CN111340728B - Point cloud denoising method and device based on 3D point cloud segmentation and storage medium - Google Patents
Point cloud denoising method and device based on 3D point cloud segmentation and storage medium Download PDFInfo
- Publication number
- CN111340728B CN111340728B CN202010120615.3A CN202010120615A CN111340728B CN 111340728 B CN111340728 B CN 111340728B CN 202010120615 A CN202010120615 A CN 202010120615A CN 111340728 B CN111340728 B CN 111340728B
- Authority
- CN
- China
- Prior art keywords
- point cloud
- point
- noise
- absolute phase
- value
- 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 47
- 230000011218 segmentation Effects 0.000 title claims abstract description 34
- 238000010587 phase diagram Methods 0.000 claims abstract description 24
- 238000013507 mapping Methods 0.000 claims abstract description 16
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 abstract description 10
- 238000005259 measurement Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012876 topography Methods 0.000 description 1
Images
Classifications
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
Abstract
The invention discloses a point cloud denoising method, a point cloud denoising device and a storage medium based on 3D point cloud segmentation, wherein each three-dimensional point in a 3D point cloud is uniquely corresponding to a two-dimensional point in an absolute phase diagram, after a three-dimensional point cloud serial number mapping image is constructed, a second point cloud value corresponding to the first point cloud value is obtained from a K neighbor lookup table, and whether the first point cloud value and the second point cloud value belong to the same point cloud is judged according to a preset distance threshold value, so that the calculation complexity in the 3D point cloud segmentation is greatly reduced; after the 3D point cloud segmentation is completed, the point cloud is divided into the noise-free point cloud and the noise point cloud according to the number of the point clouds, the noise-free absolute phase corresponding to the noise point cloud is calculated by combining with the reference absolute phase, a new noise-free point cloud is reconstructed according to the noise-free absolute phase, the 3D point cloud containing noise is recovered into the noise-free 3D point cloud, the noise point cloud in any condition can be removed, and the precision of the 3D point cloud is greatly improved.
Description
Technical Field
The invention relates to the technical field of three-dimensional topography measurement, in particular to a point cloud denoising method and device based on 3D point cloud segmentation and a storage medium.
Background
At present, the three-dimensional measurement technology is more and more widely applied in production and life, although a three-dimensional scanner can easily acquire a 3D point cloud of an object, in the acquisition process, noise point cloud is generated due to various factors, so that the precision of the 3D point cloud is reduced, and therefore how to remove the noise in the 3D point cloud is the key for improving the three-dimensional measurement precision. In the prior art, the relation between points is usually calculated when 3D point clouds are calculated so as to judge whether the point clouds are noise or not, but the method has large calculation amount and high calculation complexity, and when the noise point clouds in the conditions of scattered state, disorder state, block state and the like are encountered, the point clouds in the noise part cannot be recovered, so that the data loss is caused.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide a point cloud denoising method, a point cloud denoising device and a storage medium based on 3D point cloud segmentation, which can calculate high-precision point cloud with lower calculation amount in practical application and avoid the loss of point cloud data.
The technical scheme adopted by the invention for solving the problems is as follows: in a first aspect, the invention provides a point cloud denoising method based on 3D point cloud segmentation, which comprises the following steps:
a client acquires a 3D point cloud, numbers each three-dimensional point in the 3D point cloud and uniquely corresponds to a two-dimensional point in an absolute phase diagram, and a three-dimensional point cloud serial number mapping image is constructed;
the client reads a K neighbor lookup table and a preset distance threshold, selects a first point cloud value from the 3D point cloud, acquires a second point cloud value corresponding to the first point cloud value from the K neighbor lookup table, divides the first point cloud value and the second point cloud value into the same point cloud if the distance between most first point cloud values and the second point cloud value is less than the distance threshold, and repeatedly executes the division until all point cloud values of the 3D point cloud are divided, so as to acquire the divided 3D point cloud;
the client side counts the point cloud number of each part in the segmented 3D point cloud, and the part with the point cloud number larger than a preset number threshold is a noise-free point cloud, otherwise, the part is set as a noise point cloud;
the client acquires a two-dimensional coordinate corresponding to a noise point in the noise point cloud in the absolute phase diagram according to the three-dimensional point cloud serial number mapping image, acquires a reference absolute phase and calculates a corresponding noise-free point to obtain a noise-free absolute phase;
and the client reconstructs the noise-free absolute phase to finish the denoising of the 3D point cloud.
Further, the 3D point cloud is obtained by reconstructing an absolute phase diagram and internal and external parameters of the measuring equipment.
Further, the K-nearest neighbor look-up table is calculated from the 3D point cloud according to a K-nearest neighbor algorithm.
Further, the repeatedly executing until all point cloud values of the 3D point cloud are divided specifically includes: if the second point cloud value and the first point cloud value are divided into the same point cloud, obtaining a third point cloud value corresponding to the second point cloud value from the K neighbor lookup table for division; and if the second point cloud value and the first point cloud value are divided into different point clouds, selecting a point cloud value which is not divided from the 3D point cloud as a third point cloud value for division.
Further, the acquiring of the reference absolute phase specifically includes the following steps:
the client selects a starting point and an end point of a noise point from the two-dimensional coordinates corresponding to the noise point in the absolute phase diagram;
and the client calculates the slope and the intercept of a line segment formed by connecting a point before the starting point and a point after the ending point to obtain the reference absolute phase.
Further, the acquiring of the reference absolute phase and the calculating of the corresponding noise-free point specifically include the following steps:
the client calculates the corresponding fringe order according to the reference absolute phase;
and the client calculates an absolute phase point corresponding to the noise point according to the fringe order, wherein the absolute phase point is a noise-free point.
In a second aspect, the present invention provides an apparatus for performing a point cloud denoising method based on 3D point cloud segmentation, comprising a CPU unit for performing the following steps:
the method comprises the steps that a client side obtains a 3D point cloud, numbers each three-dimensional point in the 3D point cloud and uniquely corresponds to a two-dimensional point in an absolute phase diagram, and a three-dimensional point cloud serial number mapping image is constructed;
the client reads a K neighbor lookup table and a preset distance threshold, selects a first point cloud value from the 3D point cloud, acquires a second point cloud value corresponding to the first point cloud value from the K neighbor lookup table, divides the first point cloud value and the second point cloud value into the same point cloud if the distance between most first point cloud values and the second point cloud value is less than the distance threshold, and repeatedly executes the division until all point cloud values of the 3D point cloud are divided, so as to acquire the divided 3D point cloud;
the client counts the point cloud number of each part in the segmented 3D point cloud, and the part of which the point cloud number is greater than a preset number threshold is a noise-free point cloud, otherwise, the part is set as a noise point cloud;
the client acquires a two-dimensional coordinate corresponding to a noise point in the noise point cloud in the absolute phase diagram according to the three-dimensional point cloud serial number mapping image, acquires a reference absolute phase and calculates a corresponding noise-free point to obtain a noise-free absolute phase;
and the client reconstructs the noise-free absolute phase to complete the denoising of the 3D point cloud.
Further, the CPU unit is further configured to perform the steps of:
the client selects a starting point and an end point of a noise point from the two-dimensional coordinates corresponding to the noise point in the absolute phase diagram;
and the client calculates the slope and the intercept of a line segment formed by connecting a point before the starting point and a point after the ending point to obtain the reference absolute phase.
Further, the CPU unit is further configured to perform the steps of:
the client calculates the corresponding fringe order according to the reference absolute phase;
and the client calculates an absolute phase point corresponding to the noise point according to the fringe order, wherein the absolute phase point is a noise-free point.
In a third aspect, the present invention provides an apparatus for performing a point cloud denoising method based on 3D point cloud segmentation, comprising at least one control processor and a memory for communicative connection with the at least one control processor; the memory stores instructions executable by the at least one control processor to enable the at least one control processor to perform a point cloud denoising method based on 3D point cloud segmentation as described above.
In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions for causing a computer to perform the point cloud denoising method based on 3D point cloud segmentation as described above.
In a fifth aspect, the present invention also provides a computer program product comprising a computer program stored on a computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, cause the computer to perform the method for denoising a point cloud based on 3D point cloud segmentation as described above.
One or more technical schemes provided in the embodiment of the invention have at least the following beneficial effects: according to the method, each three-dimensional point in the 3D point cloud is uniquely corresponding to a two-dimensional point in the absolute phase diagram, after a three-dimensional point cloud serial number mapping image is constructed, a second point cloud value corresponding to the first point cloud value is obtained from a K neighbor lookup table, whether the first point cloud value and the second point cloud value belong to the same point cloud or not is judged through a preset distance threshold, and the calculation complexity in 3D point cloud segmentation is greatly reduced; after the 3D point cloud segmentation is completed, the point cloud is divided into the noise-free point cloud and the noise point cloud according to the number of the point clouds, the noise-free absolute phase corresponding to the noise point cloud is calculated by combining with the reference absolute phase, a new noise-free point cloud is reconstructed according to the noise-free absolute phase, the 3D point cloud containing noise is recovered into the noise-free 3D point cloud, the noise point cloud in any condition can be removed, and the precision of the 3D point cloud is greatly improved.
Drawings
The invention is further illustrated with reference to the following figures and examples.
Fig. 1 is a flowchart of a point cloud denoising method based on 3D point cloud segmentation according to an embodiment of the present invention;
fig. 2 is a flowchart of acquiring a reference absolute phase in a point cloud denoising method based on 3D point cloud segmentation according to an embodiment of the present invention;
fig. 3 is a flowchart of acquiring a reference absolute phase and calculating a corresponding noise-free point in the point cloud denoising method based on 3D point cloud segmentation according to the embodiment of the present invention;
fig. 4 is a schematic diagram of an apparatus for performing a point cloud denoising method based on 3D point cloud segmentation according to a second embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
It should be noted that, if not conflicted, the various features of the embodiments of the invention may be combined with each other within the scope of protection of the invention. Additionally, while functional block divisions are performed in apparatus schematics, with logical sequences shown in flowcharts, in some cases, steps shown or described may be performed in sequences other than block divisions in apparatus or flowcharts.
Referring to fig. 1, a first embodiment of the present invention provides a point cloud denoising method based on 3D point cloud segmentation, including the following steps:
s100, a client acquires 3D point clouds, numbers each three-dimensional point in the 3D point clouds and uniquely corresponds to a two-dimensional point in an absolute phase diagram, and a three-dimensional point cloud serial number mapping image is constructed;
step S200, a client reads a K neighbor lookup table and a preset distance threshold, selects a first point cloud value from the 3D point cloud, acquires a second point cloud value corresponding to the first point cloud value from the K neighbor lookup table, divides the first point cloud value and the second point cloud value into the same point cloud if the distance between most of the first point cloud values and the second point cloud values is less than the distance threshold, and repeatedly executes until all point cloud values of the 3D point cloud are divided, so as to obtain the divided 3D point cloud;
step S300, the client side counts the point cloud number of each part in the segmented 3D point cloud, and the part with the point cloud number larger than a preset number threshold is a noise-free point cloud, otherwise, the part is set as a noise point cloud;
step S400, the client acquires a two-dimensional coordinate corresponding to a noise point in the noise point cloud in the absolute phase diagram according to the three-dimensional point cloud serial number mapping image, acquires a reference absolute phase and calculates a corresponding noise-free point to obtain a noise-free absolute phase;
and S500, reconstructing the noise-free absolute phase by the client to finish the denoising of the 3D point cloud.
It should be noted that, since the three-dimensional points are reconstructed from each point of the absolute phase map in the image plane, each point of the absolute phase map in the image plane has a one-to-one correspondence relationship with the reconstructed three-dimensional points. It can be understood that, each three-dimensional point (X, Y, Z) is numbered from 1 to n as a unique serial number corresponding to the three-dimensional point, and the serial number is in one-to-one correspondence with the two-dimensional point (u, v) of the absolute phase diagram to construct a three-dimensional point cloud serial number mapping image, that is, each coordinate point (u, v) of the mapping image corresponds to the unique serial number of the three-dimensional point.
It should be noted that the first point cloud value in step S200 is a point cloud value arbitrarily selected from the 3D point cloud, and the first point cloud value is preferably marked after being selected, so that repeated selection and repeated calculation caused when the point cloud value is selected again are avoided. It will be appreciated that the distance threshold may be any value, as may be adjusted according to the accuracy required. It can be understood that if the distance between the first point cloud value and the second point cloud value is greater than or equal to the distance threshold, the distance between the two point cloud values is determined to be too large and not belong to the same point cloud. It can be understood that, in practical application, a plurality of point cloud values which conform to the condition that the distance from the first point cloud value is smaller than the distance threshold may exist, and the point cloud values are compared one by one, so that the complexity of calculation can be effectively reduced and the calculation efficiency can be improved by comparing the point cloud values with the first point cloud value through numerical values.
It should be noted that the number threshold in step S300 may be adjusted according to the requirement for precision, and the specific numerical value is not limited in this embodiment. It can be understood that the number of the segmented point clouds can be judged by setting a number threshold, and if the number of the segmented point clouds is smaller than the number threshold, the number of the point clouds is too small, and the segmented point clouds belong to noise point clouds.
It should be noted that, if a plurality of noise-free absolute phases exist in step S500, after reconstruction is performed one by one, the new noise-free point cloud obtained and the noise-free point cloud obtained in step S300 together form a denoised 3D point cloud.
Further, in another embodiment of the present invention, the 3D point cloud is reconstructed from the absolute phase map and the internal and external parameters of the measurement device.
It should be noted that the measurement device in this embodiment may be a three-dimensional measurement device such as a camera, a video camera, and the like, and the internal and external parameters are calibration parameters of the measurement device during measurement, which is not described herein again.
Further, in another embodiment of the present invention, the K-nearest neighbor look-up table is calculated from a K-nearest neighbor algorithm on the 3D point cloud.
It should be noted that, in this embodiment, a suitable K value is preferably selected in advance, the K nearest neighbor algorithm is used to calculate the 3D point cloud itself, K nearest neighbor points from each point of the 3D point cloud to other points of the point cloud are obtained, the obtained data form a K nearest neighbor lookup table, a specific K value is selected according to an actual requirement, and this embodiment does not limit a specific numerical value. Because the calculation complexity of the K neighbor algorithm is low, the embodiment only needs to calculate once to construct the K neighbor lookup table, and the 3D point cloud can be segmented only through data lookup, so that the calculation amount is greatly reduced.
Further, in another embodiment of the present invention, the repeatedly executing until all point cloud values of the 3D point cloud are divided specifically includes: if the second point cloud value and the first point cloud value are divided into the same point cloud, obtaining a third point cloud value corresponding to the second point cloud value from the K neighbor lookup table for division; and if the second point cloud value and the first point cloud value are divided into different point clouds, selecting an undivided point cloud value from the 3D point cloud as a third point cloud value for division.
It should be noted that the first point cloud value is the initial point cloud value in this embodiment, the second point cloud value divided into the same point cloud is selected as the new initial point cloud value in this embodiment, the search of the K neighbor lookup table and the distance threshold comparison are repeatedly performed, and the point cloud segmentation can be completed by performing a simple table lookup and numerical comparison on each point. It should be noted that, even if the distance between the first point cloud value and the second point cloud value does not satisfy the distance threshold, the first point cloud value is already divided, in order to avoid repeated division, the embodiment preferably selects a third point cloud value from the point cloud values that are not divided to perform the dividing step until all the points in the 3D point cloud are divided, so as to avoid that the missing point cloud values affect the point cloud precision.
Referring to fig. 2, further, in another embodiment of the present invention, the obtaining of the reference absolute phase specifically includes the following steps:
step S410, the client selects the starting point and the ending point of the noise point from the two-dimensional coordinates of the noise point corresponding to the absolute phase diagram;
in step S420, the client calculates the slope and intercept of a line segment connecting a point before the start point and a point after the end point to obtain the reference absolute phase.
The present embodiment is described below with an example:
firstly, the absolute phase containing noise points is searched to the starting point (x) of each noise point 1 ,y 1 ) And an end point (x) n ,y n ) The point (x) before the starting point of the noise point 0 ,y 0 ) And the point (x) subsequent to the termination point n+1 ,y n+1 ) Middle point (x) 0 ,y 0 ) And (x) n+1 ,y n+1 ) A noise-free point on the absolute phase; calculating point (x) 0 ,y 0 ) And (x) n+1 ,y n+1 ) Slope k and intercept b of connected line segments, where k = (y) n+1 -y 0 )/(x n+1 -x 0 )，b＝y 0 -kx 0 (ii) a Establishing a noisy point abscissa x i (i =1,2,3, \8230;, n) establishing a noiseless reference line segment y = kx i + b, (i =1,2,3, \ 8230;, n), defined as the absolute phase Φ of the reference r I.e. phi r ＝y＝kx i +b(i =1,2,3, \8230;, n), i.e. a noise-free absolute phase of a reference is established.
Referring to fig. 3, further, in another embodiment of the present invention, the step of obtaining the reference absolute phase and calculating the corresponding noise-free point specifically includes the following steps:
step S430, the client calculates the corresponding fringe order according to the reference absolute phase;
in step S440, the client calculates an absolute phase point corresponding to the noise point according to the fringe order, where the absolute phase point is a noise-free point.
The present embodiment is described below with an example:
absolute phase value containing noise points of phi err ＝φ+2πm err Wherein m is err Phi is the correct wrapped phase value for the wrong fringe order. According to the reference absolute phase phi r Recalculating fringe order m, m = (phi) corresponding to noise point ref -Φ err ) And/2 pi. Recalculating the absolute phase point corresponding to the noise point according to the new fringe order m, namely phi =2 pi m + phi, obtaining the absolute phase phi without noise, and completing the recovery of the absolute phase noise
It should be noted that the noise point is generated because the calculated fringe order m is erroneous in the process of recovering the absolute phase, which results in obtaining the absolute phase Φ err Is erroneous and the wrapped phase phi itself is correct. So with the correct reference absolute phase phi r And recalculating the fringe order m, and recalculating the absolute phase according to the new fringe order m to obtain the correct absolute phase.
Referring to fig. 4, a second embodiment of the present invention further provides an apparatus for performing a point cloud denoising method based on 3D point cloud segmentation, where the apparatus is a smart device, such as a smart phone, a computer, a tablet computer, and the like, and the embodiment is described by taking the computer as an example.
In the computer 4000 for performing a point cloud denoising method based on 3D point cloud segmentation, a CPU unit 4100 is included, the CPU unit 4100 is configured to perform the following steps:
the client acquires the 3D point cloud, numbers each three-dimensional point in the 3D point cloud and uniquely corresponds to the two-dimensional point in the absolute phase diagram, and a three-dimensional point cloud serial number mapping image is constructed;
the method comprises the steps that a client reads a K neighbor lookup table and a preset distance threshold, a first point cloud value is selected from 3D point clouds, a second point cloud value corresponding to the first point cloud value is obtained from the K neighbor lookup table, if the distance between most of the first point cloud values and the second point cloud value is smaller than the distance threshold, the first point cloud value and the second point cloud value are divided into the same point cloud, the operation is repeatedly carried out until all point cloud values of the 3D point clouds are divided, and the divided 3D point clouds are obtained;
the client side counts the point cloud number of each part in the segmented 3D point cloud, and the part with the point cloud number larger than a preset number threshold is a noise-free point cloud, otherwise, the part is set as a noise point cloud;
the client acquires a two-dimensional coordinate corresponding to a noise point in the noise point cloud in the absolute phase diagram according to the three-dimensional point cloud serial number mapping image, acquires a reference absolute phase and calculates a corresponding noise-free point to obtain a noise-free absolute phase;
and the client reconstructs the noise-free absolute phase to complete the denoising of the 3D point cloud.
Further, in another embodiment of the present invention, the CPU unit 4100 is further configured to perform the following steps:
the client selects a starting point and an ending point of the noise point from the two-dimensional coordinates corresponding to the noise point in the absolute phase diagram;
the client calculates the slope and intercept of a line segment formed by connecting a point before the starting point and a point after the ending point to obtain a reference absolute phase.
Further, in another embodiment of the present invention, the CPU unit 4100 is further configured to perform the following steps:
the client calculates the corresponding fringe order according to the reference absolute phase;
and the client calculates the absolute phase point corresponding to the noise point according to the fringe order, wherein the absolute phase point is a noise-free point.
The computer 4000 and the CPU unit 4100 may be connected by a bus or other means, and the computer 4000 further includes a memory as a non-transitory computer readable storage medium, which can be used to store a non-transitory software program, a non-transitory computer executable program, and a module, such as program instructions/modules corresponding to the apparatus for performing the point cloud denoising method based on 3D point cloud segmentation in the embodiment of the present invention. The computer 4000 controls the CPU unit 4100 to execute various functional applications and data processing for executing the point cloud denoising method based on 3D point cloud segmentation by operating non-transitory software programs, instructions, and modules stored in the memory, that is, to implement the point cloud denoising method based on 3D point cloud segmentation of the above method embodiment.
The memory may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the stored data area may store data created from the use of the CPU unit 4100, and the like. Further, the memory may include high speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory optionally includes memory located remotely from the CPU unit 4100, and such remote memory may be connected to the computer 4000 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The one or more modules are stored in the memory and, when executed by the CPU unit 4100, perform the point cloud denoising method based on 3D point cloud segmentation in the above method embodiments.
An embodiment of the present invention further provides a computer-readable storage medium, where computer-executable instructions are stored, and executed by the CPU unit 4100, so as to implement the point cloud denoising method based on 3D point cloud segmentation.
The above-described embodiments of the apparatus are merely illustrative, and the apparatuses described as separate components may or may not be physically separate, may be located in one place, or may be distributed over a plurality of network apparatuses. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
It should be noted that, since the apparatus for performing the point cloud denoising method based on 3D point cloud segmentation in the present embodiment is based on the same inventive concept as the above-mentioned point cloud denoising method based on 3D point cloud segmentation, the corresponding contents in the method embodiment are also applicable to the present apparatus embodiment, and are not described in detail here.
Through the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be implemented by software plus a general hardware platform. Those skilled in the art will appreciate that all or part of the processes in the methods of the above embodiments may be implemented by hardware related to instructions of a computer program, which may be stored in a computer-readable storage medium, and when executed, may include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read Only Memory (ROM), a Random Access Memory (RAM), or the like.
While the preferred embodiments of the present invention have been described in detail, it will be understood by those skilled in the art that the foregoing and various other changes, omissions and deviations in the form and detail thereof may be made without departing from the scope of this invention.
Claims (9)
1. A point cloud denoising method based on 3D point cloud segmentation is characterized by comprising the following steps:
a client acquires a 3D point cloud, numbers each three-dimensional point in the 3D point cloud and uniquely corresponds to a two-dimensional point in an absolute phase diagram, and a three-dimensional point cloud serial number mapping image is constructed;
the client reads a K neighbor lookup table and a preset distance threshold, selects a first point cloud value from the 3D point cloud, acquires a second point cloud value corresponding to the first point cloud value from the K neighbor lookup table, divides the first point cloud value and the second point cloud value into the same point cloud if the distance between most first point cloud values and the second point cloud value is less than the distance threshold, and repeatedly executes the division until all point cloud values of the 3D point cloud are divided, so as to acquire the divided 3D point cloud;
the client side counts the point cloud number of each part in the segmented 3D point cloud, and the part with the point cloud number larger than a preset number threshold is a noise-free point cloud, otherwise, the part is set as a noise point cloud;
the client acquires a two-dimensional coordinate corresponding to a noise point in the noise point cloud in the absolute phase diagram according to the three-dimensional point cloud serial number mapping image, acquires a reference absolute phase and calculates a corresponding noise-free point to obtain a noise-free absolute phase;
the client reconstructs the noise-free absolute phase to finish denoising the 3D point cloud;
the reference absolute phase acquisition specifically includes the following steps:
the client searches the starting point (x) of each noise point in the two-dimensional coordinates corresponding to the noise point in the absolute phase diagram 1 ，y 1 ) And an end point (x) n ，x n ) Determining a point (x) preceding the starting point of the noise point 0 ，y 0 ) And the point (x) subsequent to the termination point n+1 ，y n+1 ) Middle point (x) 0 ，y 0 ) And (x) n+1 ，y n+1 ) A noise-free point on the absolute phase;
the client computing point (x) 0 ，y 0 ) And (x) n+1 ，y n+1 ) Slope k and intercept b of connected line segments, where k = (y) n+1 -y 0 )/(x n+1 -x 0 )，b＝y 0 -kx 0 ；
Establishing a horizontal coordinate x of a noise point i (ε =1,2,3, \8230;, n) establishes a noiseless reference line segment y = kx i + b, (i =1,2,3, \ 8230;, n), which is noiseless for referenceAbsolute phase phi r Wherein is phi r ＝y＝kx i +b，(i＝1，2，3，…，n)。
2. The method of claim 1, wherein the method comprises: and reconstructing the 3D point cloud by using the absolute phase diagram and the internal and external parameters of the measuring equipment.
3. The method of claim 1, wherein the method comprises: the K-nearest neighbor look-up table is calculated by the 3D point cloud according to a K-nearest neighbor algorithm.
4. The method of claim 1, wherein the repeating until all point cloud values of the 3D point cloud are segmented comprises: if the second point cloud value and the first point cloud value are divided into the same point cloud, acquiring a third point cloud value corresponding to the second point cloud value from the K neighbor lookup table for division; and if the second point cloud value and the first point cloud value are divided into different point clouds, selecting a point cloud value which is not divided from the 3D point cloud as a third point cloud value for division.
5. The point cloud denoising method based on 3D point cloud segmentation according to claim 1, wherein the obtaining of the reference absolute phase and the calculating of the corresponding noise-free point specifically comprises the following steps:
the client calculates a corresponding fringe order according to the reference absolute phase;
and the client calculates an absolute phase point corresponding to the noise point according to the fringe order, wherein the absolute phase point is a noise-free point.
6. An apparatus for performing a point cloud denoising method based on 3D point cloud segmentation, comprising a CPU unit for performing the steps of:
a client acquires a 3D point cloud, numbers each three-dimensional point in the 3D point cloud and uniquely corresponds to a two-dimensional point in an absolute phase diagram, and a three-dimensional point cloud serial number mapping image is constructed;
the client reads a K neighbor lookup table and a preset distance threshold, selects a first point cloud value from the 3D point cloud, acquires a second point cloud value corresponding to the first point cloud value from the K neighbor lookup table, divides the first point cloud value and the second point cloud value into the same point cloud if the distance between most of the first point cloud values and the second point cloud values is smaller than the distance threshold, and repeatedly executes the steps until all point cloud values of the 3D point cloud are divided, so as to acquire the divided 3D point cloud;
the client side counts the point cloud number of each part in the segmented 3D point cloud, and the part with the point cloud number larger than a preset number threshold is a noise-free point cloud, otherwise, the part is set as a noise point cloud;
the client acquires a two-dimensional coordinate corresponding to a noise point in the noise point cloud in the absolute phase diagram according to the three-dimensional point cloud serial number mapping image, acquires a reference absolute phase and calculates a corresponding noise-free point to obtain a noise-free absolute phase;
the client reconstructs the noise-free absolute phase to complete the denoising of the 3D point cloud;
the reference absolute phase acquisition specifically includes the following steps:
the client searches the starting point (x) of each noise point in the two-dimensional coordinates corresponding to the noise point in the absolute phase diagram 1 ，y 1 ) And an end point (x) n ，x n ) Determining a point (x) preceding the starting point of the noise point 0 ，y 0 ) And the point (x) subsequent to the termination point n+1 ，y n+1 ) Middle point (x) 0 ，y 0 ) And (x) n+1 ，y n+1 ) A noise-free point on the absolute phase;
the client computing point (x) 0 ，y 0 ) And (x) n+1 ，y n+1 ) Slope k and intercept b of connected line segments, where k = (y) n+1 -y 0 )/(x n+1 -x 0 )，b＝y 0 -kx 0 ；
Establishing a noisy point abscissa x i (ε =1,2,3, \8230;, n) establishes a noiseless reference line segment y = kx i + b, (i =1,2,3, \ 8230;, n), the noiseless absolute phase Φ of which the reference is obtained r Wherein is phi r ＝y＝kx i +b，(i＝1，2，3，…，n)。
7. The apparatus of claim 6, wherein the CPU unit is further configured to perform the following steps:
the client selects a starting point and an end point of a noise point from the two-dimensional coordinates corresponding to the noise point in the absolute phase diagram;
and the client calculates the slope and the intercept of a line segment formed by connecting a point before the starting point and a point after the ending point to obtain the reference absolute phase.
8. The apparatus of claim 6, wherein the CPU unit is further configured to perform the following steps:
the client calculates the corresponding fringe order according to the reference absolute phase;
and the client calculates an absolute phase point corresponding to the noise point according to the fringe order, wherein the absolute phase point is a noise-free point.
9. A computer-readable storage medium characterized by: the computer-readable storage medium stores computer-executable instructions for causing a computer to perform a method for point cloud denoising based on 3D point cloud segmentation according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010120615.3A CN111340728B (en) | 2020-02-26 | 2020-02-26 | Point cloud denoising method and device based on 3D point cloud segmentation and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010120615.3A CN111340728B (en) | 2020-02-26 | 2020-02-26 | Point cloud denoising method and device based on 3D point cloud segmentation and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111340728A CN111340728A (en) | 2020-06-26 |
| CN111340728B true CN111340728B (en) | 2023-03-21 |
Family
ID=71187088
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010120615.3A Active CN111340728B (en) | 2020-02-26 | 2020-02-26 | Point cloud denoising method and device based on 3D point cloud segmentation and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111340728B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112508803B (en) * | 2020-11-03 | 2023-10-03 | 中山大学 | Denoising method and device for three-dimensional point cloud data and storage medium |
| CN114612598A (en) * | 2022-02-16 | 2022-06-10 | 苏州一径科技有限公司 | Point cloud processing method and device and laser radar |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108181635A (en) * | 2017-12-11 | 2018-06-19 | 中国南方电网有限责任公司超高压输电公司广州局 | A kind of laser point cloud sorting technique for transmission line of electricity scissors crossing analysis |
| CN110458772A (en) * | 2019-07-30 | 2019-11-15 | 五邑大学 | A kind of point cloud filtering method, device and storage medium based on image procossing |
-
2020
- 2020-02-26 CN CN202010120615.3A patent/CN111340728B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108181635A (en) * | 2017-12-11 | 2018-06-19 | 中国南方电网有限责任公司超高压输电公司广州局 | A kind of laser point cloud sorting technique for transmission line of electricity scissors crossing analysis |
| CN110458772A (en) * | 2019-07-30 | 2019-11-15 | 五邑大学 | A kind of point cloud filtering method, device and storage medium based on image procossing |
Non-Patent Citations (1)
| Title |
|---|
| Error analysis in the absolute phase maps recovered by fringe patterns with three different wavelengths;Jiale Long et al.;《Journal of Modern Optics 》;20171006;第65卷(第3期);第1-2页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111340728A (en) | 2020-06-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110458772B (en) | Point cloud filtering method and device based on image processing and storage medium | |
| CN107705333B (en) | Space positioning method and device based on binocular camera | |
| CN111340728B (en) | Point cloud denoising method and device based on 3D point cloud segmentation and storage medium | |
| WO2021120410A1 (en) | Hough transform-based absolute phase noise removal method and apparatus, and storage medium | |
| CN111160298B (en) | Robot and pose estimation method and device thereof | |
| CN111582054B (en) | Point cloud data processing method and device and obstacle detection method and device | |
| CN111275633B (en) | Point cloud denoising method, system, device and storage medium based on image segmentation | |
| CN108124489B (en) | Information processing method, apparatus, cloud processing device and computer program product | |
| JP2019021291A (en) | Camera pose estimating method and system | |
| WO2018107427A1 (en) | Rapid corresponding point matching method and device for phase-mapping assisted three-dimensional imaging system | |
| CN110738730A (en) | Point cloud matching method and device, computer equipment and storage medium | |
| EP3501759B1 (en) | Position and orientation estimation apparatus, position and orientation estimation method, and program | |
| CN115222889A (en) | 3D reconstruction method and device based on multi-view image and related equipment | |
| KR101593316B1 (en) | Method and apparatus for recontructing 3-dimension model using stereo camera | |
| CN111739071A (en) | Rapid iterative registration method, medium, terminal and device based on initial value | |
| Muresan et al. | A multi patch warping approach for improved stereo block matching | |
| CN103196460B (en) | A kind of method and apparatus estimating step-length | |
| CN111506679B (en) | Map element data generation and display method and device, medium and equipment | |
| CN114415698B (en) | Robot, positioning method and device of robot and computer equipment | |
| CN111160362A (en) | FAST feature homogenization extraction and IMU-based inter-frame feature mismatching removal method | |
| CN113551617B (en) | Binocular double-frequency complementary three-dimensional surface type measuring method based on fringe projection | |
| CN111508009B (en) | Binocular stereo matching preprocessing method and device | |
| CN111368675B (en) | Gesture depth information processing method, device, equipment and storage medium | |
| CN108230377B (en) | Point cloud data fitting method and system | |
| CN117437552B (en) | Method, device, equipment and storage medium for constructing visual positioning map |
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 |