CN108053443A - A kind of object point cloud pose evaluation method and system based on particle group optimizing - Google Patents
A kind of object point cloud pose evaluation method and system based on particle group optimizing Download PDFInfo
- Publication number
- CN108053443A CN108053443A CN201711157067.6A CN201711157067A CN108053443A CN 108053443 A CN108053443 A CN 108053443A CN 201711157067 A CN201711157067 A CN 201711157067A CN 108053443 A CN108053443 A CN 108053443A
- Authority
- CN
- China
- Prior art keywords
- particle
- cloud
- posture information
- cloud data
- 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
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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 present embodiments relate to a kind of object point cloud pose evaluation methods and system based on particle group optimizing, belong to robot for space visual perception technical field.Wherein, this method includes：Initialize the initial pose and translational speed of the representative of each particle in population, according to initial pose, rotation transformation is done to spatial point cloud, geometrical model where according to the cloud data after conversion and putting cloud, determine position relationship of each point with geometric surface, and then determine the closest distance of the point cloud after conversion and geometrical model, the fitness of each particle is determined according to closest distance and algorithmic rule, the posture information of a cloud is determined according to the posture information of fitness more new particle, and according to update posture information.The technical solution provided through this embodiment realizes the accurate and efficient technique effect for obtaining point cloud posture information.
Description
Technical field
The present embodiments relate to robot for space visual perception technical fields more particularly to one kind to be based on particle group optimizing
Object point cloud pose evaluation method and system.
Background technology
With the development of science and technology, machine man-based development and related application progressively develop simultaneously gradually perfect.In existing skill
In art, to obtain point cloud posture information, on the one hand, by way of being cloud data collection by model conversion, so as to fulfill a cloud
The acquisition of posture information；On the other hand, some enterprises begin to use " alignment algorithm " to improve acquisition point cloud posture information
Efficiency.
But during inventor realizes the present invention, it is found that at least there are following problems：
1st, put cloud pose and estimate that efficiency is relatively low, time complexity is higher；
2nd, it is relatively low to put cloud pose estimation precision, high-precision robot crawl demand can not be met.
The content of the invention
In order to solve the above technical problems, an embodiment of the present invention provides a kind of object point cloud poses based on particle group optimizing
Evaluation method and system.
It is according to embodiments of the present invention in a first aspect, an embodiment of the present invention provides a kind of objects based on particle group optimizing
Point cloud pose evaluation method, the described method includes：
Step S100：According to pre-set preprocessing rule to get point cloud in point initial cloud data into
Row pretreatment, obtains cloud data；
Step S200：The particle in particle populations is initialized according to pre-set initialization rule, obtains grain
The posture information of son；
Step S300：The point in described cloud is translated and revolved according to the posture information and the cloud data
Turn, the cloud data after being converted；
Step S400：The corresponding point of cloud data after ergodic transformation, after the cloud data after conversion and conversion
Geometrical model where point cloud determines position relationship of the point with the geometric surface in the geometrical model；
Step S500：According to pre-set computation rule and the position relationship determine conversion after point cloud with it is described several
The closest distance of what model；
Step S600：The adaptation of each particle is determined according to the closest distance and pre-set algorithmic rule
Degree；
Step S700：According to the fitness of each particle and the history fitness of each particle got,
Determine each optimal score of history of the particle and global optimum's score；
Step S800：According to the optimal score of history of the regular and each particle of pre-set update and global optimum
Score updates the posture information of each particle, obtains update posture information；
Step S900：When the corresponding error of the update posture information is less than pre-set error threshold, alternatively, update
Number when being equal to or more than pre-set iteration threshold, then the pose of described cloud is determined according to the update posture information
Information；
Step S1000：When the corresponding error of the update posture information be equal to or more than pre-set error threshold or
Person, when newer number is less than pre-set iteration threshold, then return to step S300.
Closest geometric surface is determined according to the position relationship of each point and the geometric surface；
Determined according to the closest geometric surface between the corresponding geometrical model of point cloud after conversion it is closest away from
From.
Further, the step S100 is specifically included：
Initial cloud data is obtained by sensor；
A cloud filtration treatment, exceptional value removal processing, cluster segmentation processing are carried out to the initial cloud data successively, is obtained
To the cloud data.
