CN106373118A - A complex curved surface part point cloud reduction method capable of effectively keeping boundary and local features - Google Patents
A complex curved surface part point cloud reduction method capable of effectively keeping boundary and local features Download PDFInfo
- Publication number
- CN106373118A CN106373118A CN201610767783.5A CN201610767783A CN106373118A CN 106373118 A CN106373118 A CN 106373118A CN 201610767783 A CN201610767783 A CN 201610767783A CN 106373118 A CN106373118 A CN 106373118A
- Authority
- CN
- China
- Prior art keywords
- point
- cloud
- curved surface
- complex curved
- point cloud
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/0006—Industrial image inspection using a design-rule based approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
Abstract
The invention belongs to the technical field of precision machining and measurement, and discloses a complex curved surface part point cloud reduction method capable of effectively keeping boundary and local features. The method comprises the following steps: generating a scanned point cloud of a complex curved surface part; obtaining a plurality of neighborhood points for each point in the point cloud and calculating each normal line vector; finding out m points within the shortest radius range with each point being as the centre of sphere, and then, calculating the average value of included angles between the normal line vectors of the points in the point cloud and the normal line vectors of the m points; setting a threshold value based on the average value of included angles, and then, carrying out feature coarse classification; carrying out secondary subdivision to finish selection of a first reduction subset, and then, calculating directional Hausdorff distance to finish selection of a second reduction subset; and finally, merging the two reduction subsets to obtain a reduced scanned point cloud. Compared with the prior art, the method can achieve higher precision and efficiency, and can effectively keep the boundary and local features of the point cloud model.
Description
Technical field
The invention belongs to Precision Machining and field of measuring technique, border drawn game can be effectively retained more particularly, to one kind
The complex curved surface parts point cloud compressing method of portion's feature.
Background technology
With the continuous development in the fields such as computer vision, pattern recognition, non-contact scanning technology is in precision component
More and more important effect is played, especially in the three-dimensional modeling of workpiece, cutter positioning, geometric profile degree in processing and detection
Have a wide range of applications in detection.Based on optical principle, non-contact scanning equipment can obtain number with ten thousand within several seconds
The three-dimensional point data of meter, and to some labyrinth curved surfaces and large-sized object, the cloud data being obtained is very huge,
Accordingly, it is difficult to directly these data are used for calculating process.Point cloud compressing technology is that the subsequent treatment of a cloud provides one kind and has
The solution of effect, during simplifying, on the one hand needs to select point as much as possible to ensure in model representation region
Model after simplifying and archetype have higher similarity；On the other hand need the quantity of point is control effectively, thus
Reach and simplify the purpose calculating.Therefore, simplifying of cloud of point is a large-scale complicated technical problem.
Point cloud compressing method of the prior art adopts traditional stochastical sampling and uniform sampling approach mostly, due to being not required to
Consider the characteristic information of model, the computational efficiency of both approaches is higher, but find both approaches not in practice
Be applied to high accuracy complex curved surface parts point cloud simplifies operation.For example, in the profile tolerance detection process of blade of aviation engine
In, need to calculate profile errors by the blade point cloud after simplifying with designing a model to mate, if adopting stochastical sampling method, often
Different testing results being obtained after secondary measurement, further, since not accounting for feature and boundary point, being obtained using traditional sampling method
To testing result can not react the true mismachining tolerance of blade.Therefore, it is necessary to propose a kind of new to part boundary and
The method for simplifying that local feature is effectively retained.
Additionally, further retrieval finds, it is quick that cn104881498a discloses a kind of out-of-core of massive point cloud
Uniformly compressing method, can be used for simplifying beyond the surface in kind sampled data hosting tolerance limit, but the method essence is using encirclement
Box, to simplify cloud data, can lose the geometric properties of partial dot cloud during simplifying；Cn102800114a discloses one kind
Based on the point cloud data compressing method of poisson-disk sampling, the method pass through sparse augment or remove with close quarters adopt
Sampling point can prevent sampled point localized clusters, but sparse relatively difficult with defining of close quarters in actual applications；
Cn104732581a discloses a kind of mobile context point cloud compressing method based on a feature histogram, and the method calculates often first
The standard deviation of the feature histogram of individual point, and be compared with default standard deviation threshold method, it is more than standard deviation threshold method by deleting
The purpose to reach simplification for the point, but due to a cloud between geometry difference very big, therefore, it is difficult to determine one general
Standard deviation threshold method proposes by a mapping table come self adaptation selection standard threshold value, but this table to set up process more multiple
Miscellaneous, time-consuming；Cn104915986a discloses a kind of solid threedimensional model method for automatic modeling, and the method is for three having built up
Dimension grid model utilizes edge contraction method, deletes point and side on three-dimensional grid model in proportion, thus setting up out the three of object
Dimension simplifies surface model, due to not accounting for local feature and boundary point, the simplified model obtained by this invention and true model
Between there is larger error.
