CN104408055B - Storage method and device for laser radar point cloud data - Google Patents
Storage method and device for laser radar point cloud data Download PDFInfo
- Publication number
- CN104408055B CN104408055B CN201410592173.7A CN201410592173A CN104408055B CN 104408055 B CN104408055 B CN 104408055B CN 201410592173 A CN201410592173 A CN 201410592173A CN 104408055 B CN104408055 B CN 104408055B
- Authority
- CN
- China
- Prior art keywords
- scan line
- laser
- laser spots
- coordinate
- detected
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000009466 transformation Effects 0.000 claims abstract description 50
- 238000001514 detection method Methods 0.000 claims abstract description 16
- 230000003139 buffering effect Effects 0.000 claims description 46
- 238000007906 compression Methods 0.000 claims description 9
- 230000006835 compression Effects 0.000 claims description 9
- 238000004321 preservation Methods 0.000 claims description 8
- 241001269238 Data Species 0.000 claims description 4
- 230000000630 rising effect Effects 0.000 claims description 2
- 238000012545 processing Methods 0.000 abstract description 10
- 230000008859 change Effects 0.000 abstract description 6
- 230000035772 mutation Effects 0.000 description 4
- 238000013500 data storage Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- ATJFFYVFTNAWJD-UHFFFAOYSA-N Tin Chemical compound [Sn] ATJFFYVFTNAWJD-UHFFFAOYSA-N 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 238000013144 data compression Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000002592 echocardiography Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 231100000350 mutagenesis Toxicity 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000010408 sweeping Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T1/00—General purpose image data processing
- G06T1/60—Memory management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/60—Rotation of whole images or parts thereof
- G06T3/604—Rotation of whole images or parts thereof using coordinate rotation digital computer [CORDIC] devices
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Optical Radar Systems And Details Thereof (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The application discloses a storage method and a device of laser radar point cloud data, wherein the storage method comprises the following steps: performing coordinate transformation on each laser point in the original laser point cloud data; detecting the laser point after coordinate transformation, and judging whether the currently detected laser point is the starting point of a new scanning line or not by utilizing the change of the coordinate value of the laser point; when the currently scanned laser point is judged to be the starting point of a new scanning line, storing the laser point cloud data on the previous scanning line into a buffer area and carrying out coordinate inverse transformation on each laser point on the previous scanning line in the buffer area; establishing a spatial index for the scanning line to which the laser point subjected to the coordinate inverse transformation belongs and storing the spatial index; and processing the next scanning line, and repeating the steps until the laser point cloud data on all the scanning lines are completely stored. The device mainly comprises a coordinate transformation unit, a detection judgment unit, a cache unit and a storage unit. Through the technical scheme in the application, the efficiency of subsequent data processing can be improved.
Description
Technical field
The application is related to laser radar data process field, more particularly to a kind of storage method of laser radar point cloud data
And device.
Background technology
Three-dimensional laser scanning technique (Light Detection And Ranging, LiDAR) is a kind of advanced full-automatic
High Precision Stereo scanning technique, research and development of the technology originating from Nasa in 1970, is sent out rapidly the 1980s
Exhibition, 20 end of the centurys, survey field have also started the research boom of three-dimensional laser scanning technique, and sweep object is more and more, application
Field is more and more wider.According to three-dimensional laser scanning system according to carrying platform divide, can be divided into airborne three-dimensional laser scanning system,
Vehicle-mounted three-dimensional laser scanning system and Three Dimensional Ground laser scanning system.Wherein, Three Dimensional Ground laser scanning system is mainly directed towards
The three-dimensional modeling of high-precision reverse-engineering and reconstruct, it can efficiently gather a large amount of three-dimensional coordinate points of ground object, will be each
Large-scale, complicated, the irregular outdoor scene three-dimensional data of kind is intactly collected in computer, so as to which quick reconfiguration goes out the three of target
Tie up point cloud model.The three dimensional point cloud gathered can be widely applied to survey and draw, measure, analyzes, emulate, simulation shows monitoring,
The fields such as virtual reality.
At present, the Three Dimensional Ground laser scanning system gathered data more by the way of line scanning, using line by line or by column
Scan mode gathered data, the three-dimensional laser point cloud data collected have certain structural relation, but at present to being gathered
To three-dimensional laser point cloud data storage method in most of pyramid piecemeal, this simple place are only carried out to massive point cloud
Reason is advantageous to the display of massive point cloud, but because the point cloud in each data tile is disorderly and unsystematic, lacks between points any
Topological relation, therefore this method is unfavorable for follow-up data processing.
The content of the invention
In order to overcome disadvantages described above of the prior art, this application provides a kind of follow-up data treatment effeciency of improving
The storage method and device of laser radar point cloud data.
In order to realize above technical purpose, the application is achieved through the following technical solutions：
This application provides a kind of storage method of laser radar point cloud data, the storage method includes：
S1, according to the original laser cloud data gathered, to the laser spots in original laser point cloud in each scan line
Coordinate transform is carried out, obtains the coordinate value of all laser spots after coordinate transform；
S2, detection carry out coordinate transform after laser spots coordinate value, by the coordinate value of the laser spots currently detected with
Whether the laser spots that the coordinate value of the laser spots formerly detected is contrasted to judge currently to be detected are rising for new scan line
Initial point；
S3, when the laser spots for judging currently to be detected are not the starting points of new scan line, then by after the coordinate transform
Present laser point coordinate data of the coordinate value as the laser spots on current scan line；
, then will be current after the coordinate transform in starting point of the laser spots for judging currently to be detected as new scan line
Coordinate data of the coordinate value of laser spots as starting point in new scan line, and by the point of all laser spots on current scan line
Cloud data are stored in buffering area, and each laser spots on current scan line in the buffering area are carried out with the inversion of the coordinate transform
Change and preserve, and using the new scan line as current scan line；
S4, continue to hold using next laser spots of the laser spots currently detected on current scan line as present laser point
Row step S2 to S3, finished until all original laser cloud datas preserve.
