A 3D vision-based outsole and upper adhesive profile scribing device and method

Abstract

The application relates to a 3D-vision-based sole and vamp adhesive contour line marking device and method, and relates to the field of shoe processing.The device comprises a fixing support, a pressing module, a 3D-vision module, a line marking module and a control system.The pressing module comprises a rotating part arranged on the fixing support, a supporting part arranged above the rotating part and a pressing part arranged on the fixing support and above the supporting part.The 3D-vision module comprises a 3D-vision sensor arranged on the fixing support.The line marking module comprises a line marking part, a first moving assembly for driving the line marking part to move in the vertical direction and a second moving assembly for driving the line marking part to move in the direction of approaching or moving away from the supporting part.The control system is used for carrying out 3D image reconstruction according to the 3D-vision module and calculating the adhesive contour line when the pressing module works, and controlling the line marking module to work according to the adhesive contour line.The application has the effects of improving the line marking accuracy and quality and realizing intelligent production.

Description

Technical Field

[0001] This application relates to the field of footwear processing, and in particular to a device and method for marking the outline of adhesive on soles and uppers based on 3D vision. Background Technology

[0002] During shoe production, the upper (shoe collar) and sole need to be tightly bonded together, and adhesive must be evenly applied to the bonding area for fixation. Therefore, the accuracy of the adhesive application directly affects the adhesion between the sole and the upper. If the adhesive extends beyond the bonding area, excess adhesive will occur; conversely, if the adhesive does not cover the bonding area, insufficient adhesive will result. Both of these situations directly affect the quality of the shoes.

[0003] Currently, shoe factories typically have workers manually bond and press the sole and upper together before applying glue. Then, a highlighter is used to mark the glue application boundary along the outline of the bonding area. This is then handed over to glue application personnel who manually apply the glue along this boundary. However, this traditional manual marking and glue application method requires a high level of experience and is prone to inaccuracies and deviations. Therefore, it often necessitates the addition of inspectors to verify the marking process. This method is not only inefficient and costly in terms of labor, but also lacks quality control, hindering the implementation of intelligent manufacturing. Summary of the Invention

[0004] To improve the accuracy and quality of marking lines and realize intelligent production, this application provides a device and method for marking the outline of adhesive on shoe soles and uppers based on 3D vision.

[0005] This application provides a 3D vision-based device and method for marking the outline of adhesive on shoe soles and uppers, employing the following technical solution:

[0006] In a first aspect, this application provides a 3D vision-based device for marking the outline of adhesive on the sole and upper of a shoe, employing the following technical solution:

[0007] A 3D vision-based device for marking the outline of adhesive on shoe soles and uppers includes a fixed bracket, a pressing module disposed on the fixed bracket, a 3D vision module, a marking module, and a control system.

[0008] The pressing module includes a rotating part disposed on the fixed bracket, a supporting part disposed above the rotating part, and a pressing part disposed on the fixed bracket and located above the supporting part. The rotating part is used to drive the supporting part to rotate in the horizontal plane, and the pressing part is used to press in the vertical direction.

[0009] The 3D vision module includes a 3D vision sensor disposed on the fixed bracket, with the 3D vision sensor facing between the support portion and the pressing portion;

[0010] The marking module includes a marking section, a first moving component disposed on the fixed bracket and used to drive the marking section to move in a vertical direction, and a second moving component connected to the first moving component and used to drive the marking section to move in a direction close to or away from the support section, wherein the marking section is disposed on the second moving component;

[0011] The control system is used to reconstruct 3D images based on the 3D vision module and calculate the adhesive outline when the pressing module is working, and to control the scribing module to work according to the adhesive outline.

[0012] Optionally, the control system controls the scribing module to operate according to the adhesive outline according to the following steps:

[0013] S1: Establish a right-handed rectangular coordinate system, denoted as M1, with the intersection of the driving direction of the first moving component and the driving direction of the second moving component as the origin O, the driving direction of the second moving component as the X-axis, and the driving direction of the first moving component as the Y-axis. The 3D vision sensor coordinate system is denoted as M2. The rotation matrix for transforming from the M2 coordinate system to the M1 coordinate system is calibrated using a calibration plate and is denoted as RT0.

[0014] S2: After the 3D image reconstruction is completed, multiply each point in the adhesive contour point cloud dataset Q by RT0, rotate it to the M1 coordinate system, and the result after rotation is Q1; denote point P(P x ,P y ,P z Let P be any point on Q1, and let P's projected coordinates in the XOZ plane be P. ， (P x ,P y ,0), passing through point P ， A line perpendicular to the X-axis intersects the X-axis at point P. ，， (P x ,0,0), vector OP ， With vector OP ，， With an included angle of θ1, the control system controls the rotating part to reverse the support part by θ1 degrees, rotating point P into the XOY plane;

[0015] S3: Calculate the rotation matrix of the contour point cloud about the Y-axis by θ1:

[0016] Multiplying point P by RT1 gives the coordinates of point P1 rotated to the XOY plane. x1 ,P y1 ,0), the control system controls the second moving component to move the scribing part along the X-axis P x1The control system controls the first moving component to move the scribing portion along the Y-axis. y1 And draw lines;

[0017] S4: Repeat steps S2 and S3 until every point in the adhesive contour point cloud dataset Q has been calculated and drawn, and the contour point cloud dataset is Q in the nth calculation. n =Q n-1 *RT n .

[0018] Optionally, the pressing part includes, from top to bottom, a cylinder vertically fixed to the fixed bracket, a bushing fixed to the cylinder piston rod, and a telescopic pressure rod vertically arranged and rotatably connected to the bushing.

[0019] Optionally, the telescopic pressure rod is coaxially arranged with the rotation axis of the rotating part.

[0020] Optionally, the marking section includes a pen holder connected to the second moving component, a marking pen disposed at the end of the pen holder, and a spring located inside the pen holder and connected to the marking pen.

[0021] Optionally, the first moving component includes a first guide rod that is vertically arranged and fixed to the fixed bracket, a first lead screw that is vertically arranged and rotatably connected to the fixed bracket, a first drive motor that is fixed to the fixed bracket and fixedly connected to one end of the first lead screw, and a first slide that is slidably arranged on the first guide rod and threadedly connected to the first lead screw.

[0022] The second moving component includes a second guide rod disposed along a direction close to or away from the support and fixed to the first slide, a second lead screw disposed parallel to the second guide rod and rotatably connected to the first slide, a second drive motor fixed to the first slide and fixedly connected to one end of the second lead screw, and a second slide slidably disposed on the second guide rod and threadedly connected to the second lead screw, the second slide being connected to the scribing portion.

[0023] Optionally, two 3D vision sensors are provided and symmetrically arranged on both sides of the pressing module.

[0024] Optionally, the control system acquires 3D visual data from two 3D vision sensors, calibrates the relationship between the two sets of 3D visual data and places them in the same coordinate system, and then reconstructs complete 3D visual data through image stitching and fusion.

[0025] Optionally, the support is made of rubber and the top surface of the support is arc-shaped.

[0026] Secondly, this application provides a method for outlining the adhesive contours of the sole and upper based on 3D vision, employing the following technical solution:

[0027] A method for marking the outline of adhesive on the sole and upper of a shoe based on 3D vision, using the aforementioned marking device, includes the following steps:

[0028] Place the sole and upper to be bonded on the support, and the control system controls the pressing part to press vertically downwards, so that the adhesive parts of the sole and upper are tightly bonded.

[0029] The 3D vision sensor reconstructs a 3D visual image of the area where the sole and upper are bonded after pressing, and the control system calculates the adhesive outline based on the 3D visual image.

[0030] The control system calculates the rotation angle of the rotating part driving the support part, the vertical movement distance of the first moving component driving the scribing part, and the movement distance of the second moving component driving the scribing part along the direction closer to or farther from the support part based on the adhesive contour line. Then, based on the calculation results, the control system controls the rotating part, the first moving component, and the second moving component to work, so that the scribing part moves to the adhesive contour line to scribing.

[0031] In summary, this application includes at least one of the following beneficial technical effects:

[0032] 1. By setting up a control system, the system can control and manage the bonding module, 3D vision module, and marking module, thereby realizing operations such as bonding of the sole and upper, 3D image reconstruction, data processing and calculation, processing trajectory conversion, and marking processing. This avoids manual operation and realizes automated and intelligent production, which not only saves labor but also greatly improves efficiency.

[0033] 2. By employing 3D vision technology and arranging 3D vision sensors on both sides of the support, 3D images of the shoe upper and sole bonding area can be reconstructed in real time. The data processing system can then quickly analyze and calculate the images, automatically extract the boundary contour lines of the bonding area, and convert the contour line data into the processing trajectory information of the scribing module. This drives the scribing module to achieve fully automatic and precise scribing, which not only reduces reliance on manual labor but also makes the scribing trajectory recognized by 3D vision more accurate, thus avoiding human scribing errors and improving product quality.

[0034] 3. The scribing module structure used in this invention is simple. It only requires controlling the rotation of the first drive motor and the second drive motor to realize the up-down and back-forward movement of the scribing pen. Combined with the rotation of the rotary motor, accurate scribing can be achieved. Compared with the traditional manual scribing method, the use of this structure improves the accuracy of scribing. Moreover, the structure is simple, has low manufacturing cost, and is suitable for mass production. Attached Figure Description

[0035] Figure 1 This is a structural schematic diagram of an embodiment of this application;

[0036] Figure 2 This is a schematic diagram of the pressing module according to an embodiment of this application;

[0037] Figure 3 This is a schematic diagram of the structure of the scribing module according to an embodiment of this application;

[0038] Figure 4 This is a coordinate system diagram of the calculation process of the control system in the embodiments of this application.

[0039] Explanation of reference numerals in the attached figures:

[0040] 1. Fixed bracket; 2. Pressing module; 21. Rotating part; 211. Rotary motor; 212. Turntable; 22. Support part; 23. Pressing part; 231. Cylinder; 232. Bushing; 233. Telescopic pressure rod; 3. 3D vision module; 31. 3D vision sensor; 4. Marking module; 41. First moving component; 411. First guide rod; 412. First lead screw; 413. First drive motor; 414. First slide; 42. Second moving component; 421. Second guide rod; 422. Second lead screw; 423. Second drive motor; 424. Second slide; 43. Marking part; 431. Marking pen; 432. Pen holder. Detailed Implementation

[0041] The following is in conjunction with the appendix Figure 1-4 This application will be described in further detail.

[0042] This application discloses a 3D vision-based device for marking the outline of adhesive on the sole and upper of a shoe.

[0043] Reference Figure 1 A 3D vision-based adhesive contour marking device for shoe soles and uppers includes a fixed bracket 1, a pressing module 2, a 3D vision module 3, a marking module 4, and a control system (not shown in the figure) disposed on the fixed bracket 1. The pressing module 2 is used to tightly bond the shoe sole and the upper. The 3D vision module 3 is used to reconstruct the 3D visual image of the bonding area between the pressed shoe sole and the upper. The control system calculates the adhesive contour line based on the 3D visual image and controls the marking module 4 to mark the adhesive contour line.

[0044] The pressing module 2 includes a rotating part 21 disposed on the fixed bracket 1, a supporting part 22 disposed above the rotating part 21, and a pressing part 23 disposed on the fixed bracket 1 and located above the supporting part 22. The rotating part 21 is used to drive the supporting part 22 to rotate in the horizontal plane, and the pressing part 23 is used to press in the vertical direction.

[0045] The rotating part 21 includes a rotary motor 211 fixedly mounted on the fixed bracket 1. The output shaft of the rotary motor 211 is vertically oriented, and a turntable 212 is fixedly mounted on the top of the rotary motor 211. The support part 22 is fixedly mounted above the turntable 212. The support part 22 is made of one or more pieces of rubber material. The top surface of the support part 22 is arc-shaped and fits against the bottom surface of the shoe sole. This ensures that the entire shoe sole is evenly stressed during the pressing process, and the elastic rubber material also provides support for the shoe sole and ensures that the shoe sole and the upper fit together fully.

[0046] The pressing part 23, from top to bottom, includes a cylinder 231 vertically fixed to the fixed bracket 1, a bushing 232 fixed to the piston rod of the cylinder 231, and a telescopic pressure rod 233 vertically arranged and rotatably connected to the bushing 232. The telescopic pressure rod 233 is coaxially arranged with the rotation axis of the rotary motor 211. The telescopic pressure rod 233 can rotate freely 360 degrees through the bushing 232, preventing damage to the shoe upper due to the fixed telescopic pressure rod 233 when the support part 22 rotates. Since the telescopic pressure rod 233 and the rotation axis of the support part 22 are on the same rotation axis, when the cylinder 231 is activated, it drives the telescopic pressure rod 233 to press down, giving the shoe upper and sole placed on the support part 22 a downward pressure, so that the shoe upper and sole are completely compacted and fully fit together.

[0047] The 3D vision module 3 includes two 3D vision sensors 31 mounted on the fixed bracket 1. These sensors are symmetrically positioned above both sides of the support 22, facing the sole and upper placed between the support 22 and the pressing part 23. After the sole and upper are pressed together, the control system activates the 3D scanning system. The two sets of 3D vision sensors 31 reconstruct 3D images of the pressed area between the sole and upper. The control system acquires 3D vision data from both sensors 31, calibrates the relationship between the two sets of 3D vision data and places them in the same coordinate system, then reconstructs complete 3D vision data through image stitching and fusion, and calculates the adhesive outline of the area where the sole and upper are pressed together using an algorithm.

[0048] The scribing module 4 is located on one side of the pressing module 2. The scribing module 4 includes a scribing part 43, a first moving component 41 disposed on the fixed bracket 1 and used to drive the scribing part 43 to move in the vertical direction, and a second moving component 42 connected to the first moving component 41 and used to drive the scribing part 43 to move in the direction close to or away from the support part 22. The scribing part 43 is disposed on the second moving component 42.

[0049] The first moving assembly 41 includes a first guide rod 411 vertically arranged and fixed to the fixed bracket 1, a first lead screw 412 vertically arranged and rotatably connected to the fixed bracket 1, a first drive motor 413 fixed to the fixed bracket 1 and fixedly connected to one end of the first lead screw 412, and a first slide 414 slidably arranged on the first guide rod 411 and threadedly connected to the first lead screw 412; the second moving assembly 42 includes a second guide rod 421 arranged along the direction close to or away from the support portion 22 and fixed to the first slide 414, and a second drive motor 413 parallel to the second guide rod 421. A second lead screw 422 rotatably connected to the first slide 414, a second drive motor 423 fixed to the first slide 414 and fixedly connected to one end of the second lead screw 422, and a second slide 424 slidably disposed on the second guide rod 421 and threadedly connected to the second lead screw 422 are provided. The second slide 424 is connected to the marking part 43. The marking part 43 includes a pen holder 432 fixedly connected to the second slide 424, a marking pen 431 disposed at the end of the pen holder 432, and a spring (not shown in the figure) located inside the pen holder 432 and connected to the marking pen 431. The spring is sleeved on the tail end of the marking pen 431, allowing the marking pen 431 to extend and retract back and forth during the marking process, thereby avoiding rigid contact between the marking pen 431 and the shoe upper, which would cause damage to the shoe upper.

[0050] The control system controls the first drive motor 413 to drive the first lead screw 412 to rotate, so that the first slide 414 drives the marking pen 431 to move in the vertical direction. The control system controls the second drive motor 423 to drive the second lead screw 422 to rotate, so that the second slide 424 drives the marking pen 431 to move in the direction close to or away from the support part 22, thereby moving the marking pen 431 to the adhesive outline of the sole and the upper of the shoe to draw lines.

[0051] After the control system calculates the adhesive outline, it controls the scribing module 4 to work according to the adhesive outline according to the following steps:

[0052] S1: Take the intersection of the driving direction of the first moving component 41 and the driving direction of the second moving component 42 as the origin O, the driving direction of the second moving component 42 as the X-axis, and the driving direction of the first moving component 41 as the Y-axis, establish a right-hand rectangular coordinate system, denoted as M1, and the 3D vision sensor coordinate system, denoted as M2. Use a calibration plate to calibrate the rotation matrix from the M2 coordinate system to the M1 coordinate system, denoted as RT0.

[0053] S2: After the 3D image reconstruction is completed, multiply each point in the adhesive contour point cloud dataset Q by RT0, rotate it to the M1 coordinate system, and the result after rotation is Q1; denote point P(P x ,P y ,P z Let P be any point on Q1, and let P's projected coordinates in the XOZ plane be P. ， (P x ,P y ,0), passing through point P ， A line perpendicular to the X-axis intersects the X-axis at point P. ，， (P x ,0,0), vector OP ， With vector OP ，， With an included angle of θ1, the control system controls the rotary motor 211 to reverse the support part 22 by θ1 / 360 turns, rotating point P into the XOY plane;

[0054] S3: Calculate the rotation matrix of the contour point cloud about the Y-axis by θ1:

[0055] Multiplying point P by RT1 gives the coordinates of point P1 rotated to the XOY plane. x1 ,P y1 ,0), the control system controls the second moving component 42 to move the scribing part 43 along the X-axis P x1 The control system controls the first moving component 41 to move the scribing part 43 along the Y-axis. y1 And draw lines;

[0056] Let the initial values ​​of the ends of the first lead screw 412 and the second lead screw 422 be V respectively. y V x The leads of the first lead screw 412 and the second lead screw 422 are g respectively. Y g X If (P) x1 -V x If )>0, then the second drive motor 423 rotates forward (|P) x1 -V x |) / (g X (360) circles, and vice versa (|P) circles. x1 -V x |) / (g X *360) circles; similarly, if (P y1 -V y If )>0, then the first drive motor 413 rotates forward (|P) y1 -V y |) / (g Y(360) circles, and vice versa (|P) circles. y1 -V y |) / (g Y Make a 360-degree circle so that the end of the pen reaches point P1.

[0057] S4: Repeat steps S2 and S3 until every point in the adhesive contour point cloud dataset Q has been calculated and drawn, and the contour point cloud dataset is Q in the nth calculation. n =Q n-1 *RT n .

[0058] This application also discloses a method for outlining the adhesive contours of the sole and upper based on 3D vision.

[0059] A method for marking the outline of adhesive on the sole and upper of a shoe based on 3D vision, using the aforementioned marking device, includes the following steps:

[0060] Place the sole and upper to be bonded on the support part 22. The control system controls the cylinder 231 to start, driving the telescopic pressure rod 233 to press down, giving the sole and upper placed on the support part 22 a downward pressure, so that the sole and upper are completely compacted and fully bonded.

[0061] After the sole and upper are pressed together, the control system starts the 3D scanning system. Two sets of 3D vision sensors 31 reconstruct 3D images of the area where the sole and upper are bonded together. The control system collects 3D vision data from the two 3D vision sensors 31 respectively, calibrates the relationship between the two sets of 3D vision data and places them in the same coordinate system, and then reconstructs complete 3D vision data through image stitching and fusion. The adhesive outline of the area where the sole and upper are bonded together is calculated by the algorithm.

[0062] The control system calculates the rotation angle of the rotating part driving the support part 22, the vertical movement distance of the first moving component 41 driving the scribing part 43, and the movement distance of the second moving component 42 driving the scribing part 43 along the direction closer to or away from the support part 22 based on the adhesive contour line. Then, based on the calculation results, it controls the rotating part, the first moving component 41, and the second moving component 42 to operate, causing the scribing part 43 to move onto the adhesive contour line to scribing. Specifically, this includes:

[0063] S1: Take the intersection of the driving direction of the first moving component 41 and the driving direction of the second moving component 42 as the origin O, the driving direction of the second moving component 42 as the X-axis, and the driving direction of the first moving component 41 as the Y-axis, establish a right-hand rectangular coordinate system, denoted as M1, and the 3D vision sensor coordinate system, denoted as M2. Use a calibration plate to calibrate the rotation matrix from the M2 coordinate system to the M1 coordinate system, denoted as RT0.

[0064] S2: After the 3D image reconstruction is completed, multiply each point in the adhesive contour point cloud dataset Q by RT0, rotate it to the M1 coordinate system, and the result after rotation is Q1; denote point P(P x ,P y ,P z Let P be any point on Q1, and let P's projected coordinates in the XOZ plane be P. ， (P x ,P y ,0), passing through point P ， A line perpendicular to the X-axis intersects the X-axis at point P. ，， (P x ,0,0), vector OP ， With vector OP ，， With an included angle of θ1, the control system controls the rotary motor 211 to reverse the support part 22 by θ1 / 360 turns, rotating point P into the XOY plane;

[0065] S3: Calculate the rotation matrix of the contour point cloud about the Y-axis by θ1:

[0066] Multiplying point P by RT1 gives the coordinates of point P1 rotated to the XOY plane. x1 ,P y1 ,0), the control system controls the second moving component 42 to move the scribing part 43 along the X-axis P x1 The control system controls the first moving component 41 to move the scribing part 43 along the Y-axis. y1 And draw lines;

[0067] Let the initial values ​​of the ends of the first lead screw 412 and the second lead screw 422 be V respectively. y V x The leads of the first lead screw 412 and the second lead screw 422 are g respectively. Y g X If (P) x1 -V x If )>0, then the second drive motor 423 rotates forward (|P) x1 -V x |) / (g X (360) circles, and vice versa (|P) circles. x1 -V x |) / (g X *360) circles; similarly, if (P y1 -V y If )>0, then the first drive motor 413 rotates forward (|P) y1 -V y |) / (g Y (360) circles, and vice versa (|P) circles.y1 -V y |) / (g Y Make a 360-degree circle so that the end of the pen reaches point P1.

[0068] S4: Repeat steps S2 and S3 until every point in the adhesive contour point cloud dataset Q has been calculated and drawn, and the contour point cloud dataset is Q in the nth calculation. n =Q n-1 *RT n .

[0069] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A 3D vision-based device for marking the outline of adhesive on shoe soles and uppers, characterized in that, It includes a fixed bracket (1), a pressing module (2) disposed on the fixed bracket (1), a 3D vision module (3), a scribing module (4), and a control system; The pressing module (2) includes a rotating part (21) disposed on the fixed bracket (1), a supporting part (22) disposed above the rotating part (21), and a pressing part (23) disposed on the fixed bracket (1) and located above the supporting part (22). The rotating part (21) is used to drive the supporting part (22) to rotate in the horizontal plane, and the pressing part (23) is used to press in the vertical direction. The 3D vision module (3) includes a 3D vision sensor (31) disposed on the fixed bracket (1), and the 3D vision sensor (31) faces between the support part (22) and the pressing part (23); The marking module (4) includes a marking part (43), a first moving component (41) disposed on the fixed bracket (1) and used to drive the marking part (43) to move in the vertical direction, and a second moving component (42) connected to the first moving component (41) and used to drive the marking part (43) to move in the direction close to or away from the support part (22), wherein the marking part (43) is disposed on the second moving component (42); The control system is used to reconstruct the 3D image based on the 3D vision module (3) and calculate the adhesive outline when the pressing module (2) is working, and to control the scribing module (4) to work according to the adhesive outline. The control system controls the scribing module (4) to work according to the adhesive outline according to the following steps: S1: Take the intersection of the driving direction of the first moving component (41) and the driving direction of the second moving component (42) as the origin O, the driving direction of the second moving component (42) as the X-axis, and the driving direction of the first moving component (41) as the Y-axis, establish a right-hand rectangular coordinate system, denoted as M1, and the coordinate system of the 3D vision sensor (31), denoted as M2. Use a calibration plate to calibrate the rotation matrix from the M2 coordinate system to the M1 coordinate system, which is RT0. S2: After the 3D image reconstruction is completed, multiply each point in the adhesive contour point cloud dataset Q by RT0, rotate it to the M1 coordinate system, and the result after rotation is Q1; denote point P (P x ,P y ,P z Let P be any point on Q1, and let P's projected coordinates in the XOZ plane be P. ， (P) x ,P y ,0),passing through point P ， A line perpendicular to the X-axis intersects the X-axis at point P. ，， (P) x (,0,0), vector OP ， With vector OP ，， With an included angle of θ1, the control system controls the rotating part (21) to reverse the support part (22) by θ1 degrees and rotate point P into the XOY plane; S3: Calculate the rotation matrix of the contour point cloud about the Y-axis by θ1: RT1=( Multiplying point P by RT1 gives the coordinates of point P rotated to point P1 in the XOY plane. x1 ,P y1 ,0), the control system controls the second moving component (42) to move the scribing part (43) along the X-axis P x1 The control system controls the first moving component (41) to move the scribing part (43) along the Y-axis. y1 And draw lines; S4: Repeat steps S2 and S3 until every point in the adhesive contour point cloud dataset Q has been calculated and drawn, and the contour point cloud dataset is Q in the nth calculation. n =Q n-1 *RT n .

2. The 3D vision-based adhesive contour marking device for shoe soles and uppers according to claim 1, characterized in that, The pressing part (23) includes, from top to bottom, a cylinder (231) vertically fixed to the fixed bracket (1), a bushing (232) fixed to the piston rod of the cylinder (231), and a telescopic pressure rod (233) vertically arranged and rotatably connected to the bushing (232).

3. The 3D vision-based adhesive contour marking device for shoe soles and uppers according to claim 2, characterized in that, The telescopic pressure rod (233) is coaxially arranged with the rotation axis of the rotating part (21).

4. The 3D vision-based adhesive contour marking device for shoe soles and uppers according to claim 3, characterized in that, The marking section (43) includes a pen holder (432) connected to the second moving component (42), a marking pen (431) disposed at the end of the pen holder (432), and a spring located inside the pen holder (432) and connected to the marking pen (431).

5. The 3D vision-based adhesive contour marking device for shoe soles and uppers according to claim 1, characterized in that, The first moving component (41) includes a first guide rod (411) that is vertically arranged and fixed to the fixed bracket (1), a first lead screw (412) that is vertically arranged and rotatably connected to the fixed bracket (1), a first drive motor (413) that is fixed to the fixed bracket (1) and fixedly connected to one end of the first lead screw (412), and a first slide (414) that is slidably arranged on the first guide rod (411) and threadedly connected to the first lead screw (412); The second moving component (42) includes a second guide rod (421) arranged along the direction close to or away from the support (22) and fixed to the first slide (414), a second lead screw (422) arranged parallel to the second guide rod (421) and rotatably connected to the first slide (414), a second drive motor (423) fixed to the first slide (414) and fixedly connected to one end of the second lead screw (422), and a second slide (424) slidably arranged on the second guide rod (421) and threadedly connected to the second lead screw (422), the second slide (424) being connected to the scribing part (43).

6. The 3D vision-based adhesive contour marking device for shoe soles and uppers according to claim 1, characterized in that, Two 3D vision sensors (31) are provided and are symmetrically arranged on both sides of the pressing module (2).

7. The 3D vision-based adhesive contour marking device for shoe soles and uppers according to claim 6, characterized in that, The control system collects 3D visual data from two 3D vision sensors (31), calibrates the relationship between the two sets of 3D visual data and places them in the same coordinate system, and then reconstructs the complete 3D visual data through image stitching and fusion.

8. The 3D vision-based adhesive contour marking device for shoe soles and uppers according to claim 1, characterized in that, The support part (22) is made of rubber, and the top surface of the support part (22) is arc-shaped.

9. A method for marking the outline of adhesive on shoe soles and uppers based on 3D vision, using the marking device according to any one of claims 1-8, characterized in that, Includes the following: Place the sole and upper to be bonded on the support (22), and the control system controls the pressing part (23) to press vertically downwards, so that the adhesive parts of the sole and upper are tightly bonded; The 3D vision sensor (31) performs 3D vision image reconstruction on the part of the shoe sole and shoe upper after pressing, and the control system calculates the adhesive outline based on the 3D vision image. The control system calculates the rotation angle of the rotating part driving the support part (22), the vertical movement distance of the first moving component (41) driving the scribing part (43), and the movement distance of the second moving component (42) driving the scribing part (43) along the direction closer to or further away from the support part (22) based on the adhesive contour line. Then, based on the calculation results, the control system controls the rotating part, the first moving component (41), and the second moving component (42) to work, so that the scribing part (43) moves to the adhesive contour line to scribing. The control system controls the scribing module (4) to work according to the adhesive outline according to the following steps: S1: Take the intersection of the driving direction of the first moving component (41) and the driving direction of the second moving component (42) as the origin O, the driving direction of the second moving component (42) as the X-axis, and the driving direction of the first moving component (41) as the Y-axis, establish a right-hand rectangular coordinate system, denoted as M1, and the coordinate system of the 3D vision sensor (31), denoted as M2. Use a calibration plate to calibrate the rotation matrix from the M2 coordinate system to the M1 coordinate system, which is RT0. S2: After the 3D image reconstruction is completed, multiply each point in the adhesive contour point cloud dataset Q by RT0, rotate it to the M1 coordinate system, and the result after rotation is Q1; denote point P (P x ,P y ,P z Let P be any point on Q1, and let P's projected coordinates in the XOZ plane be P. ， (P) x ,P y ,0),passing through point P ， A line perpendicular to the X-axis intersects the X-axis at point P. ，， (P) x (,0,0), vector OP ， With vector OP ，， With an included angle of θ1, the control system controls the rotating part (21) to reverse the support part (22) by θ1 degrees and rotate point P into the XOY plane; S3: Calculate the rotation matrix of the contour point cloud about the Y-axis by θ1: RT1=( Multiplying point P by RT1 gives the coordinates of point P rotated to point P1 in the XOY plane. x1 ,P y1 ,0), the control system controls the second moving component (42) to move the scribing part (43) along the X-axis P x1 The control system controls the first moving component (41) to move the scribing part (43) along the Y-axis. y1 And draw lines; S4: Repeat steps S2 and S3 until every point in the adhesive contour point cloud dataset Q has been calculated and drawn, and the contour point cloud dataset is Q in the nth calculation. n =Q n-1 *RT n .

Images