CN117994272B - Point cloud segmentation method and system applied to industrial disordered soft packet stack - Google Patents
Point cloud segmentation method and system applied to industrial disordered soft packet stack Download PDFInfo
- Publication number
- CN117994272B CN117994272B CN202410406582.7A CN202410406582A CN117994272B CN 117994272 B CN117994272 B CN 117994272B CN 202410406582 A CN202410406582 A CN 202410406582A CN 117994272 B CN117994272 B CN 117994272B
- Authority
- CN
- China
- Prior art keywords
- point
- point cloud
- normal
- segmentation
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 230000011218 segmentation Effects 0.000 title claims abstract description 58
- 230000012010 growth Effects 0.000 claims abstract description 26
- 238000004458 analytical method Methods 0.000 claims abstract description 7
- 239000013598 vector Substances 0.000 claims description 33
- 230000008859 change Effects 0.000 claims description 17
- 238000004422 calculation algorithm Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 238000010845 search algorithm Methods 0.000 claims description 3
- 238000000638 solvent extraction Methods 0.000 claims 4
- 230000008569 process Effects 0.000 description 10
- 238000004590 computer program Methods 0.000 description 6
- 238000004891 communication Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000003860 storage Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 239000000463 material Substances 0.000 description 3
- 230000002093 peripheral effect Effects 0.000 description 3
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000004568 cement Substances 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003631 expected effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000035040 seed growth Effects 0.000 description 1
Landscapes
- Image Generation (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a point cloud segmentation method and a system applied to industrial disordered soft packet stacks, wherein the method comprises the following steps: s1, extracting a concave region of a point cloud of a target soft package by adopting a method combining super voxel segmentation and local geometric feature analysis; s2, performing curve fitting on the extracted concave area; s3, selecting a minimum curvature point of the non-concave area as an initial seed point; s4, the initial seed points perform region growth based on normal constraint and curve constraint fitted by the nearest concave region, so that accurate point cloud segmentation of the target soft package is achieved. Compared with the prior art, the method has the advantages that the segmentation accuracy is greatly improved, and the under-segmentation rate is reduced. The method can not only easily distinguish the target soft packages of different layers, but also well divide the target soft packages which are tightly attached, the dividing accuracy and stability of the method are far better than those of the prior method, and simultaneously the problem of underdividing on the adjacent soft package division can be well solved.
Description
Technical Field
The invention belongs to the technical field of point cloud segmentation, and particularly relates to a point cloud segmentation method and system applied to industrial unordered soft package stack placement.
Background
At present, in industrial production, soft pack stacking is a common scene. At present, for the material package unstacking of a soft package stack such as a cement soft package stack, the material package unstacking is mainly performed manually or the material package stack is orderly stacked, so that a robot walks a fixed track to unstack, and the real-time performance is not realized. The unstacking of the soft package stacks in a disordered manner is difficult to obtain the coordinates of the soft packages, namely the soft package stacks in a disordered manner cannot be segmented, so that the specific coordinates of each soft package are obtained, and the unstacking is a main difficulty in current research. When the robot is used for unstacking, three-dimensional coordinates of the soft package are required to be obtained, a three-dimensional point cloud technology is required to be adopted for processing, when the soft package stack is processed through the three-dimensional point cloud technology, due to the characteristic that soft packages are easy to deform, the soft package is tightly attached, so that the limit is unclear, and when the point cloud segmentation is carried out on the adjacent soft package stack through the three-dimensional point cloud technology, the problem of serious undersection exists.
How to solve the problem of under-segmentation in the process of neighbor soft packet point cloud segmentation is a problem which needs to be solved in the prior art.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, provides a point cloud segmentation method and a point cloud segmentation system applied to industrial disordered soft packet pile placement, and increases the limiting conditions based on normal limiting conditions of an area growth algorithm to enhance the limit on seed growth in the area growth process so as to realize accurate segmentation of neighbor soft packet point clouds, thereby overcoming the problem of under segmentation in the process of neighbor soft packet point cloud segmentation.