Preferably, the laser spots in the point cloud to original laser in each scan line, which carry out coordinate transform, includes：
The three-dimensional rectangular coordinate (x, y, z) of each laser spots in original laser point cloud is converted to 3 D laser scanning
Spherical coordinate (r, θ, φ) centered on instrument, wherein r are distance of the laser spots to scanner, and θ is the folder of laser rays and horizontal plane
Angle, φ are to horizontally rotate angle when scanner works.
Preferably, the coordinate value of laser spots of the coordinate value by the laser spots currently detected with formerly being detected enters
Whether row contrast is that the starting point of new scan line includes the laser spots that judge currently to be detected：
The coordinate value of laser spots of the coordinate value of the laser spots currently detected with formerly being detected is contrasted to sentence
Whether the disconnected laser spots currently detected can be as the pre- reconnaissances of the starting point of new scan line；
After the laser spots for judging currently to be detected can be as the pre- reconnaissance of the starting point of new scan line, determine whether
The laser spots currently detected whether be new scan line starting point.
Preferably, the coordinate value of laser spots of the coordinate value by the laser spots currently detected with formerly being detected enters
Row contrast is come the laser spots that judge currently to be detected to include as the starting point of new scan line：
By regarding the laser point cloud data of same scan line as stable ergodic time series, with operator mould
Plate detects the magnitude relationship between the θ values of the laser spots currently detected and the θ values of a upper laser spots to judge current institute
Whether the laser spots of detection can be as the pre- reconnaissances of the starting point of new scan line；
After the laser spots for judging currently to be detected can be as the pre- reconnaissance of the starting point of new scan line, determine whether
The laser spots currently detected whether be new scan line starting point.
Preferably, whether the laser spots for determining whether currently to be detected are that the starting point of new scan line includes：
Judge to work as by the magnitude relationship between the θ values for the follow-up N number of laser spots for detecting the laser spots currently detected
Preceding detected laser spots, and/or, by by all laser spots on the φ values of the laser spots currently detected and current scan line
φ values the average value laser spots that compare to judge currently to be detected whether be new scan line starting point, wherein N is just
Integer.
Preferably, the cloud data deposit buffering area of all laser spots by current scan line, to the buffering
Each laser spots in area on current scan line carry out the inverse transformation of the coordinate transform and preservation includes：
Coordinate data in laser point cloud data on current scan line and attribute data are stored in buffering area respectively, then
The inverse transformation of the coordinate transform is carried out to the coordinate data of each laser spots on current scan line in the buffering area；
It is current after the inverse transformation of the coordinate data progress coordinate transform to each laser spots on current scan line
Scan line is established spatial index table and preserved.
Preferably, the coordinate data of each laser spots on current scan line carries out the inversion of the coordinate transform
After changing, establishing spatial index table for current scan line and carry out preservation includes：
Coordinate data after the inverse transformation of the coordinate transform is carried out in buffering area and attribute data are compressed；
After being compressed to the coordinate data in buffering area and attribute data, encoded to current scan line, currently swept
Retouch line and correspond to a unique numbering；
Spatial index table is established for the current scan line after being numbered, the keyword of the spatial index table is currently to sweep
The numbering of line is retouched, index entry value corresponding with keyword is laser point cloud data on current scan line in the spatial index table
Storage address；
After spatial index table is established to current scan line, by the number of coordinates of the buffering area after being compressed on current scan line
Preserved according to the storage address write back with attribute data and in spatial index table as corresponding to current scan line.
Present invention also provides a kind of storage device of laser radar point cloud data, the storage device includes,
Coordinate transformation unit, for according to the original laser cloud data gathered, will each be swept in original laser point cloud
The laser spots retouched on line carry out coordinate transform, obtain the coordinate value of all laser spots after coordinate transform；
Judging unit is detected, for detecting the coordinate value of the laser spots after carrying out coordinate transform, utilizes what is currently detected
Difference between laser point coordinates value and the laser point coordinates value formerly detected judges whether is the laser spots that are currently detected
For the starting point of new scan line；
Buffer unit, for the laser point cloud data in scan line to be stored in into buffering area and in scan line in buffering area
Each laser spots carry out the inverse transformation of the coordinate transform；
Memory cell, for being protected after carrying out the inverse transformation of the coordinate transform to the laser point cloud data in scan line
Deposit.
Preferably, the coordinate transformation unit includes coordinate transform subelement, and the coordinate transform subelement is used for original
The three-dimensional rectangular coordinate (x, y, z) of each laser spots in beginning laser point cloud data is converted to centered on three-dimensional laser scanner
Spherical coordinate (r, θ, φ), wherein r is the distance that laser spots arrive scanner, and θ is the angle of laser rays and horizontal plane, and φ is to sweep
Horizontally rotate angle when retouching instrument work.
Preferably, the detection judging unit includes the first subelement and the second subelement, and first subelement is used for
Pass through the laser spots changed to judge currently to be detected of the θ values for the follow-up N number of laser spots for detecting the laser spots currently detected
Whether be new scan line starting point, wherein N is positive integer；Second subelement is used for the laser by will currently be detected
The average value of the φ values of point and the φ values of all laser spots on current scan line compares come the laser spots for judging currently to be detected
Whether be new scan line starting point.
Preferably, the buffer unit includes：
Coordinate data buffer unit, the coordinate data of all laser spots on cache sweep line；
Attribute data buffer unit, the attribute data of all laser spots on cache sweep line；
Coordinate inversion unit, for the coordinate of all laser spots in coordinate data buffer unit to be carried out into coordinate inversion
Change.
