Search

CN-122008338-A - Correction calibration method for cutting indentation cutter through indentation identification and cutting machine

CN122008338ACN 122008338 ACN122008338 ACN 122008338ACN-122008338-A

Abstract

The invention discloses a correction calibration method for cutting an indentation cutter through indentation identification and a cutting machine, and relates to the field of image processing; the method comprises the steps of receiving a deviation correcting instruction, controlling an indentation cutter to press a first indentation on paper along an x-axis, moving a preset distance, rotating for 180 degrees, pressing a second indentation, and forming a plurality of groups of indentations at preset intervals and marking. The method comprises the steps of identifying the same group of indentations with smaller spacing, obtaining a standard sequence value according to the standard sequence value, regulating an indentation cutter to perform accurate indentation, controlling the indentation cutter to press a first indentation and a second indentation on paper by receiving a deviation correcting instruction to form a group of indentations, repeating operation to form a plurality of groups of indentations, identifying and comparing the same group of indentations with smaller spacing, determining the standard sequence of the same group of indentations, regulating the indentation cutter according to the standard sequence value, realizing accurate indentation, and improving indentation quality and efficiency.

Inventors

  • CHEN HONG

Assignees

  • 合肥润杰数控设备制造有限公司

Dates

Publication Date
20260512
Application Date
20260202

Claims (10)

  1. 1. The utility model provides a cut indentation sword correction calibration method through indentation discernment, the cutting machine includes dolly, blade control machine and rotary cutter, the blade control machine is including detecting adjustment motor and x axle motor, detect adjustment motor with x axle motor all with the dolly links to each other, be provided with the camera on the dolly, detect adjustment motor and drive the dolly and control the removal, contain the lead screw on the blade control machine, the lead screw with rotary cutter's benchmark axle center is in same x coordinate 0 point position, x axle motor is provided with the indentation sword, its characterized in that, the method includes: firstly, controlling an indentation cutter on an x-axis motor to press a first indentation on the x-axis of the paper, controlling the x-axis motor to move a preset distance along the radial direction of the first indentation after the first indentation operation is finished, rotating for 180 degrees, and pressing a second indentation along the opposite direction of the first indentation; presetting an indentation interval in a database, defining the first indentation and the second indentation along the x-axis direction as a group of indentations, controlling a detection and adjustment motor to drive a trolley to move left and right, extruding a preset number of groups of indentations on the x-axis of paper according to the preset indentation interval, and carrying out positive and negative value marking on each group of indentations; Step three, carrying out identification comparison on a plurality of groups of indentations along the x-axis direction, and identifying identification results of the same group of indentation sequences with small spacing, wherein the identification results comprise positive and negative value sequences between a first indentation and a second indentation in each group of indentation images of a preset number of groups; And step four, acquiring a specific sequence value of the positive and negative value sequence, and regulating and controlling the indentation cutter to carry out indentation operation according to the specific sequence value.
  2. 2. The method for correcting and calibrating a cutting and creasing blade through crease identification according to claim 1, wherein the on-machine device is configured to: the indentation tool is configured to carry out indentation operation according to the control signal: An x-axis motor configured to drive the creasing blade to move in an x-axis direction to perform creasing operation; The detection and adjustment motor is configured to adjust the position of the trolley according to a preset indentation interval, so as to adjust the indentation interval of the indentation cutter; The image recognition module is configured to process the indentation image and recognize indentation deviation; and the control system is configured to adjust the position of the indentation cutter according to the indentation deviation parameter so as to realize accurate indentation.
  3. 3. The method for correcting and calibrating a cutting and creasing blade through crease identification according to claim 1, wherein controlling the x-axis motor to crease comprises: The reverse indentation process of the first indentation and the second indentation is realized through the bidirectional driving of the same x-axis motor, and after the motor control unit receives the pulse signal, the motor control unit firstly rotates forward for a set number of turns to drive the indentation cutter to complete the first indentation, and then rotates reversely for the same or different numbers of turns to complete the second indentation.
  4. 4. The method for correction calibration of a cutting and creasing blade by crease identification according to claim 1, wherein the identification of a plurality of sets of crease along the x-axis comprises: Obtaining an original image of an indentation, uniformly dividing the original image into a plurality of non-overlapping pixel blocks, and calculating a gradient matrix for a target pixel block, wherein the target pixel block is any one of the pixel blocks; Calculating the feature vector of a target pixel block according to the gradient matrix and the gray data, and inputting the feature vector of each pixel block into a classification model to output a classification label of each pixel block, wherein the classification label comprises a simple pixel block and a non-simple pixel block; Constructing a label matrix according to the classified labels, and calculating a gray mapping table according to the label matrix and gray data, wherein the gray mapping table comprises a simple region gray mapping table and a non-simple region gray mapping table; Calculating a correction mapping table of each pixel block through the label matrix and the gray mapping table, and filling the correction mapping table of each pixel block into a corresponding position of the pixel block in the initial matrix to obtain a mapping table matrix; And calculating according to the mapping table matrix and the gray data to obtain an enhanced image.
  5. 5. The method for calibration of trimming and creasing blade calibration by crease identification according to claim 4, wherein calculating the feature vector of the target pixel block from the gradient matrix and the gray data comprises: Uniformly dividing the original image into a plurality of pixel blocks with non-overlapping size of M multiplied by N; Respectively applying a horizontal operator and a vertical operator to the target pixel block to obtain a horizontal gradient and a vertical gradient of the target pixel block through convolution operation; combining the horizontal gradient and the vertical gradient to obtain gradient values, and arranging the gradient values in a one-to-one correspondence manner according to coordinates of the gradient values in the pixel blocks to obtain a gradient matrix; Calculating the feature vector of the target pixel block according to the gradient matrix and the gray data, wherein the feature vector comprises the following steps: Averaging gradient values in a gradient matrix of the target pixel block to obtain a gradient average value, and calculating a gradient standard deviation according to the gradient average value; calculating a first gray average value according to gray data, and calculating a first gray standard deviation according to the first gray average value; And carrying out feature normalization processing on the gradient average value, the gradient standard deviation and the first gray standard deviation to obtain feature vectors.
  6. 6. The correction calibration method for a cutting and creasing blade by crease identification according to claim 4, wherein before inputting the feature vector of each pixel block into the classification model to output the classification label of each pixel block, the training process of the classification model comprises: acquiring a training data set, inputting the training data set into a preset model for training to obtain model updating parameters, and updating the model parameters in the preset model according to the model updating parameters to obtain a classification model; Acquiring a verification data set, inputting the verification data into a classification model to obtain a verification tag, and calculating the similarity between the verification tag and a real tag to obtain a deviation value; If the deviation value is larger than the deviation threshold value, judging that the classification model training is unqualified and re-executing the iterative training until the preset condition is met, otherwise, judging that the classification model training is qualified, wherein the verification data set comprises verification data and real labels, and the verification data corresponds to the real labels one by one.
  7. 7. The method for calibration of correction of a cutting and creasing blade by crease identification according to claim 4, wherein calculating gray mapping tables from the label matrix and gray data, respectively, comprises: Constructing an all-zero matrix, traversing each pixel block, setting the element at the corresponding position in the all-zero matrix as 1 if the pixel block is classified as a simple pixel block, and keeping 0 unchanged if the pixel block is classified as a non-simple pixel block, so as to obtain a label matrix; Scanning all pixel blocks, and if a pixel block label is a simple area label and a pixel block with a label being a simple area label exists in 8-neighborhood of the pixel block label, marking the pixel block as a non-simple area label to obtain a composite area containing labels L and-L, wherein the simple area label is L, and the non-simple area label is-L; extracting gray values of all pixel blocks in the composite area, calculating a second gray average value according to the gray values of all pixel blocks in the composite area, and calculating a second gray standard deviation through the second gray average value; calculating a gray upper limit and a gray lower limit according to the second gray mean value and the second gray standard deviation, and obtaining a minimum gray value and a maximum gray value of pixel blocks in the composite region; and adjusting the gray scale range according to the gray scale upper limit, the gray scale lower limit, the minimum gray scale value and the maximum gray scale value, and calculating a gray scale mapping table through the gray scale range.
  8. 8. The method for calibration of trimming and creasing blade calibration by crease identification according to claim 4, wherein calculating a correction map for each pixel block from the label matrix and gray scale map comprises: Constructing a gray vector according to the bit width of an original image, calculating a mapping gray value for each gray value in the gray vector according to a gray range, and judging according to the mapping gray value to obtain a simple region gray mapping table; counting the occurrence times of all pixels of each non-simple pixel block on each gray level to obtain a histogram of the non-simple pixel block; calculating a gray mapping function according to a preset mapping value and an original histogram, taking the gray mapping function as an average background brightness of a corresponding gray level to diagnose a JND threshold value, and calculating a clipping threshold value of the histogram of each gray level k according to the JND threshold value; Judging the histogram of the non-simple pixel block and the clipping threshold value, and taking the histogram as a non-simple region gray mapping table if the preset condition is met; Traversing the label matrix, identifying pixel blocks with negative numbers of all labels as transition pixel blocks, and recording the positions of the transition pixel blocks in the label matrix; extracting 3X 3 neighborhood of each transition pixel block, and collecting gray mapping tables of all pixel blocks in the neighborhood; and calculating a correction mapping table of the transition pixel block according to the gray mapping tables of all pixel blocks in the 3×3 neighborhood.
  9. 9. The method for calibration of correction of a cutting and creasing blade by crease identification according to claim 4, wherein the step of calculating an enhanced image from the mapping table matrix and gray data comprises: for the gray value of each pixel block in the original image, determining 4 adjacent pixel blocks around the gray value, and acquiring a gray mapping table corresponding to the 4 pixel blocks in a mapping table matrix; calculating according to the gray mapping table corresponding to the 4 pixel blocks in the mapping table matrix to obtain the interpolation gray of the pixel blocks; performing 3×3 mean filtering on each pixel block in the original image to obtain a filtered gray value, and taking the gray value as the gray value of the original image at the position; and traversing and splicing the gray value of each pixel block according to the interpolation gray and the filtered gray value to obtain an enhanced image.
  10. 10. The method for correcting and calibrating the cutting and creasing blade through creasing identification according to claim 1, wherein the method for regulating and controlling the creasing blade to perform creasing operation according to the specific sequence value comprises the following steps: calculating a deviation compensation value of the indentation cutter in the x-axis direction according to the specific standard sequence value of the positive and negative standard sequences; the deviation compensation value is converted into a pulse compensation value of the x-axis motor, wherein the pulse compensation value is obtained through calculation of a PID control algorithm, and the input of the PID control algorithm is the difference value between the specific sequence value and a preset target value; And controlling the x-axis motor to adjust the moving distance of the indentation cutter according to the pulse compensation value so as to realize the correction and calibration of the indentation cutter.

Description

Correction calibration method for cutting indentation cutter through indentation identification and cutting machine Technical Field The invention belongs to the technical field of image processing, and particularly relates to a correction calibration method for a cutting indentation cutter through indentation identification and a cutting machine. Background The indentation cutter is also called indentation line or beer cutter, is one of key components on the die cutter plate, is not used for cutting, but utilizes the strong stamping pressure of the indentation cutter on a die cutter to press accurate grooves or marks on materials such as paper board, leather or plastics, and the like, and has the core effect of providing preset and structural weakening lines for subsequent accurate bending, thereby ensuring that products such as packaging boxes, greeting cards, notebook covers and the like can be folded into straight, flat and sharp folding angles along preset positions, and greatly improving the attractiveness and the forming stability of the products. In the step of obtaining the indentation image, the definition of the photographed indentation image is affected due to the fact that the surface glossiness of the packaging materials in different batches is different, the characteristics are not obvious due to the fact that multiple groups of indentations are arranged together, for example, some materials with high glossiness reflect light, the indentation image photographed by a camera has a reflecting area, part of indentation details are covered, the subsequent characteristic extraction is affected, so that the cutting position of an indentation cutter is inaccurate due to the fact that the texture characteristics of the image are unclear, the extraction of indentation deviation values is inaccurate, and the correction efficiency of the indentation cutter is low. Disclosure of Invention The invention aims to solve the problem of inaccurate correction of the cutting position of an indentation cutter caused by insufficient extraction capability of texture features of an indentation image, and provides a correction and calibration method for cutting the indentation cutter through indentation identification and a cutting machine. In a first aspect of the present invention, a method for correcting and calibrating a cutting and creasing knife by creasing identification is provided, where the cutting machine includes a trolley, a blade controller and a rotary knife, the blade controller includes a detection adjustment motor and an x-axis motor, the detection adjustment motor and the x-axis motor are both connected with the trolley, a camera is disposed on the trolley, the detection adjustment motor drives the trolley to move left and right, the blade controller includes a screw rod, the screw rod and a reference axis of the rotary knife are located at the same x-coordinate 0 point, and the x-axis motor is provided with a creasing knife, and the method is characterized in that: firstly, controlling an indentation cutter on an x-axis motor to press a first indentation on the x-axis of the paper, controlling the x-axis motor to move a preset distance along the radial direction of the first indentation after the first indentation operation is finished, rotating for 180 degrees, and pressing a second indentation along the opposite direction of the first indentation; presetting an indentation interval in a database, defining the first indentation and the second indentation along the x-axis direction as a group of indentations, controlling a detection and adjustment motor to drive a trolley to move left and right, extruding a preset number of groups of indentations on the x-axis of paper according to the preset indentation interval, and carrying out positive and negative value marking on each group of indentations; Step three, carrying out identification comparison on a plurality of groups of indentations along the x-axis direction, and identifying identification results of the same group of indentation sequences with small spacing, wherein the identification results comprise positive and negative value sequences between a first indentation and a second indentation in each group of indentation images of a preset number of groups; And step four, acquiring a specific sequence value of the positive and negative value sequence, and regulating and controlling the indentation cutter to carry out indentation operation according to the specific sequence value. Optionally, the on-cutter device is configured to: the indentation tool is configured to carry out indentation operation according to the control signal: An x-axis motor configured to drive the creasing blade to move in an x-axis direction to perform creasing operation; The detection and adjustment motor is configured to adjust the position of the trolley according to a preset indentation interval, so as to adjust the indentation interval of the indentation cutter; The