It provides through this embodiment：After carrying out a series of processing to the initial cloud data got, a cloud is obtained
The technical solution of data realizes the accuracy and reliability for obtaining cloud data, and then realizes quick and accurately determine
The technique effect of point cloud pose.
Further, the sensor is depth camera.
Further, the step S200 is specifically included：
The pose vector sum velocity vector of the particle is initialized respectively；
The posture information is determined according to the pose vector sum velocity vector after initialization.
Further, the step S400 is specifically included：
It is determined a little in the subpoint of the geometric surface according to point and the geometric surface；
The a little position relationship with the geometric surface is determined according to the subpoint.
Further, the step S500 is specifically included：
Closest geometric surface is determined according to the position relationship and the computation rule；
Determined according to the closest geometric surface between the corresponding geometrical model of point cloud after conversion it is closest away from
From.
Other side according to embodiments of the present invention, an embodiment of the present invention provides a kind of objects based on particle group optimizing
Body point cloud pose estimating system, the system comprises：
Preprocessing module：For according to pre-set preprocessing rule to get point cloud in point initial point cloud
Data are pre-processed, and obtain cloud data；
Initialization module：For being initialized according to pre-set initialization rule to the particle in particle populations,
Obtain the posture information of particle；
Conversion module：For the point in described cloud is carried out according to the posture information and the cloud data translation and
Rotation, the cloud data after being converted；
Spider module：For the corresponding point of the cloud data after ergodic transformation, according to the cloud data after conversion and conversion
The geometrical model where point cloud afterwards, determines position relationship of the point with the geometric surface in the geometrical model；
Search module：For determining the point Yun Yusuo after conversion according to pre-set computation rule and the position relationship
The closest distance of geometrical model is stated, each particle is determined according to the closest distance and pre-set algorithmic rule
Fitness, according to the fitness of each particle and the history fitness of each particle got, determine each
The optimal score of history and global optimum's score of the particle；
Update module：For the optimal score of history and the overall situation according to the regular and each particle of pre-set update
Optimal score updates the posture information of each particle, obtains update posture information；
Described search module is additionally operable to：When the corresponding error of the update posture information is less than pre-set error threshold
Value alternatively, when newer number is equal to or more than pre-set iteration threshold, then determines institute according to the update posture information
State the posture information of a cloud；
Iteration module：It is equal to or more than pre-set error threshold for working as the corresponding error of the update posture information
Value alternatively, when newer number is less than pre-set iteration threshold, then returns to conversion module.
Further, the preprocessing module is specifically used for：
Initial cloud data is obtained by sensor；
A cloud filtration treatment, exceptional value removal processing, cluster segmentation processing are carried out to the initial cloud data successively, is obtained
To the cloud data.
Further, the sensor is depth camera.
Further, the initialization module is specifically used for：Respectively to the pose vector sum velocity vector of the particle into
Row initialization；
The posture information is determined according to the pose vector sum velocity vector after initialization.
Further, described search module is specifically used for：
Closest geometric surface is determined according to the position relationship and the computation rule；
Determined according to the closest geometric surface between the corresponding geometrical model of point cloud after conversion it is closest away from
From.
The advantageous effect of the embodiment of the present invention is, after conversion is determined according to posture information and cloud data
Cloud data, according to the cloud data after conversion and the geometrical model where point cloud, determine the position of each point and geometric surface
Relation, and then determine the closest distance of the point cloud after conversion and geometrical model, it is determined according to closest distance and algorithmic rule
The fitness of each particle according to the posture information of fitness more new particle, and determines according to update posture information the position of cloud
The technical solution of appearance information, the technical issues of avoiding local solution in the prior art, realize quick and accurately determine a point cloud
The technique effect of pose.
Description of the drawings
Fig. 1 is a kind of flow of the object point cloud pose evaluation method based on particle group optimizing provided in an embodiment of the present invention
Schematic diagram；
A kind of Fig. 2 structures of the object point cloud pose estimating system based on particle group optimizing provided in an embodiment of the present invention are shown
It is intended to.