Preferably, the memory cell includes：
Compression unit, for being compressed to the coordinate data and attribute data that are stored in buffering area；
Coding unit, after being compressed for the coordinate data to buffering area and attribute data, carried out to each scan line
Coding, each scan line correspond to a unique numbering；
Indexing units, for establishing spatial index table for each scan line after being numbered, the spatial index table
Keyword is the numbering of scan line, and index entry numerical value corresponding with keyword is to swash in each scan line in the spatial index table
The storage address of light cloud data；
Writeback unit, after establishing spatial index table to each scan line, after being compressed in each scan line
The coordinate data and attribute data of buffering area are write back and the storage in the spatial index table as corresponding to each scan line
Location is preserved.
Compared with prior art, the application has the advantages that：
Original laser cloud data is stored by scan line for elementary cell, this improves the effect of follow-up data processing
Rate.
Brief description of the drawings
, below will be to embodiment or existing in order to illustrate more clearly of the embodiment of the present application or technical scheme of the prior art
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments described in application, for those of ordinary skill in the art, do not paying the premise of creative labor
Under, other accompanying drawings can also be obtained according to these accompanying drawings.
Fig. 1 is the schematic flow sheet of the storage method of laser radar point cloud data in the embodiment of the present application.
Fig. 2 is the schematic flow sheet of the storage method of laser radar point cloud data in the embodiment of the present application one.
Fig. 3 is three-dimensional rectangular coordinate and the graph of a relation of spherical coordinate.
Fig. 4 is the θ Distribution value figures of laser spots on different scanning line.
Fig. 5 A-5B are respectively the two kinds of operator template schematic diagrames defined.
Fig. 6 is the φ Distribution value figures of laser spots on different scanning line.
Fig. 7 is the structural representation of the storage device of laser radar point cloud data in the embodiment of the present application two.
Embodiment
In order that those skilled in the art more fully understand the technical scheme in the application, it is real below in conjunction with the application
The accompanying drawing in example is applied, the technical scheme in the embodiment of the present application is clearly and completely described, it is clear that described implementation
Example only some embodiments of the present application, rather than whole embodiments.It is common based on the embodiment in the application, this area
The every other embodiment that technical staff is obtained under the premise of creative work is not made, it should all belong to the application protection
Scope.
The application uses the technical scheme of scan line detection, and original laser cloud data is entered by scan line for elementary cell
Row storage, as shown in figure 1, comprising the following steps：
S1, according to the original laser cloud data gathered, to the laser spots in original laser point cloud in each scan line
Coordinate transform is carried out, obtains the coordinate value of all laser spots after coordinate transform；
S2, detection carry out coordinate transform after laser spots coordinate value, by the coordinate value of the laser spots currently detected with
Whether the laser spots that the coordinate value of the laser spots formerly detected is contrasted to judge currently to be detected are in new scan line
Starting point；
S3, when the laser spots for judging currently to be detected are not the starting points in new scan line, then by the coordinate transform
Coordinate data of the coordinate value of present laser point afterwards as the laser spots on current scan line；
, then will be current after the coordinate transform in laser spots of the laser spots for judging currently to be detected as new scan line
Coordinate data of the coordinate value of laser spots as starting point in new scan line, and by the point of all laser spots on current scan line
Cloud data are stored in buffering area, and each laser spots on current scan line in the buffering area are carried out with the inversion of the coordinate transform
Change and preserve, and using the new scan line as current scan line；
S4, the current next laser spots for detecting laser spots on current scan line are continued executing with as present laser point
Step S2 to S3, finished until all original laser cloud datas preserve.
Compared with the cloud data storage method of routine, the sequential storage cloud data of scan line is pressed in the application, is kept
The spatial structural form of point cloud, is also convenient for follow-up data processing.
Describe the specific implementation of the embodiment of the present application in detail with specific example below.
Embodiment one
Fig. 2 be the present embodiment in laser radar point cloud data storage method schematic flow sheet, the storage method institute pin
To be three-dimensional coordinate point that Three Dimensional Ground laser scanning system gathers using scan mode line by line or by column ground object,
These three-dimensional coordinate points are that progress space arrangement, the storage method comprise the following steps in the form of line scanning element cloud：
Step S101, according to the original laser cloud data collected, in original laser point cloud in each scan line
Laser spots carry out coordinate transform, obtain carry out coordinate transform after each laser spots coordinate value.
In the present embodiment, by the coordinate of the laser spots in the original laser point cloud that is collected in each scan line is
Three-dimensional rectangular coordinate (x, y, z), it is advantageous to the three-dimensional rectangular coordinate (x, y, z) by each laser spots be converted to three-dimensional swash
Spherical coordinate (r, θ, φ) centered on photoscanner, wherein r are distance of the laser spots to scanner, and θ is laser rays and level
The angle in face, φ are to horizontally rotate angle when scanner works, as shown in Figure 3.The spherical coordinate (r, θ, φ) and three of laser spots
The transformational relation tieed up between rectangular co-ordinate (x, y, z) is as follows：
Above-mentioned coordinate transform is not limited to the conversion between three-dimensional rectangular coordinate and spherical coordinate, here by the three of laser spots
Dimension rectangular coordinates transformation is that spherical coordinate is due to the end point of current scan line and next scan line in spherical coordinate
Very big difference is had between the θ values of starting point, the laser spots for being easy to judge currently to be detected using the mutation of θ values belong to which bar is swept
Retouch line.The coordinate of laser spots is transformed to spherical coordinate it will be apparent that being not intended to limit in the present embodiment, as long as being readily susceptible to judge currently
The laser spots detected belong to which bar scan line, such as cylindrical coordinates.
Step S102, detection carries out the coordinate value of the laser spots after coordinate transform, by the seat of the laser spots currently detected
Scale value and the coordinate value of first laser spots, which are contrasted, comes whether the preliminary laser spots for judging currently to be detected are new scan line
The pre- reconnaissance of starting point.
When a new scan line starts, thereon the coordinate value of laser spots relative to laser spots in a upper scan line seat
Scale value can difference, the coordinate value in one of direction can produce a very big mutation, therefore can utilize this feature,
The coordinate value of the laser spots currently detected and the coordinate value of a upper laser spots are contrasted, it is current so as to tentatively to judge
The laser spots detected whether be starting point in new scan line pre- reconnaissance.Here starting point refers in this scan line
First laser spots.
Coordinate for laser spots is this situation of spherical coordinate, when a new scan line starts, the θ of laser spots thereon
Value can produce mutation relative to the θ values of laser spots in a upper scan line, using this feature, the laser spots that will be currently detected
The θ values of θ values and a upper laser spots contrasted, whether be newly to scan so as to carry out the preliminary laser spots for judging currently to be detected
The pre- reconnaissance of the starting point of line.By taking the 1st article of scan line and the 2nd article of scan line shown in Fig. 4 as an example, figure 4, it is seen that
The θ values of the starting point of 2nd article of scan line generate mutation relative to the θ values of end point in the 1st article of scan line, will currently be detected
Laser spots θ values compared with the θ values of the upper laser spots detected, can be tentatively to judge the laser currently detected
Point whether be the 2nd article of scan line starting point pre- reconnaissance.
The θ value changes of detection laser spots have a variety of methods, are continuous by three-dimensional laser scanner in the present embodiment
Caused data regard an ergodic stable time series as during work, will return the problem of finding scan line starting point
The problem of becoming the catastrophe point of hunting time sequence, become using a kind of method for sliding operator template to detect the θ values of laser spots
Change, the definition of operator template continuously slips over the θ values of laser point cloud sequence using operator template defined in Fig. 5 A referring to Fig. 5 A.
Define the θ values that θ (A), θ (B), θ (C) are respectively tri- opening positions of A, B, C in operator template.Can from Fig. 3
Go out the linear decline trend of θ values of laser spots in same scan line, and the θ values of starting point can be much larger than in next scan line
The θ values of end point on current scan line, starting point and end point here refer to first laser spots in next scan line respectively
And last laser spots on current scan line.Therefore, if 3 points of A, B, C is in same scan line, and this 3 points
It is arranged in order from front to back, then when operator template is slided in this scan line, θ (B)-θ (A) should be negative value, and θ
(C)-θ (B) also should be negative value.And if 2 points of A, B in same scan line, the position of A points is before B points, and C points
In next scan line, then θ (B)-θ (A) is negative value, and θ (C)-θ (B) be on the occasion of, and | θ (C)-θ (B) |>>|θ
(B)-θ (A) |, wherein, symbol ">>" expression " being far longer than ".That is, when operator template is slided along scan line, it is such as full
Foot state formula (2) then can principium identification operator template slided into the end of current scan line, the position corresponding to C points is
The pre- reconnaissance of the starting point of new scan line.
Wherein, k is constant, represents the saltus step degree of θ values.
If in the step when the laser spots for judging currently to be detected are not the pre- reconnaissances of starting point of new scan line,
Coordinate data using the coordinate value of present laser point as the laser spots on current scan line, then proceedes to detect next laser
Point, until that can judge that detected laser spots can be as the pre- reconnaissance of the starting point in new scan line, then perform following step
Rapid S103.
Step S103, after pre- reconnaissance of the preliminary laser spots for judging currently to be detected as the starting point of new scan line, enter
One step determines whether the laser spots can be as the starting points of new scan line.
By not only having laser spots in the original laser cloud data that is collected, also there are because the reasons such as hardware are made
Into noise spot, so whether the laser spots that only can not also determine currently to be detected completely by above-mentioned detection are new scan line
Starting point, therefore, after mutation occur in the θ values for detecting present laser point, it is also necessary to further determine that that is currently detected swashs
Whether luminous point is noise spot, if noise spot, then rejects the noise spot, continues to detect next laser spots；If not
Noise spot, then be assured that the starting point that the laser spots currently detected are new scan line, this can be by following two
Method is realized：
(1) follow-up N number of laser spots of present laser point are detected, N is positive integer, in order to ensure follow-up N number of laser spots are equal
In same scan line, so N value can not be too big.If present laser point is the starting point of next scan line, and
It is not noise spot, then the θ values of follow-up N number of laser spots should be " almost " monotone decreasing.In the present embodiment, with Fig. 5 B
Defined operator template detects follow-up N number of laser spots (N 5), calculates the θ differences between two adjacent laser points,
Under the operator template, the θ differences between 10 pairs of laser spots can be calculated, in the ideal case, the θ differences of this 10 pairs of laser spots are all
Negative is should be, but due to the presence of noise spot, as long as the θ differences of most of " put to " is negative, defined formula (3) is such as
Under：
In this 5 adjacent point, one shares 10 pairs of points by ascending order arrangement to (Ni,Nj|j>I), calculated with formula (3)
This 10 pairs of points.Preferably, if f (Ni,Nj) for 1 number it is more than or equal to 7, it is possible to think that follow-up 5 points " almost " are single
Tune successively decreases, and at this moment may determine that present laser point is not noise spot, so as to the starting point using the point as new scan line, if f
(Ni,Nj) for 1 number be less than 7, then it is considered that the laser spots of current detection are noise spots, reject the noise spot.Here with
f(Ni,Nj) relation between 1 number and 7 is measurement index, simply consider for experience, be not limited to this in the application
Value.Declare herein, present laser point refers both to the laser spots currently detected in this application.
(2) the φ values of present laser point are detected, and by all laser in the φ values of present laser point and a upper scan line
The average value of the φ values of point compares, if the scan line detected is the 1st article of scan line, then just by the φ of present laser point
The average value of value and the φ values of all laser spots previously detected compares, so as to determine that the laser spots currently detected are
It is no can be as the starting point of new scan line.In the ideal case, in same scan line, the level of three-dimensional laser scanner turns
Dynamic angle φ should be consistent, but often due to hardware reason, φ values are simultaneously non-critical completely the same, but assume that same is swept
The φ values for retouching laser spots on line are to meet the random sequence of normal distribution, as shown in Figure 6.If present laser point is new scan line
Starting point, then the φ values of the current laser spots should beNear, wherein φoldFor the φ values of a upper scan line
Average value,For the average value of the horizontal sextant angle of adjacent two scan lines.When the φ values of present laser point meet following formula
(4) during condition, then the starting point that present laser point is next scan line is judged, otherwise as noise points deleting.