Content of the invention
Disadvantages described above for prior art or Improvement requirement, the invention provides one kind can be effectively retained border and local
The complex curved surface parts point cloud compressing method of feature, wherein passes through the structure with reference to complex curved surface parts itself and its point cloud model
Characteristic, and build and specifically classify and simplify algorithm and processed, accordingly compared with prior art not only possesses high accuracy, efficiently
The features such as rate and versatility are good, and border and the local feature of point cloud model can be effectively retained, it is therefore particularly suitable for example
Point cloud compressing application scenario as the large complicated carved part of blade of aviation engine etc.
For achieving the above object, it is proposed, according to the invention, provide a kind of complicated song being effectively retained border and local feature
Surface parts point cloud compressing method is it is characterised in that the method comprises the following steps:
A () executes scanning to complex curved surface parts, obtain multiple three-dimensional measurement points and generate corresponding scanning to be simplified
Point cloud p, wherein p={ pi| i=1,2 ..., np},piFor represent in scanning element cloud p each point and with the same coordinate system
X, y, z coordinate value representing, npRepresent the total quantity of the point in scanning element cloud p；
B () is directed to each point p in scanning element cloud pi, each its multiple neighborhood point p of sampling acquisitionikAnd generate corresponding neighbour
Domain point set { pi1,pi2,…,pik, wherein k represents the total quantity of neighborhood point, then calculates each in reflection scanning element cloud p
Individual point piLocal feature normal line vector v (pi)；
C () is respectively with each point piFor the centre of sphere, find out m point in the shortest radius of this point, then obtain point pi
Described normal line vector v (pi) and the normal line vector v (p corresponding to this m pointj) between angle thetaij, and this angle is taken absolutely
To being worth angle meansigma methodss and this meansigma methods σpi∈[0,π]；
D () is directed to described angle meansigma methodss predetermined lower threshold value t respectively1With upper limit threshold t2, then according to following equation
(1) to a cloud execution feature rough sort, being derived from three class rough sort subsets is non-feature point set z1, transition point set z2, feature
Point set z3:
E () adopts clustering procedure cluster centre quantitative value k different to three rough sort subset allocation respectively1, k2, k3To enter
Row subdivision, and retain its cluster centre coordinate, thus complete first and simplify subset pfSelection；
F () selects an initial point from scanning element cloud p, calculate the orientation between this initial point and other each points successively
Hausdorff distance, and retain the point meeting position relationship, so far complete second and simplify subset pbSelection；
G () simplifies subset p to by first selected by step (e)fWith by second essence selected by step (f)
Simple subset pbMerge, delete simultaneously and repeat a little, to be derived from the scanning element cloud after required simplifying.
As it is further preferred that in step (b), it is preferred to use following equation (two) is calculating described normal line vector v
(pi):
Wherein,Represent and point piThe central point of corresponding neighborhood point set, and with this neighborhood point set seat a little
Mark meansigma methodss to represent；For representing with all neighborhood point pikWith central pointCoordinate difference collectively as matrix element
Matrix constructed by element, t is used for representing the transposition to this matrix.
As it is further preferred that in step (c), m value is preferably 10.
As it is further preferred that in step (d), described lower threshold t1Value be preferably π/6, described upper limit threshold
t2Value be preferably pi/2.
As it is further preferred that in step (e), described cluster centre quantitative value k1, k2, k3Preferably according to following public affairs
Formula (three) is calculating acquisition:
Wherein, y (x) expression carries out round numbers operation, n to xnewRepresent that expectation executes the target after simplifying to scanning element cloud p
Quantity.