In order to achieve the expected effect, the invention adopts the following technical scheme:
The invention discloses a point cloud segmentation method applied to industrial disordered soft package stacks, which comprises the following steps:
s1, extracting a concave region of a point cloud of a target soft package by adopting a method combining super voxel segmentation and local geometric feature analysis;
S2, performing curve fitting on the extracted concave area;
s3, selecting a minimum curvature point of the non-concave area as an initial seed point;
S4, the initial seed points perform region growth based on normal constraint and curve constraint fitted by the nearest concave region, so that accurate point cloud segmentation of the target soft package is achieved.
Further, the step S1 includes:
S1.1, aiming at point clouds of a target soft package, subdividing original point cloud data into a plurality of small block areas by applying an ultra-voxel segmentation algorithm;
S1.2, screening out the point with the largest high curvature value in each small area through calculation of the curvature of the point cloud surface;
S1.3, judging the convexity of the small area where the high curvature value is located through the normal angle of the point with the maximum curvature value;
And S1.4, when the normal included angle is larger than a first threshold value, the small area where the point with the largest high curvature value is located is a concave area.
Further, the method further comprises the following steps: s1.5, further judging the convexity and convexity of the small block area through the normal change speed of the point cloud.
Further, when the normal line change speed of the point cloud in the small block area exceeds the second threshold value, the small block area is considered to be a concave area.
Further, the S1.3 specifically includes:
s1.3.1, obtaining a point P1 in the point cloud set P and a normal vector V1 thereof;
S1.3.2, searching a neighborhood point of a point P1 through a KD tree search algorithm, obtaining a coordinate index of a nearest preset point through neighbor searching, and generating a first space plane by utilizing the preset point;
s1.3.3, projecting the first space plane by using a point P1 to obtain a projection point P2, wherein the point P2 and the point P1 form a detection vector V2;
And S1.3.4, calculating an included angle between V1 and V2, and when the included angle is smaller than a third threshold value, locating the point P1 in the concave area.
Further, the step S2 includes: and fitting the extracted concave area by adopting an optimized B spline curve.
Further, the fitting the extracted concave region by using the optimized B-spline curve includes:
selecting key points representing the main shape of the concave area as control points;
according to the distribution characteristics of the point cloud, adjusting the node vector to optimize the smoothness and fitting precision of the fitted curve;
And (3) distributing different weights to the control points, so that the fitted curve accords with the real concave area as much as possible.
Further, the normal constraint includes: the maximum normal variation tolerance inside the recessed area is reflected by setting a normal angle threshold.
Further, the curve constraint fitted by the nearest recessed region includes: the nearest recessed area is determined by calculating the shortest distance of the initial seed point to the fitted curve.
The invention also discloses a point cloud segmentation system applied to the industrial disordered soft package stack, which comprises the following steps:
the acquisition module is used for acquiring point cloud data of the target soft package;
And the point cloud segmentation module is used for carrying out point cloud segmentation according to the method.
Compared with the prior art, the invention has the beneficial effects that: the invention provides a point cloud segmentation method and a point cloud segmentation system applied to an industrial disordered soft package stack, which aim at improving a point cloud region growth algorithm. The method and the device realize accurate point cloud segmentation of the soft package objects which are randomly placed in the industrial soft package stack based on the three-dimensional point cloud technology, and overcome the phenomenon of undersegmentation.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings described below are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a point cloud segmentation method applied to an industrial unordered soft packet stack according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of calculation of curvature of a point cloud surface according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of screening out a point with a maximum high curvature value in each small area according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of preliminary judgment of convexity through normal angle according to an embodiment of the present invention.
Fig. 5 is a schematic diagram of constraint conditions of a region growing algorithm according to an embodiment of the present invention, where (a) in fig. 5 is a normal constraint, and (b) in fig. 5 is a curve constraint fitted to a nearest concave region.