ε is the stochastic variable for meeting normal distribution in above formula, i.e. ε~N (0, σ2)。
In order to ensure the real cancelling noise point of energy, it is necessary to which above two method is used in combination, but can also be according to reality
Situation, only determine starting point that the laser spots that are currently detected are new scan line from one of which method.
The application is can be seen that from step S103 while laser spots are detected and also eliminates noise spot, that is, realizes filter
Wave energy, so in follow-up Data processing, avoid the need for being filtered processing again, so as to improve follow-up data processing
Efficiency.
If it is not the starting point of new scan line by the laser spots for finding currently to be detected after the step, then just ought
Coordinate data of the coordinate value of preceding detected laser spots as current laser spots on current scan line, and continue to detect next
Laser spots are untill it can determine that the starting point that detected laser spots are new scan line, then perform following step S104.
S104, when it is determined that the laser spots currently detected are the starting point of new scan line, then by the seat of present laser point
Coordinate data of the scale value as starting point in new scan line, and the cloud data of all laser spots on current scan line is stored in
Buffering area, each laser spots on current scan line in buffering area are carried out with the inverse transformation of coordinate transform.
It is determined that the laser spots currently detected for new scan line starting point after, by all laser on current scan line
The cloud data deposit buffering area of point, in the present embodiment, cloud data is primarily referred to as coordinate data and attribute data, number of coordinates
According to the coordinate value for laser spots, attribute data includes laser intensity, color value, number of echoes etc.；Then institute in step S101 is utilized
The inverse transformation of coordinate transform is stated, the spherical coordinate transformation of all laser spots on current scan line is returned into three-dimensional rectangular coordinate.
S105, after each laser spots in buffering area on current scan line carry out the inverse transformation of coordinate transform, to slow
The cloud data for rushing laser spots on current scan line in area is compressed, and is then established spatial index table for current scan line and is write
Return and preserve.
After the Coordinate Conversion of laser spots is returned into three-dimensional rectangular coordinate, to the point of laser spots on current scan line in buffering area
Cloud data are compressed.In the present embodiment, it is that cloud data is compressed using zlib, zlib is a kind of offer increased income
The software library of compression function, it provides very easily interface, is new slow by output after the cloud data compression in buffering area
Area's data are rushed, while also return to data length and data address after compression.Cloud data is entered using zlib in this example
Row compression detailed process is as follows：First, by the coordinate data of laser spots and attribute data separate on current scan line in buffering area,
It is respectively present in two single buffering areas, then calls the function in zlib storehouses to be compressed coordinate data and attribute data,
New coordinate data and attribute data are exported, while returns to the data length and data of coordinate data and attribute data after compression
Address.
After the cloud data of laser spots is compressed on current scan line in buffering area, established for current scan line empty
Between concordance list and write back preservation.The spatial index table to establish process as follows：First, encoded to current scan line, this is swept
Retouch corresponding one unique numbering of line；Then concordance list is established, the keyword of the concordance list is the numbering of current scan line, index
Index entry numerical value corresponding with keyword is the storage address of the cloud data of laser spots on current scan line in table.
After spatial index table is established for current scan line, by the cloud data of laser spots on the current scan line after compression
Write back and the storage address in the spatial index table as corresponding to each scan line is preserved.In the present embodiment, it is main
If the coordinate data of laser spots on current scan line and attribute data are carried out respectively to write back preservation.
By establishing spatial index table for scan line, only needed in follow-up data processing in two adjacent scan lines
Carry out triangle structure, so as to avoid establish irregular triangular mesh (TIN) with Delaunay algorithms during institute face
Face in dispersion point cloud search closest to the problem of, and then improve follow-up data processing efficiency.
S106, next scan line is handled, above step S102 to S105 is repeated, until all in original laser point cloud
Scan line is disposed.
Specifically, it is using new scan line as current scan line, by next laser spots of starting point in new scan line
As present laser point, then proceed to perform step S102 to S105, that is, detect present laser point and judge that present laser point is
The no starting point for new scan line, it is determined that after the starting point of new scan line, and by the point of all laser spots on current scan line
Cloud data deposit buffering area carries out the inverse transformation of coordinate transform, and then the cloud data in buffering area is compressed, is current
Scan line establishes spatial index table, and finally the cloud data after compression is carried out to write back preservation.Repeat above step S102 extremely
S105 is until the preservation of all original laser cloud datas finishes.
In this example, it is to have detected to enter the cloud data in this scan line after all laser spots on current scan line
Row preserves, and then detects next scan line again, if needed can also be while cloud data on preserving current scan line
Detect next scan line.
, can be easily according to space digital product institute using the storage method of laser radar point cloud data in the present embodiment
The precision specified, carry out interlacing extraction or vacuated every M rows, M is the positive integer more than 1, so as to improve at follow-up data
The efficiency of reason.In addition, by adding spatial structural form to discrete laser point cloud, so as to improve follow-up three-dimensional modeling
Efficiency.
Embodiment two
The embodiment of the present application additionally provides a kind of storage device of laser radar point cloud data, as shown in fig. 7, the storage fills
Put including：Coordinate transformation unit 201, detect judging unit 202, buffer unit 203 and memory cell 204.Wherein, coordinate becomes
Unit 201 is changed, for according to the original laser cloud data gathered, by the laser in original laser point cloud in each scan line
Point carries out coordinate transform, obtains the coordinate value of all laser spots after coordinate transform.The coordinate transformation unit 201 also includes coordinate and become
Subelement 2011 is changed, the coordinate transform subelement 2011 is used for the three-dimensional straight of each laser spots in original laser cloud data
Angular coordinate (x, y, z) is converted to the spherical coordinate (r, θ, φ) centered on three-dimensional laser scanner, and wherein r is laser spots to sweeping
The distance of instrument is retouched, θ is the angle of laser rays and horizontal plane, and φ is to horizontally rotate angle when scanner works.