As it is further preferred that in step (f), described initial point is preferably the focus point of scanning element cloud p.
As it is further preferred that in step (f), calculating the orientation between described initial point and other each points
The process of hausdorff distance is preferably according to equation below (four):
Wherein, h (a, b) represents the orientation hausdorff distance between described initial point and other each points；A represents to first
Initial point executes the renewal point set obtaining respectively after multiple bearing hausdorff distance calculates, and b represents scanning element cloud p；a、b
It is respectively the sample point updating in point set and scanning element cloud, and d (a, b) represents the Euclidean distance calculating a, b point-to-point transmission.
As it is further preferred that in step (g), simplifying subset p for described secondbQuantity k4Preferably according to following
Formula (five) is setting:
k4=nnew-(k1+k2+k3) (five)
Wherein, k1、k2And k3Represent respectively to described three rough sort subsets z1、z2And z3The different cluster centres being distributed
Quantitative value, nnewRepresent that expectation executes the destination number after simplifying to scanning element cloud p.
As it is further preferred that described complex curved surface parts are preferably blade of aviation engine.
In general, by the contemplated above technical scheme of the present invention compared with prior art, not only by being directed to
Property simplify to operate to be effectively retained the local feature of three-dimensional point cloud by the thick characteristic point cloud arriving essence is specific, and additionally use band
The orientation hausdorff distance of adaptive weighting characteristic, to be effectively retained the border of three-dimensional point cloud, in this way, on the whole can
Significantly improve precision and the efficiency of complex curved surface parts point cloud compressing, and the versatility of technique is good, in whole process no longer
Need to build polygon operation to a cloud, be therefore particularly suitable for the such as blade of aviation engine noncontact of large complicated carved part
Point cloud compressing application scenario in formula detection.
Brief description
Fig. 1 is the complex curved surface parts point cloud compressing of the be effectively retained border contemplated according to the present invention and local feature
The basic flow sheet of process；
Fig. 2 is the normal vector angle schematic diagram for each point in exemplary display point cloud；
Fig. 3 is the schematic diagram for exemplary display each point rough sort；
Fig. 4 is the original blade point cloud chart for 2500 points of exemplary display；
Fig. 5 is the schematic diagram of the blade point cloud rough sort for exemplary display to 2500 points；
Fig. 6 is the schematic diagram of the blade point cloud essence classification for exemplary display to 2500 points；
Fig. 7 is the schematic diagram calculating for exemplary display orientation hausdorff distance；
Fig. 8 is the schematic diagram of the original blade point cloud for 10075 points of exemplary display；
Fig. 9 is that used time exemplary display carries out 5000 points of essences according to the blade point cloud to 10075 each points that flow process Fig. 1 obtains
The result figure of letter；
Figure 10 is to carry out for exemplary display at 3000 points according to the blade point cloud to 10075 points that flow process Fig. 1 obtains
The result figure simplified..
Specific embodiment
In order that the objects, technical solutions and advantages of the present invention become more apparent, below in conjunction with drawings and Examples, right
The present invention is further elaborated.It should be appreciated that specific embodiment described herein is only in order to explain the present invention, and
It is not used in the restriction present invention.As long as additionally, involved technical characteristic in each embodiment of invention described below
The conflict of not constituting each other just can be mutually combined.
In the prior art, for being sufficiently reserved the raw information of model, the method for some the intuitively aspect of model such as each points
Line and curvature information have a wide range of applications during simplifying.Additionally, for improving sampling precision, some user-defined spies
That levies that index is also utilized for a cloud simplifies offer reference information, but these self-defining characteristic indexs generally require more to count
Evaluation time and bigger memory space.In many methods, compared to characteristic point, the selection to boundary point is often not given to enough
Many considerations, however, being chosen in the profile tolerance detection of complex curved surface parts of boundary point has highly important effect.Separately
Outward, during the simplifying an of cloud, on the one hand need feature and the boundary information selecting point as much as possible to carry out descriptive model；Separately
On the one hand need to replace original model using point as few as possible, simplify, to reach, the purpose calculating.And existing method is big
Only considered one of aspect more, there is different degrees of defect.
Therefore, simplify problem it is contemplated that right in practical application for three-dimensional point cloud especially complex curved surface parts point cloud
Simplify the requirement of precision and efficiency.The present invention starts with from the feature of a cloud and boundary point, by carrying out to a cloud by slightly to essence
Tagsort, thus enough reservation archetype characteristic points, additionally, using orientation hausdorff distance from each coordinate points
Position relationship start with a cloud simplified, the method is particularly effective to the selection of boundary point.Method meter proposed by the present invention
Calculate efficiency high, simplify after model there is good precision and Jian Du, the simplifying of suitable high accuracy complex curved surface parts point cloud
Effect.
Fig. 1 is the basic flow sheet of the complex curved surface parts point cloud compressing method contemplated according to the present invention.With reference to Fig. 1,
We will think to be specifically described as a example certain model blade of aviation engine point cloud compressing below.
First, treat the blade of aviation engine point cloud execution local feature assessment simplified to obtain the normal direction at each point
Amount, for ease of observing, taking the two-dimensional points cloud in Fig. 2 as a example, wherein, arrow represents the normal vector of each point calculated.
Then, for example respectively with the p1 in Fig. 2, p2, p3, p4, p5 find for the center of circle and make apart from their two nearest points
For neighborhood point, as shown in the broken circle in Fig. 2, and calculate the normal vector angle value at neighborhood point and centre of sphere point respectively, to angle
Value takes absolute value, and calculates the average angle value of neighborhood point at the centre of sphere.
Then, two threshold values of setting such as assignment t1=π/6, t2=pi/2, in conjunction with the average angle value of circle centre position, big
In zero and less than or equal to t1Centre point classify as non-characteristic point, more than t1And it is less than or equal to t2Centre point classify as transition
Point, more than t2And the centre point less than or equal to π classifies as characteristic point, repeat above procedure, can obtain non-feature point set z1,
Transition point set z2, feature point set z3, so far complete a rough sort for cloud feature, and result shown in Fig. 3 is and carries out spy to Fig. 2 each point
Levy the result after rough sort, in conjunction with the concrete instance of blade, we choose leaf model such as Fig. 4 institute with 2500 three-dimensional point
Show, can be obtained in Figure 5 to blade point cloud feature rough sort result according to above step.
Then, according to simplifying target, try to achieve and simplify cloud number nnew, according to nnewK- is executed to three rough sort point sets
Means cluster operation, expects to retain point as much as possible to feature point set in cluster process, and non-feature point set is then only needed
Retain a small amount of point, therefore, give 0.1 × n respectively to subset z1, z2, z3new、0.2×nnew、0.3×nnewCluster centre
Quantity, carries out feature essence classification based on above step, selects n here to the blade point cloud shown in Fig. 5new=1500, essence classification knot
Fruit is as shown in fig. 6, so far complete first to simplify subset pfSelection.
Then, calculate the focus point of original point cloud p, and using this point as initial point, calculate reconnaissance and original point successively
Orientation hausdorff distance value between each point in cloud, orients the definition of hausdorff distance, with shown in Fig. 7 for simple and clear description
As a example point cloud, wherein, it is original point cloud m from the point of view of all in Fig. 7, i.e. m=[a1, a2, a3, b1, b2], from the point of view of b1 and b2
Doing is subset n obtaining of sampling from m, i.e. n=[b1, b2], intends now selecting one from m using orientation hausdorff distance
Individual new point, as sampled point, concretely comprises the following steps: calculate b1 to a1, a2, a3 first respectively, distance, and retain minima
D11, then calculates b2 to a1, the distance of a2, a3 respectively, and retains minima d23, finally compare the value of d11 and d23 and retain
Maximum, this it appears that d23 > d11 from this example, therefore point a3 is selected as next sampled point, and that is, son point cloud n is more
It is newly n=[b1, b2, a3], we are simplified to blade point cloud using the method in an embodiment of the present invention, orientation
Hausdorff distance essence is to distribute different weighted values according to the distance between point, additionally, the information of reconnaissance will affect
The selection of new point, this process belongs to region growth method, because the boundary point relative interior point in point cloud has more maximum probability to count
Calculate the ultimate range obtaining putting apart from certain, therefore the method can effectively retain to border point, can obtain based on above step
Simplify subset p to secondb.
Then, merge two having obtained and simplify subset pfAnd pb, and delete and repeat a little.
Finally, determine number t of the point being repeated selection, recalculate t sampled point using orientation hausdorff distance
And be added to completed simplify in a cloud, so execute, until it reaches simplify target number, in conjunction with the above step to such as
The blade point cloud with 10075 points shown in Fig. 8 carries out simplifying of 5000 points and 3000 points, and last simplifies result such as
Shown in Fig. 9 and Figure 10.
To sum up, method proposed by the present invention can complete complex curved surface parts point cloud is simplified in effective time.Should
Method robustness is good, and computational efficiency is high, and the feature to a cloud and border can be effectively retained, thus are applied to complex curved surface parts
Contactless matching detection makees process.Additionally, this reduction techniques can be used for 3d modeling, target recognition, the related neck such as image segmentation
Domain.
As it will be easily appreciated by one skilled in the art that the foregoing is only presently preferred embodiments of the present invention, not in order to
Limit the present invention, all any modification, equivalent and improvement made within the spirit and principles in the present invention etc., all should comprise
Within protection scope of the present invention.
Claims (9)
1. a kind of complex curved surface parts point cloud compressing method being effectively retained border and local feature is it is characterised in that the party
Method comprises the following steps:
A () executes scanning to complex curved surface parts, obtain multiple three-dimensional measurement points and generate corresponding scanning element cloud to be simplified
P, wherein p={ pi| i=1,2 ..., np},piFor represent in scanning element cloud p each point and with the x in the same coordinate system, y,
Z coordinate value representing, npRepresent the total quantity of the point in scanning element cloud p；
B () is directed to each point p in scanning element cloud pi, each its multiple neighborhood point p of sampling acquisitionikAnd generate corresponding neighborhood point
Set { pi1,pi2,…,pik, wherein k represents the total quantity of neighborhood point, then calculates each point in reflection scanning element cloud p
piLocal feature normal line vector v (pi)；
C () is respectively with each point piFor the centre of sphere, find out m point in the shortest radius of this point, then obtain point piInstitute
State normal line vector v (pi) and the normal line vector v (p corresponding to this m pointj) between angle thetaij, and this angle is taken absolute value
Draw angle meansigma methodss and this meansigma methods σpi∈[0,π]；
D () is directed to described angle meansigma methodss predetermined lower threshold value t respectively1With upper limit threshold t2, then right according to following equation ()
Point cloud execution feature rough sort, being derived from three class rough sort subsets is non-feature point set z1, transition point set z2, feature point set z3:
E () adopts clustering procedure cluster centre quantitative value k different to three rough sort subset allocation respectively1, k2, k3To carry out two
Secondary subdivision, and retain its cluster centre coordinate, thus complete first and simplify subset pfSelection；
F () selects an initial point from scanning element cloud p, calculate the orientation between this initial point and other each points successively
Hausdorff distance, and retain the point meeting position relationship, so far complete second and simplify subset pbSelection；
G () simplifies subset p to by first selected by step (e)fSimplify son with by second selected by step (f)
Collection pbMerge, delete simultaneously and repeat a little, to be derived from the scanning element cloud after required simplifying.
2. a kind of complex curved surface parts point cloud compressing side being effectively retained border and local feature as claimed in claim 1
Method is it is characterised in that in step (b), it is preferred to use following equation (two) is calculating described normal line vector v (pi):
Wherein,Represent and point piThe central point of corresponding neighborhood point set, and with this neighborhood point set coordinate a little put down
Average is representing；For representing with all neighborhood point pikWith central pointCoordinate difference collectively as matrix element institute
The matrix building, t is used for representing the transposition to this matrix.
3. a kind of complex curved surface parts point cloud compressing being effectively retained border and local feature as claimed in claim 1 or 2
Method is it is characterised in that in step (c), m value is preferably 10.
4. a kind of complex curved surface parts point being effectively retained border and local feature as described in claim 1-3 any one
Cloud compressing method it is characterised in that in step (d), described lower threshold t1Value be preferably π/6, described upper limit threshold t2's
Value is preferably pi/2.
5. a kind of complex curved surface parts point being effectively retained border and local feature as described in claim 1-4 any one
Cloud compressing method it is characterised in that in step (e), described cluster centre quantitative value k1, k2, k3Preferably according to following equation
(3) calculating acquisition:
Wherein, y (x) expression carries out round numbers operation, n to xnewRepresent that expectation executes the number of targets after simplifying to scanning element cloud p
Amount.
6. a kind of complex curved surface parts point cloud compressing side being effectively retained border and local feature as claimed in claim 5
Method is it is characterised in that in step (f), described initial point is preferably the focus point of scanning element cloud p.
7. a kind of complex curved surface parts point being effectively retained border and local feature as described in claim 1-6 any one
Cloud compressing method is it is characterised in that in step (f), calculate orientation hausdorff between described initial point and other each points
The process of distance is preferably according to equation below (four):
Wherein, h (a, b) represents the orientation hausdorff distance between described initial point and other each points；A represents to initial point
The renewal point set that execution multiple bearing hausdorff distance obtains after calculating respectively, b represents scanning element cloud p；A, b are respectively
For updating the sample point in point set and scanning element cloud, and d (a, b) represents the Euclidean distance calculating a, b point-to-point transmission.
8. a kind of complex curved surface parts point cloud compressing side being effectively retained border and local feature as claimed in claim 5
Method is it is characterised in that in step (g), simplify subset p for described secondbQuantity k4Preferably to set according to following equation (five)
Fixed:
k4=nnew-(k1+k2+k3) (five)
Wherein, k1、k2And k3Represent respectively to described three rough sort subsets z1、z2And z3The different cluster centre quantity distributed
Value, nnewRepresent that expectation executes the destination number after simplifying to scanning element cloud p.
9. a kind of complex curved surface parts point cloud compressing side being effectively retained border and local feature as described in claim 1-8
Method is it is characterised in that described complex curved surface parts are preferably blade of aviation engine.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610767783.5A CN106373118B (en) | 2016-08-30 | 2016-08-30 | The complex curved surface parts point cloud compressing method of border and local feature can be effectively retained |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610767783.5A CN106373118B (en) | 2016-08-30 | 2016-08-30 | The complex curved surface parts point cloud compressing method of border and local feature can be effectively retained |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106373118A true CN106373118A (en) | 2017-02-01 |
| CN106373118B CN106373118B (en) | 2017-09-22 |
Family
ID=57901773
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610767783.5A Active CN106373118B (en) | 2016-08-30 | 2016-08-30 | The complex curved surface parts point cloud compressing method of border and local feature can be effectively retained |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106373118B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107194998A (en) * | 2017-05-23 | 2017-09-22 | 哈尔滨工业大学 | A kind of method of multi-layer three-dimension point cloud single-layered |
| CN108198244A (en) * | 2017-12-20 | 2018-06-22 | 中国农业大学 | A kind of Apple Leaves point cloud compressing method and device |
| CN108759669A (en) * | 2018-05-31 | 2018-11-06 | 武汉中观自动化科技有限公司 | A kind of self-positioning 3-D scanning method and system in interior |
| CN108765434A (en) * | 2018-01-15 | 2018-11-06 | 中国人民解放军陆军装甲兵学院 | The contour extraction method of point cloud model is remanufactured based on increasing material |
| CN109726509A (en) * | 2019-01-21 | 2019-05-07 | 南京航空航天大学 | A kind of part geometry feature representation model and construction method towards aircraft assembly |
| CN110111430A (en) * | 2019-04-11 | 2019-08-09 | 暨南大学 | One kind extracting quadric method from three-dimensional point cloud |
| CN110340738A (en) * | 2019-06-21 | 2019-10-18 | 武汉理工大学 | A kind of robot wire drawing high-speed rail white body workpiece method for precisely marking based on PCA |
| CN111080653A (en) * | 2019-11-06 | 2020-04-28 | 广西大学 | Method for simplifying multi-view point cloud by using region segmentation and grouping random simplification method |
| CN111652855A (en) * | 2020-05-19 | 2020-09-11 | 西安交通大学 | Point cloud simplification method based on survival probability |
| CN112102397A (en) * | 2020-09-10 | 2020-12-18 | 敬科(深圳)机器人科技有限公司 | Method, equipment and system for positioning multilayer part and readable storage medium |
| CN113111548A (en) * | 2021-03-27 | 2021-07-13 | 西北工业大学 | Product three-dimensional feature point extraction method based on cycle angle difference |
| WO2021159838A1 (en) * | 2020-10-12 | 2021-08-19 | 平安科技（深圳）有限公司 | Method and apparatus for simplifying point cloud data, and storage medium and electronic device |
| CN113744389A (en) * | 2021-08-24 | 2021-12-03 | 武汉理工大学 | Point cloud simplification method for complex part curved surface feature retention |
| CN114485664A (en) * | 2021-12-30 | 2022-05-13 | 广州极飞科技股份有限公司 | Boundary simplifying method, path planning method, device, equipment and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100296705A1 (en) * | 2007-11-07 | 2010-11-25 | Krysztof Miksa | Method of and arrangement for mapping range sensor data on image sensor data |
| CN102254097A (en) * | 2011-07-08 | 2011-11-23 | 普建涛 | Method for identifying fissure on lung CT (computed tomography) image |
| CN103745459A (en) * | 2013-12-26 | 2014-04-23 | 西安交通大学 | Detection method of an unstructured point cloud feature point and extraction method thereof |
-
2016
- 2016-08-30 CN CN201610767783.5A patent/CN106373118B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100296705A1 (en) * | 2007-11-07 | 2010-11-25 | Krysztof Miksa | Method of and arrangement for mapping range sensor data on image sensor data |
| CN102254097A (en) * | 2011-07-08 | 2011-11-23 | 普建涛 | Method for identifying fissure on lung CT (computed tomography) image |
| CN103745459A (en) * | 2013-12-26 | 2014-04-23 | 西安交通大学 | Detection method of an unstructured point cloud feature point and extraction method thereof |
Non-Patent Citations (3)
| Title |
|---|
| 杨小青: "基于法向量的三位点云配准方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 武敬民等: "基于法向夹角与Hausdorff距离的点云精简方法研究", 《微电子学与计算机》 * |
| 陈志扬等: "基于图形制导的复杂曲面最佳匹配的一种算法", 《航空学报》 * |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107194998B (en) * | 2017-05-23 | 2019-01-08 | 哈尔滨工业大学 | A kind of method of multi-layer three-dimension point cloud single-layered |
| CN107194998A (en) * | 2017-05-23 | 2017-09-22 | 哈尔滨工业大学 | A kind of method of multi-layer three-dimension point cloud single-layered |
| CN108198244A (en) * | 2017-12-20 | 2018-06-22 | 中国农业大学 | A kind of Apple Leaves point cloud compressing method and device |
| CN108198244B (en) * | 2017-12-20 | 2020-11-10 | 中国农业大学 | Apple leaf point cloud simplification method and device |
| CN108765434B (en) * | 2018-01-15 | 2022-05-24 | 中国人民解放军陆军装甲兵学院 | Contour extraction method based on additive remanufacturing point cloud model |
| CN108765434A (en) * | 2018-01-15 | 2018-11-06 | 中国人民解放军陆军装甲兵学院 | The contour extraction method of point cloud model is remanufactured based on increasing material |
| CN108759669B (en) * | 2018-05-31 | 2020-07-21 | 武汉中观自动化科技有限公司 | Indoor self-positioning three-dimensional scanning method and system |
| CN108759669A (en) * | 2018-05-31 | 2018-11-06 | 武汉中观自动化科技有限公司 | A kind of self-positioning 3-D scanning method and system in interior |
| CN109726509A (en) * | 2019-01-21 | 2019-05-07 | 南京航空航天大学 | A kind of part geometry feature representation model and construction method towards aircraft assembly |
| CN110111430A (en) * | 2019-04-11 | 2019-08-09 | 暨南大学 | One kind extracting quadric method from three-dimensional point cloud |
| CN110111430B (en) * | 2019-04-11 | 2023-03-10 | 暨南大学 | Method for extracting quadric surface from three-dimensional point cloud |
| CN110340738A (en) * | 2019-06-21 | 2019-10-18 | 武汉理工大学 | A kind of robot wire drawing high-speed rail white body workpiece method for precisely marking based on PCA |
| CN111080653A (en) * | 2019-11-06 | 2020-04-28 | 广西大学 | Method for simplifying multi-view point cloud by using region segmentation and grouping random simplification method |
| CN111080653B (en) * | 2019-11-06 | 2022-09-20 | 广西大学 | Method for simplifying multi-view point cloud by using region segmentation and grouping random simplification method |
| CN111652855A (en) * | 2020-05-19 | 2020-09-11 | 西安交通大学 | Point cloud simplification method based on survival probability |
| CN112102397A (en) * | 2020-09-10 | 2020-12-18 | 敬科(深圳)机器人科技有限公司 | Method, equipment and system for positioning multilayer part and readable storage medium |
| WO2021159838A1 (en) * | 2020-10-12 | 2021-08-19 | 平安科技（深圳）有限公司 | Method and apparatus for simplifying point cloud data, and storage medium and electronic device |
| CN113111548A (en) * | 2021-03-27 | 2021-07-13 | 西北工业大学 | Product three-dimensional feature point extraction method based on cycle angle difference |
| CN113744389A (en) * | 2021-08-24 | 2021-12-03 | 武汉理工大学 | Point cloud simplification method for complex part curved surface feature retention |
| CN113744389B (en) * | 2021-08-24 | 2023-10-10 | 武汉理工大学 | Point cloud simplifying method for complex part curved surface feature preservation |
| CN114485664A (en) * | 2021-12-30 | 2022-05-13 | 广州极飞科技股份有限公司 | Boundary simplifying method, path planning method, device, equipment and system |
| CN114485664B (en) * | 2021-12-30 | 2022-12-27 | 广州极飞科技股份有限公司 | Boundary simplifying method, path planning method, device, equipment and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106373118B (en) | 2017-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106373118B (en) | The complex curved surface parts point cloud compressing method of border and local feature can be effectively retained | |
| CN111299815B (en) | Visual detection and laser cutting trajectory planning method for low-gray rubber pad | |
| CN110287873B (en) | Non-cooperative target pose measurement method and system based on deep neural network and terminal equipment | |
| JP5705147B2 (en) | Representing 3D objects or objects using descriptors | |
| CN106023298B (en) | Point cloud Rigid Registration method based on local Poisson curve reestablishing | |
| Keselman et al. | Many-to-many graph matching via metric embedding | |
| Ji et al. | A novel simplification method for 3D geometric point cloud based on the importance of point | |
| CN107798696A (en) | A kind of three-dimensional point cloud method for registering based on guarantor office PCA | |
| CN111028327A (en) | Three-dimensional point cloud processing method, device and equipment | |
| El‐Sayed et al. | Plane detection in 3D point cloud using octree‐balanced density down‐sampling and iterative adaptive plane extraction | |
| CN110807781B (en) | Point cloud simplifying method for retaining details and boundary characteristics | |
| CN109685080A (en) | Multiple dimensioned plane extracting method based on Hough transformation and region growing | |
| Biasotti et al. | SHREC'18 track: Recognition of geometric patterns over 3D models | |
| CN111524168B (en) | Point cloud data registration method, system and device and computer storage medium | |
| CN105930859B (en) | Radar Signal Sorting Method based on linear manifold cluster | |
| Moitra et al. | Cluster-based data reduction for persistent homology | |
| Zafari et al. | Segmentation of partially overlapping convex objects using branch and bound algorithm | |
| CN112529945A (en) | Registration method for multi-view three-dimensional ISAR scattering point set | |
| Liu et al. | Method for extraction of airborne LiDAR point cloud buildings based on segmentation | |
| CN114332172A (en) | Improved laser point cloud registration method based on covariance matrix | |
| An et al. | Extracting statistical signatures of geometry and structure in 2D occupancy grid maps for global localization | |
| Li et al. | Primitive fitting using deep geometric segmentation | |
| CN109345571B (en) | Point cloud registration method based on extended Gaussian image | |
| Omidalizarandi et al. | Segmentation and classification of point clouds from dense aerial image matching | |
| Song et al. | Image classification based on BP neural network and sine cosine algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 |