CN110400374A - The method for building up of panorama point cloud data and establish system - Google Patents
The method for building up of panorama point cloud data and establish system Download PDFInfo
- Publication number
- CN110400374A CN110400374A CN201810392772.2A CN201810392772A CN110400374A CN 110400374 A CN110400374 A CN 110400374A CN 201810392772 A CN201810392772 A CN 201810392772A CN 110400374 A CN110400374 A CN 110400374A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- cloud data
- panorama
- data
- building
- 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
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000006243 chemical reaction Methods 0.000 claims abstract description 39
- 239000011159 matrix material Substances 0.000 claims abstract description 28
- 241001269238 Data Species 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 12
- 238000003384 imaging method Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 3
- 102000003729 Neprilysin Human genes 0.000 description 14
- 108090000028 Neprilysin Proteins 0.000 description 14
- 238000010586 diagram Methods 0.000 description 13
- 102100022005 B-lymphocyte antigen CD20 Human genes 0.000 description 10
- 101000897405 Homo sapiens B-lymphocyte antigen CD20 Proteins 0.000 description 10
- 230000004888 barrier function Effects 0.000 description 3
- 102100034112 Alkyldihydroxyacetonephosphate synthase, peroxisomal Human genes 0.000 description 1
- 101000799143 Homo sapiens Alkyldihydroxyacetonephosphate synthase, peroxisomal Proteins 0.000 description 1
- 238000000848 angular dependent Auger electron spectroscopy Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005034 decoration Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/003—Navigation within 3D models or images
Abstract
A kind of method for building up of panorama point cloud data and establish system.The method for building up of panorama point cloud data includes the following steps.Geometric object is referred to the overlapping region shooting at least one of these spatial digitizers with two adjacent spatial digitizers, to obtain two reference point clouds data.Fixed reference feature plane is obtained from each reference point clouds data.According to these fixed reference feature planes, coordinate conversion matrix is obtained.It is shot with these spatial digitizers, to obtain multiple instant point cloud datas.According to the coordinate conversion matrix, coordinate conversion is carried out to these instant point cloud datas, and combine these instant point cloud datas, to obtain panorama point cloud data.
Description
Technical field
The invention relates to a kind of method for building up of panorama point cloud data and establish system.
Background technique
Automotive safety is gradually taken seriously in recent years, and government regulations mandate new listing vehicle money need to install it is advanced drive it is auxiliary
Auxiliary system (Advanced Driver Assistance Systems, ADAS), with guarantee driving safety.In addition, with
The fast development of artificial intelligence, the application for automating mobile carrier and robot are also more and more extensive.These applications are also required to adopt
With auxiliary system is moved, to help mobile carrier or robot to be moved.
Current driving/movement auxiliary system is usually using ultrasonic radar, and using bidimensional image come additional barrier object
Judgement.However, the range of ultrasonic radar is smaller, and the detection for needing that vehicle periphery could be eliminated using more sensors is dead
Angle.In addition, also there is driving/movement auxiliary system using infrared radar sensor.However, the sensing of infrared radar sensor
Angle is lower, when raining or thick fog weather, is easy misidentification barrier.In addition, also there is driving/movement auxiliary system using two dimension
Image sensor.However, 2 D image sensor can not learn object far and near distance with single-sensor, in night, rain or
When thick fog weather, it is easy misidentification barrier, can be failed when serious.Furthermore these drive/move auxiliary system can not be 0.5
Accurately three-dimensional measuring information is obtained in~5 meters of short distance, so that safety is had a greatly reduced quality.
Summary of the invention
The present invention about a kind of panorama point cloud data method for building up and establish system, using point cloud data be aligned journey
Sequence carries out a cloud alignment, and carries out point cloud data and capture program immediately, to obtain panorama point cloud data.The standard of panorama point cloud data
Exactness is high, and can obtain 0.5~5 meter of short distance situation, is quite suitably applied in driving/movement auxiliary system.
An embodiment according to the present invention proposes a kind of method for building up of panorama point cloud data.Panorama point cloud data is built
Cube method includes the following steps.At least one is shot with overlapping region of the two adjacent spatial digitizers to these spatial digitizers
With reference to geometric object, to obtain two reference point clouds data.Fixed reference feature plane is obtained from each reference point clouds data.According to these
Fixed reference feature plane obtains coordinate conversion matrix.It is shot with these spatial digitizers, to obtain multiple i.e. time point cloud numbers
According to.According to the coordinate conversion matrix, coordinate conversion is carried out to these instant point cloud datas, and combine these instant point cloud datas,
To obtain panorama point cloud data.
According to another embodiment of the present invention, propose a kind of panorama point cloud data establishes system.Panorama point cloud data
Establishing system includes two spatial digitizer characteristic processing units and coordinates process unit.These spatial digitizers be disposed adjacent and
Geometric object is referred to the overlapping region shooting at least one to these spatial digitizers, to obtain two reference point clouds data.
This feature processing unit is to obtain fixed reference feature plane from each reference point clouds data.The coordinates process unit is according to these references
Characteristic plane obtains coordinate conversion matrix.These spatial digitizers are also to be shot, to obtain multiple i.e. time point cloud numbers
According to.The coordinates process unit carries out coordinate conversion according to the coordinate conversion matrix, to these instant point cloud datas, and combines these
Instant point cloud data, to obtain panorama point cloud data.
More preferably understand to have to above-mentioned and other aspect of the invention, special embodiment below, and cooperates attached drawing detailed
Carefully it is described as follows:
Detailed description of the invention
Fig. 1 illustrates the schematic diagram of mobile carrier and multiple spatial digitizers.
The schematic diagram of the overlapping region of Fig. 2 graphical three-dimensional scanner.
Fig. 3 illustrates the schematic diagram for establishing system of panorama point cloud data according to the embodiment.
Fig. 4 illustrates the flow chart of the method for building up of panorama point cloud data according to the embodiment.
The schematic diagram of each step of the pictorial image of Fig. 5~9 4.
The schematic diagram for establishing system of the panorama point cloud data of Figure 10 diagram according to another embodiment.
Specific embodiment
Fig. 1~2 is please referred to, Fig. 1 illustrates mobile carrier 800 and multiple spatial digitizers 110 (0), 110 (1), 110
(2) ..., the schematic diagram of 110 (N-2), 110 (N-1), 110 (N), the overlapping of Fig. 2 graphical three-dimensional scanner 110 (0), 110 (1)
The schematic diagram of region OL.In the present embodiment, mobile carrier 800 is, for example, vehicle, transfer mechanism or robot.Three-dimensional is swept
Retouch instrument 110 (0), 110 (1), 110 (2) ..., 110 (N-2), 110 (N-1), 110 (N) be, for example, ranging (Time of when flying
Flight, ToF) sensor or infrared sensor.These spatial digitizers 110 (0), 110 (1), 110 (2) ..., 110 (N-
2), 110 (N-1), 110 (N) shooting distance be 0.5~5 meter.In the range of 0.5~5 meter, including pit-hole, vehicle,
Human body can be detected accurately.
The disclosure by multiple spatial digitizers 110 (0), 110 (1), 110 (2) ..., 110 (N-2), 110 (N-1), 110
(N) instant point cloud data is captured, the overlapping of overlapped region OL constitutes panorama point cloud data.In the present embodiment, adjacent
Spatial digitizer 110 (0), 110 (1), 110 (2) ..., the shooting angle of 110 (N-2), 110 (N-1), 110 (N) there is weight
Folded region OL, this overlapping region OL can be 5~15 degree (being e.g. greater than 10 degree).
Due to each spatial digitizer 110 (0), 110 (1), 110 (2) ..., 110 (N-2), 110 (N-1), 110 (N)
Coordinate system C0, C1, C2 ..., the coordinate system C800 of CN-2, CN-1, CN and mobile carrier 800 it is not consistent, therefore the disclosure
Cloud alignment (point cloud registration) is carried out by coordinate switch technology, allows each spatial digitizer 110
(0), 110 (1), 110 (2) ..., a plurality of instant point cloud data being capable of group acquired by 110 (N-2), 110 (N-1), 110 (N)
It closes.
For example, the coordinate system C1 of spatial digitizer 110 (1) can be converted by coordinate conversion matrix T1 to seat
Mark system C0.The coordinate system C2 of spatial digitizer 110 (2) can by coordinate conversion matrix T2 and coordinate conversion matrix T1,
It converts to coordinate system C0.The coordinate system C3 of spatial digitizer 110 (3) can be converted by coordinate conversion matrix T3, coordinate
Matrix T2 and coordinate conversion matrix T1, conversion to coordinate system C0.Similarly, the coordinate system CN-1 of spatial digitizer 110 (N-1)
Can by coordinate conversion matrix TN-1, TN-2 ..., conversion is to coordinate system C0.The coordinate system of spatial digitizer 110 (N)
CN can by coordinate conversion matrix TN, TN-1 ..., conversion is to coordinate system C0.In this way, multiple spatial digitizers 110
(0), 110 (1), 110 (2) ..., the instant point cloud data that is captured of 110 (N-2), 110 (N-1), 110 (N), turn by coordinate
After changing, that is, constitute panorama point cloud data.In the example in fig 1, each spatial digitizer 110 (0), 110 (1), 110
(2) ..., 110 (N-2), 110 (N-1), 110 (N) pass through two retrospect paths for origin coordinate system transform to coordinate system C0.In
In another embodiment, each spatial digitizer 110 (0), 110 (1), 110 (2) ..., 110 (N-2), 110 (N-1), 110 (N) also
Path road can be traced by a circulation, circumferentially by origin coordinate system transform to coordinate system C0.For example, spatial digitizer 110
(N) coordinate system CN can by coordinate conversion matrix TN, TN-1 ..., T2, T1, conversion is to coordinate system C0.
Referring to figure 3., the schematic diagram for establishing system 100 of the panorama point cloud data according to an embodiment is illustrated.It establishes
System 100 include at least two spatial digitizers 110 (0), 110 (1) ..., 110 (N), characteristic processing unit 120, coordinate processing
Unit 130 and database 140.Characteristic processing program of the characteristic processing unit 120 to carry out a cloud, coordinates process unit 130
To carry out the calculation procedure of coordinate conversion.Characteristic processing unit 120 and coordinates process unit 130 are, for example, chip, circuit, electricity
The storage device of road plate or storage arrays program code.Database 140 is to store data, e.g. memory, hard disk, cloud
Data center or CD.Also the running of every component of system 100 is established in collocation flow chart, detailed description below.
Referring to figure 4.~flow chart of 9, Fig. 4 diagram according to the method for building up of the panorama point cloud data of an embodiment, Fig. 5
The schematic diagram of each step of~9 pictorial images 4.The method for building up of the panorama point cloud data of the present embodiment includes point cloud data alignment journey
Sequence S1 and point cloud data capture program S2 immediately.In the step S110 of point cloud data alignment procedure S1, example as shown in Figure 5, with
Adjacent two spatial digitizers 110 (0), 110 (1) to the overlapping region OL of spatial digitizer 110 (0), 110 (1) shoot to
Few one refers to geometric object 900, to obtain reference point clouds data CD10, CD20 respectively.It is, for example, cube with reference to geometric object 900
Body, plate or sphere.It can be one or greater than two with reference to the quantity of geometric object 900.As shown in fig. 6, it illustrates reference point
Cloud data CD10.In reference point clouds data CD10, each point of visible object in three-dimensional space is showed.
Then, in the step S120 of point cloud data alignment procedure S1,120 self-reference point cloud data of characteristic processing unit
CD10, CD20 obtain fixed reference feature plane PL.
Step S120 includes multiple detailed procedures, is now obtained from reference point clouds data CD10 with reference to special by the explanation of Fig. 6~9
Levy the process of plane PL.As shown in figs. 6-7, reference point clouds data of the noise stripper 121 of characteristic processing unit 120 to Fig. 6
CD10 filtering noise information, to obtain the reference point clouds data CD11 of Fig. 7.In the reference point clouds data CD11 of Fig. 7, noise is filtered
It removes.
Then, the separator 122 of characteristic processing unit 120 is with region growing algorithm (region growing
Algorithm), according to feature similarity degree, by the imaging point of the reference point clouds data CD11 of Fig. 7 divide group be multiple group G1,
G2。
Then, the separator 122 of characteristic processing unit 120 takes from group G1, G2 of the reference point clouds data CD11 of Fig. 7
Group G1 is as foreground object FO out.Fig. 8 is the reference point clouds data CD12 for retaining foreground object FO.
Then, as shown in Fig. 8~9, the maximum that the withdrawal device 123 of characteristic processing unit 120 extracts each foreground object FO is flat
Face is fixed reference feature plane PL.As shown in figure 8, the imaging point of foreground object FO is transformed into hough space by withdrawal device 123
(hough space), and with the maximum value of sliding window algorithm (Sliding window) search hough space, to obtain maximum
Plane.
Then, in the step S130 of point cloud data alignment procedure S1, coordinates process unit 130 is according to fixed reference feature plane
PL obtains coordinate conversion matrix T1.In this step, the iteration closest approach algorithm of K-D tree framework (K-D tree) can be used
(Iterative Closest Points Algorithm, ICP Algorithm), obtains coordinate conversion matrix T1.
Coordinate conversion matrix T1 can be stored in database 140, be used so that point cloud data captures program S2 immediately.
Then, in step S140, judge whether to obtain all coordinate conversion matrixs.If not yet obtaining all coordinates
Transition matrix is then back to step S110, again execution point cloud data alignment program S1；If having obtained all coordinate conversion squares
Battle array then enters point cloud data and captures program S2 immediately.
In the step S210 that point cloud data captures program S2 immediately, these spatial digitizers 110 (0), 110 (1), 110
(2) ..., 110 (N-2), 110 (N-1), 110 (N) around mobile carrier 800 to shooting, to obtain multiple i.e. time point clouds
Data CD10 ', CD20 ' ....
Then, in the step S220 that point cloud data captures program S2 immediately, coordinates process unit 130 is converted according to coordinate
Matrix TC, T1, T2, T3 ..., TN-1, TN, to these instant point cloud data CD10 ', CD20 ' ... carry out coordinate conversion, and tie
Close instant point cloud data CD10 ', CD20 ' ..., to obtain panorama point cloud data M1 '.
According to above-described embodiment, when establishing system 100 and being mounted on mobile carrier 800 of panorama point cloud data, Ke Yili
A cloud alignment is carried out with point cloud data alignment procedure S1, and carries out point cloud data and captures program S2 immediately, to obtain panorama point cloud
Data M1 '.The accuracy of panorama point cloud data M1 ' is high, and can obtain 0.5~5 meter of short distance situation, works as mobile carrier
When 800 movement speed is 5~10 kilometers or so per hour, quite it is suitably applied in driving/movement auxiliary system.
In another embodiment, coordinate conversion matrix TC, T1, T2, T3 ..., in the acquired situation of TN-1, TN, can
Program S2 is captured immediately directly to carry out point cloud data, and without execution point cloud data alignment program S1 again.Figure 10 is please referred to,
The schematic diagram for establishing system 200 of the panorama point cloud data of diagram according to another embodiment.Establishing system 200 includes multiple three-dimensionals
Scanner 210 (0), 210 (1) ..., 210 (N), coordinate transformation unit 230 and database 240.Coordinate conversion matrix TC, T1,
T2, T3 ..., TN-1, TN be pre-stored in database 240, and without execution point cloud data alignment program S1.Establish system
200 can directly carry out point cloud data captures program S2 immediately, coordinates process unit 230 according to coordinate conversion matrix T1 ..., it is right
These instant point cloud data CD10 ", CD20 " ' ... carry out coordinate conversion, and combine these instant point cloud data CD10 ",
CD20 " ..., to obtain panorama point cloud data M1 ".The accuracy of panorama point cloud data M1 " is high, and can obtain 0.5~5 meter
Short distance situation, when the movement speed of mobile carrier 800 be 5~10 kilometers or so per hour when, be quite suitably applied and drive
It sails/moves in auxiliary system.
Although however, it is not to limit the invention in conclusion the present invention is disclosed as above with embodiment.Institute of the present invention
Belong in technical field and have usually intellectual, without departing from the spirit and scope of the present invention, when various change and profit can be made
Decorations.Therefore, protection scope of the present invention is subject to view appended claims protection scope institute defender.
[symbol description]
100,200: establishing system
110(0)、110(1)、110(2)、110(3)、110(N-1)、110(N-2)、110(N)、210(0)、210(1)、
210 (N): spatial digitizer
120: characteristic processing unit
121: noise stripper
122: separator
123: withdrawal device
130,230: coordinates process unit
140,240: database
800: mobile carrier
900: referring to geometric object
C0, C1, C2, C3, CN-1, CN, C800: coordinate system
CD10, CD20, CD11, CD12: reference point clouds data
CD10 ', CD20 ', CD10 ", CD20 ": instant point cloud data
FO: foreground object
G1, G2: group
M1 ', M1 ": panorama point cloud data
OL: overlapping region
PL: fixed reference feature plane
S110, S120, S130, S140, S210, S220: step
S1: point cloud data alignment procedure
S2: point cloud data captures program immediately
T1, T2, T3, TN-1, TN, TC: coordinate conversion matrix
Claims (18)
1. a kind of method for building up of panorama point cloud data, comprising:
Geometric object is referred to the overlapping region shooting at least one of these spatial digitizers with two adjacent spatial digitizers, with
Obtain two reference point clouds data；
Fixed reference feature plane is obtained from each reference point clouds data；
According to these fixed reference feature planes, coordinate conversion matrix is obtained；
It is shot with these spatial digitizers, to obtain multiple instant point cloud datas；And
According to the coordinate conversion matrix, coordinate conversion is carried out to these instant point cloud datas, and combine these instant point cloud datas,
To obtain panorama point cloud data.
2. the method for building up of panorama point cloud data as described in claim 1, wherein obtaining the reference from each reference point clouds data
The step of characteristic plane includes:
To each reference point clouds data filtering noise information；
Isolate the foreground object of each reference point clouds data；And
The maximum planes for extracting each foreground object are the fixed reference feature plane.
3. the method for building up of panorama point cloud data as claimed in claim 2, wherein isolate each reference point clouds data this before
Scape object includes:
With region growing algorithm, according to feature similarity degree, group is divided to be multiple groups multiple imaging points；And
The foreground object is taken out from these groups.
4. the method for building up of panorama point cloud data as claimed in claim 2, wherein extracting the maximum planes of each foreground object
Include: for the step of fixed reference feature plane
Multiple imaging points of each foreground object are transformed into hough space；And
The maximum value of the hough space is searched for, with sliding window algorithm to obtain the maximum planes.
5. the method for building up of panorama point cloud data as described in claim 1, wherein this refers to geometric object and is cube, puts down
Plate or sphere.
6. the method for building up of panorama point cloud data as described in claim 1, wherein this at least one with reference to geometric object quantity
Greater than two.
7. the method for building up of panorama point cloud data as described in claim 1, wherein the overlapping angle of these spatial digitizers is big
In 10 degree.
8. the method for building up of panorama point cloud data as described in claim 1, wherein the shooting distance of each spatial digitizer is 0.5
~5 meters.
9. the method for building up of panorama point cloud data as described in claim 1, wherein the spatial digitizer system is that ranging passes when flying
Sensor or infrared sensor.
10. a kind of panorama point cloud data establishes system, comprising:
Two spatial digitizers, these spatial digitizers are disposed adjacent and to the overlapping region shootings to these spatial digitizers
At least one refers to geometric object, to obtain two reference point clouds data；
Characteristic processing unit, to obtain fixed reference feature plane from each reference point clouds data；And
Coordinates process unit obtains coordinate conversion matrix, wherein these spatial digitizers are also used according to these fixed reference feature planes
To be shot, to obtain multiple instant point cloud datas, and the coordinates process unit is according to the coordinate conversion matrix, to these
Instant point cloud data carries out coordinate conversion, and combines these instant point cloud datas, to obtain panorama point cloud data.
11. panorama point cloud data as claimed in claim 10 establishes system, wherein this feature processing unit includes:
Noise stripper, to each reference point clouds data filtering noise information；
Separator, to isolate the foreground object of each reference point clouds data；And
Withdrawal device, the maximum planes to extract each foreground object are the fixed reference feature plane.
12. panorama point cloud data as claimed in claim 11 establishes system, wherein the separator is with region growing algorithm, according to
According to feature similarity degree, group is divided to be multiple groups multiple imaging points, and take out the foreground object from these groups.
13. panorama point cloud data as claimed in claim 11 establishes system, wherein the withdrawal device system is by each foreground object
Multiple imaging points are transformed into hough space, and the maximum value of the hough space is searched for sliding window algorithm, to obtain the maximum
Plane.
14. panorama point cloud data as claimed in claim 10 establishes system, wherein this refers to geometric object and is cube, puts down
Plate or sphere.
15. panorama point cloud data as claimed in claim 10 establishes system, wherein the number of at least one reference geometric object
Amount is greater than two.
16. panorama point cloud data as claimed in claim 10 establishes system, wherein the overlapping angle of these spatial digitizers
Greater than 10 degree.
17. panorama point cloud data as claimed in claim 10 establishes system, wherein the shooting distance of each spatial digitizer is
0.5~5 meter.
18. panorama point cloud data as claimed in claim 10 establishes system, wherein the spatial digitizer system is ranging when flying
Sensor or infrared sensor.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW107113877A TWI667529B (en) | 2018-04-24 | 2018-04-24 | Building system and building method for panorama point cloud |
| TW107113877 | 2018-04-24 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110400374A true CN110400374A (en) | 2019-11-01 |
| CN110400374B CN110400374B (en) | 2023-06-13 |
Family
ID=68316166
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810392772.2A Active CN110400374B (en) | 2018-04-24 | 2018-04-27 | Method and system for establishing cloud data of scenic spots |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110400374B (en) |
| TW (1) | TWI667529B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113890955A (en) * | 2021-12-08 | 2022-01-04 | 天远三维（天津）科技有限公司 | Scanning method, device and system of multiple sets of photographing scanners |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI753344B (en) * | 2019-12-30 | 2022-01-21 | 奇景光電股份有限公司 | Hybrid depth estimation system |
| US11132804B2 (en) | 2020-01-07 | 2021-09-28 | Himax Technologies Limited | Hybrid depth estimation system |
| TWI806294B (en) | 2021-12-17 | 2023-06-21 | 財團法人工業技術研究院 | 3d measuring equipment and 3d measuring method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110316963A1 (en) * | 2008-12-30 | 2011-12-29 | Huawei Device Co., Ltd. | Method and device for generating 3d panoramic video streams, and videoconference method and device |
| US20140043436A1 (en) * | 2012-02-24 | 2014-02-13 | Matterport, Inc. | Capturing and Aligning Three-Dimensional Scenes |
| US9269155B2 (en) * | 2012-04-05 | 2016-02-23 | Mediatek Singapore Pte. Ltd. | Region growing method for depth map/color image |
| CN106407947A (en) * | 2016-09-29 | 2017-02-15 | 百度在线网络技术（北京）有限公司 | Target object recognition method and device applied to unmanned vehicle |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004242048A (en) * | 2003-02-06 | 2004-08-26 | Hitachi Ltd | Wide-view angle and high resolution image photographing apparatus |
| US7499586B2 (en) * | 2005-10-04 | 2009-03-03 | Microsoft Corporation | Photographing big things |
| CN101442618A (en) * | 2008-12-31 | 2009-05-27 | 葛晨阳 | Method for synthesizing 360 DEG ring-shaped video of vehicle assistant drive |
| US20120300020A1 (en) * | 2011-05-27 | 2012-11-29 | Qualcomm Incorporated | Real-time self-localization from panoramic images |
| CN103150715B (en) * | 2013-03-13 | 2016-10-19 | 腾讯科技（深圳）有限公司 | Image mosaic processing method and processing device |
| CN104966062B (en) * | 2015-06-17 | 2021-03-23 | 浙江大华技术股份有限公司 | Video monitoring method and device |
| TWI613106B (en) * | 2016-05-05 | 2018-02-01 | 威盛電子股份有限公司 | Method and apparatus for processing surrounding images of vehicle |
| TWI588685B (en) * | 2016-08-31 | 2017-06-21 | 宅妝股份有限公司 | System for building a virtual reality and an augmented reality and method thereof |
| TWI615808B (en) * | 2016-12-16 | 2018-02-21 | 旺玖科技股份有限公司 | Image processing method for immediately producing panoramic images |
-
2018
- 2018-04-24 TW TW107113877A patent/TWI667529B/en active
- 2018-04-27 CN CN201810392772.2A patent/CN110400374B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110316963A1 (en) * | 2008-12-30 | 2011-12-29 | Huawei Device Co., Ltd. | Method and device for generating 3d panoramic video streams, and videoconference method and device |
| US20140043436A1 (en) * | 2012-02-24 | 2014-02-13 | Matterport, Inc. | Capturing and Aligning Three-Dimensional Scenes |
| US9269155B2 (en) * | 2012-04-05 | 2016-02-23 | Mediatek Singapore Pte. Ltd. | Region growing method for depth map/color image |
| CN106407947A (en) * | 2016-09-29 | 2017-02-15 | 百度在线网络技术（北京）有限公司 | Target object recognition method and device applied to unmanned vehicle |
Non-Patent Citations (1)
| Title |
|---|
| 贺辉等: "基于窗口Hough变换与阈值分割的矩形识别算法", 《计算机系统应用》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113890955A (en) * | 2021-12-08 | 2022-01-04 | 天远三维（天津）科技有限公司 | Scanning method, device and system of multiple sets of photographing scanners |
Also Published As
| Publication number | Publication date |
|---|---|
| TWI667529B (en) | 2019-08-01 |
| TW201945824A (en) | 2019-12-01 |
| CN110400374B (en) | 2023-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110411441B (en) | System and method for multi-modal mapping and localization | |
| Cesetti et al. | A visual global positioning system for unmanned aerial vehicles used in photogrammetric applications | |
| Asvadi et al. | 3D object tracking using RGB and LIDAR data | |
| CN110322511B (en) | Semantic SLAM method and system based on object and plane features | |
| US10694103B2 (en) | Building system and building method for panorama point cloud | |
| CN110400374A (en) | The method for building up of panorama point cloud data and establish system | |
| Liu et al. | Automated vehicle extraction and speed determination from QuickBird satellite images | |
| Jiang et al. | Unmanned Aerial Vehicle-Based Photogrammetric 3D Mapping: A survey of techniques, applications, and challenges | |
| CN108917753B (en) | Aircraft position determination method based on motion recovery structure | |
| US11430199B2 (en) | Feature recognition assisted super-resolution method | |
| CN109341686B (en) | Aircraft landing pose estimation method based on visual-inertial tight coupling | |
| CN113256731A (en) | Target detection method and device based on monocular vision | |
| CN114689030A (en) | Unmanned aerial vehicle auxiliary positioning method and system based on airborne vision | |
| Sakai et al. | Large-scale 3D outdoor mapping and on-line localization using 3D-2D matching | |
| JP6155091B2 (en) | Mosaic image generation apparatus, generation method, and mosaic image generation program | |
| KR102249381B1 (en) | System for generating spatial information of mobile device using 3D image information and method therefor | |
| CN116508071A (en) | System and method for annotating automotive radar data | |
| CN111024063A (en) | Star map recognition algorithm under large-mobility condition based on star point re-extraction | |
| KR20220062709A (en) | System for detecting disaster situation by clustering of spatial information based an image of a mobile device and method therefor | |
| Tang et al. | Automatic geo‐localization framework without GNSS data | |
| JP7468254B2 (en) | Position and orientation calculation method and program | |
| KR102249380B1 (en) | System for generating spatial information of CCTV device using reference image information | |
| CN117523461B (en) | Moving target tracking and positioning method based on airborne monocular camera | |
| CN117649619B (en) | Unmanned aerial vehicle visual navigation positioning recovery method, system, device and readable storage medium | |
| Yang et al. | Rotated fusion network (RFN) algorithm for aerial vehicle detection |
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 |