Specific embodiment
In being described below, in order to illustrate rather than in order to limit, it is proposed that such as particular system structure, interface, technology it
The detail of class understands the present invention to cut thoroughly.However, it will be clear to one skilled in the art that there is no these specifically
The present invention can also be realized in the other embodiments of details.In other situations, omit to well-known system, circuit and
The detailed description of method, in case unnecessary details interferes description of the invention.
An embodiment of the present invention provides a kind of object point cloud pose evaluation methods and system based on particle group optimizing.
One side according to embodiments of the present invention, an embodiment of the present invention provides a kind of objects based on particle group optimizing
Point cloud pose evaluation method.
Referring to Fig. 1, Fig. 1 is the flow signal of the evaluation method of a kind of cloud posture information provided in an embodiment of the present invention
Figure.
As shown in Figure 1, this method includes：
Step S100：According to pre-set preprocessing rule to get point cloud in point initial cloud data into
Row pretreatment, obtains cloud data；
Step S200：The particle in particle populations is initialized according to pre-set initialization rule, obtains grain
The posture information of son；
Step S300：The point in cloud is translated and rotated according to posture information and cloud data, after obtaining conversion
Cloud data；
Step S400：The corresponding point of cloud data after ergodic transformation, after the cloud data after conversion and conversion
Geometrical model where point cloud determines position relationship of the point with the geometric surface in geometrical model；
Step S500：Point cloud and geometrical model after conversion are determined according to pre-set computation rule and position relationship
Closest distance；
Step S600：The fitness of each particle is determined according to closest distance and pre-set algorithmic rule；
Step S700：According to the fitness of each particle and the history fitness of each particle got, determine each
The optimal score of history and global optimum's score of particle；
Step S800：It is obtained according to the optimal score of history of the pre-set regular and each particle of update and global optimum
Point, the posture information of each particle is updated, obtains update posture information；
Step S900：It is less than pre-set error threshold when updating the corresponding error of posture information, alternatively, newer time
When number is equal to or more than pre-set iteration threshold, then the posture information of a cloud is determined according to update posture information；
Step S1000：It is equal to or more than pre-set error threshold when updating the corresponding error of posture information, alternatively,
When newer number is less than pre-set iteration threshold, then return to step S300.
The technical solution provided through this embodiment realizes quick and precisely the technology that posture information is determined is imitated
Fruit.The time has been saved, efficiency has been improved, and improves accuracy.
In a kind of technical solution in the cards, step S100 is specifically included：
Initial cloud data is obtained by sensor；
A cloud filtration treatment, exceptional value removal processing, cluster segmentation processing are carried out to initial cloud data successively, is obtained a little
Cloud data.
In the present embodiment, initial cloud data is obtained by sensor, after getting initial cloud data, and
It is indirect that the initial cloud data is applied, but preferentially the initial cloud data is pre-processed.
Wherein, pretreatment specifically includes：Initial cloud data is filtered successively, carries out exceptional value removal after filtering again
Processing after the different removal of exceptional value, then is handled by cluster segmentation, to obtain cloud data.To ensure the accurate of cloud data
Property.
Wherein, sensor is depth camera.
The technical solution provided through this embodiment realizes technique effect that is quick and accurately obtaining cloud data,
And then realize technique effect that is quick and being precisely determined to posture information.
In a kind of technical solution in the cards, step S200 is specifically included：
The pose vector sum velocity vector of particle is initialized respectively；
Posture information is determined according to the pose vector sum velocity vector after initialization.
In the present embodiment, first particle is initialized according to formula 1, to obtain the initial bit confidence of the first particle
Breath, initializes the first particle according to formula 2, to obtain the initial velocity information of the first particle；
Formula 1：X1=[x1,y1,z1,φ1,ψ1,γ1]
Formula 2：
Wherein, X1For the initial position message of first particle, V1For the initial velocity information of first particle, x1,y1,z1
Position vector in the six-vector of respectively first particle；φ1、ψ1、γ1In the six-vector of respectively first particle
Pose vector,Velocity vector in the six-vector of respectively first particle.
In a kind of technical solution in the cards, step S400 is specifically included：
It is determined a little in the subpoint of geometric surface according to point and geometric surface；
The a little position relationship with geometric surface is determined according to subpoint.
