Intelligent positioning control system for foam buffer part die cutting process

Abstract

The invention relates to the technical field of processing positioning, and discloses an intelligent positioning control system for a foam buffer piece die cutting process, and the system comprises an acquisition module which is used for acquiring a first surface gray level image; the calculation module is used for positioning a first MARK point center coordinate and a second MARK point center coordinate based on the first surface grayscale image and calculating a first X-direction deviation value, a first Y-direction deviation value and a first rotation angle deviation value; the compensation motion module is used for executing compensation motion; and the circulation module is used for re-collecting a second surface gray level image of the foam material which is compensated in place and calculating residual deviation values, when the residual deviation values are all smaller than a target deviation value, a punching instruction is sent, and otherwise, the compensation motion process is re-executed. The problems of position offset and rotation deflection angle caused by flexible deformation in the feeding process of foam materials are effectively solved, and the positioning precision and the production efficiency of foam buffer part die cutting are improved.

Description

Technical Field

[0001] This invention relates to the field of processing and positioning technology, and in particular to an intelligent positioning control system for the die-cutting process of foam cushioning components. Background Technology

[0002] Foam cushioning components are key materials for packaging and shock absorption in electronic products, and the precision of their die-cutting directly affects the product assembly quality. Foam materials are soft and easily deformable, inevitably resulting in positional offsets in the X and Y directions and rotational angle deviations during high-speed roll-to-roll feeding. This leads to misalignment between the die-cutting position and the die, causing quality problems such as dimensional errors or edge burrs.

[0003] Traditional die-cutting equipment relies mainly on manual visual adjustment or mechanical limits for positioning, resulting in low positioning accuracy and inefficiency, making it difficult to meet the ±0.1mm precision requirements of mass production. Furthermore, the porous structure of the foam surface blurs the edges of pre-printed positioning marks, making it difficult for traditional image recognition methods to accurately extract the center coordinates of the marks, further affecting the accuracy of positioning compensation. Summary of the Invention

[0004] This invention provides an intelligent positioning control system for the die-cutting process of foam cushioning components. This invention effectively solves the problems of positional offset and rotation angle caused by flexible deformation of foam materials during feeding, improves the positioning accuracy and production efficiency of foam cushioning component die-cutting, and realizes the intelligentization and automation of the foam die-cutting process.

[0005] In a first aspect, the present invention provides an intelligent positioning control system for the die-cutting process of foam cushioning components, the intelligent positioning control system for the die-cutting process of foam cushioning components comprising: The acquisition module is used to send the foam material to the die-cutting station through the material pulling mechanism and acquire the grayscale image of the first surface if the current working mode is CCD automatic positioning mode. The calculation module is used to locate the center coordinates of the first MARK point and the center coordinates of the second MARK point based on the first surface grayscale image and to calculate the first X-direction deviation value, the first Y-direction deviation value and the first rotation angle deviation value; The compensation motion module is used to perform compensation motion based on the first X-direction deviation value, the first Y-direction deviation value, and the first rotation angle deviation value; The loop module is used to re-acquire the grayscale image of the second surface of the compensated foam material and calculate the residual deviation value. When the residual deviation value is less than the target deviation value, a punching command is sent; otherwise, the compensation motion process is re-executed.

[0006] In conjunction with the first aspect, in the first implementation of the first aspect of the present invention, the acquisition module is specifically used for: If the current working mode is CCD automatic positioning mode, the material pulling mechanism is activated to pull the foam material from the unwinding mechanism to the die-cutting station. The material pulling length is measured in real time using an encoder; When the measured length of the material being pulled reaches the preset length value, the material pulling mechanism stops, so that the two cross-shaped MARK markers are within the field of view of the CCD camera; The CCD camera is activated and the ring LED light source is turned on to illuminate the surface of the foam material. The area containing the two cross-shaped MARK markers is captured to obtain a first surface grayscale image.

[0007] In conjunction with the first aspect, in a second implementation of the first aspect of the present invention, the calculation module further includes: Extraction unit is used to extract the gray value of the first pixel in the first surface grayscale image and mark the first pixel with the gray value lower than the preset grayscale threshold as the second pixel. A connected component labeling unit is used to locate the first MARK point region and the second MARK point region based on the second pixel. The center coordinate calculation unit is used to calculate the center coordinates of the first MARK point in the first MARK point region and the center coordinates of the second MARK point in the second MARK point region. The deviation calculation unit is used to calculate the first X-direction deviation value and the first Y-direction deviation value based on the center coordinates of the first MARK point and the center coordinates of the second MARK point, and to calculate the first rotation angle deviation value by connecting the center coordinates of the first MARK point and the center coordinates of the second MARK point.

[0008] In conjunction with the first aspect, in a third implementation of the first aspect of the present invention, the connected component labeling unit is specifically used for: The second pixel is labeled with connected components to obtain multiple connected components, and the area of ​​each connected component is calculated. Select the two connected components with the largest areas from the plurality of connected components, mark the connected component with the largest area as the first MARK point region, and mark the connected component with the second largest area as the second MARK point region.

[0009] In conjunction with the first aspect, in the fourth implementation of the first aspect of the present invention, the deviation calculation unit is specifically used for: The first coordinate difference is obtained by subtracting the center coordinates of the first MARK point from the preset first standard coordinates; the second coordinate difference is obtained by subtracting the center coordinates of the second MARK point from the preset second standard coordinates. The first coordinate difference and the second coordinate difference are averaged in the X and Y directions, respectively, to obtain the first X-direction deviation value and the first Y-direction deviation value. The first connecting angle is calculated based on the center coordinates of the first MARK point and the center coordinates of the second MARK point. The second connecting angle is calculated based on the first standard coordinates and the second standard coordinates. The difference between the first connecting angle and the second connecting angle is used to obtain the first rotation angle deviation value.

[0010] In conjunction with the first aspect, in the fifth implementation of the first aspect of the present invention, the motion compensation module further includes: The conversion coefficient query unit is used to query the first conversion coefficient of the X-axis and the second conversion coefficient of the Y-axis. The displacement compensation unit is used to multiply the first X-direction deviation value by the first conversion coefficient to obtain the first compensation displacement, and to multiply the first Y-direction deviation value by the second conversion coefficient of the Y-axis to obtain the second compensation displacement. The pulse calculation unit is used to divide the first compensation displacement by the X-axis pulse equivalent to obtain the first pulse number, and divide the second compensation displacement by the Y-axis pulse equivalent to obtain the second pulse number; The drive unit is configured to perform compensating motion based on the first pulse count, the second pulse count, and the first rotation angle deviation value.

[0011] In conjunction with the first aspect, in a sixth implementation of the first aspect of the present invention, the driving unit further includes: The compensation start subunit is used to send the first pulse count to the X-axis servo driver to drive the X-axis servo motor to perform compensation motion, send the second pulse count to the Y-axis servo driver to drive the Y-axis servo motor to perform compensation motion, and simultaneously drive the θ-axis rotation mechanism to perform angle compensation according to the first rotation angle deviation value. The real-time feedback subunit is used to provide real-time feedback of the actual displacement of the X-axis and the actual displacement of the Y-axis through the X-axis encoder and the Y-axis encoder, and to calculate the first displacement difference between the actual displacement of the X-axis and the first compensation displacement, and the second displacement difference between the actual displacement of the Y-axis and the second compensation displacement, respectively. The compensation termination unit is used to determine that compensation is in place when both the first displacement difference and the second displacement difference are less than a preset displacement difference.

[0012] In conjunction with the first aspect, in the seventh implementation of the first aspect of the present invention, the real-time feedback subunit is specifically used for: The cumulative pulse count of the X-axis servo motor and the Y-axis servo motor is collected in real time by the X-axis encoder and the Y-axis encoder. The cumulative pulse count is multiplied by the X-axis pulse equivalent to obtain the actual X-axis displacement. The cumulative pulse count is then multiplied by the Y-axis pulse equivalent to obtain the actual Y-axis displacement. The first displacement difference is obtained by subtracting the actual displacement of the X-axis from the first compensation displacement, and the second displacement difference is obtained by subtracting the actual displacement of the Y-axis from the second compensation displacement.

[0013] In conjunction with the first aspect, in the eighth implementation of the first aspect of the present invention, the loop module further includes: The coordinate comparison unit is used to re-acquire the grayscale image of the second surface of the foam material after compensation is in place, and identify the corresponding center coordinates of the third MARK point and the center coordinates of the fourth MARK point. Based on the center coordinates of the third MARK point and the center coordinates of the fourth MARK point, the second X-direction deviation value, the second Y-direction deviation value, and the second rotation angle deviation value are calculated. The loop output unit is used to send a punching command to the die-cutting mechanism to perform die-cutting when the second X-direction deviation value, the second Y-direction deviation value, and the second rotation angle deviation value are all less than the target deviation value; otherwise, the compensation motion process is re-executed.

[0014] In conjunction with the first aspect, in the ninth implementation of the first aspect of the present invention, the loop output unit is specifically used for: When the second X-direction deviation value, the second Y-direction deviation value and the second rotation angle deviation value are all less than the target deviation value, a punching command is sent to the die-cutting mechanism to drive the die to press down and perform foam die-cutting. If any one of the second X-direction deviation value, the second Y-direction deviation value, or the second rotation angle deviation value exceeds the target deviation value, the compensation motion process is re-executed based on the second X-direction deviation value, the second Y-direction deviation value, and the second rotation angle deviation value.

[0015] The technical solution provided by this invention binarizes and marks connected components in the grayscale image of the foam surface, and uses a connected component area filtering method to effectively eliminate noise interference generated by the porous surface of the foam, accurately identifying the center coordinates of two MARK points. This solves the technical problem of blurred edges and difficulty in positioning of marker points caused by the porous structure of the foam surface. By configuring dual MARK points, the X-direction position deviation, Y-direction position deviation, and rotation angle deviation are calculated simultaneously, achieving comprehensive perception of the three-dimensional spatial posture of the foam material. The image coordinate system deviation is converted into a mechanical coordinate system compensation amount through a pre-calibrated transformation coefficient, and then converted into servo motor pulse commands to drive the XYθ platform to perform compensation motion. The actual displacement is compared with the target displacement in real time through the encoder to ensure that the compensation accuracy meets the preset requirements. After compensation, the image is re-acquired, the residual deviation is calculated, and iterative compensation is performed to ensure that the final positioning accuracy meets the die-cutting requirements. This effectively solves the problem of position offset and rotation angle caused by flexible deformation of the foam material during feeding, improves the positioning accuracy and production efficiency of foam buffer die-cutting, and realizes the intelligence and automation of the foam die-cutting process. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the first embodiment of the intelligent positioning control system for the die-cutting process of foam cushioning components in this invention. Figure 2 This is a schematic diagram of the second embodiment of the intelligent positioning control system for the die-cutting process of foam cushioning components in this invention. Figure 3 This is a schematic diagram of the die-cutting and positioning control process of foam buffer components based on MARK point image recognition and XYθ platform servo compensation. Detailed Implementation

[0018] This invention provides an intelligent positioning control system for the die-cutting process of foam cushioning components. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, systems, products, or devices.

[0019] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the intelligent positioning control system for the die-cutting process of foam cushioning components in this invention includes: The acquisition module 101 is used to send the foam material to the die-cutting station through the material pulling mechanism and acquire the grayscale image of the first surface if the current working mode is CCD automatic positioning mode. The calculation module 102 is used to locate the center coordinates of the first MARK point and the center coordinates of the second MARK point based on the grayscale image of the first surface, and to calculate the first X-direction deviation value, the first Y-direction deviation value and the first rotation angle deviation value. The compensation motion module 103 is used to perform compensation motion based on the first X-direction deviation value, the first Y-direction deviation value, and the first rotation angle deviation value; The loop module 104 is used to re-acquire the grayscale image of the second surface of the compensated foam material and calculate the residual deviation value. When the residual deviation value is less than the target deviation value, a punching command is sent; otherwise, the compensation motion process is re-executed.

[0020] In one specific embodiment, the acquisition module 101 is specifically used for: If the current working mode is CCD automatic positioning mode, the material pulling mechanism will be activated to pull the foam material from the unwinding mechanism to the die-cutting station. The material pulling length is measured in real time using an encoder; When the measured length of the material being pulled reaches the preset length value, the material pulling mechanism stops, so that the two cross-shaped MARK markers are within the field of view of the CCD camera; The CCD camera is activated and the ring LED light source is turned on to illuminate the surface of the foam material. The area containing two cross-shaped MARK markers is captured to obtain the first surface grayscale image.

[0021] Specifically, the controller determines whether it is in CCD automatic positioning mode via a working mode switching command. When the result is true, the controller sends a start command to the feeding mechanism, causing the traction roller and the unwinding mechanism to work together to pull the foam material from the unwinding shaft to the front end of the die-cutting platform at a constant linear speed. During the feeding process, the encoder module is activated simultaneously. The encoder is mounted on the guide roller shaft, and its output pulse count is linearly related to the roller shaft rotation angle. Combined with the roller diameter, this can be converted into the real-time feeding length, which is continuously accumulated in millimeters. The controller periodically reads the cumulative pulse count from the encoder and multiplies it by a pre-calibrated displacement corresponding to a unit pulse to obtain the feeding distance. To ensure that the two cross-shaped MARK points are within the image acquisition field of view, a set of feeding termination length parameters is preset. The length value is based on the longitudinal spacing L0 of the MARK points, the CCD camera field of view height, and the foam leading edge margin, and remains constant in each feeding cycle. When the encoder feedback indicates that the material pull length is equal to or slightly exceeds the preset value, the controller immediately issues a stop command to the material pull driver, causing the foam material to stop moving below the die-cutting station. At this point, the two pre-printed cross-shaped MARK marks on the surface of the foam material are located within the imaging area of ​​the CCD camera mounted directly above the die-cutting platform. The controller then sequentially activates the CCD camera and its matching ring LED light source. The LED light source is set to oblique illumination with an incident angle of approximately 45° to enhance the grayscale difference between the MARK marks and the foam body, improving the edge sharpness and structural integrity of the image. The LED brightness is set to 70% to 80% of the maximum rated operating current to balance imaging brightness and thermal stability. The CCD camera's imaging resolution is set to 2048×1536 pixels, with a field of view covering an area of ​​approximately 300mm×225mm on the foam surface. The lens focal length has been adjusted during system calibration to ensure that the cross-shaped MARK marks are clearly focused in the image. The CCD camera acquires an image of the foam material surface in its current static state upon triggering a command, and outputs it to the image processing controller as an 8-bit grayscale image. The grayscale value of each pixel in the image ranges from 0 to 255. The resulting image is the first surface grayscale image containing two cross-shaped MARK markers.

[0022] In one specific embodiment, the calculation module 102 further includes: Figure 2This is a schematic diagram of the intelligent positioning control system for the die-cutting process of foam buffer components according to the present invention. It consists of four main modules: an acquisition module, a calculation module, a compensation motion module, and a loop module. The calculation module includes four sub-units: an extraction unit, a connected component marking unit, a center coordinate calculation unit, and a deviation calculation unit. It is responsible for identifying MARK points from the grayscale image and calculating the deviation value. The compensation motion module includes a conversion coefficient query unit, a displacement compensation unit, a pulse calculation unit, and a drive unit. The drive unit further includes three sub-units: a compensation start sub-unit, a real-time feedback sub-unit, and a compensation end sub-unit. It is responsible for converting the deviation value into a servo motor control signal. The loop module includes a coordinate comparison unit and a loop output unit. It is responsible for determining the compensation accuracy and deciding whether to send a punching command or provide feedback for re-compensation.

[0023] An extraction unit is used to extract the gray value of a first pixel in a first surface grayscale image and mark the first pixel with a gray value lower than a preset grayscale threshold as a second pixel. Connected component labeling unit, used to locate the first MARK point region and the second MARK point region based on the second pixel; The center coordinate calculation unit is used to calculate the center coordinates of the first MARK point in the first MARK point region and the center coordinates of the second MARK point in the second MARK point region. The deviation calculation unit is used to calculate the first X-direction deviation value and the first Y-direction deviation value based on the center coordinates of the first MARK point and the center coordinates of the second MARK point, and to calculate the first rotation angle deviation value by connecting the center coordinates of the first MARK point and the center coordinates of the second MARK point.

[0024] Specifically, input the first surface grayscale image into the image buffer of the image processing controller. The first surface grayscale image is an 8-bit grayscale image with pixel values ranging from 0 to 255. Scan each pixel of the entire image row by row and column by column, and extract the grayscale value G(x,y) of each first pixel point in the image, where (x,y) is the two-dimensional coordinate index of the pixel point in the image coordinate system. To effectively distinguish the cross-shaped MARK points from the background area of the foam body, calculate the grayscale mean Gb of the background area based on the pixel grayscale distribution in the entire image. The background area is defined as the set of pixel points with grayscale values greater than 150. Then multiply the background grayscale mean by the attenuation coefficient 0.7 and set it as the preset grayscale threshold Tg, that is, Tg = 0.7×Gb. During the pixel point scanning process, for pixel points that satisfy G(x,y) < Tg, mark them as second pixel points, that is, consider them as the candidate area pixels of the MARK points, and assign the value 1 to such pixels in the binary image, and assign the value 0 to the remaining pixels to generate a binary image. The connected component labeling unit performs eight-neighbor connectivity analysis on the binary image to determine whether each second pixel point forms a connected set with the adjacent pixels in the eight surrounding directions. Classify all connected 1-value pixel points in the image through the standard connected component labeling algorithm, and assign a unique label number to each group of connected pixel regions, and at the same time count the area S of each connected component, that is, the number of pixel points it contains. Since there are random noise holes on the foam surface, the connected components formed by them have a small area, less than 50 pixel points, while the connected component area of the real MARK point region is generally greater than 200 pixel points due to the clear cross-shaped structure. Therefore, sort the connected component areas from largest to smallest and select the first two largest connected components as the first MARK point region and the second MARK point region. The center coordinate calculation unit processes these two connected component regions respectively, extracts the coordinate set {(x i ,y j )} of all pixel points in each connected component, and calculates its geometric center coordinates. For the first MARK point region, the center abscissa is xc1 = Σx i / N1, and the center ordinate is yc1 = Σy i / N1, where N1 is the total number of pixel points in this region; similarly, for the second MARK point region, calculate its center abscissa xc2 = Σx j / N2 and ordinate yc2 = Σy j / N2 to obtain the center coordinates (xc1,yc1) and (xc2,yc2) of the two MARK points. The deviation calculation unit compares the two center coordinates with the standard MARK point coordinates (x 01 ,y 01 ) and (x 02 ,y 02 ) pre-stored in the controller, and calculates the first X-direction deviation value ΔX = (xc1 – x 01 + xc2 – x02 ) / 2, the first Y-direction deviation value ΔY=(yc1–y 01 +yc2–y 02 The positional deviation of the foam material in the horizontal and vertical directions is obtained by calculating the slope of the line connecting the two center coordinates, α1 = arctan[(yc2–yc1) / (xc2–xc1)], and the angle α0 = arctan[(yc2–yc1) / (xc2–xc1)] is drawn from this line to the standard MARK point. 02 –y 01 ) / (x 02 -x 01 By comparing the values, the first rotation angle deviation value Δθ = α1 – α0 is obtained, which reflects the deflection trend of the foam material rotating around the center on the die-cutting platform.

[0025] Figure 3 This is a schematic diagram of the die-cutting and positioning control process of foam buffer based on MARK point image recognition and XYθ platform servo compensation. Figure 3 The process includes the pre-compensation state, where foam material is placed on a moving platform as the test object. Two positioning reference points are pre-printed on the surface of the test object. Due to positional offset and angular deflection during the feeding process, the two positioning reference points deviate from the preset standard position, exhibiting deviations in the X direction, Y direction, and rotation angle. The process also includes the post-compensation state, where a CCD camera identifies the actual position of the two positioning reference points and calculates the deviation. The control system drives the XYθ platform to perform compensation motion based on the deviation. The moving platform moves the test object to adjust its spatial position, ultimately aligning the two positioning reference points with the preset standard position, thus completing precise positioning and providing an accurate processing benchmark for subsequent die-cutting processes.

[0026] Before marking pixels with gray values ​​lower than a preset gray value threshold as candidate pixels for MARK points, the process includes an adaptive gray value threshold setting step: The gray value of all pixels in the foam surface grayscale image is extracted by scanning rows and columns; pixels with gray values ​​greater than a first preset value are counted as background area pixels; the mean gray value and gray value variance of the background area pixels are calculated; the pore density level of the foam surface is determined based on the gray value variance; when the gray value variance is greater than a second preset value, it is determined to be high-density pore foam, and a gray value threshold coefficient is set as the first coefficient value; when the gray value variance is less than or equal to the second preset value, it is determined to be low-density pore foam, and a gray value threshold coefficient is set as the second coefficient value; the mean gray value of the background is multiplied by the gray value threshold coefficient to obtain the adaptively adjusted preset gray value threshold; the grayscale image of the foam surface is binarized using the adaptively adjusted preset gray value threshold, and pixels with gray values ​​lower than the adaptively adjusted preset gray value threshold are marked as candidate pixels for MARK points, thus achieving adaptive threshold segmentation for foam materials with different pore densities.

[0027] In one specific embodiment, the connected component labeling unit is specifically used for: The second pixel is labeled with connected components to obtain multiple connected components, and the area of ​​each connected component is calculated. Select the two connected components with the largest areas from the multiple connected components. Mark the connected component with the largest area as the first MARK point region and the connected component with the second largest area as the second MARK point region.

[0028] Specifically, the image processing controller enables a connected component labeling algorithm. It scans each second pixel with a value of 1 row-by-row and column-by-column, using the eight-neighbor connectivity criterion to determine if the pixel is connected to its left, top, top-left, bottom-left, top-right, bottom-right, top, and right neighboring pixels. If any neighboring pixel in any direction has a value of 1, it is considered to belong to the same connected component. A unique number is assigned to each set of connected pixels, and the pixel index corresponding to each number is continuously tracked until all second pixels in the entire image are assigned to a certain label number, forming several discrete connected components. An area calculation operation is performed on each connected component, that is, the total number of pixels assigned to each number is counted to obtain the area value of each connected component, and the area value is stored in a temporary array structure. Because the surface of the foam material contains many small, irregularly shaped pseudo-connected components due to factors such as micropores, reflective defects, or ink scattering, to ensure that the identified MARK point region has a complete cross-shaped structure and a large pixel coverage area, all connected components are sorted according to the area parameter. The sorting is done in descending order, and the two connected components with the largest areas correspond to the two preset MARK points. The numbers of the two connected components with the largest areas are read, and the connected component with the largest area is marked as the first MARK point region, and the connected component with the second largest area is marked as the second MARK point region.

[0029] The process includes marking pixels with grayscale values ​​below a preset grayscale threshold as MARK candidate pixels, and before performing connected component labeling on the MARK candidate pixels, a MARK point edge continuity processing step is included: acquiring a binarized image formed by the MARK candidate pixels, setting the structure element size for morphological closing operations based on the average diameter of the foam surface pores, setting the structure element to a first-size circular kernel when the average diameter of the foam pores is less than a third preset value, and setting the structure element to a second-size circular kernel when the average diameter of the foam pores is greater than or equal to the third preset value; and applying a process to the binarized image. The structuring element performs morphological dilation, expanding the candidate pixels of the MARK point outward to fill the small hole region inside the MARK point, resulting in a dilated binarized image. The same structuring element is then used to perform morphological erosion on the dilated binarized image, shrinking the dilated MARK point region back to its original size while maintaining the closed state of the filled small hole region, resulting in a continuous edge MARK point region. Pixels with a grayscale value of 1 in the continuous edge MARK point region are updated to the processed MARK point candidate pixels, eliminating the discontinuous edge phenomenon of the MARK points caused by the porous structure of the foam.

[0030] The step of extracting the coordinates of all pixels within the first MARK point region and calculating the mean coordinates to obtain the center coordinates of the first MARK point includes a distance-weighted center coordinate calculation step: extracting the coordinates of all pixels within the first MARK point region; summing the x-coordinates of all pixels within the first MARK point region and dividing by the total number of pixels to obtain the initial x-coordinate centroid; summing the y-coordinates of all pixels and dividing by the total number of pixels to obtain the initial y-coordinate centroid, thus forming the initial centroid coordinates; calculating the Euclidean distance from each pixel within the first MARK point region to the initial centroid coordinates to obtain the distance value corresponding to each pixel; calculating the weight coefficient of each pixel based on the distance value; and using the sum of the preset unit values ​​divided by the square of the distance value as the weight coefficient. Pixels closer to the initial centroid receive a larger weight coefficient, while pixels farther from the initial centroid receive a smaller weight coefficient. The x-coordinate of each pixel within the first MARK point region is multiplied by its corresponding weight coefficient, summed, and then divided by the sum of all weight coefficients to obtain the weighted corrected x-coordinate of the first MARK point center. Similarly, the y-coordinate of each pixel is multiplied by its corresponding weight coefficient, summed, and then divided by the sum of all weight coefficients to obtain the weighted corrected y-coordinate of the first MARK point center. The same method is used to calculate the weighted corrected center coordinate of the second MARK point region. Distance weighting suppresses interference from edge pixels on the center coordinate calculation, improving the noise resistance of the center coordinate calculation.

[0031] In this embodiment, before selecting the two connected components with the largest area from each connected component, a connected component screening step based on shape features is included: calculating shape feature parameters for each connected component, extracting the minimum bounding rectangle of each connected component, calculating the ratio of the long side to the short side of the minimum bounding rectangle as the aspect ratio feature value, and calculating the ratio of the connected component area to the area of ​​the minimum bounding rectangle as the density feature value. Since the aspect ratio of the real cross-shaped MARK marker points is close to the preset ratio range and the density is high, while the shape of the connected components formed by the aggregation of foam holes is irregular, the aspect ratio and density deviate from the preset range; setting the aspect ratio discrimination threshold range and the density discrimination threshold lower limit, traversing each connected component to determine whether its aspect ratio feature value is within the aspect ratio discrimination threshold range and whether the density feature value is greater than the density discrimination threshold lower limit, the connected component that meets both conditions is marked as a candidate MARK point connected component, and the connected component that does not meet the conditions is marked as a candidate MARK point connected component. The connected components of the condition are marked as noisy connected components and removed. The number of connected components of candidate MARK points is counted. If the number of connected components of candidate MARK points is equal to 2, the two connected components of candidate MARK points are directly marked as the first MARK point region and the second MARK point region, respectively. If the number of connected components of candidate MARK points is greater than 2, the two connected components with the largest areas from the candidate MARK point connected components are selected and marked as the first MARK point region and the second MARK point region, respectively. If the number of connected components of candidate MARK points is less than 2, the aspect ratio discrimination threshold range is relaxed and the lower limit of the density discrimination threshold is lowered, and the connected component screening is re-executed until at least two candidate MARK point connected components are obtained. Before area sorting, the irregularly shaped noisy connected components formed by the aggregation of foam holes are removed by shape feature discrimination to avoid false connected components being misjudged as MARK point regions due to their large area, thereby improving the accuracy of MARK point recognition.

[0032] In one specific embodiment, the deviation calculation unit is specifically used for: The first coordinate difference is obtained by subtracting the center coordinates of the first MARK point from the preset first standard coordinates. The second coordinate difference is obtained by subtracting the center coordinates of the second MARK point from the preset second standard coordinates. The first coordinate difference and the second coordinate difference are averaged in the X and Y directions respectively to obtain the first X-direction deviation value and the first Y-direction deviation value. The first connecting angle is calculated based on the center coordinates of the first MARK point and the center coordinates of the second MARK point. The second connecting angle is calculated based on the first standard coordinates and the second standard coordinates. The difference between the first connecting angle and the second connecting angle is used to obtain the first rotation angle deviation value.

[0033] Specifically, the center coordinates (xc1, yc1) of the first MARK point are respectively compared with the first standard coordinates (xc1, yc1). 01 , y 01The first coordinate difference Δx1 = xc1 is obtained by subtracting the values ​​in the X and Y directions. x 01 Δy1= yc1 y 01 ; Compare the center coordinates (xc2, yc2) of the second MARK point with the second standard coordinates (x 02 , y 02 The difference between the X and Y directions is used to obtain the second coordinate difference Δx2 = xc2. x 02 Δy2= yc2 y 02 To eliminate random errors caused by local detection fluctuations and improve the stability of displacement estimation, the coordinate differences between the two MARK points in each direction are averaged. The calculation formulas are: first X-direction deviation ΔX = (Δx1 + Δx2) / 2, first Y-direction deviation ΔY = (Δy1 + Δy2) / 2. The obtained ΔX and ΔY are the overall positional offset of the foam material in the X and Y axes of the die-cutting platform, expressed in pixels. To assess whether the foam material has a rotational deviation around the vertical axis of the platform in the current positioning state, the angle α1 of the line connecting the center coordinates of the two actual MARK points is calculated based on the slope of the line connecting the center coordinates of the two actual MARK points, i.e., α1 = arctan[(yc2) / 2]. yc1) / (xc2 [xc1)], this angle reflects the actual orientation of the two MARK points in the image coordinate system. Then, for the standard coordinate pair (x... 01 , y 01 ) and (x 02 , y 02 Performing the same calculation, we obtain the standard connecting angle α0 = arctan[(y 02 y 01 ) / (x 02 x 01 This angle represents the ideal alignment obtained during the equipment calibration phase. The difference between these angles is Δθ = α1. α0 yields the first rotation angle deviation value Δθ, which characterizes the rotational deviation of the foam material on the die-cutting platform caused by feed skew or uneven tension. The angle deviation is expressed in degrees, with positive values ​​indicating clockwise deflection and negative values ​​indicating counterclockwise deflection.

[0034] In one specific embodiment, the motion compensation module 103 further includes: The conversion coefficient query unit is used to query the first conversion coefficient of the X-axis and the second conversion coefficient of the Y-axis. The displacement compensation unit is used to multiply the first X-direction deviation value by a first conversion coefficient to obtain a first compensation displacement, and to multiply the first Y-direction deviation value by a second conversion coefficient of the Y-axis to obtain a second compensation displacement. The pulse calculation unit is used to divide the first compensation displacement by the X-axis pulse equivalent to obtain the first pulse count, and divide the second compensation displacement by the Y-axis pulse equivalent to obtain the second pulse count; The drive unit is used to perform compensating motion based on the first pulse count, the second pulse count, and the first rotation angle deviation value.

[0035] Specifically, the conversion coefficient query unit extracts calibration data related to the correspondence between the image coordinate system and the platform coordinate system from the system calibration database. The calibration data includes a first conversion coefficient Kpx for the X-axis and a second conversion coefficient Kpy for the Y-axis, representing the actual physical displacement of each image pixel in the X and Y directions, respectively, in millimeters per pixel. These two sets of coefficients are obtained through the platform calibration process, i.e., by driving the XYθ platform to move a known distance (e.g., 10 mm) in the X and Y directions, and recording the displacement of the corresponding MARK point in the image at the pixel coordinates. This allows for the reverse calculation of the actual length represented by each pixel. Multiplying the first X-direction deviation value ΔX by the first conversion coefficient Kpx yields the first compensation displacement Lx = ΔX × Kpx; multiplying the first Y-direction deviation value ΔY by the second conversion coefficient Kpy yields the second compensation displacement Ly = ΔY × Kpy. The resulting Lx and Ly are the two-dimensional translation compensation values ​​that the foam material needs to undergo on the actual physical platform. The pulse calculation unit queries the servo pulse equivalent parameters for each axis in the system, namely the X-axis pulse equivalent δx and the Y-axis pulse equivalent δy. These are defined as the distance the platform moves for each pulse emitted in the X-axis and Y-axis directions, respectively, in millimeters per pulse. The first pulse count Px required by the X-axis servo motor is obtained by dividing the compensation displacement Lx by δx, i.e., Px = Lx / δx; similarly, the second pulse count Py required by the Y-axis is obtained by dividing Ly by δy, i.e., Py = Ly / δy. After receiving the first pulse count Px, the second pulse count Py, and the first rotation angle deviation value Δθ, the drive unit simultaneously controls the three-axis execution structure of the XYθ platform. The X-axis servo driver drives the platform to move precisely Lx distance along the X-axis according to the first pulse count Px, and the Y-axis servo driver synchronously drives the platform to move Ly distance along the Y-axis according to the second pulse count Py. The rotational actuator (such as a rotary motor or cylinder) receives the angle command –Δθ and performs a reverse rotation compensation operation to correct the posture of the foam material. Throughout the compensation process, the encoder provides real-time feedback on the motion status and implements closed-loop position control to ensure that the deviation between the actual displacement or rotation angle of each axis and the command value is within the allowable error range.

[0036] In this embodiment, before calling the pre-calibrated and stored X-axis pixel-to-millimeter conversion coefficients and Y-axis pixel-to-millimeter conversion coefficients, the X-axis pixel-to-millimeter conversion coefficients and Y-axis pixel-to-millimeter conversion coefficients are obtained through a nine-point calibration method, including: fixing the calibration target on the XYθ platform worktable during the calibration stage; pre-setting nine calibration points on the surface of the calibration target in a three-row, three-column evenly distributed manner; acquiring the initial image of the calibration target and identifying the initial pixel coordinates of the nine calibration points in the image coordinate system; controlling the X-axis servo motor of the XYθ platform to drive the worktable to move a first preset distance in the positive X direction; acquiring the image of the calibration target after the movement and identifying the first pixel coordinates of the nine calibration points after the movement; calculating the pixel change of each calibration point in the X direction; averaging the pixel changes of the nine calibration points to obtain the average pixel change in the X direction; and dividing the first preset distance by the average pixel change in the X direction to obtain the X-axis... The first conversion coefficient is obtained by controlling the X-axis servo motor of the XYθ platform to drive the worktable to move backward along the X direction by a second preset distance, acquiring the image of the calibration target after the movement, identifying the pixel coordinates of the nine calibration points after the second movement, calculating the pixel change in the X direction and averaging it, dividing the second preset distance by the average value to obtain the second X-axis conversion coefficient, and averaging the first and second X-axis conversion coefficients to obtain the final X-axis pixel-to-millimeter conversion coefficient; the same method is used to drive the Y-axis servo motor to move forward and backward along the Y direction by a third and a fourth preset distance, respectively, and calculating the pixel change in the Y direction of the nine calibration points to obtain the final Y-axis pixel-to-millimeter conversion coefficient. The nine-point distribution covers the entire working field of view to eliminate the influence of lens distortion and installation tilt, and the bidirectional movement eliminates mechanical clearance error, thereby improving the calibration accuracy of the pixel-to-millimeter conversion coefficient.

[0037] In one specific embodiment, the driving unit further includes: The compensation start subunit is used to send the first pulse count to the X-axis servo driver to drive the X-axis servo motor to perform compensation motion, send the second pulse count to the Y-axis servo driver to drive the Y-axis servo motor to perform compensation motion, and at the same time drive the θ-axis rotation mechanism to perform angle compensation according to the first rotation angle deviation value. The real-time feedback subunit is used to provide real-time feedback of the actual displacement of the X-axis and the actual displacement of the Y-axis through the X-axis encoder and the Y-axis encoder, and to calculate the first displacement difference between the actual displacement of the X-axis and the first compensation displacement, and the second displacement difference between the actual displacement of the Y-axis and the second compensation displacement, respectively. The compensation termination unit is used to determine that compensation is in place when both the first displacement difference and the second displacement difference are less than the preset displacement difference.

[0038] Specifically, the compensation initiation subunit sends the first and second pulse counts to the corresponding servo drives for the X and Y axes, respectively, thereby initiating the displacement compensation execution actions of the X-axis and Y-axis servo motors. The X-axis servo motor performs linear movement along the X direction based on the received first pulse count, causing the platform to generate a translational movement corresponding to the first compensation displacement. The Y-axis servo motor performs compensation movement along the Y direction in the same manner based on the second pulse count, thereby guiding the foam material to the expected position matching the image coordinate system in the platform coordinate system. Simultaneously, based on the first rotation angle deviation value Δθ, the rotation actuator (such as a servo motor or cylinder) of the θ-axis is controlled to rotate, achieving attitude compensation around the vertical axis. The compensation direction is determined by the sign of Δθ, and the compensation angle is equal to –Δθ, thus achieving reverse correction. During the compensation action, a high-precision incremental encoder installed at the tail of the X-axis and Y-axis servo motors samples the rotation angle of the current motor output shaft in real time. The cumulative pulse count output by the encoder is multiplied by the pulse equivalent parameters δx and δy of the servo system, and converted into the actual displacement dx of the X-axis and the actual displacement dy of the Y-axis, respectively. These two real-time displacement values ​​are then compared with the first compensation displacement Lx and the second compensation displacement Ly to calculate the first displacement difference ΔLx = |dx – Lx| for the X-axis and the second displacement difference ΔLy = |dy – Ly| for the Y-axis, reflecting the execution error of the current platform's compensation accuracy. The compensation termination unit then determines whether the compensation has reached the preset accuracy standard based on these differences. If the first displacement difference ΔLx is less than or equal to the displacement tolerance threshold set for the X-axis (e.g., 0.05 mm), and the second displacement difference ΔLy is less than or equal to the tolerance threshold for the Y-axis (e.g., 0.05 mm), it indicates that the displacement compensation for both the X-axis and Y-axis has been completed. Combined with the angle feedback signal from the θ-axis rotary actuator, if the angle error between the current rotation angle and –Δθ is less than or equal to the rotation angle tolerance (e.g., 0.1 degrees), the system determines that the compensation for all three axes is in place. The compensation completion unit sends a compensation completion signal, and the system status switches to "waiting for image re-inspection" or "preparing for punching" mode.

[0039] Before sending the target pulse count of the X-axis servo motor to the X-axis servo driver to drive the X-axis servo motor to perform compensated motion, the process includes an XY-axis coupling compensation correction step: This involves calling pre-calibrated and stored XY-axis cross-coupling coefficients, which include a first coupling coefficient from the X-axis to the Y-axis and a second coupling coefficient from the Y-axis to the X-axis. The first coupling coefficient represents the ratio of the coupled displacement in the Y-axis direction caused by X-axis motion to the amount of X-axis motion, and the second coupling coefficient represents the ratio of the coupled displacement in the X-axis direction caused by Y-axis motion to the amount of Y-axis motion. An XY-axis coupling compensation matrix is ​​then established based on the first and second coupling coefficients. The first row and first column elements of the coupling compensation matrix are... 1. The element in the first row and second column is the negative value of the second coupling coefficient, the element in the second row and first column is the negative value of the first coupling coefficient, and the element in the second row and second column is 1. The target pulse number of the X-axis servo motor and the target pulse number of the Y-axis servo motor are used to form the original pulse vector. The original pulse vector is multiplied by the coupling compensation matrix to obtain the corrected target pulse number of the X-axis servo motor and the corrected target pulse number of the Y-axis servo motor. The corrected target pulse number of the X-axis servo motor is sent to the X-axis servo driver, and the corrected target pulse number of the Y-axis servo motor is sent to the Y-axis servo driver to eliminate the cross-interference error caused by the mechanical coupling of the XY axis and realize the precise coordinated compensation motion of the XY axis and the θ axis.

[0040] In one specific embodiment, the real-time feedback subunit is specifically used for: The cumulative pulse count of the X-axis servo motor and the Y-axis servo motor is collected in real time by the X-axis encoder and the Y-axis encoder. The cumulative pulse count is multiplied by the X-axis pulse equivalent to obtain the actual X-axis displacement. The cumulative pulse count is then multiplied by the Y-axis pulse equivalent to obtain the actual Y-axis displacement. The first displacement difference is obtained by subtracting the actual displacement on the X-axis from the first compensation displacement, and the second displacement difference is obtained by subtracting the actual displacement on the Y-axis from the second compensation displacement.

[0041] Specifically, the cumulative pulse counts of the X-axis and Y-axis servo motors are collected using X-axis and Y-axis encoders, respectively. The X-axis encoder is installed at the end of the rotating shaft of the X-axis servo motor, and its output signal provides feedback on the motor's rotation angle information in the form of incremental pulses. The Y-axis encoder works similarly, reflecting the motor's operation in the form of high-frequency pulses. The controller collects the encoder output values ​​of both axes at fixed time intervals and records the total cumulative pulse count, denoted as Nx and Ny, respectively, to represent the cumulative displacement of the motor since self-start compensation. The cumulative X-axis pulse count Nx is multiplied by the pre-calibrated X-axis pulse equivalent δx to convert it into the actual X-axis displacement dx, i.e., dx = Nx × δx; the cumulative Y-axis pulse count Ny is multiplied by the Y-axis pulse equivalent δy to convert it into the actual Y-axis displacement dy, i.e., dy = Ny × δy, where δx and δy represent the actual displacement length corresponding to a single pulse in the X-axis and Y-axis directions, respectively, in millimeters per pulse. The difference between the actual displacement dx and the set first compensation displacement Lx is used to obtain the first displacement difference value ΔLx = |dx|. Lx|, and subtract dy from the second compensation displacement Ly to obtain the second displacement difference ΔLy = |dy|. Ly|. These two displacement differences are expressed in absolute value form and are used to quantitatively assess the compensation execution error. When ΔLx and ΔLy are less than or equal to the accuracy tolerance thresholds of the X-axis and Y-axis (e.g., 0.05mm), respectively, it is determined that the current displacement compensation action has achieved the closed-loop control objective; otherwise, the controller will continue to maintain the real-time feedback process until the compensation error converges within the tolerance.

[0042] In one specific embodiment, the loop module 104 further includes: The coordinate comparison unit is used to re-acquire the grayscale image of the second surface of the foam material after compensation is in place, and identify the corresponding center coordinates of the third MARK point and the fourth MARK point. Based on the center coordinates of the third MARK point and the fourth MARK point, the second X-direction deviation value, the second Y-direction deviation value, and the second rotation angle deviation value are calculated. The loop output unit is used to send a punching command to the die-cutting mechanism to perform die-cutting when the second X-direction deviation value, the second Y-direction deviation value, and the second rotation angle deviation value are all less than the target deviation value; otherwise, the compensation motion process is re-executed.

[0043] Specifically, after completing the first servo compensation motion and reaching the initial compensation positioning, the coordinate comparison unit starts, and the controller restarts the CCD image acquisition process to perform a second image acquisition on the surface of the foam material, which is currently stationary at the die-cutting station, obtaining a second surface grayscale image. The acquisition parameters of the second surface grayscale image are consistent with those of the initial acquisition, including image resolution, field of view, light source brightness and angle, ensuring consistency in image contrast and recognition stability. The image processing module performs image analysis on the second surface grayscale image, using a fixed grayscale threshold strategy and a connected component extraction algorithm to identify the actual connected regions of the current MARK point in the image. Based on the area sorting principle, the two connected regions with the largest areas are selected as the third and fourth MARK point regions in the current compensation state, and their geometric center coordinates are calculated respectively, obtaining the center coordinates of the third MARK point (xc3, yc3) and the center coordinates of the fourth MARK point (xc4, yc4). The coordinate comparison unit performs difference analysis with the standard MARK point coordinates based on these two center coordinates. The center coordinates of the third MARK point (xc3, yc3) are compared with the first standard coordinates (x...). 01 , y 01 Subtract the coordinates of the center of the fourth MARK point (xc4, yc4) from the coordinates of the second standard coordinate (xc4, yc4). 02 , y 02 Subtracting the two values, we calculate the coordinate differences Δx3 and Δx4 in the X direction and Δy3 and Δy4 in the Y direction for the second stage. Averaging these two sets of differences along the X and Y axes, we obtain the second X-direction deviation ΔX′ = (Δx3 + Δx4) / 2 and the second Y-direction deviation ΔY′ = (Δy3 + Δy4) / 2. Simultaneously, we recalculate the actual angle α′1 of the line connecting the third and fourth MARK points α′1 = arctan[(yc4) / 2]. yc3) / (xc4 xc3)], and the angle α0 = arctan[(y 02 y 01 ) / (x 02 x 01 Subtracting the two values, we get the second rotation angle deviation value Δθ′ = α′1. α0 is used to quantify the residual attitude offset after the first compensation. The loop output unit compares the second X-direction deviation value ΔX′, the second Y-direction deviation value ΔY′, and the second rotation angle deviation value Δθ′ with the target deviation threshold preset by the controller. The target threshold is set to be less than or equal to 0.1 mm in the X and Y directions and less than or equal to 0.2° in the rotation angle. If all three deviation values ​​meet the threshold conditions, the loop output unit sends a punching start command to the die-cutting actuator, causing the die-cutting actuator to perform the die-cutting action to complete the die-cutting process; otherwise, if any deviation still exceeds the target threshold, the system determines that the compensation is unqualified, immediately returns to the aforementioned compensation process, recalculates the compensation displacement and rotation angle based on the residual deviation amount in the second stage, and re-executes pulse sending and motion control to achieve closed-loop iteration of compensation. The loop output unit synchronously updates the current compensation iteration number and compares it with the maximum iteration number. If the number of iterations exceeds the system's allowed upper limit (e.g., three times) and the accuracy requirement is still not met, the controller issues an alarm signal and interrupts the subsequent die-cutting operation, thereby avoiding the output of unqualified products due to positioning errors.

[0044] In one specific embodiment, the loop output unit is specifically used for: When the second X-direction deviation value, the second Y-direction deviation value and the second rotation angle deviation value are all less than the target deviation value, a punching command is sent to the die-cutting mechanism to drive the die to press down and perform foam die-cutting. If any one of the second X-direction deviation value, the second Y-direction deviation value, or the second rotation angle deviation value exceeds the target deviation value, the compensation motion process is re-executed based on the second X-direction deviation value, the second Y-direction deviation value, and the second rotation angle deviation value.

[0045] Specifically, the second X-direction deviation value, the second Y-direction deviation value, and the second rotation angle deviation value are compared with the preset target deviation values ​​in the controller. The target deviation thresholds for the X and Y directions are set to no more than ±0.1 mm, and the target threshold for the rotation angle deviation is set to no more than ±0.2°. The controller determines that the actual positioning state of the foam material fully meets the die-cutting accuracy requirements only when the absolute values ​​of ΔX′, ΔY′, and Δθ′ are all less than or equal to the corresponding target thresholds. At this point, the system immediately sends a punching start command to the die-cutting mechanism and drives the die-cutting die to press down. The vertical drive component presses the die-cutting die of the set shape along the Z-axis to contact the foam material and perform the punching operation, thereby completing the high-precision die-cutting of the current foam workpiece. During the die-cutting process, punching depth control and pressure closed-loop feedback are executed simultaneously to ensure processing stability and die life. If any of the above three deviation values ​​exceeds the corresponding target deviation threshold, for example, ΔX′>0.1mm, ΔY′>0.1mm, or Δθ′>0.2°, it is determined that the current material positioning state still has unacceptable residual error and cannot immediately enter the punching stage. At this time, the controller uses the currently identified second X-direction deviation value, second Y-direction deviation value, and second rotation angle deviation value as the input parameters for a new round of compensation control. It calls the pixel-to-displacement conversion coefficient again and recalculates the X-axis compensation displacement Lx′, Y-axis compensation displacement Ly′, and θ-axis rotation angle compensation Δθ′. Then, it converts them into the corresponding X-axis pulse count Px′ and Y-axis pulse count Py′ and resends them to the servo drive system to drive the XYθ platform to perform precise position and attitude compensation again. Simultaneously update the current compensation iteration count and determine whether the preset maximum iteration count (e.g., 3 times) has been reached. If the limit has not been exceeded, the compensation process proceeds normally. If the upper limit has been reached and the process still cannot converge to the target deviation range, an alarm signal is issued to prompt the operator to check the material alignment status or the operating accuracy of the die-cutting equipment, and the current die-cutting task is automatically paused to prevent the generation of waste.

[0046] Through the collaborative efforts of the aforementioned components, the grayscale image of the foam surface is binarized and connected component marking is performed. A connected component area filtering method is used to effectively eliminate noise interference from the porous surface of the foam, accurately identifying the center coordinates of the two MARK points. This solves the technical problem of blurred marker point edges and difficulty in positioning caused by the porous structure of the foam surface. By configuring dual MARK points, the X-direction position deviation, Y-direction position deviation, and rotation angle deviation are calculated simultaneously, achieving comprehensive perception of the three-dimensional spatial posture of the foam material. The image coordinate system deviation is converted into a mechanical coordinate system compensation amount using pre-calibrated transformation coefficients, and then converted into servo motor pulse commands to drive the XYθ platform to perform compensation motion. The encoder provides real-time feedback of the actual displacement and compares it with the target displacement to ensure that the compensation accuracy meets the preset requirements. After compensation, the image is re-acquired, residual deviations are calculated, and iterative compensation is performed to ensure that the final positioning accuracy meets the die-cutting requirements. This effectively solves the problem of positional offset and rotation angle caused by flexible deformation of the foam material during feeding, improving the positioning accuracy and production efficiency of foam buffer die-cutting, and realizing the intelligent and automated nature of the foam die-cutting process.

[0047] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing system embodiments, and will not be repeated here.

[0048] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the system described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0049] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention 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 of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent positioning control system for the die-cutting process of foam cushioning components, characterized in that, include: The acquisition module is used to send the foam material to the die-cutting station through the material pulling mechanism and acquire the grayscale image of the first surface if the current working mode is CCD automatic positioning mode. The calculation module is used to locate the center coordinates of the first MARK point and the center coordinates of the second MARK point based on the first surface grayscale image and to calculate the first X-direction deviation value, the first Y-direction deviation value and the first rotation angle deviation value; The compensation motion module is used to perform compensation motion based on the first X-direction deviation value, the first Y-direction deviation value, and the first rotation angle deviation value; The loop module is used to re-acquire the grayscale image of the second surface of the compensated foam material and calculate the residual deviation value. When the residual deviation value is less than the target deviation value, a punching command is sent; otherwise, the compensation motion process is re-executed.

2. The intelligent positioning control system for the die-cutting process of foam cushioning components according to claim 1, characterized in that, The data acquisition module is specifically used for: If the current working mode is CCD automatic positioning mode, the material pulling mechanism is activated to pull the foam material from the unwinding mechanism to the die-cutting station. The material pulling length is measured in real time using an encoder; When the measured length of the material being pulled reaches the preset length value, the material pulling mechanism stops, so that the two cross-shaped MARK markers are within the field of view of the CCD camera; The CCD camera is activated and the ring LED light source is turned on to illuminate the surface of the foam material. The area containing the two cross-shaped MARK markers is captured to obtain a first surface grayscale image.

3. The intelligent positioning control system for the die-cutting process of foam cushioning components according to claim 1, characterized in that, The calculation module also includes: Extraction unit is used to extract the gray value of the first pixel in the first surface grayscale image and mark the first pixel with the gray value lower than the preset grayscale threshold as the second pixel. A connected component labeling unit is used to locate the first MARK point region and the second MARK point region based on the second pixel. The center coordinate calculation unit is used to calculate the center coordinates of the first MARK point in the first MARK point region and the center coordinates of the second MARK point in the second MARK point region. The deviation calculation unit is used to calculate the first X-direction deviation value and the first Y-direction deviation value based on the center coordinates of the first MARK point and the center coordinates of the second MARK point, and to calculate the first rotation angle deviation value by connecting the center coordinates of the first MARK point and the center coordinates of the second MARK point.

4. The intelligent positioning control system for the die-cutting process of foam cushioning components according to claim 3, characterized in that, Connected component labeling unit, specifically used for: The second pixel is labeled with connected components to obtain multiple connected components, and the area of ​​each connected component is calculated. Select the two connected components with the largest areas from the plurality of connected components, mark the connected component with the largest area as the first MARK point region, and mark the connected component with the second largest area as the second MARK point region.

5. The intelligent positioning control system for the die-cutting process of foam cushioning components according to claim 3, characterized in that, The deviation calculation unit is specifically used for: The first coordinate difference is obtained by subtracting the center coordinates of the first MARK point from the preset first standard coordinates; the second coordinate difference is obtained by subtracting the center coordinates of the second MARK point from the preset second standard coordinates. The first coordinate difference and the second coordinate difference are averaged in the X and Y directions, respectively, to obtain the first X-direction deviation value and the first Y-direction deviation value. The first connecting angle is calculated based on the center coordinates of the first MARK point and the center coordinates of the second MARK point. The second connecting angle is calculated based on the first standard coordinates and the second standard coordinates. The difference between the first connecting angle and the second connecting angle is used to obtain the first rotation angle deviation value.

6. The intelligent positioning control system for the die-cutting process of foam cushioning components according to claim 1, characterized in that, The motion compensation module also includes: The conversion coefficient query unit is used to query the first conversion coefficient of the X-axis and the second conversion coefficient of the Y-axis. The displacement compensation unit is used to multiply the first X-direction deviation value by the first conversion coefficient to obtain the first compensation displacement, and to multiply the first Y-direction deviation value by the second conversion coefficient of the Y-axis to obtain the second compensation displacement. The pulse calculation unit is used to divide the first compensation displacement by the X-axis pulse equivalent to obtain the first pulse number, and divide the second compensation displacement by the Y-axis pulse equivalent to obtain the second pulse number; The drive unit is configured to perform compensating motion based on the first pulse count, the second pulse count, and the first rotation angle deviation value.

7. The intelligent positioning control system for the die-cutting process of foam cushioning components according to claim 6, characterized in that, The drive unit also includes: The compensation start subunit is used to send the first pulse count to the X-axis servo driver to drive the X-axis servo motor to perform compensation motion, send the second pulse count to the Y-axis servo driver to drive the Y-axis servo motor to perform compensation motion, and simultaneously drive the θ-axis rotation mechanism to perform angle compensation according to the first rotation angle deviation value. The real-time feedback subunit is used to provide real-time feedback of the actual displacement of the X-axis and the actual displacement of the Y-axis through the X-axis encoder and the Y-axis encoder, and to calculate the first displacement difference between the actual displacement of the X-axis and the first compensation displacement, and the second displacement difference between the actual displacement of the Y-axis and the second compensation displacement, respectively. The compensation termination unit is used to determine that compensation is in place when both the first displacement difference and the second displacement difference are less than a preset displacement difference.

8. The intelligent positioning control system for the die-cutting process of foam cushioning components according to claim 7, characterized in that, The real-time feedback subunit is specifically used for: The cumulative pulse count of the X-axis servo motor and the Y-axis servo motor is collected in real time by the X-axis encoder and the Y-axis encoder. The cumulative pulse count is multiplied by the X-axis pulse equivalent to obtain the actual X-axis displacement. The cumulative pulse count is then multiplied by the Y-axis pulse equivalent to obtain the actual Y-axis displacement. The first displacement difference is obtained by subtracting the actual displacement of the X-axis from the first compensation displacement, and the second displacement difference is obtained by subtracting the actual displacement of the Y-axis from the second compensation displacement.

9. The intelligent positioning control system for the die-cutting process of foam cushioning components according to claim 1, characterized in that, The loop module also includes: The coordinate comparison unit is used to re-acquire the grayscale image of the second surface of the foam material after compensation is in place, and identify the corresponding center coordinates of the third MARK point and the center coordinates of the fourth MARK point. Based on the center coordinates of the third MARK point and the center coordinates of the fourth MARK point, the second X-direction deviation value, the second Y-direction deviation value, and the second rotation angle deviation value are calculated. The loop output unit is used to send a punching command to the die-cutting mechanism to perform die-cutting when the second X-direction deviation value, the second Y-direction deviation value, and the second rotation angle deviation value are all less than the target deviation value; otherwise, the compensation motion process is re-executed.

10. The intelligent positioning control system for the die-cutting process of foam cushioning components according to claim 9, characterized in that, The loop output unit is specifically used for: When the second X-direction deviation value, the second Y-direction deviation value and the second rotation angle deviation value are all less than the target deviation value, a punching command is sent to the die-cutting mechanism to drive the die to press down and perform foam die-cutting. If any one of the second X-direction deviation value, the second Y-direction deviation value, or the second rotation angle deviation value exceeds the target deviation value, the compensation motion process is re-executed based on the second X-direction deviation value, the second Y-direction deviation value, and the second rotation angle deviation value.

Images