Fig. 6 is a schematic view of region growth provided in an embodiment of the present invention, where (a) in fig. 6 is a schematic view of an initial seed point selected, (b) in fig. 6 is a schematic view of a process of region growth of the initial seed point, (c) in fig. 6 is a schematic view of a nearest concave curve calculated during the initial seed point growth process, and (d) in fig. 6 is a schematic view of completion of the initial seed point growth.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1 to 6, the invention discloses a point cloud segmentation method applied to industrial disordered soft packet stacks, which comprises the following steps:
s1, extracting a concave region of a point cloud of a target soft package by adopting a method combining super voxel segmentation and local geometric feature analysis;
specifically, the method and the device can efficiently and accurately extract the concave area in the target soft package point cloud through the curvature and the normal line information of the target soft package point cloud.
In one embodiment, the S1 includes:
s1.1, aiming at point clouds of a target soft package, subdividing original point cloud data into a plurality of small block areas by applying an ultra-voxel segmentation algorithm; specifically, each patch region represents a set of points having similar geometric characteristics.
S1.2, screening out the point with the largest high curvature value in each small area through calculation of the curvature of the point cloud surface;
Specifically, as shown in FIG. 2, for a set of points in a point cloud The neighborhood point set is;/>Is the i-th neighborhood point; n is the normal vector of any point P in the point set P,/>Is/>Normal vector sets of (2). As shown in FIG. 3, the local orthogonal coordinate system L { p, X, Y, Z },/>And/>Is an orthogonal unit vector,/>Is the vector N and/>An angle therebetween; beta is the vector N and/>An angle therebetween. The surface curvature of p can be represented by a closed circle of p, namely:
;
S1.3, judging the convexity of the small area where the high curvature value is located through the normal angle of the point with the maximum curvature value;
In one embodiment, the step S1.3 specifically includes:
s1.3.1, obtaining a point P1 in the point cloud set P and a normal vector V1 thereof;
S1.3.2, searching a neighborhood point of a point P1 through a KD tree search algorithm, obtaining a coordinate index of a nearest preset point through neighbor searching, and generating a first space plane by utilizing the preset point;
s1.3.3, projecting the first space plane by using a point P1 to obtain a projection point P2, wherein the point P2 and the point P1 form a detection vector V2;
And S1.3.4, calculating an included angle between V1 and V2, and when the included angle is smaller than a third threshold value, locating the point P1 in the concave area.
And S1.4, when the normal included angle is larger than a first threshold value, the small area where the point with the largest high curvature value is located is a concave area.
Specifically, a schematic diagram of preliminary judgment of convexity by normal angle is shown in fig. 4.
In one embodiment, further comprising: s1.5, further judging the convexity and convexity of the small block area through the normal change speed of the point cloud.
In one embodiment, a patch region is considered to be a recessed region when the point cloud normal change speed in the patch region exceeds a second threshold.
It is worth noting that in the industrial unordered soft package stacking scene, the soft package attaching is relatively compact, so that the normal included angle is not obvious, and the change degree of the normal vector of the point cloud midpoint in the neighborhood of the point cloud is measured through the normal change speed besides the initial judgment of the normal angle.
In particular, for pointsAnd its normal vector is/>The normal change speed can be calculated/>Normal vector of (2) and its neighborhood point/>Is estimated by the average angle difference between the normal vectors of (a), the normal change speed calculation formula can be expressed as:
;
combining the normal angle and the normal change speed delta Point/>Can be judged by the following rule: if the normal angle exceeds 90 degrees and delta/>Beyond the normal change threshold TΔN, then point/>Is considered as a concave point. Conversely, if none of these conditions is met, then the point/>Possibly in a convex or relatively flat region.
S2, performing curve fitting on the extracted concave area;
Further, the step S2 includes: and fitting the extracted concave area by adopting an optimized B spline curve. The curve fitting of the concave area of the target soft package is complex, and in order to enable the fitted curve to effectively represent the shape of the concave area, the optimized B-spline curve is adopted for fitting during the curve fitting.
Further, the fitting the extracted concave region by using the optimized B-spline curve includes:
Selecting key points representing the main shape of the concave area as control points by analyzing the point cloud data;
according to the distribution characteristics of the point cloud, adjusting the node vector to optimize the smoothness and fitting precision of the fitted curve;
To counteract the effects of noise, the control points are assigned different weights so that the fitted curve corresponds as far as possible to the true pit area.
Specifically, the method adopts the B spline curve fitting method to perform curve fitting on the extracted concave area to obtain peripheral limiting conditions, and the B spline curve can effectively fit the three-dimensional curve. The B-spline curve is a curve defined by a sequence of control points, and the mathematical expression of the p-th order B-spline curve is:
;
Wherein C (u) is a point on the curve, Is a control point,/>And p (u) is a p-th order B-spline basis function defined based on the node vector, u being a parameterized variable.
S3, selecting a minimum curvature point of the non-concave area as an initial seed point;
specifically, selecting the initial seed point from the smoothest region (non-recessed region) can improve the segmentation efficiency. Therefore, a method using the minimum curvature priority is studied to calculate the curvature of each non-concave region point, and a point having the minimum curvature is selected as an initial seed point. The calculation of the local curvature K can be performed by the following formula:
;
In this formula: v represents a point in the point cloud. Δs (v) represents the change in normal vector of the surface at point v, which can be approximated by the difference in normal vector between adjacent points. Deltav is the vector difference between point v and its neighbors. I deltav i is the right. Is the vector delta the square of the modulus of v, representing the square of the distance between the point v and its neighboring points.
By calculating the local curvature and setting a curvature threshold value, the point with the local curvature lower than the threshold value can be screened out from the point cloud to be used as an initial seed point. This approach helps to ensure that the seed point is located inside the recessed area, rather than on the edge or noise point.
S4, the initial seed points perform region growth based on normal constraint and curve constraint fitted by the nearest concave region, so that accurate point cloud segmentation of the target soft package is achieved.
In one embodiment, the normal constraints include: the maximum normal variation tolerance inside the recessed area is reflected by setting a normal angle threshold.
In another embodiment, the curve constraint fitted to the nearest recessed region comprises: the nearest recessed area is determined by calculating the shortest distance of the initial seed point to the fitted curve.
Specifically, the initial seed point is subjected to region growth under the constraint condition of the peripheral concave region, the growth of the initial seed point ensures the applicability of the segmentation result through region growth, meanwhile, the growth region also ensures the robustness of the segmentation result to the adjacent objects through calculating the distance between the seed and each concave region and then determining the nearest concave region, and the growth region of the initial seed point is limited by the nearest concave region.
In fig. 5 (a) is the normal constraint of the region growing algorithm, the non-recessed region point p is the initial seed point,Is the nearest neighbor point searched by the KD tree algorithm; x is the normal vector of p; y is the normal vector of q; θ is the normal vector angle between X and Y; θn is the normal vector angle threshold. The normal constraint of the region growing algorithm mainly reflects the maximum normal vector variation tolerance in the concave region by setting a normal vector included angle threshold value so as to perform region growing on the initial seed point. By ensuring that the seed point growth is in a normal vector direction similar to the recessed region, erroneous crossing of object boundaries can be effectively avoided.
In addition to the judgment of the normal angle threshold, the final segmentation of each point needs to be determined by the nearest concave region. In fig. 5 (B), for the nearest peripheral recess constraint, the distance from the initial seed point to each fitted curve is calculated, and the shortest distance from the initial seed point to the B-spline curve: the shortest distance L between the initial seed point s and the B-spline curve is calculated by minimizing the distance of the point to each point on the curve. This distance can be expressed as:
;
Wherein: u is a parameter of the B-spline curve, and x (u), y (u), and z (u) are points on the curve defined by the parameter u. 、/>、Is the initial seed point s coordinate. Distance set/>Including the distance from the seed point to each concave region point cloud fitted curve. Minimum in L/>A curve (RA) was fitted to the nearest recessed area.
It is noted that each determination of the growth region requires a complete segmentation of the current region by normal constraints and nearest recessed region constraints, see fig. 6 for a specific region growth process. In fig. 6, (a) is the selected initial seed point, in fig. 6, (b) is the process of initial seed point region growth, in fig. 6, (c) is the process of initial seed point growth, the latest concave curve is calculated, region growth is performed under the limitation of the latest concave curve, and in fig. 6, (d) is the process of initial seed point growth, namely, the point cloud segmentation process is completed.