The coordinate value that judging unit 202 is used to detect the laser spots after carrying out coordinate transform is detected, is detected using current
Laser point coordinates value and the laser point coordinates value that is formerly detected between difference be to judge the laser spots currently detected
The no laser spots in new scan line.The detection judging unit 202 includes the first subelement 2021, and first subelement 2021 is used
In the laser changed to judge currently to be detected of the θ values of follow-up N number of laser spots by detecting the laser spots currently detected
Whether point can be the starting point of new scan line, and wherein N is positive integer.The second son can also also be included by detecting judging unit 202
Unit 2022, second subelement 2022 are used for by by institute in the φ values of the laser spots currently detected and a upper scan line
Whether the laser spots that the average value for having the φ values of laser spots compares to judge currently to be detected can be the starting of new scan line
Point.
Buffer unit 203 is used to the laser point cloud data in scan line being stored in buffering area and to scan line in buffering area
Each laser spots carry out coordinate transformation unit 201 in coordinate transform inverse transformation.The buffer unit 203 includes following three sons
Unit：Coordinate data buffer unit 2031, attribute data buffer unit 2032 and coordinate inversion unit 2033, these three sons
Unit is respectively used to the attribute of the coordinate data of all laser spots on cache sweep line, all laser spots on cache sweep line
Data and the coordinate progress coordinate inversion by all laser spots in coordinate data buffer unit.
Memory cell 204, for being protected after carrying out the inverse transformation of coordinate transform to the laser point cloud data in scan line
Deposit.The memory cell 204 includes following four subelement：
Compression unit 2041, for being compressed to the coordinate data and attribute data that are stored in buffering area；
Coding unit 2042, after being compressed for the coordinate data to buffering area and attribute data, to each scan line
Encoded, each scan line corresponds to a unique numbering；
Indexing units 2043, for establishing spatial index table for each scan line after being numbered, the spatial index table
Keyword be scan line numbering, index entry numerical value corresponding with keyword is to swash in each scan line in the spatial index table
The storage address of light cloud data；
Writeback unit 2044, after establishing spatial index table to each scan line, it will be compressed in each scan line
The coordinate data and attribute data of buffering area afterwards are write back and depositing in the spatial index table as corresponding to each scan line
Storage address is preserved.
Although depicting the application by embodiment, it will be appreciated by the skilled addressee that the application have it is many deformation and
Change is without departing from spirit herein, it is desirable to which appended claim includes these deformations and changed without departing from the application's
Spirit.
Claims (12)
- A kind of 1. storage method of laser radar point cloud data, it is characterised in that including：S1, according to the original laser cloud data gathered, the laser spots in original laser point cloud in each scan line are carried out Three-dimensional coordinate transformation, obtain the coordinate value of all laser spots after three-dimensional coordinate transformation；S2, detection carry out three-dimensional coordinate transformation after laser spots coordinate value, by the coordinate value of the laser spots currently detected with Whether the laser spots that the coordinate value of the laser spots formerly detected is contrasted to judge currently to be detected are rising for new scan line Initial point；S3, when the laser spots for judging currently to be detected are not the starting points of new scan line, then by after the three-dimensional coordinate transformation Present laser point coordinate data of the coordinate value as the laser spots on current scan line；, then will be current after the three-dimensional coordinate transformation in starting point of the laser spots for judging currently to be detected as new scan line Coordinate data of the coordinate value of laser spots as starting point in new scan line, and by the point of all laser spots on current scan line Cloud data are stored in buffering area, and the three-dimensional coordinate transformation is carried out to each laser spots on current scan line in the buffering area Inverse transformation simultaneously preserves, and using the new scan line as current scan line；S4, step is continued executing with using next laser spots of the laser spots currently detected on current scan line as present laser point Rapid S2 to S3, is finished until all original laser cloud datas preserve.
- 2. the storage method of laser radar point cloud data according to claim 1, it is characterised in that described to original laser Laser spots in point cloud in each scan line, which carry out three-dimensional coordinate transformation, to be included：By the three-dimensional rectangular coordinate (x, y, z) of each laser spots in original laser point cloud be converted to using three-dimensional laser scanner as The spherical coordinate (r, θ, φ) at center, wherein r are the distance that laser spots arrive scanner, and θ is the angle of laser rays and horizontal plane, φ Horizontally rotate angle when being worked for scanner.
- 3. the storage method of laser radar point cloud data according to claim 2, it is characterised in that described currently to be examined The coordinate value of the laser spots of survey and the coordinate value of the laser spots formerly detected are contrasted the laser to judge currently to be detected Whether point is that the starting point of new scan line includes：The coordinate value of laser spots of the coordinate value of the laser spots currently detected with formerly being detected is contrasted to judge to work as Whether preceding detected laser spots can be as the pre- reconnaissances of the starting point of new scan line；After the laser spots for judging currently to be detected can be as the pre- reconnaissance of the starting point of new scan line, determine whether current The laser spots detected whether be new scan line starting point.
- 4. the storage method of laser radar point cloud data according to claim 3, it is characterised in that described currently to be examined The coordinate value of the laser spots of survey and the coordinate value of the laser spots formerly detected are contrasted the laser to judge currently to be detected Point includes for the starting point as new scan line：By regarding the laser point cloud data of same scan line as stable ergodic time series, with operator template come The magnitude relationship between the θ values of laser spots and the θ values of a upper laser spots currently detected is detected to judge currently to be detected Laser spots whether be new scan line starting point pre- reconnaissance；After the laser spots for judging currently to be detected is the pre- reconnaissances of the starting point of new scan line, determine whether currently to be detected Laser spots whether be new scan line starting point.