In a kind of technical solution in the cards, step S500 is specifically included：
Closest geometric surface is determined according to the position relationship and computation rule；
The closest distance between the corresponding geometrical model of point cloud after conversion is determined according to closest geometric surface.
It is understood that all geometrical models can be subdivided by geometric surface (in this application, geometry one by one
Face refers to triangular facet) composition.Specifically, the triangular facet data of geometrical model can be obtained based on the CAD model of STL forms.
Determine subpoint of the point on target triangular facet, wherein, target triangular facet is any one in multiple triangular facets；
When subpoint falls in target triangular facet or subpoint falls on any side of target triangular facet, then it is European away from
From between cloud and subpoint with a distance from line；
When subpoint falls outside the side of target triangular facet, then Euclidean distance for point cloud and target triangular facet three sides it
Between distance in shortest distance.
Closest geometric surface is determined according to Euclidean distance.
According to closest geometric surface, the closest distance between the corresponding geometrical model of point cloud after conversion is determined.
In a kind of technical solution in the cards, the posture information of cloud is determined according to formula 3 and formula 4,
Formula 3：
Formula 4：
Wherein, the speed of first particle be+1 iteration of first particle kth velocity information, the position of first particle
It is set to the location information of+1 iteration of first particle kth, w is inertial factor, c1、c1For Studying factors, r1、r2For first
The random factor of+1 iteration of particle kth, k are the integer more than 1,For top score in first particle k times,
gbestkFor all particle k Ci Zhong global optimums score.
Other side according to embodiments of the present invention, an embodiment of the present invention provides with the corresponding one kind of the above method
The estimating system of point cloud posture information.
Referring to Fig. 2, Fig. 2 is a kind of object point cloud pose estimation based on particle group optimizing provided in an embodiment of the present invention
The structure diagram of system.
As shown in Fig. 2, the system includes：
Preprocessing module：For according to pre-set preprocessing rule to get point cloud in point initial point cloud
Data are pre-processed, and obtain cloud data；
Initialization module：For being initialized according to pre-set initialization rule to the particle in particle populations,
Obtain the posture information of particle；
Conversion module：For the point in cloud to be translated and rotated according to posture information and cloud data, become
Cloud data after changing；
Spider module：For the corresponding point of the cloud data after ergodic transformation, according to the cloud data after conversion and conversion
The geometrical model where point cloud afterwards, determines position relationship of the point with the geometric surface in geometrical model；
Search module：For the point cloud after conversion and what model to be determined according to pre-set computation rule and position relationship
Closest distance, the fitness of each particle is determined according to closest distance and pre-set algorithmic rule, according to each
The fitness of particle and the history fitness of each particle got, determine each particle the optimal score of history and it is global most
Excellent score；
Update module：For the optimal score of history according to the pre-set regular and each particle of update and global optimum
Score updates the posture information of each particle, obtains update posture information；
Search module is additionally operable to：It is less than pre-set error threshold when updating the corresponding error of posture information, alternatively, more
When new number is equal to or more than pre-set iteration threshold, then the posture information of a cloud is determined according to update posture information；
Iteration module：For when the corresponding error of update posture information be equal to or more than pre-set error threshold or
Person when newer number is less than pre-set iteration threshold, then returns to conversion module.
In a kind of technical solution in the cards, preprocessing module is specifically used for：
Initial cloud data is obtained by sensor；
A cloud filtration treatment, exceptional value removal processing, cluster segmentation processing are carried out to initial cloud data successively, is obtained a little
Cloud data.
In a kind of technical solution in the cards, sensor is depth camera.
In a kind of technical solution in the cards, initialization module is specifically used for：Respectively to the pose vector sum of particle
Velocity vector is initialized；
Posture information is determined according to the pose vector sum velocity vector after initialization.
In a kind of technical solution in the cards, search module is specifically used for：
Closest geometric surface is determined according to position relationship and computation rule；
The closest distance between the corresponding geometrical model of point cloud after conversion is determined according to closest geometric surface.
Reader should be understood that in the description of this specification, reference term " one embodiment ", " some embodiments ", " show
The description of example ", " specific example " or " some examples " etc. mean to combine the specific features of the embodiment or example description, structure,
Material or feature are contained at least one embodiment of the present invention or example.In the present specification, above-mentioned term is shown
The statement of meaning property need not be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described
It may be combined in any suitable manner in any one or more of the embodiments or examples.In addition, without conflicting with each other, this
The technical staff in field can be by the different embodiments described in this specification or example and different embodiments or exemplary spy
Sign is combined and combines.
It is apparent to those skilled in the art that for convenience of description and succinctly, the dress of foregoing description
The specific work process with unit is put, may be referred to the corresponding process in preceding method embodiment, details are not described herein.
In several embodiments provided herein, it should be understood that disclosed apparatus and method can pass through it
Its mode is realized.For example, the apparatus embodiments described above are merely exemplary, for example, the division of unit, is only
A kind of division of logic function, can there is an other dividing mode in actual implementation, for example, multiple units or component can combine or
Person is desirably integrated into another system or some features can be ignored or does not perform.
The unit illustrated as separating component may or may not be physically separate, be shown as unit
Component may or may not be physical location, you can be located at a place or can also be distributed to multiple networks
On unit.Some or all of unit therein can be selected to realize the mesh of the embodiment of the present invention according to the actual needs
's.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it can also
It is that unit is individually physically present or two or more units integrate in a unit.It is above-mentioned integrated
The form that hardware had both may be employed in unit is realized, can also be realized in the form of SFU software functional unit.
If integrated unit is realized in the form of SFU software functional unit and is independent production marketing or in use, can
To be stored in a computer read/write memory medium.Based on such understanding, technical scheme substantially or
Saying all or part of the part contribute to the prior art or the technical solution can be embodied in the form of software product
Out, which is stored in a storage medium, is used including some instructions so that a computer equipment
(can be personal computer, server or the network equipment etc.) performs all or part of each embodiment method of the present invention
Step.And foregoing storage medium includes：It is USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random
Access various Jie that can store program code such as memory (RAM, Random Access Memory), magnetic disc or CD
Matter.
It should also be understood that in various embodiments of the present invention, the size of the sequence number of above-mentioned each process is not meant to execution sequence
Priority, the execution sequence of each process should determine with its function and internal logic, the implementation without tackling the embodiment of the present invention
Journey forms any restriction.
More than, it is only specific embodiment of the invention, but protection scope of the present invention is not limited thereto, and it is any to be familiar with
Those skilled in the art in the technical scope disclosed by the present invention, can readily occur in various equivalent modifications or substitutions,
These modifications or substitutions should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be wanted with right
Subject to the protection domain asked.
Claims (10)
1. a kind of object point cloud pose evaluation method based on particle group optimizing, which is characterized in that the described method includes：
Step S100：The initial cloud data for putting the point in cloud got is carried out according to pre-set preprocessing rule pre-
Processing, obtains cloud data；
Step S200：The particle in particle populations is initialized according to pre-set initialization rule, obtains particle
Posture information；
Step S300：The point in described cloud is translated and rotated according to the posture information and the cloud data, is obtained
Cloud data after to conversion；
Step S400：The corresponding point of cloud data after ergodic transformation, according to the point cloud after the cloud data after conversion and conversion
The geometrical model at place determines position relationship of the point with the geometric surface in the geometrical model；
Step S500：Point cloud after variation and the geometry mould are determined according to pre-set computation rule and the position relationship
The closest distance of type；
Step S600：The fitness of each particle is determined according to the closest distance and pre-set algorithmic rule；
Step S700：According to the fitness of each particle and the history fitness of each particle got, determine
Each optimal score of history of the particle and global optimum's score；
Step S800：It is obtained according to the optimal score of history of the regular and each particle of pre-set update and global optimum
Point, the posture information of each particle is updated, obtains update posture information；
Step S900：When the corresponding error of the update posture information is less than pre-set error threshold, alternatively, newer time
When number is equal to or more than pre-set iteration threshold, then determine that the pose of described cloud is believed according to the update posture information
Breath；
Step S1000：When the corresponding error of the update posture information is equal to or more than pre-set error threshold, alternatively,
When newer number is less than pre-set iteration threshold, then return to step S300.
2. a kind of object point cloud pose evaluation method based on particle group optimizing according to claim 1, which is characterized in that
The step S100 is specifically included：
Initial cloud data is obtained by sensor；
A cloud filtration treatment, exceptional value removal processing, cluster segmentation processing are carried out to the initial cloud data successively, obtains institute
State cloud data.
3. a kind of object point cloud pose evaluation method based on particle group optimizing according to claim 2, which is characterized in that
The sensor is depth camera.
4. a kind of object point cloud pose evaluation method based on particle group optimizing according to claim 1, which is characterized in that
The step S200 is specifically included：
The pose vector sum velocity vector of the particle is initialized respectively；
The posture information is determined according to the pose vector sum velocity vector after initialization.
5. a kind of object point cloud pose evaluation method based on particle group optimizing according to claim 1, which is characterized in that
The step S400 is specifically included：
It is determined a little in the subpoint of the geometric surface according to point and the geometric surface；
The a little position relationship with the geometric surface is determined according to the subpoint.
6. a kind of object point cloud pose evaluation method based on particle group optimizing according to any one of claim 1-5,
It is characterized in that, the step S500 is specifically included：
Closest geometric surface is determined according to the position relationship and the computation rule；
The closest distance between the corresponding geometrical model of point cloud after conversion is determined according to the closest geometric surface.
7. a kind of object point cloud pose estimating system based on particle group optimizing, which is characterized in that the system comprises：
Preprocessing module：For according to pre-set preprocessing rule to get point cloud in point initial cloud data
It is pre-processed, obtains cloud data；
Initialization module：For being initialized according to pre-set initialization rule to the particle in particle populations, obtain
The posture information of particle；
Conversion module：For the point in described cloud to be translated and revolved according to the posture information and the cloud data
Turn, the cloud data after being converted；
Spider module：For each point in the cloud data after ergodic transformation, after the cloud data after conversion and conversion
Point cloud where geometrical model, determine point and the geometric surface in the geometrical model position relationship；
Search module：For according to pre-set computation rule and the position relationship determine conversion after point cloud with it is described several
The closest distance of what model determines the suitable of each particle according to the closest distance and pre-set algorithmic rule
Response according to the fitness of each particle and the history fitness of each particle got, determines each described
The optimal score of history and global optimum's score of particle；
Update module：For the optimal score of history according to the regular and each particle of pre-set update and global optimum
Score updates the posture information of each particle, obtains update posture information；
Described search module is additionally operable to：When the corresponding error of the update posture information be less than pre-set error threshold or
Person when newer number is equal to or more than pre-set iteration threshold, then determines the point according to the update posture information
The posture information of cloud；
Iteration module：For work as the corresponding error of the update posture information be equal to or more than pre-set error threshold or
Person when newer number is less than pre-set iteration threshold, then returns to conversion module.
8. a kind of object point cloud pose estimating system based on particle group optimizing according to claim 7, which is characterized in that
The preprocessing module is specifically used for：
Initial cloud data is obtained by sensor；
A cloud filtration treatment, exceptional value removal processing, cluster segmentation processing are carried out to the initial cloud data successively, obtains institute
State cloud data.
9. a kind of object point cloud pose estimating system based on particle group optimizing according to claim 8, which is characterized in that
The sensor is depth camera.
10. a kind of object point cloud pose estimating system based on particle group optimizing according to any one of claim 7-9,
It is characterized in that,
The initialization module is specifically used for：The pose vector sum velocity vector of the particle is initialized respectively；
The posture information is determined according to the pose vector sum velocity vector after initialization.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711157067.6A CN108053443B (en) | 2017-11-20 | 2017-11-20 | A kind of object point cloud pose evaluation method and system based on particle group optimizing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711157067.6A CN108053443B (en) | 2017-11-20 | 2017-11-20 | A kind of object point cloud pose evaluation method and system based on particle group optimizing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108053443A true CN108053443A (en) | 2018-05-18 |
| CN108053443B CN108053443B (en) | 2019-08-02 |
Family
ID=62119990
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711157067.6A Active CN108053443B (en) | 2017-11-20 | 2017-11-20 | A kind of object point cloud pose evaluation method and system based on particle group optimizing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108053443B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110310325A (en) * | 2019-06-28 | 2019-10-08 | Oppo广东移动通信有限公司 | A kind of virtual measurement method, electronic equipment and computer readable storage medium |
| CN111033299A (en) * | 2018-07-02 | 2020-04-17 | 北京嘀嘀无限科技发展有限公司 | Vehicle navigation system based on point cloud utilization pose estimation |
| CN111079208A (en) * | 2019-11-20 | 2020-04-28 | 杭州电子科技大学 | Particle swarm optimization algorithm-based method for identifying surface correspondence between CAD models |
| WO2020237996A1 (en) * | 2019-05-30 | 2020-12-03 | 魔门塔(苏州)科技有限公司 | Vehicle pose correction method and device |
| CN112084937A (en) * | 2020-09-08 | 2020-12-15 | 清华大学 | Dynamic vehicle detection method |
| CN113433532A (en) * | 2021-07-07 | 2021-09-24 | 清华大学苏州汽车研究院（吴江） | Laser radar attitude calibration method and device based on particle swarm algorithm |
| CN114355415A (en) * | 2022-01-06 | 2022-04-15 | 上海安亭地平线智能交通技术有限公司 | Pose information determining method and device, electronic equipment and storage medium |
| CN117496074A (en) * | 2023-12-29 | 2024-02-02 | 中国人民解放军国防科技大学 | Efficient three-dimensional scene reconstruction method suitable for rapid movement of camera |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102446354A (en) * | 2011-08-29 | 2012-05-09 | 北京建筑工程学院 | Integral registration method of high-precision multisource ground laser point clouds |
| CN103412565A (en) * | 2013-05-17 | 2013-11-27 | 浙江中控研究院有限公司 | robot with global location rapid estimating capability and positioning method thereof |
| CN105551015A (en) * | 2015-12-02 | 2016-05-04 | 南京邮电大学 | Scattered-point cloud image registering method |
| US20160253807A1 (en) * | 2015-02-26 | 2016-09-01 | Mitsubishi Electric Research Laboratories, Inc. | Method and System for Determining 3D Object Poses and Landmark Points using Surface Patches |
| CN106951860A (en) * | 2017-03-20 | 2017-07-14 | 河南腾龙信息工程有限公司 | A kind of three-dimensional data intelligent identification Method based on a cloud |
| US20170258391A1 (en) * | 2016-03-14 | 2017-09-14 | Sensors Unlimited, Inc. | Multimodal fusion for object detection |
-
2017
- 2017-11-20 CN CN201711157067.6A patent/CN108053443B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102446354A (en) * | 2011-08-29 | 2012-05-09 | 北京建筑工程学院 | Integral registration method of high-precision multisource ground laser point clouds |
| CN103412565A (en) * | 2013-05-17 | 2013-11-27 | 浙江中控研究院有限公司 | robot with global location rapid estimating capability and positioning method thereof |
| US20160253807A1 (en) * | 2015-02-26 | 2016-09-01 | Mitsubishi Electric Research Laboratories, Inc. | Method and System for Determining 3D Object Poses and Landmark Points using Surface Patches |
| CN105551015A (en) * | 2015-12-02 | 2016-05-04 | 南京邮电大学 | Scattered-point cloud image registering method |
| US20170258391A1 (en) * | 2016-03-14 | 2017-09-14 | Sensors Unlimited, Inc. | Multimodal fusion for object detection |
| CN106951860A (en) * | 2017-03-20 | 2017-07-14 | 河南腾龙信息工程有限公司 | A kind of three-dimensional data intelligent identification Method based on a cloud |
Non-Patent Citations (3)
| Title |
|---|
| YUQIN GE 等: "A point cloud registration method combining enhanced particle swarm optimization and iterative closest point method", 《2016 CHINESE CONTROL AND DECISION CONFERENCE (CCDC)》 * |
| 翟欢乐: "基于几何特征的点云拼合研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 韩贤权 等: "散乱点云数据精配准的粒子群优化算法", 《武汉大学学报信息科学版》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111033299B (en) * | 2018-07-02 | 2023-06-16 | 北京嘀嘀无限科技发展有限公司 | Vehicle navigation system based on point cloud utilization pose estimation |
| CN111033299A (en) * | 2018-07-02 | 2020-04-17 | 北京嘀嘀无限科技发展有限公司 | Vehicle navigation system based on point cloud utilization pose estimation |
| WO2020237996A1 (en) * | 2019-05-30 | 2020-12-03 | 魔门塔(苏州)科技有限公司 | Vehicle pose correction method and device |
| CN110310325A (en) * | 2019-06-28 | 2019-10-08 | Oppo广东移动通信有限公司 | A kind of virtual measurement method, electronic equipment and computer readable storage medium |
| CN111079208A (en) * | 2019-11-20 | 2020-04-28 | 杭州电子科技大学 | Particle swarm optimization algorithm-based method for identifying surface correspondence between CAD models |
| CN111079208B (en) * | 2019-11-20 | 2024-01-23 | 杭州电子科技大学 | Particle swarm algorithm-based CAD model surface corresponding relation identification method |
| CN112084937A (en) * | 2020-09-08 | 2020-12-15 | 清华大学 | Dynamic vehicle detection method |
| CN112084937B (en) * | 2020-09-08 | 2021-03-19 | 清华大学 | Dynamic vehicle detection method |
| CN113433532A (en) * | 2021-07-07 | 2021-09-24 | 清华大学苏州汽车研究院（吴江） | Laser radar attitude calibration method and device based on particle swarm algorithm |
| WO2023131048A1 (en) * | 2022-01-06 | 2023-07-13 | 上海安亭地平线智能交通技术有限公司 | Position and attitude information determining method and apparatus, electronic device, and storage medium |
| CN114355415A (en) * | 2022-01-06 | 2022-04-15 | 上海安亭地平线智能交通技术有限公司 | Pose information determining method and device, electronic equipment and storage medium |
| CN117496074A (en) * | 2023-12-29 | 2024-02-02 | 中国人民解放军国防科技大学 | Efficient three-dimensional scene reconstruction method suitable for rapid movement of camera |
| CN117496074B (en) * | 2023-12-29 | 2024-03-22 | 中国人民解放军国防科技大学 | Efficient three-dimensional scene reconstruction method suitable for rapid movement of camera |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108053443B (en) | 2019-08-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108053443A (en) | A kind of object point cloud pose evaluation method and system based on particle group optimizing | |
| CN105741355B (en) | A kind of block dividing method of triangle grid model | |
| KR102068576B1 (en) | Convolutional neural network based image processing system and method | |
| CN112991447B (en) | Visual positioning and static map construction method and system in dynamic environment | |
| US11636306B2 (en) | Implementing traditional computer vision algorithms as neural networks | |
| CN112328715B (en) | Visual positioning method, training method of related model, related device and equipment | |
| CN108062543A (en) | A kind of face recognition method and device | |
| CN111627050B (en) | Training method and device for target tracking model | |
| CN111324145B (en) | Unmanned aerial vehicle autonomous landing method, device, equipment and storage medium | |
| CN109146817B (en) | Noise processing method for non-iterative single object scattered point cloud data | |
| CN110991513A (en) | Image target recognition system and method with human-like continuous learning capability | |
| CN110599506A (en) | Point cloud segmentation method for three-dimensional measurement of complex special-shaped curved surface robot | |
| CN112101073B (en) | Face image processing method, device, equipment and computer storage medium | |
| CN105868767B (en) | Face feature point positioning method and device | |
| CN111915657A (en) | Point cloud registration method and device, electronic equipment and storage medium | |
| Spiller | Object Localization Using Deformable Templates | |
| CN110992310A (en) | Method and device for determining partition where mediastinal lymph node is located | |
| CN108876776A (en) | A kind of method of generating classification model, eye fundus image classification method and device | |
| CN110533781B (en) | Automatic labeling method for multi-class three-dimensional model components | |
| CN110378352B (en) | High-precision anti-interference two-dimensional filtering navigation data denoising method in complex underwater environment | |
| CN116228850A (en) | Object posture estimation method, device, electronic equipment and readable storage medium | |
| CN115147433A (en) | Point cloud registration method | |
| CN112435342A (en) | Region division method and system for special-shaped curved surface | |
| CN108830860B (en) | Binocular image target segmentation method and device based on RGB-D constraint | |
| CN106570914B (en) | Method for transplanting roles in two-dimensional image |
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 |