Table 1 is the comparison data of the segmentation rate of the point cloud segmentation result, and compared with the prior art, the method has the advantages that the segmentation accuracy is greatly improved, and the under-segmentation rate is reduced. The method can not only easily distinguish the target soft packages of different layers, but also well divide the target soft packages which are tightly attached, the dividing accuracy and stability of the method are far better than those of the prior method, and simultaneously the problem of underdividing on the adjacent soft package division can be well solved.
TABLE 1
The invention also discloses a point cloud segmentation system applied to the industrial disordered soft package stack, which comprises the following steps:
the acquisition module is used for acquiring point cloud data of the target soft package;
And the point cloud segmentation module is used for carrying out point cloud segmentation according to the method.
The system embodiments may be implemented in one-to-one correspondence with the foregoing method embodiments, and are not described herein.
Based on the same thought, the invention also discloses electronic equipment, which can comprise: the device comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are in communication with each other through the communication bus. The processor may call logic instructions in the memory to perform a point cloud segmentation method for an industrial unordered placement of soft packet stacks, comprising:
s1, extracting a concave region of a point cloud of a target soft package by adopting a method combining super voxel segmentation and local geometric feature analysis;
S2, performing curve fitting on the extracted concave area;
s3, selecting a minimum curvature point of the non-concave area as an initial seed point;
S4, the initial seed points perform region growth based on normal constraint and curve constraint fitted by the nearest concave region, so that accurate point cloud segmentation of the target soft package is achieved.
Further, the logic instructions in the memory described above may be implemented in the form of software functional units and stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, an embodiment of the present invention further provides a computer program product, where the computer program product includes a computer program stored on a non-transitory computer readable storage medium, where the computer program includes program instructions, when the program instructions are executed by a computer, enable the computer to execute a point cloud segmentation method applied to an industrial unordered soft packet stack provided in the foregoing method embodiments, where the method includes:
s1, extracting a concave region of a point cloud of a target soft package by adopting a method combining super voxel segmentation and local geometric feature analysis;
S2, performing curve fitting on the extracted concave area;
s3, selecting a minimum curvature point of the non-concave area as an initial seed point;
S4, the initial seed points perform region growth based on normal constraint and curve constraint fitted by the nearest concave region, so that accurate point cloud segmentation of the target soft package is achieved.
In still another aspect, an embodiment of the present invention further provides a non-transitory computer readable storage medium, on which a computer program is stored, where the computer program is implemented when executed by a processor to perform a point cloud segmentation method applied to an industrial unordered placement soft packet stack provided in the foregoing embodiments, where the method includes:
s1, extracting a concave region of a point cloud of a target soft package by adopting a method combining super voxel segmentation and local geometric feature analysis;
S2, performing curve fitting on the extracted concave area;
s3, selecting a minimum curvature point of the non-concave area as an initial seed point;
S4, the initial seed points perform region growth based on normal constraint and curve constraint fitted by the nearest concave region, so that accurate point cloud segmentation of the target soft package is achieved.
It should be understood that, although the steps in the flowcharts of the figures are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited in order and may be performed in other orders, unless explicitly stated herein. Moreover, at least some of the steps in the flowcharts of the figures may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, the order of their execution not necessarily being sequential, but may be performed in turn or alternately with other steps or at least a portion of the other steps or stages.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (7)
1. The point cloud segmentation method applied to the industrial disordered soft package stack is characterized by comprising the following steps of:
s1, extracting a concave region of a point cloud of a target soft package by adopting a method combining super voxel segmentation and local geometric feature analysis;
The S1 comprises the following steps:
S1.1, aiming at point clouds of a target soft package, subdividing original point cloud data into a plurality of small block areas by applying an ultra-voxel segmentation algorithm;
S1.2, screening out the point with the largest high curvature value in each small area through calculation of the curvature of the point cloud surface;
S1.3, judging the convexity of the small area where the high curvature value is located through the normal angle of the point with the maximum curvature value;
S1.4, when the normal included angle is larger than a first threshold value, the small area where the point with the largest high curvature value is located is a concave area;
S1.5, further judging the convexity and convexity of the small block area through the normal change speed of the point cloud; when the change speed of the point cloud normal line in the small block area exceeds a second threshold value, the small block area is considered to be a concave area; the method specifically comprises the following steps:
For points And its normal vector is/>The normal change speed is calculated/>Normal vector of (2) and its neighborhood pointIs estimated by the average angle difference between the normal vectors of (a), the normal change speed is calculated by the following formula:
;
combining the normal angle and the normal change speed delta Point/>Is judged by the following rule: if the normal angle exceeds 90 degrees and delta/>Beyond the normal change threshold TΔN, then point/>Is considered as a concave point; conversely, if none of these conditions is met, then the point/>In convex or relatively flat areas;
S2, performing curve fitting on the extracted concave area;
s3, selecting a minimum curvature point of the non-concave area as an initial seed point;
S4, the initial seed points perform region growth based on normal constraint and curve constraint fitted by the nearest concave region, so that accurate point cloud segmentation of the target soft package is realized; the curve constraint fitted by the nearest recessed region includes: the nearest recessed area is determined by calculating the shortest distance of the initial seed point to the fitted curve.
2. The method for partitioning point cloud applied to industrial disordered soft packet stacks according to claim 1, wherein the step S1.3 specifically comprises:
s1.3.1, obtaining a point P1 in the point cloud set P and a normal vector V1 thereof;
S1.3.2, searching a neighborhood point of a point P1 through a KD tree search algorithm, obtaining a coordinate index of a nearest preset point through neighbor searching, and generating a first space plane by utilizing the preset point;
s1.3.3, projecting the first space plane by using a point P1 to obtain a projection point P2, wherein the point P2 and the point P1 form a detection vector V2;
And S1.3.4, calculating an included angle between V1 and V2, and when the included angle is smaller than a third threshold value, locating the point P1 in the concave area.
3. The method for partitioning the point cloud applied to the industrial disordered soft packet stack according to claim 1, wherein the step S2 comprises: and fitting the extracted concave area by adopting an optimized B spline curve.
4. The method for partitioning point cloud for industrial disordered soft packet stacks according to claim 3, wherein said fitting said extracted recessed area with an optimized B-spline curve comprises:
selecting key points representing the main shape of the concave area as control points;
according to the distribution characteristics of the point cloud, adjusting the node vector to optimize the smoothness and fitting precision of the fitted curve;
And (3) distributing different weights to the control points, so that the fitted curve accords with the real concave area as much as possible.
5. The method for partitioning a point cloud for industrial disordered soft packet stacks according to claim 1, wherein said normal constraints include: the maximum normal variation tolerance inside the recessed area is reflected by setting a normal angle threshold.
6. The method of claim 1, wherein the curve constraint fitted by the nearest concave region comprises: the nearest recessed area is determined by calculating the shortest distance of the initial seed point to the fitted curve.
7. Point cloud segmentation system applied to industrial disordered soft package stacks is characterized by comprising:
the acquisition module is used for acquiring point cloud data of the target soft package;
the point cloud segmentation module is configured to perform point cloud segmentation according to the method of any one of claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410406582.7A CN117994272B (en) | 2024-04-07 | 2024-04-07 | Point cloud segmentation method and system applied to industrial disordered soft packet stack |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410406582.7A CN117994272B (en) | 2024-04-07 | 2024-04-07 | Point cloud segmentation method and system applied to industrial disordered soft packet stack |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117994272A CN117994272A (en) | 2024-05-07 |
CN117994272B true CN117994272B (en) | 2024-06-18 |
Family
ID=90901090
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410406582.7A Active CN117994272B (en) | 2024-04-07 | 2024-04-07 | Point cloud segmentation method and system applied to industrial disordered soft packet stack |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117994272B (en) |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6984211B2 (en) * | 2003-01-03 | 2006-01-10 | Mayo Foundation For Medical Education And Research | Detection of tumor halos in ultrasound images |
DE102009006416B3 (en) * | 2009-01-28 | 2010-08-26 | Siemens Aktiengesellschaft | Method and branch determination device for determining a branch point and a center line within a hollow organ together with associated computer program |
CN106600622B (en) * | 2016-12-06 | 2019-11-08 | 西安电子科技大学 | A kind of point cloud data segmentation method based on super voxel |
CN108520525B (en) * | 2018-04-12 | 2021-11-02 | 重庆理工大学 | Spinal cord segmentation method based on convex constraint seed region growth |
CN109816664B (en) * | 2018-12-25 | 2023-05-09 | 西安中科天塔科技股份有限公司 | Three-dimensional point cloud segmentation method and device |
CN111091062B (en) * | 2019-11-21 | 2023-07-04 | 东南大学 | Robot out-of-order target sorting method based on 3D visual clustering and matching |
CN113012161B (en) * | 2021-03-05 | 2023-04-28 | 华南理工大学 | Stacked scattered target point cloud segmentation method based on convex region growth |
-
2024
- 2024-04-07 CN CN202410406582.7A patent/CN117994272B/en active Active
Non-Patent Citations (2)
Title |
---|
"3D Mesh Segmentation by Region Growing based on Discrete Curvature";Mohamed Yaghmorasan Benzian etc;《2020 1st International Conference on Communications, Control Systems and Signal Processing (CCSSP)》;20200517;全文 * |
融合3D视觉的机器人自主更换变压器熔断器***;赵云涛;唐千;王晓飞;曹念;王龙祥;;自动化与仪表;20200325(第03期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN117994272A (en) | 2024-05-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109541997B (en) | Spraying robot rapid intelligent programming method for plane/approximate plane workpiece | |
CN111932688B (en) | Indoor plane element extraction method, system and equipment based on three-dimensional point cloud | |
CN107123164B (en) | Three-dimensional reconstruction method and system for keeping sharp features | |
CN107369161A (en) | A kind of workpiece point cloud segmentation method at random based on the European cluster of improvement | |
CN114170279A (en) | Point cloud registration method based on laser scanning | |
CN114943736B (en) | Production quality detection method and system for automobile radiating fins | |
CN114972377B (en) | 3D point cloud segmentation method and device based on mobile least square method and super-voxel | |
CN113012161B (en) | Stacked scattered target point cloud segmentation method based on convex region growth | |
CN113409332B (en) | Building plane segmentation method based on three-dimensional point cloud | |
CN113628263A (en) | Point cloud registration method based on local curvature and neighbor characteristics thereof | |
CN110766719A (en) | Target tracking method, device and storage medium | |
CN111508073A (en) | Method for extracting roof contour line of three-dimensional building model | |
CN113640826B (en) | Obstacle identification method and system based on 3D laser point cloud | |
CN115205690B (en) | Method and device for extracting street tree in monomer mode based on MLS point cloud data | |
CN111783722B (en) | Lane line extraction method of laser point cloud and electronic equipment | |
CN110910435B (en) | Building point cloud extraction method and device, computer equipment and readable storage medium | |
CN112257721A (en) | Image target region matching method based on Fast ICP | |
CN117994272B (en) | Point cloud segmentation method and system applied to industrial disordered soft packet stack | |
CN114325760A (en) | Road tunnel quality inspection obstacle avoidance method and system based on multi-line laser radar | |
CN109815763A (en) | Detection method, device and the storage medium of two dimensional code | |
CN107122782B (en) | Balanced semi-dense stereo matching method | |
CN111179327B (en) | Depth map calculation method | |
CN116469092A (en) | Self-adaptive Alpha Shapes contour extraction method based on DBSCAN algorithm | |
CN113486741B (en) | Stock yard stock pile point cloud step identification method | |
CN111815637A (en) | Method and device for extracting 3D image edge point cloud and computer storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |