Sole glue spraying track generation method and system based on 3D contour

Abstract

The invention discloses a shoe sole glue spraying track generation method based on a 3D contour. The method comprises the following steps: acquiring point cloud data containing a shoe sole; segmenting the point cloud data into foreground point cloud and background point cloud, wherein the sole point cloud is the foreground point cloud; determining the main direction of the sole point cloud, and adjusting the main direction of the sole point cloud to be the same as the direction of the assembly line; re-sampling the sole point cloud to obtain a re-sampled secondary sole point cloud; de-noising processing is carried out on the secondary sole point cloud; dividing the denoised secondary sole point cloud into a toe cap section point cloud, a middle shoe section point cloud and a heel section point cloud according to the direction from the sole heel to the sole tip; glue spraying point positions are searched, the toe cap section point cloud, the middle shoe section point cloud and the heel section point cloud are searched, and toe cap glue spraying point positions of the toe cap section point cloud, middle shoe section glue spraying point positions of the middle shoe section point cloud and heel section glue spraying point positions of the heel section point cloud are obtained. The shoe sole glue spraying track is intelligently generated, track data are provided for glue spraying of the mechanical arm, manual participation is not needed, and the production efficiency is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of shoe sole glue spraying, and in particular to a method and system for generating a shoe sole glue spraying trajectory based on a 3D contour. Background Art

[0002] Industrial automation equipment has been popularized in the field of shoe production, and the intelligence level of related equipment has gradually improved. In particular, the introduction of machine vision has provided a foundation for the intelligent production of special-shaped products such as shoes. Robotic flexible glue spraying operations guided by machine vision have also appeared in the industry. Among the existing technologies, there are two main ways of robot automated glue spraying:

[0003] The first method is to manually teach the robot and generate the glue spraying trajectory. This method programs the robot to generate the trajectory of glue spraying for different styles and sizes of shoes. This process requires the participation of professionals, which is cumbersome and highly dependent on the experience of construction workers. In addition, due to process reasons, there will be a certain range of deformation in the shoe production process, which will lead to the known trajectory being not accurate enough, which will eventually affect the glue spraying effect.

[0004] The second method is to model the sole based on machine vision 3D imaging technology, and use the sole model data to generate the glue path trajectory of the sole, thereby guiding the robot to spray glue; but the current technology is generally not adaptable, and professionals are required to post-process, manually edit and optimize the 3D model, etc., with a lot of manual participation and a low degree of intelligence. Summary of the invention

[0005] The purpose of the present invention is to overcome the deficiencies of the above-mentioned prior art and to provide a method for generating a shoe sole glue spraying trajectory based on 3D contours.

[0006] A method for generating a shoe sole glue spraying trajectory based on a 3D contour comprises the following steps:

[0007] Obtain point cloud data containing the sole;

[0008] Segmentation processing: segment the obtained point cloud data into foreground point cloud and background point cloud, and the sole point cloud is the foreground point cloud;

[0009] Direction processing: determine the main direction of the sole point cloud and adjust the main direction of the sole point cloud to the same direction as the assembly line;

[0010] Resampling processing: resampling the sole point cloud that has been adjusted in direction to obtain the resampled secondary sole point cloud;

[0011] De-noising: De-noising the secondary sole point cloud;

[0012] Segment processing: the denoised secondary sole point cloud is divided into the toe segment point cloud, the midsole segment point cloud, and the heel segment point cloud according to the direction from the heel to the tip of the sole;

[0013] Search for glue spraying points, respectively search the toe segment point cloud, the mid-shoe segment point cloud and the heel segment point cloud, and obtain the toe glue spraying points of the toe segment point cloud, the mid-shoe segment glue spraying points of the mid-shoe segment point cloud, and the heel segment glue spraying points of the heel segment point cloud;

[0014] Connect the glue spraying points at the toe, the glue spraying points at the middle of the shoe and the glue spraying points at the heel in sequence to form a glue spraying track.

[0015] The segmentation process is processed using a random sampling consensus algorithm.

[0016] As a further improvement, the direction processing is performed using principal component analysis. If the main direction of the sole point cloud is different from the assembly line direction, the sole point cloud is rotated to keep the main direction of the sole point cloud the same as the assembly line direction.

[0017] As a further improvement, the denoising process is specifically as follows:

[0018] Set the search radius SR and the number of adjacent points threshold NP, traverse each point of the secondary sole point cloud, search for points within the radius SR around the current point, and count the number of searched points TP. If TP is less than NP, the current point is a noise point and the noise point needs to be removed. After traversal, the denoised sole point cloud is obtained.

[0019] As a further improvement, the point cloud of the toe segment is searched in the following way:

[0020] There is a shoe tip mid-shoe edge line between the shoe tip section and the shoe mid-shoe section, with the midpoint of the shoe tip mid-shoe edge line as the circle center TC, and the circle center TC divides the shoe tip mid-shoe edge line into a starting ray TL0 on the left and a termination ray TLn on the right. The position with the largest jump along the positive direction of the starting ray TL0 is found as the quasi-glue spraying point TP_0, and the point with a height lower than the quasi-glue spraying point TP_0 and a height difference less than TH is searched in the reverse direction of the starting ray TL0 as the quasi-glue spraying point TP_1, and the quasi-glue spraying point TP_1 is set to be the glue spraying point TP0. The starting ray TL0 rotates counterclockwise around the circle center TC by a preset angle to form a ray TL1, and the search process along the starting ray TL0 is repeated to obtain the glue spraying point TP1, which is rotated in sequence by the preset angle to the termination ray TLn on the right side of the circle center TC to obtain all the glue spraying point positions {TP0, TP1, ...TPn} of the shoe tip section point cloud;

[0021] The jump refers to the height difference between the quasi-spraying point found and the center TC, the positive direction refers to the direction along the starting ray TLO from the center TC, and the reverse direction refers to the direction from the outside of the starting ray TL0 toward the center TC.

[0022] As a further improvement, the heel segment point cloud is searched in the following way:

[0023] There is a heel-shoe middle edge line between the heel section and the mid-shoe section, with the midpoint of the heel-shoe middle edge line as the circle center BC, and the circle center BC divides the heel-shoe middle edge line into a starting ray BL0 on the left and a termination ray BLn on the right. The position with the largest jump change along the positive direction of the starting ray BL0 is found as the quasi-glue spraying point BP_0, and the point with a lower height than BP_0 and a height difference less than BH is searched in the reverse direction of the starting ray BL0 as the quasi-glue spraying point BP_1, and the quasi-glue spraying point BP_1 is set to be the glue spraying point BP0. The starting ray BL0 rotates clockwise around the circle center BC by a preset angle to form a ray BL1, and the search process along the starting ray BL0 is repeated to obtain the glue spraying point BP1, and the preset angle is rotated to the termination ray BLn on the right side of the circle center BC in turn to obtain all the glue spraying points {BP0, BP1, ...BPn} of the heel section point cloud;

[0024] The positive direction of the starting ray BL0 refers to the direction along the starting ray BL0 from the center BC of the circle, and the negative direction of the starting ray BL0 refers to the direction from the outside of the starting ray BL0 toward the center BC of the circle.

[0025] As a further improvement, the mid-shoe point cloud is searched in the following way:

[0026] The shoe tip mid-shoe edge line between the shoe tip section and the shoe mid-shoe section is used as the starting line ML0, the shoe heel mid-shoe edge line is used as the ending line MLn, the midpoint of all point clouds on the starting line ML0 is used as the search starting point MB0, the position with the largest jump change is found to the left along ML0 as the quasi-glue spraying point MPL_0, and the point with a height lower than MPL_0 and a height difference less than MH is searched to the right along ML0 as the quasi-glue spraying point MPL_1, and the quasi-glue spraying point MPL_1 is set to be the glue spraying point MPL0;

[0027] Along ML0 to the right, search for the position with the largest jump as the quasi-glue point MPR_0, and along ML0 to the left, search for a point whose height is lower than MPR_0 and whose height difference is less than MH as the quasi-glue point MPR_1. Set the quasi-glue point MPR_1 to be the glue point MPR0, and translate the starting line ML0 along its own orthogonal direction toward the side of the heel and shoe midline by a preset distance to form line ML1. Repeat the search process along the starting line ML0 to obtain the glue point MPL1 on the left and the glue point MPR1 on the right, and translate them in sequence by the preset distance to the termination line MLn to obtain all the glue points {MPL0, MPL1, ...MPLn}, {MPR0, MPR1, ...MPRn} of the mid-shoe point cloud.

[0028] A shoe sole glue spraying trajectory generation system based on 3D contour, comprising:

[0029] A collection unit, used to collect point cloud data of the sole;

[0030] A segmentation unit, used for segmenting the point cloud data into a foreground point cloud and a background point cloud;

[0031] A rotation unit, used for rotating the foreground point cloud;

[0032] De-noising unit, used to process the noise of the sole point cloud;

[0033] A segmentation unit, used to divide the sole point cloud into a toe segment point cloud, a midsole segment point cloud, and a heel segment point cloud;

[0034] The search unit is used to search for glue spraying points on the toe segment point cloud, the midsole segment point cloud and the heel segment point cloud.

[0035] Compared with the prior art, the present invention has the following beneficial technical effects:

[0036] By adopting a specific search method for glue spraying points, accurate glue spraying points can be obtained, and the glue spraying trajectory of the sole can be intelligently generated, providing trajectory data for the robot to spray glue without manual intervention, thereby improving production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the overall process of the present invention;

[0038] Figure 2 This is a flow chart of the algorithm for searching the glue spraying points in the toe section of the present invention;

[0039] Figure 3 This is a flow chart of the algorithm for searching glue spraying points in the heel segment of the present invention. DETAILED DESCRIPTION

[0040] In order to further understand the features, technical means, specific objectives and functions of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0041] like Figure 1 As shown, a method for generating a shoe sole glue spraying trajectory based on a 3D contour comprises the following steps:

[0042] Obtain point cloud data including the sole, and use a 3D camera to obtain point cloud data of the area where the sole is located, including the sole and environmental data.

[0043] Segmentation processing, the RANSAC random sampling consensus algorithm is used to segment the obtained point cloud data into foreground point cloud and background point cloud, and the sole point cloud is the foreground point cloud, thereby obtaining the segmented sole point cloud and background point cloud, which is convenient for subsequent separate processing of the sole point cloud.

[0044] Direction processing, the PCA principal component analysis method is used to determine the main direction of the sole point cloud, and the main direction of the sole point cloud is adjusted to the same direction as the assembly line. If the main direction of the sole point cloud is different from the direction of the assembly line, the sole point cloud is rotated to ensure that the main direction of the sole point cloud remains the same as the assembly line.

[0045] Resampling processing: resampling the sole point cloud that has been adjusted in direction to obtain the resampled secondary sole point cloud;

[0046] De-noising: De-noising the secondary sole point cloud;

[0047] Segment processing, the denoised secondary sole point cloud is divided into toe segment point cloud, mid-shoe segment point cloud, and heel segment point cloud according to the direction from the heel to the tip of the sole. According to the direction from the heel to the toe of the sole, it is divided into toe segment, mid-shoe segment, and heel segment. The toe segment is roughly the position from the toe to the forefoot of the shoe, the heel segment is the semicircular part of the heel of the sole, and the middle part of the sole except the toe segment and the heel segment is the mid-shoe segment. The toe segment and the heel segment are roughly semicircular, and the mid-shoe segment is roughly linear.

[0048] Search for glue spraying points, respectively search the toe segment point cloud, the mid-shoe segment point cloud and the heel segment point cloud, and obtain the toe glue spraying points of the toe segment point cloud, the mid-shoe segment glue spraying points of the mid-shoe segment point cloud, and the heel segment glue spraying points of the heel segment point cloud;

[0049] The glue spraying points at the toe, the middle and the heel are connected in sequence to form a glue spraying track. Glue is sprayed according to the glue spraying track to achieve effective bonding of the sole, provide trajectory data for the robot to spray glue, without manual intervention, and improve production efficiency.

[0050] The denoising process is specifically as follows:

[0051] Set the search radius SR and the number of adjacent points threshold NP, traverse each point of the secondary sole point cloud, search for points within the radius SR around the current point, and count the number of searched points TP. If TP is less than NP, the current point is a noise point and the noise point needs to be removed. After traversal, the denoised sole point cloud is obtained.

[0052] Search the toe point cloud in the following way:

[0053] like Figure 2As shown in the figure, there is a shoe tip mid-shoe edge line between the shoe tip section and the shoe mid-shoe section, and the midpoint of the shoe tip mid-shoe edge line is taken as the circle center TC. The circle center TC divides the shoe tip mid-shoe edge line into the starting ray TL0 on the left and the ending ray TLn on the right. The position with the largest jump change along the positive direction of the starting ray TL0 is found as the quasi-glue spraying point TP_0, and the point with a height lower than the quasi-glue spraying point TP_0 and a height difference less than TH is searched in the reverse direction of the starting ray TL0 as the quasi-glue spraying point TP_1. The quasi-glue spraying point TP_1 is set to be the glue spraying point TP0, and the starting ray TL0 is reversed around the circle center TC. The clockwise direction rotates a preset angle to form a ray TL1, and the search process along the starting ray TL0 is repeated to obtain the glue spraying point TP1. The preset angle is rotated to the terminating ray TLn on the right side of the center TC in sequence to obtain all the glue spraying points {TP0, TP1, ...TPn} of the toe segment point cloud, where n is a natural number; the jump refers to the height difference between the quasi-glue spraying point found and the center TC. The positive direction refers to the direction along the starting ray TLO from the center TC, and the reverse direction refers to the direction from the outside of the starting ray TL0 toward the center TC. For the preset angle, for example, it can be rotated 3 degrees, 5 degrees, or according to other set angles at a time. The height difference TH is a set constant value. Through the above search, an optimal glue spraying point can be found on a ray. Start the search from the left side of the center TC and rotate to the right side of the center TC to complete the complete search of the toe segment.

[0054] Search the heel segment point cloud in the following way:

[0055] like Figure 3 As shown, there is a heel-shoe middle edge line between the heel section and the mid-shoe section, with the midpoint of the heel-shoe middle edge line as the center BC, and the center BC divides the heel-shoe middle edge line into a starting ray BL0 on the left and an ending ray BLn on the right. The position with the largest jump along the positive direction of the starting ray BL0 is found as the quasi-glue point BP_0, and the point with a height lower than BP_0 and a height difference less than BH is searched in the reverse direction of the starting ray BL0 as the quasi-glue point BP_1. The quasi-glue point BP_1 is set to be the glue point BP0, and the starting ray BL0 rotates clockwise around the center BC by a preset angle to form a ray BL1. The search process along the starting ray BL0 is repeated to obtain the glue point BP1, and the preset angle is rotated to the ending ray BLn on the right side of the center BC in turn to obtain all the glue points {BP0, BP1, ...BPn} of the heel segment point cloud, where n is a natural number. The positive direction of the starting ray BL0 refers to the direction along the starting ray BLO from the center BC, and the negative direction of the starting ray BL0 refers to the direction from the outside of the starting ray BL0 toward the center BC. BH is a preset value.

[0056] The search method for the heel segment and the toe segment is roughly the same, both of which are to search from one side of the center of the circle first, then rotate the preset angle and turn to the other side to complete the overall search. Of course, you can also start the search from the right side of the center of the circle TC or the center of the circle BC, and then rotate to the left.

[0057] Search the point cloud of the middle part of the shoe in the following way:

[0058] The toe shoe mid-edge line between the toe section and the middle section of the shoe is used as the starting line ML0, the heel shoe mid-edge line is used as the ending line MLn, and the midpoint of all point clouds on the starting line ML0 is used as the search starting point MB0. The position with the largest jump is found to the left along ML0 as the quasi-glue spraying point MPL_0, and the point with a height lower than MPL_0 and a height difference less than MH is searched to the right along ML0 as the quasi-glue spraying point MPL_1. The quasi-glue spraying point MPL_1 is set to be the glue spraying point MPL0, and MH is the preset value.

[0059] Along ML0 to the right, search for the position with the largest jump as the quasi-glue point MPR_0, and along ML0 to the left, search for a point whose height is lower than MPR_0 and whose height difference is less than MH as the quasi-glue point MPR_1. Set the quasi-glue point MPR_1 to be the glue point MPR0, and translate the starting line ML0 along its own orthogonal direction toward the side of the heel and shoe midline by a preset distance to form line ML1. Repeat the search process along the starting line ML0 to obtain the glue point MPL1 on the left and the glue point MPR1 on the right, and translate them in sequence by the preset distance to the termination line MLn to obtain all the glue points {MPL0, MPL1, ...MPLn}, {MPR0, MPR1, ...MPRn} of the mid-shoe point cloud.

[0060] For searching the middle section of the shoe, since the middle section of the shoe is roughly a straight line, the search line needs to be translated longitudinally to achieve an all-round search.

[0061] By searching the toe section, the middle section and the heel section respectively, the glue spraying point can be accurately positioned.

[0062] In addition, a shoe sole glue spraying trajectory generation system based on 3D contours includes:

[0063] A collection unit, used to collect point cloud data of the sole;

[0064] A segmentation unit, used for segmenting the point cloud data into a foreground point cloud and a background point cloud;

[0065] A rotation unit, used for rotating the foreground point cloud;

[0066] De-noising unit, used to process the noise of the sole point cloud;

[0067] A segmentation unit, used to divide the sole point cloud into a toe segment point cloud, a midsole segment point cloud, and a heel segment point cloud;

[0068] The search unit is used to search for glue spraying points on the toe segment point cloud, the midsole segment point cloud and the heel segment point cloud.

[0069] It should be noted that the above description is not a limitation of the present invention. Without departing from the creative concept of the present invention, any obvious replacement is within the protection scope of the present invention.

Claims

1. A method for generating a shoe sole glue spraying trajectory based on 3D contour, characterized in that: The following steps are involved: Obtain point cloud data containing the sole; Segmentation processing: segment the obtained point cloud data into foreground point cloud and background point cloud, and the sole point cloud is the foreground point cloud; Direction processing: determine the main direction of the sole point cloud and adjust the main direction of the sole point cloud to the same direction as the assembly line; Resampling processing: resampling the sole point cloud that has been adjusted in direction to obtain the resampled secondary sole point cloud; De-noising: De-noising the secondary sole point cloud; Segment processing: the denoised secondary sole point cloud is divided into the toe segment point cloud, the midsole segment point cloud, and the heel segment point cloud according to the direction from the heel to the tip of the sole; Search for glue spraying points, respectively search the toe segment point cloud, the mid-shoe segment point cloud and the heel segment point cloud, and obtain the toe glue spraying points of the toe segment point cloud, the mid-shoe segment glue spraying points of the mid-shoe segment point cloud, and the heel segment glue spraying points of the heel segment point cloud; Connect the glue spraying points at the toe, the glue spraying points at the middle of the shoe and the glue spraying points at the heel in sequence to form a glue spraying track.

2. The method for generating a shoe sole glue spraying trajectory based on 3D contour according to claim 1, characterized in that: The segmentation process is processed using a random sampling consensus algorithm.

3. The method for generating a shoe sole glue spraying trajectory based on 3D contour according to claim 1, characterized in that: The direction processing is performed using a principal component analysis method. If the main direction of the sole point cloud is different from the assembly line direction, the sole point cloud is rotated so that the main direction of the sole point cloud remains the same as the assembly line direction.

4. The method for generating a shoe sole glue spraying trajectory based on 3D contour according to claim 1, characterized in that: The denoising process is specifically as follows: Set the search radius SR and the number of adjacent points threshold NP, traverse each point of the secondary sole point cloud, search for points within the radius SR around the current point, and count the number of searched points TP. If TP is less than NP, the current point is a noise point and the noise point needs to be removed. After traversal, the denoised sole point cloud is obtained.

5. The method for generating a shoe sole glue spraying trajectory based on 3D contour according to claim 1, characterized in that: Search the toe point cloud in the following way: There is a shoe tip mid-shoe edge line between the shoe tip section and the shoe mid-shoe section, with the midpoint of the shoe tip mid-shoe edge line as the circle center TC, and the circle center TC divides the shoe tip mid-shoe edge line into a starting ray TL0 on the left and a termination ray TLn on the right. The position with the largest jump along the positive direction of the starting ray TL0 is found as the quasi-glue spraying point TP_0, and the point with a height lower than the quasi-glue spraying point TP_0 and a height difference less than TH is searched in the reverse direction of the starting ray TL0 as the quasi-glue spraying point TP_1, and the quasi-glue spraying point TP_1 is set to be the glue spraying point TP0. The starting ray TL0 rotates counterclockwise around the circle center TC by a preset angle to form a ray TL1, and the search process along the starting ray TL0 is repeated to obtain the glue spraying point TP1, which is rotated in sequence by the preset angle to the termination ray TLn on the right side of the circle center TC to obtain all the glue spraying point positions {TP0, TP1, ...TPn} of the shoe tip section point cloud; The jump refers to the height difference between the quasi-spraying point found and the center TC. The positive direction of the starting ray TL0 refers to the direction along the starting ray TLO from the center TC, and the reverse direction of the starting ray TL0 refers to the direction from the outside of the starting ray TL0 toward the center TC.

6. The method for generating a shoe sole glue spraying trajectory based on 3D contour according to claim 5, characterized in that: Search the heel segment point cloud in the following way: There is a heel-shoe middle edge line between the heel section and the mid-shoe section, with the midpoint of the heel-shoe middle edge line as the circle center BC, and the circle center BC divides the heel-shoe middle edge line into a starting ray BL0 on the left and a termination ray BLn on the right. The position with the largest jump change along the positive direction of the starting ray BL0 is found as the quasi-glue spraying point BP_0, and the point with a lower height than BP_0 and a height difference less than BH is searched in the reverse direction of the starting ray BL0 as the quasi-glue spraying point BP_1, and the quasi-glue spraying point BP_1 is set to be the glue spraying point BP0. The starting ray BL0 rotates clockwise around the circle center BC by a preset angle to form a ray BL1, and the search process along the starting ray BL0 is repeated to obtain the glue spraying point BP1, and the preset angle is rotated to the termination ray BLn on the right side of the circle center BC in turn to obtain all the glue spraying points {BP0, BP1, ...BPn} of the heel section point cloud; The positive direction of the starting ray BL0 refers to the direction along the starting ray BL0 from the center BC of the circle, and the negative direction of the starting ray BL0 refers to the direction from the outside of the starting ray BL0 toward the center BC of the circle.

7. The method for generating a shoe sole glue spraying trajectory based on 3D contour according to claim 6, characterized in that: Search the point cloud of the middle part of the shoe in the following way: The shoe tip mid-shoe edge line between the shoe tip section and the shoe mid-shoe section is used as the starting line ML0, the shoe heel mid-shoe edge line is used as the ending line MLn, the midpoint of all point clouds on the starting line ML0 is used as the search starting point MB0, the position with the largest jump change is found to the left along ML0 as the quasi-glue spraying point MPL_0, and the point with a height lower than MPL_0 and a height difference less than MH is searched to the right along ML0 as the quasi-glue spraying point MPL_1, and the quasi-glue spraying point MPL_1 is set to be the glue spraying point MPL0; Along ML0 to the right, search for the position with the largest jump as the quasi-glue point MPR_0, and along ML0 to the left, search for a point whose height is lower than MPR_0 and whose height difference is less than MH as the quasi-glue point MPR_1. Set the quasi-glue point MPR_1 to be the glue point MPR0, and translate the starting line ML0 along its own orthogonal direction toward the side of the heel and shoe midline by a preset distance to form line ML1. Repeat the search process along the starting line ML0 to obtain the glue point MPL1 on the left and the glue point MPR1 on the right, and translate them in sequence by the preset distance to the termination line MLn to obtain all the glue points {MPL0, MPL1, ...MPLn}, {MPR0, MPR1, ...MPRn} of the mid-shoe point cloud.

8. A shoe sole glue spraying trajectory generation system based on 3D contour, characterized in that: include: A collection unit, used to collect point cloud data of the sole; A segmentation unit, used for segmenting the point cloud data into a foreground point cloud and a background point cloud; A rotation unit, used for rotating the foreground point cloud; De-noising unit, used to process the noise of the sole point cloud; A segmentation unit, used to divide the sole point cloud into a toe segment point cloud, a midsole segment point cloud, and a heel segment point cloud; The search unit is used to search for glue spraying points on the toe segment point cloud, the midsole segment point cloud and the heel segment point cloud.

Images