- 5. the storage method of the laser radar point cloud data according to claim 3 or 4, it is characterised in that described further Judge whether the laser spots that are currently detected are that the starting point of new scan line includes：Current institute is judged by the magnitude relationship between the θ values for the follow-up N number of laser spots for detecting the laser spots currently detected The laser spots of detection, and/or, by by the φ of all laser spots on the φ values of the laser spots currently detected and current scan line The laser spots that the average value of value compares to judge currently to be detected whether be new scan line starting point, wherein N is positive integer.
- 6. the storage method of laser radar point cloud data according to claim 1 or 2, it is characterised in that described by currently The cloud data deposit buffering area of all laser spots in scan line, to each laser on current scan line in the buffering area Point carries out the inverse transformation of the three-dimensional coordinate transformation and preservation includes：Coordinate data in laser point cloud data on current scan line and attribute data are stored in buffering area respectively, then to institute The coordinate data for stating each laser spots in buffering area on current scan line carries out the inverse transformation of the three-dimensional coordinate transformation；It is current after the inverse transformation of the coordinate data progress three-dimensional coordinate transformation to each laser spots on current scan line Scan line is established spatial index table and preserved.
- 7. the storage method of laser radar point cloud data according to claim 6, it is characterised in that described to Current Scan After the coordinate data of each laser spots on line carries out the inverse transformation of the three-dimensional coordinate transformation, space is established for current scan line Concordance list simultaneously carries out preservation and included：Coordinate data after the inverse transformation of the three-dimensional coordinate transformation is carried out in buffering area and attribute data are compressed；After being compressed to the coordinate data in buffering area and attribute data, encoded to current scan line, current scan line A corresponding unique numbering；Spatial index table is established for the current scan line after being numbered, the keyword of the spatial index table is current scan line Numbering, index entry value corresponding with keyword is the storage of laser point cloud data on current scan line in the spatial index table Address；After spatial index table is established to current scan line, by the coordinate data of the buffering area after being compressed on current scan line and Attribute data is write back and the storage address in the spatial index table as corresponding to current scan line is preserved.
- A kind of 8. storage device of laser radar point cloud data, it is characterised in that including,Coordinate transformation unit, for according to the original laser cloud data gathered, by each scan line in original laser point cloud On laser spots carry out three-dimensional coordinate transformation, obtain the coordinate value of all laser spots after three-dimensional coordinate transformation；Judging unit is detected, for detecting the coordinate value of the laser spots after carrying out three-dimensional coordinate transformation, utilizes what is currently detected Difference between laser point coordinates value and the laser point coordinates value formerly detected judges whether is the laser spots that are currently detected For the starting point of new scan line；Buffer unit, for the laser point cloud data in scan line to be stored in into buffering area and to each in scan line in buffering area Laser spots carry out the inverse transformation of the three-dimensional coordinate transformation；Memory cell, for being protected after carrying out the inverse transformation of the three-dimensional coordinate transformation to the laser point cloud data in scan line Deposit.
- 9. the storage device of laser radar point cloud data according to claim 8, it is characterised in that the coordinate transform list Member includes coordinate transform subelement, and the coordinate transform subelement is used for each laser spots in original laser cloud data Three-dimensional rectangular coordinate (x, y, z) is converted to the spherical coordinate (r, θ, φ) centered on three-dimensional laser scanner, and wherein r is laser Point is to the distance of scanner, and θ is the angle of laser rays and horizontal plane, and φ is to horizontally rotate angle when scanner works.
- 10. the storage device of laser radar point cloud data according to claim 9, it is characterised in that the detection judges Unit includes the first subelement and the second subelement, and first subelement is used for the laser spots currently detected by detection The laser spots for changing to judge currently to be detected of the θ values of follow-up N number of laser spots whether be new scan line starting point, wherein N For positive integer；Second subelement is used for by swashing all in the φ values of the laser spots currently detected and current scan line The laser spots that the average value of the φ values of luminous point compares to judge currently to be detected whether be new scan line starting point.
- 11. the storage device of the laser radar point cloud data according to claim any one of 8-10, it is characterised in that described Buffer unit includes：Coordinate data buffer unit, the coordinate data of all laser spots on cache sweep line；Attribute data buffer unit, the attribute data of all laser spots on cache sweep line；Coordinate inversion unit, for the coordinate of all laser spots in coordinate data buffer unit to be carried out into coordinate inversion.
- 12. the storage device of the laser radar point cloud data according to claim any one of 8-10, it is characterised in that described Memory cell includes：Compression unit, for being compressed to the coordinate data and attribute data that are stored in buffering area；Coding unit, after being compressed for the coordinate data to buffering area and attribute data, encoded to each scan line, Each scan line corresponds to a unique numbering；Indexing units, for establishing spatial index table, the key of the spatial index table for each scan line after being numbered Word is the numbering of scan line, and index entry numerical value corresponding with keyword is laser spots in each scan line in the spatial index table The storage address of cloud data；Writeback unit, after establishing spatial index table to each scan line, by the buffering after being compressed in each scan line The coordinate data and attribute data in area are write back and the storage address in the spatial index table as corresponding to each scan line is entered Row preserves.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410592173.7A CN104408055B (en) | 2014-10-29 | 2014-10-29 | Storage method and device for laser radar point cloud data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410592173.7A CN104408055B (en) | 2014-10-29 | 2014-10-29 | Storage method and device for laser radar point cloud data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104408055A CN104408055A (en) | 2015-03-11 |
| CN104408055B true CN104408055B (en) | 2018-03-13 |
Family
ID=52645687
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410592173.7A Active CN104408055B (en) | 2014-10-29 | 2014-10-29 | Storage method and device for laser radar point cloud data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104408055B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105319536B (en) * | 2015-11-03 | 2018-03-16 | 深圳中智科创机器人有限公司 | A kind of radar three-dimensional scan control method and system |
| EP3315906B1 (en) * | 2016-10-27 | 2023-05-31 | Leica Geosystems AG | Method for processing scan data |
| CN107798657B (en) * | 2017-10-30 | 2019-01-04 | 武汉珞珈新空科技有限公司 | A kind of vehicle-mounted laser point cloud filtering method based on circular cylindrical coordinate |
| CN109684297B (en) * | 2018-04-19 | 2024-01-09 | 河北雄安兴诺科技服务有限公司 | Three-dimensional laser scanning mass point data storage method |
| CN108549711A (en) * | 2018-04-20 | 2018-09-18 | 广东工业大学 | A kind of method, apparatus, equipment and the storage medium of storage big data |
| CN108765584B (en) * | 2018-05-31 | 2023-07-14 | 深圳市易成自动驾驶技术有限公司 | Laser point cloud data set augmentation method, device and readable storage medium |
| CN108804714A (en) * | 2018-06-30 | 2018-11-13 | 浙江同创空间技术有限公司 | Point cloud data storage method and device |
| CN109781010B (en) * | 2019-01-21 | 2020-11-10 | 珠海博明软件有限公司 | Point cloud data calculation method, device and system for large-range data splicing |
| CN111696159B (en) * | 2020-06-15 | 2021-06-11 | 湖北亿咖通科技有限公司 | Feature storage method of laser odometer, electronic device and storage medium |
| CN115601272B (en) * | 2022-12-16 | 2023-04-11 | 海纳云物联科技有限公司 | Point cloud data processing method, device and equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5912676A (en) * | 1996-06-14 | 1999-06-15 | Lsi Logic Corporation | MPEG decoder frame memory interface which is reconfigurable for different frame store architectures |
| US6134022A (en) * | 1995-07-14 | 2000-10-17 | Kabushiki Kaisha Toshiba | Color image printing system capable of correcting density deviation on image and system for detecting color deviation on image |
| CN1864410A (en) * | 2003-10-04 | 2006-11-15 | 皇家飞利浦电子股份有限公司 | Method and apparatus for processing image data |
| CN102253821A (en) * | 2011-04-12 | 2011-11-23 | 深圳市蓝韵实业有限公司 | Data transmission processing method in ultrasonic diagnostic equipment |
-
2014
- 2014-10-29 CN CN201410592173.7A patent/CN104408055B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6134022A (en) * | 1995-07-14 | 2000-10-17 | Kabushiki Kaisha Toshiba | Color image printing system capable of correcting density deviation on image and system for detecting color deviation on image |
| US5912676A (en) * | 1996-06-14 | 1999-06-15 | Lsi Logic Corporation | MPEG decoder frame memory interface which is reconfigurable for different frame store architectures |
| CN1864410A (en) * | 2003-10-04 | 2006-11-15 | 皇家飞利浦电子股份有限公司 | Method and apparatus for processing image data |
| CN102253821A (en) * | 2011-04-12 | 2011-11-23 | 深圳市蓝韵实业有限公司 | Data transmission processing method in ultrasonic diagnostic equipment |
Non-Patent Citations (2)
| Title |
|---|
| "基于激光扫描数据的隧道曲面重建技术研究";白廷义;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20120215(第02期);正文第2.2.1部分、第5.3.3部分 * |
| "数据点云预处理及特征角点检测算法研究";温银放;《中国优秀硕士学位论文全文数据库 信息科技辑》;20071015(第04期);正文第1.3部分 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104408055A (en) | 2015-03-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104408055B (en) | Storage method and device for laser radar point cloud data | |
| CN108537876B (en) | Three-dimensional reconstruction method, device, equipment and storage medium | |
| CN105976312B (en) | Point cloud autoegistration method based on point feature histogram | |
| CN104200212B (en) | A kind of building external boundary line drawing method based on airborne LiDAR data | |
| US20100021068A1 (en) | System and method for extracting boundary elements of an object | |
| CN104851127B (en) | It is a kind of based on interactive building point cloud model texture mapping method and device | |
| CN111260668B (en) | Power line extraction method, system and terminal | |
| CN103258345A (en) | Method for extracting parameters of tree branches based on ground laser radar three-dimensional scanning | |
| CN108804714A (en) | Point cloud data storage method and device | |
| CN104050709B (en) | A kind of three dimensional image processing method and electronic equipment | |
| CN105631939B (en) | A kind of three-dimensional point cloud distortion correction method and its system based on curvature filtering | |
| CN105654483A (en) | Three-dimensional point cloud full-automatic registration method | |
| CN106600700B (en) | Three-dimensional model data processing system | |
| CN110070567A (en) | A kind of ground laser point cloud method for registering | |
| CN108986024B (en) | Grid-based laser point cloud rule arrangement processing method | |
| CN108447124A (en) | A kind of ancient building contour extraction method and system based on cloud | |
| CN104751479A (en) | Building extraction method and device based on TIN data | |
| CN109031234A (en) | The method of quick obtaining radar reflectivity data three-dimensional contour surface | |
| CN116721228A (en) | Building elevation extraction method and system based on low-density point cloud | |
| CN111724436A (en) | Tunnel injection square quantity estimation method based on point cloud data processing | |
| CN106803266A (en) | A kind of hull complexity outside plate point cloud costal line extracting method and device | |
| CN104573144A (en) | System and method for simulating offline point cloud of measuring equipment | |
| US20230196719A1 (en) | Method for cargo counting, computer equipment, and storage medium | |
| CN110458780A (en) | 3D point cloud data de-noising method, apparatus, computer equipment and readable storage medium storing program for executing | |
| CN103808646A (en) | Image analyzer and method for measuring line fractal dimension of granular material |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |