hole position processing center is aligned with the actual edge contour curve.

2. The method of claim 1, wherein, The method for obtaining the morphological feature information of the edge of the paper stack to be processed comprises the following steps: Obtain the original image data; Extract the region of interest in the original image data based on the gray gradient threshold value; Perform adaptive binarization processing and edge detection on the region of interest to obtain a set of edge feature point coordinates, and take the set of edge feature point coordinates as the morphological feature information.

3. The method of claim 1, wherein the method further comprises: The method for determining the reference line and the actual edge contour curve according to the morphological feature information comprises the following steps: Iteratively sample the morphological feature information based on a random sample consensus algorithm, remove outlier noise points, and retain the inlier point set; Obtain an initial pixel contour in an image pixel coordinate system based on the inlier point set; Obtain the calibration parameters of the vision system; Map the initial pixel contour to the physical processing coordinate system by using the calibration parameters, and fit the actual edge contour curve in the physical processing coordinate system; Obtain the mechanical zero-point parameters of the automatic punching equipment, and generate the reference line in the physical processing coordinate system based on the mechanical zero-point parameters.

4. The method of claim 1, wherein, The method for determining the pose deviation vector according to the reference line and the actual edge contour curve based on the deviation mapping model comprises the following steps: In the physical processing coordinate system, calculate the included angle between the normal vector of the actual edge contour curve and the normal vector of the reference line, and take the included angle as the rotation deviation component; In the physical processing coordinate system, calculate the perpendicular Euclidean distance from the midpoint of the actual edge contour curve to the reference line, and take the perpendicular Euclidean distance as the translation deviation component; Construct the pose deviation vector based on the rotation deviation component and the translation deviation component.

5. The adaptive positioning method for machining the hole positions of the inner pages of a loose-leaf notebook according to claim 3, characterized in that, The method for obtaining the calibration parameters of the vision system comprises the following steps: Control the vision system to collect a calibration board image, wherein the calibration board image contains a calibration board placed on a processing table, and the calibration board surface is drawn with a specific geometric pattern arranged according to a preset rule; Obtain the unit physical size of the specific geometric pattern to determine a plurality of key feature points, and establish the physical coordinates of the plurality of key feature points in the physical processing coordinate system according to the preset rule arrangement; Perform image processing on the calibration board image to extract pixel coordinates corresponding to the plurality of key feature points one by one; According to a mapping relationship between a plurality of corresponding pixel coordinates and the physical coordinates, a homography matrix is solved by using a least square method or a direct linear transformation algorithm, and the homography matrix is taken as a calibration parameter.

6. The adaptive positioning method for machining the hole positions of the inner pages of a loose-leaf notebook as described in claim 1, characterized in that, The adaptive compensation algorithm is used to generate a positioning adjustment instruction according to the pose deviation vector, and the positioning adjustment instruction comprises: A predicted deviation is determined according to a historical deviation vector and a historical compensation gain of a previous processing period; The pose deviation vector and the predicted deviation are input into a position closed-loop control model to determine a target compensation amount, wherein the position closed-loop control model is constructed based on a fuzzy PID control algorithm, and the fuzzy PID control algorithm dynamically adjusts a proportional coefficient, an integral coefficient and a differential coefficient according to a size of the pose deviation vector; The target compensation amount is converted into a multi-axis driving signal of the positioning mechanism as the positioning adjustment instruction.

7. The adaptive positioning method for machining the hole positions of the inner pages of a loose-leaf notebook according to claim 6, characterized in that, After the target compensation amount is converted into the multi-axis driving signal of the positioning mechanism as the positioning adjustment instruction, the method further comprises: The positioning mechanism executes the positioning adjustment instruction; Secondary shape feature information is acquired; A residual deviation vector is calculated according to the secondary shape feature information; If the residual deviation vector is greater than a preset precision allowable threshold, a secondary fine adjustment instruction is generated based on the residual deviation vector until the residual deviation vector converges within the precision allowable threshold.

8. The method of claim 1, wherein, Before the adaptive compensation algorithm is used to generate the positioning adjustment instruction according to the pose deviation vector, the method further comprises: It is judged whether a module length of the pose deviation vector exceeds a preset safety boundary threshold; If the module length exceeds the safety boundary threshold, an abnormal alarm signal is generated and the positioning process is interrupted, wherein the abnormal alarm signal indicates that a stacking state of the paper stack to be processed is seriously out of order or a vision system calibration fails.

9. An adaptive positioning system for machining the hole positions of inner pages in a loose-leaf notebook, characterized in that, The system is used to implement the method according to any one of claims 1-8, comprising: An acquisition module is configured to acquire shape feature information of an edge of a paper stack to be processed; A determination module is configured to determine a reference line and an actual edge contour curve according to the shape feature information; A deviation calculation module is configured to determine a pose deviation vector according to the reference line and the actual edge contour curve based on a deviation mapping model; An execution control module is configured to generate a positioning adjustment instruction according to the pose deviation vector by using an adaptive compensation algorithm, and drive a positioning mechanism to correct a pose of the paper stack to be processed.

10. An electronic device, comprising: A computer program is stored in a memory and executed by a processor, and the computer program is used to implement the adaptive positioning method for processing holes in inner pages of a notebook according to any one of claims 1-8.

Images