CN117808703A - Multi-scale large-scale component assembly gap point cloud filtering method - Google Patents
Multi-scale large-scale component assembly gap point cloud filtering method Download PDFInfo
- Publication number
- CN117808703A CN117808703A CN202410224098.2A CN202410224098A CN117808703A CN 117808703 A CN117808703 A CN 117808703A CN 202410224098 A CN202410224098 A CN 202410224098A CN 117808703 A CN117808703 A CN 117808703A
- Authority
- CN
- China
- Prior art keywords
- point cloud
- point
- scale
- filtering
- points
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 86
- 238000001914 filtration Methods 0.000 title claims abstract description 63
- 238000004364 calculation method Methods 0.000 claims description 11
- 239000013598 vector Substances 0.000 claims description 7
- 238000009499 grossing Methods 0.000 claims description 6
- 238000007619 statistical method Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 abstract description 4
- 230000000694 effects Effects 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 6
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000004590 computer program Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks
-
- 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
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a multi-scale large-scale component assembly gap point cloud filtering method, which comprises the following steps: the ROI region is segmented for the initial point cloud P of the assembly gap of the large-scale component, and the point cloud density is obtained through extraction of a point cloud density sensorMinimum ROI region point cloudThereby preserving effective point cloud data; for ROI region point cloudFiltering invalid pointsFiltering and filtering large-scale outliers to obtain filtered point cloudsThe method comprises the steps of carrying out a first treatment on the surface of the Extraction by region growing clustering methodClustering all the point clouds to form a continuous point cloud cluster, and dividing and filtering the gap irrelevant point clouds based on a weight voting method to obtain an effective point cloudThe accuracy of the gap-related point cloud is improved; by aiming at the effective point cloudPerforming small-scale point cloud fairing to solve the problem of rough point cloud in local area and obtain final filtered point cloud. The invention realizes rapid and comprehensive filtering treatment of the point cloud by utilizing a multi-scale filtering method.
Description
Technical Field
The invention relates to an algorithm for three-dimensional data processing, in particular to a multi-scale large-scale component assembly gap point cloud filtering method.
Background
The gap is an important index for large-scale component assembly, so that the gap of a vehicle body is very necessary to measure, the point cloud of a large-scale component assembly gap structure can be completely obtained through a three-dimensional scanning technology, and the size of the large-scale component assembly gap can be measured by further carrying out feature analysis according to the obtained point cloud, so that the process of large-scale component assembly is guided.
However, due to environmental interference, vibration and other problems, the obtained point cloud may have outliers, invalid points, mixing of point clouds in irrelevant areas, rough point clouds and other conditions, which all cause interference to subsequent feature extraction and calculation processes. Because the cloud states of noise points are different, the noise points cannot be well filtered by using a single traditional filtering algorithm. In addition, the existence of invalid point clouds and irrelevant area point clouds can influence the result and also need to be filtered.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a multi-scale large-scale component assembly gap point cloud filtering method, which solves the problems that a single traditional filtering algorithm cannot accurately and effectively filter point clouds with outlier points, invalid points, irrelevant area point cloud mixing and rough point clouds, and the like. The segmentation process can identify areas with higher density in the point cloud, so that effective point cloud data can be reserved. And then, filtering invalid points and large-scale outliers from the segmented ROI region point cloud to improve the accuracy of subsequent processing. And then, clustering the filtered point cloud by adopting a region growing method, so as to form a point cloud cluster with continuity. Secondly, dividing and filtering the clustered point cloud data based on a weight voting method; judging whether the point belongs to the point cloud irrelevant to the gap or not by analyzing the weight of each point, and further segmenting and filtering the operation so as to improve the accuracy of the point cloud relevant to the gap; finally, aiming at the situation of rough point cloud of a local area, small-scale point cloud fairing processing is carried out on the point cloud; the processing can smooth the point cloud data, so that the point cloud data is more continuous and finer, and the quality of the point cloud data is improved; the invention utilizes a multi-scale filtering method to rapidly filter the point cloud of the assembly gap of the large-scale component, thereby realizing rapid and comprehensive filtering treatment of the point cloud.
In order to solve the technical problems, the invention provides the following technical scheme: a multi-scale large-scale component assembly clearance point cloud filtering method comprises the following steps:
s1, segmenting an ROI (region of interest) region of an original point cloud P of a large component assembly gap, and extracting by a point cloud density sensor to obtain a point cloud densityMinimum ROI region Point cloud->;
S2, for the ROI region point cloudFiltering the invalid points and the large-scale outlier points to obtain filtered point clouds;
S3, extracting by using a region growing clustering methodClustering all point clouds, and dividing and filtering the gap irrelevant point clouds based on a weight voting method to obtain effective point clouds +.>;
S4, through effective point cloudPerforming small-scale point cloud fairing to solve the problem of rough point cloud in a local area and obtaining final filtered point cloud +.>。
Further, in the step S1, the initial point cloud P of the assembly gap of the large-scale component is segmented into ROI areas, and the point cloud density is obtained through extraction of a point cloud density sensorMinimum ROI region Point cloud->The specific process comprises the following steps:
s11, initializing a minimum coordinate set of three dimensions of the point cloud X, Y, ZThe middle element is infinity and maximum coordinate set +.>The middle element is infinitely small, all point clouds are traversed through a point cloud iterator, and new ++is finally obtained through continuous iterative updating>And->Constructing an integral bounding box B of the point cloud according to the two sets of parameters;
s12, constructing a point cloud density sensor, wherein、/>、/>Is the length, width and height of the point cloud density sensor,,/>is 2 times the gap width based on prior, +.>;
S13, in the point cloud integral bounding box B, performing iterative search through a point cloud density sensor to perform density calculation, and performing point cloud density calculation,/>For the number of points in the current density sensor range, determining the point cloud density by searching>Minimum ROI region Point cloud->。
Further, in the step S2, the ROI region point cloud is obtainedPerforming ineffective spot filteringFiltering the filtered points by removing large-scale outlier filtering to obtain filtered point cloud +.>The specific process comprises the following steps:
s21, performing constraint on the object by a multi-condition constraint methodFiltering the invalid point cloud to make constraint condition +.>Co (all ]>Whether the individual condition judging point cloud is an invalid point cloud or not, and judging the condition +.>,/>For a point in the current point cloud, when +.>When the point cloud is an invalid point cloud, the point cloud after filtering the invalid point cloud is +.>;
S22, point cloudIs>Defining a local neighborhood->For every point->Curvature of->And its office-basedPart neighborhood->Curvature characteristics of->Based on dot->Curvature of->And local curvature characteristics thereof>Filtering the outlier by the standard deviation statistical method to obtain filtered point cloud +.>。
Further, the step S3 is extracted by a region growing clustering methodClustering all point clouds, and dividing and filtering the gap irrelevant point clouds based on a weight voting method to obtain effective point clouds +.>The specific process comprises the following steps:
s31, clustering point clouds in a mode based on region growth, and selectingThe inner single point is used as a seed point to be used as a clustering starting point to construct k-near neighborhood +.>Cosine similarity-based->The way the metrics are determined determines the similarity between points,/-between points>The value of the product is [ -1,1]Within the range of>Representing the current seed point, ++>Representing a current neighborhoodThe points in the tree are clustered according to cosine similarity of seed points, the process is repeated until no point meeting the condition is generated, and then the tree is clustered again from +.>Selecting seed points from the remaining point cloud and repeating the above process until +.>Ending without point cloud;
s32, calculating the average Hausdorff distance of each clustering point cloud relative to other point cloudsAnd normal +/of each cluster point cloud>And the number of clouds per cluster +.>And similarity with template point cloud +.>,/>;
S33, carrying out weight voting on each cluster in a weight voting based mode, and finally obtaining the score of each cluster as followsThe calculation formula is +.>Wherein,/>,/>,/>Is->Fitting the normal of the line, +.>For the modular length of the two vectors, will +.>The two clusters with the maximum value are extracted to be used as inner point clouds, and the point clouds are treated as outer point clouds to obtain effective point clouds +.>。
Further, the step S4 is implemented through a point cloudPerforming small-scale point cloud fairing to solve the problem of rough point cloud in a local area and obtaining final filtered point cloud +.>The specific process comprises the following steps:
s41, determining a proper small-scale smooth radius, wherein the radius is used for defining a smooth operation range, and the smooth radius is selected according to the characteristics of point clouds and application requirements;
s42, for the effective point cloudEach point of->Determining its neighborhood according to the smooth radius>Neighborhood, method of generating a neighborhood, and computer program productIs an adjacent point contained within a smooth radius range;
s43, for neighborhoodCalculating a normal vector thereof by using a least square method;
s44, for each pointUtilize neighborhood->The normal line information in the neighborhood is subjected to smoothing operation, a weighted average method is used, the new position of the point is determined by the weighted average of the points in the neighborhood, the weight is calculated according to the consistency of the point and the normal line directions of other points in the neighborhood, the processes S41-S44 are repeated until the whole point cloud is subjected to smoothing treatment, and the final filtered point cloud is obtained>。
By means of the technical scheme, the invention provides a multi-scale large-scale component assembly gap point cloud filtering method, which has at least the following beneficial effects:
firstly, the original point cloud is segmented into the ROI areas based on the point cloud density sensing method, and the segmentation process can identify areas with higher density in the point cloud, so that effective point cloud data are reserved. And then, filtering invalid points and large-scale outliers from the segmented ROI region point cloud to improve the accuracy of subsequent processing. Then, clustering operation is carried out on the filtered point cloud by adopting a region growing method so as to form a point cloud cluster with continuity, then, the clustered point cloud data is segmented and filtered by adopting a weight voting-based method, and whether the point cloud belongs to the point cloud irrelevant to the gap is judged by analyzing the weight of each point, and then the segmentation and filtering operation is carried out so as to improve the accuracy of the point cloud relevant to the gap. And finally, aiming at the situation of rough point cloud of the local area, carrying out small-scale point cloud fairing processing on the point cloud. The method can smooth the point cloud data, so that the point cloud data is more continuous and finer, and the quality of the point cloud data is improved; the invention realizes rapid and comprehensive filtering treatment of the point cloud by utilizing a multi-scale filtering method.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:
FIG. 1 is a flow chart of a multi-scale large part assembly gap point cloud filtering method provided by the invention;
FIG. 2 is a schematic diagram of an overall bounding box of a point cloud constructed in accordance with the present invention;
FIG. 3 is a schematic diagram of the effect of the large-scale filtering and extraneous point filtering of the three-dimensional point cloud of the assembly gap of the large-scale component of the present invention;
fig. 4 is a schematic diagram of the final filtering effect of the filtering of the three-dimensional point cloud of the large part assembly gap of the present invention.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description. Therefore, the implementation process of how to apply the technical means to solve the technical problems and achieve the technical effects can be fully understood and implemented.
Those of ordinary skill in the art will appreciate that all or a portion of the steps in a method of implementing an embodiment described above may be implemented by a program to instruct related hardware, and thus the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Referring to fig. 1-4, a specific implementation manner of the present embodiment is shown, first, an ROI region is segmented from an original point cloud by a method based on point cloud density sensing. The segmentation process can identify areas with higher density in the point cloud, so that effective point cloud data can be reserved. And then, filtering invalid points and large-scale outliers from the segmented ROI region point cloud to improve the accuracy of subsequent processing. And then, clustering the filtered point cloud by adopting a region growing method, so as to form a point cloud cluster with continuity. And secondly, dividing and filtering the clustered point cloud data based on a weight voting method. And judging whether the point belongs to the point cloud irrelevant to the gap or not by analyzing the weight of each point, and further performing segmentation and filtering operations so as to improve the accuracy of the point cloud relevant to the gap. And finally, aiming at the situation of rough point cloud of the local area, carrying out small-scale point cloud fairing processing on the point cloud. The processing can smooth the point cloud data, so that the point cloud data is more continuous and finer, and the quality of the point cloud data is improved. The invention realizes rapid and comprehensive filtering treatment of the point cloud by utilizing a multi-scale filtering method.
Referring to fig. 1, the present embodiment provides a multi-scale large-scale component assembly gap point cloud filtering method, which includes the following steps:
s1, segmenting an ROI (region of interest) region of an original point cloud P of a large component assembly gap, and extracting by a point cloud density sensor to obtain a point cloud densityMinimum ROI region Point cloud->;
As a preferred embodiment of step S1, the specific procedure comprises the steps of:
s11, initializing point cloud X, Y,Minimum coordinate set of three dimensions ZThe middle element is infinity and maximum coordinate set +.>The middle element is infinitely small, all point clouds are traversed through a point cloud iterator, and new ++is finally obtained through continuous iterative updating>And->And constructing an integral bounding box B of the point cloud according to the two sets of parameters, wherein the effect is shown in figure 2;
s12, constructing a point cloud density sensor, wherein、/>、/>Is the length, width and height of the point cloud density sensor,,/>is 2 times the gap width based on prior, +.>;
S13, in the point cloud integral bounding box B, performing iterative search through a point cloud density sensor to perform density calculation, and performing point cloud density calculation,/>For the current density sensor rangeThe number of interior points, the point cloud density is determined by searching>Minimum ROI region Point cloud->。
In this embodiment, the ROI area is segmented for the original point cloud by a method based on the point cloud density sensing. The segmentation process can identify areas with higher density in the point cloud, so that effective point cloud data can be reserved.
S2, for the ROI region point cloudFiltering the invalid points and the large-scale outlier points to obtain filtered point clouds;
As a preferred embodiment of step S2, the specific procedure comprises the steps of:
s21, performing constraint on the object by a multi-condition constraint methodFiltering the invalid point cloud to make constraint condition +.>Co (all ]>Whether the individual condition judging point cloud is an invalid point cloud or not, and judging the condition +.>,/>For a point in the current point cloud, when +.>When the point cloud is an invalid point cloud, the point cloud after filtering the invalid point cloud is +.>;
S22, point cloudIs>Defining a local neighborhood->For every point->Curvature of->And based on local neighborhood->Curvature characteristics of->Based on dot->Curvature of->And local curvature characteristics thereof>Filtering the outlier by the standard deviation statistical method to obtain filtered point cloud +.>;
Wherein,,/>、/>is a dot->Is>;
;
Is a local neighborhood->The number of points in>Is->Neighborhood point of->Is provided for the curvature of the lens.
In this embodiment, the invalid points and large-scale outliers of the segmented ROI region point cloud are filtered to improve accuracy of subsequent processing.
S3, extracting by using a region growing clustering methodClustering all point clouds, and dividing and filtering the gap irrelevant point clouds based on a weight voting method to obtain effective point clouds +.>;
As a preferred embodiment of step S3, the specific procedure comprises the steps of:
s31, clustering point clouds in a mode based on region growth, and selectingThe inner single point is used as a seed point to be used as a clustering starting point to construct k-near neighborhood +.>Cosine similarity-based->The way the metrics are determined determines the similarity between points,/-between points>The value of the product is [ -1,1]Within the range of>Representing the current seed point, ++>Representing a current neighborhoodThe points in the tree are clustered according to cosine similarity of seed points, the process is repeated until no point meeting the condition is generated, and then the tree is clustered again from +.>Selecting seed points from the remaining point cloud and repeating the above process until +.>Ending without point cloud;
s32, calculating the average Hausdorff distance of each clustering point cloud relative to other point cloudsAnd normal +/of each cluster point cloud>And the number of clouds per cluster +.>And is similar to a template point cloudDegree->,/>;
Wherein:,/>,/>and->Is made of->Clustering of middle divisions;
s33, carrying out weight voting on each cluster in a weight voting based mode, and finally obtaining the score of each cluster as followsThe calculation formula is +.>Wherein,/>,/>,/>Is->Fitting the normal of the line, +.>For the modular length of the two vectors, will +.>The two clusters with the maximum value are extracted to be used as inner point clouds, and the point clouds are treated as outer point clouds to obtain effective point clouds +.>The effect is shown in fig. 3.
In this embodiment, a region growing method is used to perform clustering operation on the filtered point cloud, so as to form a point cloud cluster with continuity. And secondly, dividing and filtering the clustered point cloud data based on a weight voting method. And judging whether the point belongs to the point cloud irrelevant to the gap or not by analyzing the weight of each point, and further segmenting and filtering the operation so as to improve the accuracy of the point cloud relevant to the gap.
S4, through effective point cloudPerforming small-scale point cloud fairing to solve the problem of rough point cloud in a local area and obtaining final filtered point cloud +.>。
As a preferred embodiment of step S4, the specific process comprises the steps of:
s41, determining a proper small-scale smooth radius, wherein the radius is used for defining a smooth operation range, and the smooth radius is selected according to the characteristics of point clouds and application requirements;
s42, for the effective point cloudEach point of->Determining its neighborhood according to the smooth radius>Neighborhood, method of generating a neighborhood, and computer program productIs an adjacent point contained within a smooth radius range;
s43, for neighborhoodCalculating a normal vector thereof by using a least square method;
s44, for each pointUtilize neighborhood->The normal line information in the neighborhood is subjected to smoothing operation, a weighted average method is used, the new position of the point is determined by the weighted average of the points in the neighborhood, the weight is calculated according to the consistency of the point and the normal line directions of other points in the neighborhood, the processes S41-S44 are repeated until the whole point cloud is subjected to smoothing treatment, and the final filtered point cloud is obtained>The effect is shown in fig. 4.
Correction of sample point position using weighted average of neighboring sample points is expressed asWherein->For original point cloud->For filtered point cloud->Is a filtering factor, n is->A normal of the location;
;
calculation ofIs->And->Is>Is->At->Distance in the normal direction,/>Is->Is (are) neighborhood points->Representing the vector product>、/>Is Gaussian kernel function, representing the adjacent point pair +.>The weight is affected. />Is->Distance to each nearest point, +.>Is->To each of the adjacent points to->The projection distance on the normal has an impact weight on this point.
In this embodiment, aiming at the situation that the local area point cloud is rough, small-scale point cloud fairing processing is performed on the point cloud. The processing can smooth the point cloud data, so that the point cloud data is more continuous and finer, and the quality of the point cloud data is improved.
In conclusion, the method utilizes a multi-scale filtering method to rapidly filter the point cloud of the assembly gap of the large-scale component, and achieves rapid and comprehensive filtering treatment of the point cloud.
The above is only a preferred embodiment of the present invention, and the scope of the present invention is not limited to the above embodiment, and all technical solutions belonging to the concept of the present invention are within the scope of the present invention. It should be noted that modifications and adaptations to the invention without departing from the principles thereof are intended to be within the scope of the invention as set forth in the following claims.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions.
The foregoing embodiments have been presented in a detail description of the invention, and are presented herein with a particular application to the understanding of the principles and embodiments of the invention, the foregoing embodiments being merely intended to facilitate an understanding of the method of the invention and its core concepts; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present invention, the present description should not be construed as limiting the present invention in view of the above.
Claims (5)
1. The multi-scale large-scale component assembly gap point cloud filtering method is characterized by comprising the following steps of:
s1, segmenting an ROI (region of interest) region of an original point cloud P of a large component assembly gap, and extracting by a point cloud density sensor to obtain a point cloud densityMinimum ROI region Point cloud->;
S2, for the ROI region point cloudFiltering the invalid points and the large-scale outlier points to obtain filtered point cloud +.>;
S3, extracting by using a region growing clustering methodClustering of all point cloudsDividing and filtering the gap irrelevant point cloud based on a weight voting method to obtain an effective point cloud +.>;
S4, through effective point cloudPerforming small-scale point cloud fairing to solve the problem of rough point cloud in a local area and obtaining final filtered point cloud +.>。
2. The multi-scale large component assembly gap point cloud filtering method according to claim 1, wherein in S1, the large component assembly gap original point cloud P is segmented into ROI areas, and the point cloud density is obtained by extracting through a point cloud density sensorMinimum ROI region Point cloud->The specific process comprises the following steps:
s11, initializing a minimum coordinate set of three dimensions of the point cloud X, Y, ZThe middle element is infinity and maximum coordinate set +.>The middle element is infinitely small, all point clouds are traversed through a point cloud iterator, and new ++is finally obtained through continuous iterative updating>And->Constructing an integral bounding box B of the point cloud according to the two sets of parameters;
s12, constructing a point cloud density sensor, wherein、/>、/>Length, width and height of the point cloud density sensor are +.>,Is 2 times the gap width based on prior, +.>;
S13, in the point cloud integral bounding box B, performing iterative search through a point cloud density sensor to perform density calculation, and performing point cloud density calculation,/>For the number of points in the current density sensor range, determining the point cloud density by searching>Minimum ROI region Point cloud->。
3. The method for multi-scale large part assembly gap point cloud filtering according to claim 1, which is characterized in thatCharacterized in that the point cloud of the ROI area is obtained in the S2Filtering the invalid points and the large-scale outlier points to obtain filtered point cloud +.>The specific process comprises the following steps:
s21, performing constraint on the object by a multi-condition constraint methodFiltering the invalid point cloud to make constraint condition +.>Co (all ]>Whether the individual condition judging point cloud is an invalid point cloud or not, and judging the condition +.>,/>For a point in the current point cloud, when +.>When the point cloud is an invalid point cloud, the point cloud after filtering the invalid point cloud is +.>;
S22, point cloudIs>Defining a local neighborhood->For every point->Curvature of->And based on local neighborhood->Curvature characteristics of->Based on dot->Curvature of->And local curvature characteristics thereof>Filtering the outlier by the standard deviation statistical method to obtain filtered point cloud +.>。
4. The multi-scale large part assembly gap point cloud filtering method according to claim 1, wherein the step of extracting in the step of S3 is performed by a region growing clustering methodClustering all point clouds, and dividing and filtering the gap irrelevant point clouds based on a weight voting method to obtain effective point clouds +.>The specific process comprises the following steps:
S31、clustering point clouds by a mode based on region growth, and selectingThe inner single point is used as a seed point to be used as a clustering starting point to construct k-near neighborhood +.>Cosine similarity-based->The way the metrics are determined determines the similarity between points,/-between points>The value of the product is [ -1,1]Within the range of>Representing the current seed point, ++>Representing the current neighborhood->The points in the tree are clustered according to cosine similarity of seed points, the process is repeated until no point meeting the condition is generated, and then the tree is clustered again from +.>Selecting seed points from the remaining point cloud and repeating the above process until +.>Ending without point cloud;
s32, calculating the average Hausdorff distance of each clustering point cloud relative to other point cloudsAnd normal +/of each cluster point cloud>And the number of clouds per cluster +.>And similarity with template point cloud +.>,/>;
S33, carrying out weight voting on each cluster in a weight voting based mode, and finally obtaining the score of each cluster as followsThe calculation formula is +.>Wherein,/>,/>,/>Is->Fitting the normal of the line, +.>For the modular length of the two vectors, will +.>The two clusters with the largest value are extracted asTreating the point cloud as an external point cloud to obtain an effective point cloud>。
5. The multi-scale large part assembly gap point cloud filtering method according to claim 1, wherein the step S4 is performed by using point cloudPerforming small-scale point cloud fairing to solve the problem of rough point cloud in a local area and obtaining final filtered point cloud +.>The specific process comprises the following steps:
s41, determining a proper small-scale smooth radius, wherein the radius is used for defining a smooth operation range, and the smooth radius is selected according to the characteristics of point clouds and application requirements;
s42, for the effective point cloudEach point of->Determining its neighborhood according to the smooth radius>Neighborhood->Is an adjacent point contained within a smooth radius range;
s43, for neighborhoodCalculating a normal vector thereof by using a least square method;
s44, for each pointUtilize neighborhood->The normal line information in the neighborhood is subjected to smoothing operation, a weighted average method is used, the new position of the point is determined by the weighted average of the points in the neighborhood, the weight is calculated according to the consistency of the point and the normal line directions of other points in the neighborhood, the processes S41-S44 are repeated until the whole point cloud is subjected to smoothing treatment, and the final filtered point cloud is obtained>。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410224098.2A CN117808703B (en) | 2024-02-29 | 2024-02-29 | Multi-scale large-scale component assembly gap point cloud filtering method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410224098.2A CN117808703B (en) | 2024-02-29 | 2024-02-29 | Multi-scale large-scale component assembly gap point cloud filtering method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117808703A true CN117808703A (en) | 2024-04-02 |
CN117808703B CN117808703B (en) | 2024-05-10 |
Family
ID=90422173
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410224098.2A Active CN117808703B (en) | 2024-02-29 | 2024-02-29 | Multi-scale large-scale component assembly gap point cloud filtering method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117808703B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118392054A (en) * | 2024-06-27 | 2024-07-26 | 南京航空航天大学 | Method for measuring thickness of aerospace wallboard based on point cloud data and Haosdorf measurement |
Citations (45)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103049751A (en) * | 2013-01-24 | 2013-04-17 | 苏州大学 | Improved weighting region matching high-altitude video pedestrian recognizing method |
KR101612779B1 (en) * | 2014-11-03 | 2016-04-15 | 계명대학교 산학협력단 | Method of detecting view-invariant, partially occluded human in a plurality of still images using part bases and random forest and a computing device performing the method |
US20170016870A1 (en) * | 2012-06-01 | 2017-01-19 | Agerpoint, Inc. | Systems and methods for determining crop yields with high resolution geo-referenced sensors |
WO2017011595A1 (en) * | 2015-07-13 | 2017-01-19 | Agerpoint, Inc. | Modular systems and methods for determining crop yields with high resolution geo-referenced sensors |
WO2017087415A1 (en) * | 2015-11-17 | 2017-05-26 | The Board Of Trustees Of The Leland Stanford Junior University | Profiling of pathology images for clinical applications |
CN106909902A (en) * | 2017-03-01 | 2017-06-30 | 北京航空航天大学 | A kind of remote sensing target detection method based on the notable model of improved stratification |
CN107369161A (en) * | 2017-07-19 | 2017-11-21 | 无锡信捷电气股份有限公司 | A kind of workpiece point cloud segmentation method at random based on the European cluster of improvement |
CN109409437A (en) * | 2018-11-06 | 2019-03-01 | 安徽农业大学 | A kind of point cloud segmentation method, apparatus, computer readable storage medium and terminal |
WO2019073252A1 (en) * | 2017-10-13 | 2019-04-18 | The Chancellor, Masters And Scholars Of The University Of Oxford | Methods and systems for analysing time ordered image data |
WO2019242174A1 (en) * | 2018-06-21 | 2019-12-26 | 华南理工大学 | Method for automatically detecting building structure and generating 3d model based on laser radar |
US20200240794A1 (en) * | 2019-01-28 | 2020-07-30 | Uatc, Llc | Scaffolds for globally consistent maps |
WO2020154967A1 (en) * | 2019-01-30 | 2020-08-06 | Baidu.Com Times Technology (Beijing) Co., Ltd. | Map partition system for autonomous vehicles |
WO2020154966A1 (en) * | 2019-01-30 | 2020-08-06 | Baidu.Com Times Technology (Beijing) Co., Ltd. | A rgb point clouds based map generation system for autonomous vehicles |
WO2020154964A1 (en) * | 2019-01-30 | 2020-08-06 | Baidu.Com Times Technology (Beijing) Co., Ltd. | A point clouds registration system for autonomous vehicles |
WO2021097618A1 (en) * | 2019-11-18 | 2021-05-27 | 深圳市大疆创新科技有限公司 | Point cloud segmentation method and system, and computer storage medium |
WO2021129317A1 (en) * | 2019-12-26 | 2021-07-01 | 华南理工大学 | Point cloud smoothing filtering method based on normal vector |
CN113269791A (en) * | 2021-04-26 | 2021-08-17 | 西安交通大学 | Point cloud segmentation method based on edge judgment and region growth |
US11128636B1 (en) * | 2020-05-13 | 2021-09-21 | Science House LLC | Systems, methods, and apparatus for enhanced headsets |
WO2021184757A1 (en) * | 2020-03-14 | 2021-09-23 | 苏州艾吉威机器人有限公司 | Robot vision terminal positioning method and device, and computer-readable storage medium |
US20210331815A1 (en) * | 2020-04-27 | 2021-10-28 | Nanjing University Of Aeronautics And Astronautics | Method for controlling gap distribution of wing-fuselage joining based on measured data |
US20210374391A1 (en) * | 2020-05-28 | 2021-12-02 | Science House LLC | Systems, methods, and apparatus for enhanced cameras |
US20210373676A1 (en) * | 2020-06-01 | 2021-12-02 | Science House LLC | Systems, methods, and apparatus for enhanced presentation remotes |
CN114155355A (en) * | 2021-12-07 | 2022-03-08 | 亿嘉和科技股份有限公司 | Point cloud-based electrified main wire detection method and device |
US20220122317A1 (en) * | 2020-10-15 | 2022-04-21 | Nanjing University Of Aeronautics And Astronautics | Method for measuring a seam on aircraft skin based on large-scale point cloud |
CN114492619A (en) * | 2022-01-22 | 2022-05-13 | 电子科技大学 | Point cloud data set construction method and device based on statistics and concave-convex property |
WO2022099528A1 (en) * | 2020-11-12 | 2022-05-19 | 深圳元戎启行科技有限公司 | Method and apparatus for calculating normal vector of point cloud, computer device, and storage medium |
US20220183208A1 (en) * | 2020-10-16 | 2022-06-16 | Verdant Robotics, Inc. | Autonomous detection and control of vegetation |
WO2022193335A1 (en) * | 2021-03-15 | 2022-09-22 | 深圳大学 | Point cloud data processing method and apparatus, and computer device and storage medium |
CN115147433A (en) * | 2021-03-29 | 2022-10-04 | 苏州杰锐思智能科技股份有限公司 | Point cloud registration method |
CN115294293A (en) * | 2022-10-08 | 2022-11-04 | 速度时空信息科技股份有限公司 | Method for automatically compiling high-precision map road reference lines based on low-altitude aerial photography results |
CN115578398A (en) * | 2022-10-25 | 2023-01-06 | 华南理工大学 | Weld point cloud segmentation method based on region growing method |
CN115601269A (en) * | 2022-11-01 | 2023-01-13 | 南京工业大学(Cn) | Multi-scale filtering method for structural point cloud data considering environmental dynamic influence |
WO2023024482A1 (en) * | 2021-08-23 | 2023-03-02 | 奥比中光科技集团股份有限公司 | Interior structured reconstruction method and apparatus, and computer-readable storage medium |
RU2791416C1 (en) * | 2021-11-12 | 2023-03-07 | Цзянсуская корпорация по ядерной энергетике | Method for three-dimensional reconstruction of the thread of the holes for the studs of the main connector of the reactor pressure vessel and automatic identification of defects |
CN116012600A (en) * | 2023-01-30 | 2023-04-25 | 西湾智慧(广东)信息科技有限公司 | Method for extracting point cloud local curved surface features based on morphology |
WO2023066231A1 (en) * | 2021-10-18 | 2023-04-27 | 北京魔鬼鱼科技有限公司 | Vehicle point cloud recognition imaging method, system, computer device, and storage medium |
CN116503419A (en) * | 2023-03-31 | 2023-07-28 | 云南电网有限责任公司曲靖供电局 | Line point cloud tree segmentation method based on ground point removal and density guide filtering |
CN116612132A (en) * | 2023-05-25 | 2023-08-18 | 重庆茂侨科技有限公司 | 3D point cloud target segmentation method based on aggregate characteristics |
CN116738339A (en) * | 2023-06-09 | 2023-09-12 | 北京航空航天大学 | Multi-classification deep learning recognition detection method for small-sample electric signals |
CN116883754A (en) * | 2023-07-20 | 2023-10-13 | 成都市勘察测绘研究院(成都市基础地理信息中心) | Building information extraction method for ground LiDAR point cloud |
CN116977679A (en) * | 2023-08-09 | 2023-10-31 | 王子静 | Image acquisition method and system based on image recognition |
CN117115012A (en) * | 2023-07-25 | 2023-11-24 | 山东科技大学 | Road surface point cloud marking segmentation denoising method |
CN117218343A (en) * | 2023-09-11 | 2023-12-12 | 电子科技大学 | Semantic component attitude estimation method based on deep learning |
CN117292181A (en) * | 2023-09-20 | 2023-12-26 | 吉林大学 | Sheet metal part hole group classification and full-size measurement method based on 3D point cloud processing |
CN117408913A (en) * | 2023-12-11 | 2024-01-16 | 浙江托普云农科技股份有限公司 | Method, system and device for denoising point cloud of object to be measured |
-
2024
- 2024-02-29 CN CN202410224098.2A patent/CN117808703B/en active Active
Patent Citations (45)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170016870A1 (en) * | 2012-06-01 | 2017-01-19 | Agerpoint, Inc. | Systems and methods for determining crop yields with high resolution geo-referenced sensors |
CN103049751A (en) * | 2013-01-24 | 2013-04-17 | 苏州大学 | Improved weighting region matching high-altitude video pedestrian recognizing method |
KR101612779B1 (en) * | 2014-11-03 | 2016-04-15 | 계명대학교 산학협력단 | Method of detecting view-invariant, partially occluded human in a plurality of still images using part bases and random forest and a computing device performing the method |
WO2017011595A1 (en) * | 2015-07-13 | 2017-01-19 | Agerpoint, Inc. | Modular systems and methods for determining crop yields with high resolution geo-referenced sensors |
WO2017087415A1 (en) * | 2015-11-17 | 2017-05-26 | The Board Of Trustees Of The Leland Stanford Junior University | Profiling of pathology images for clinical applications |
CN106909902A (en) * | 2017-03-01 | 2017-06-30 | 北京航空航天大学 | A kind of remote sensing target detection method based on the notable model of improved stratification |
CN107369161A (en) * | 2017-07-19 | 2017-11-21 | 无锡信捷电气股份有限公司 | A kind of workpiece point cloud segmentation method at random based on the European cluster of improvement |
WO2019073252A1 (en) * | 2017-10-13 | 2019-04-18 | The Chancellor, Masters And Scholars Of The University Of Oxford | Methods and systems for analysing time ordered image data |
WO2019242174A1 (en) * | 2018-06-21 | 2019-12-26 | 华南理工大学 | Method for automatically detecting building structure and generating 3d model based on laser radar |
CN109409437A (en) * | 2018-11-06 | 2019-03-01 | 安徽农业大学 | A kind of point cloud segmentation method, apparatus, computer readable storage medium and terminal |
US20200240794A1 (en) * | 2019-01-28 | 2020-07-30 | Uatc, Llc | Scaffolds for globally consistent maps |
WO2020154967A1 (en) * | 2019-01-30 | 2020-08-06 | Baidu.Com Times Technology (Beijing) Co., Ltd. | Map partition system for autonomous vehicles |
WO2020154966A1 (en) * | 2019-01-30 | 2020-08-06 | Baidu.Com Times Technology (Beijing) Co., Ltd. | A rgb point clouds based map generation system for autonomous vehicles |
WO2020154964A1 (en) * | 2019-01-30 | 2020-08-06 | Baidu.Com Times Technology (Beijing) Co., Ltd. | A point clouds registration system for autonomous vehicles |
WO2021097618A1 (en) * | 2019-11-18 | 2021-05-27 | 深圳市大疆创新科技有限公司 | Point cloud segmentation method and system, and computer storage medium |
WO2021129317A1 (en) * | 2019-12-26 | 2021-07-01 | 华南理工大学 | Point cloud smoothing filtering method based on normal vector |
WO2021184757A1 (en) * | 2020-03-14 | 2021-09-23 | 苏州艾吉威机器人有限公司 | Robot vision terminal positioning method and device, and computer-readable storage medium |
US20210331815A1 (en) * | 2020-04-27 | 2021-10-28 | Nanjing University Of Aeronautics And Astronautics | Method for controlling gap distribution of wing-fuselage joining based on measured data |
US11128636B1 (en) * | 2020-05-13 | 2021-09-21 | Science House LLC | Systems, methods, and apparatus for enhanced headsets |
US20210374391A1 (en) * | 2020-05-28 | 2021-12-02 | Science House LLC | Systems, methods, and apparatus for enhanced cameras |
US20210373676A1 (en) * | 2020-06-01 | 2021-12-02 | Science House LLC | Systems, methods, and apparatus for enhanced presentation remotes |
US20220122317A1 (en) * | 2020-10-15 | 2022-04-21 | Nanjing University Of Aeronautics And Astronautics | Method for measuring a seam on aircraft skin based on large-scale point cloud |
US20220183208A1 (en) * | 2020-10-16 | 2022-06-16 | Verdant Robotics, Inc. | Autonomous detection and control of vegetation |
WO2022099528A1 (en) * | 2020-11-12 | 2022-05-19 | 深圳元戎启行科技有限公司 | Method and apparatus for calculating normal vector of point cloud, computer device, and storage medium |
WO2022193335A1 (en) * | 2021-03-15 | 2022-09-22 | 深圳大学 | Point cloud data processing method and apparatus, and computer device and storage medium |
CN115147433A (en) * | 2021-03-29 | 2022-10-04 | 苏州杰锐思智能科技股份有限公司 | Point cloud registration method |
CN113269791A (en) * | 2021-04-26 | 2021-08-17 | 西安交通大学 | Point cloud segmentation method based on edge judgment and region growth |
WO2023024482A1 (en) * | 2021-08-23 | 2023-03-02 | 奥比中光科技集团股份有限公司 | Interior structured reconstruction method and apparatus, and computer-readable storage medium |
WO2023066231A1 (en) * | 2021-10-18 | 2023-04-27 | 北京魔鬼鱼科技有限公司 | Vehicle point cloud recognition imaging method, system, computer device, and storage medium |
RU2791416C1 (en) * | 2021-11-12 | 2023-03-07 | Цзянсуская корпорация по ядерной энергетике | Method for three-dimensional reconstruction of the thread of the holes for the studs of the main connector of the reactor pressure vessel and automatic identification of defects |
CN114155355A (en) * | 2021-12-07 | 2022-03-08 | 亿嘉和科技股份有限公司 | Point cloud-based electrified main wire detection method and device |
CN114492619A (en) * | 2022-01-22 | 2022-05-13 | 电子科技大学 | Point cloud data set construction method and device based on statistics and concave-convex property |
CN115294293A (en) * | 2022-10-08 | 2022-11-04 | 速度时空信息科技股份有限公司 | Method for automatically compiling high-precision map road reference lines based on low-altitude aerial photography results |
CN115578398A (en) * | 2022-10-25 | 2023-01-06 | 华南理工大学 | Weld point cloud segmentation method based on region growing method |
CN115601269A (en) * | 2022-11-01 | 2023-01-13 | 南京工业大学(Cn) | Multi-scale filtering method for structural point cloud data considering environmental dynamic influence |
CN116012600A (en) * | 2023-01-30 | 2023-04-25 | 西湾智慧(广东)信息科技有限公司 | Method for extracting point cloud local curved surface features based on morphology |
CN116503419A (en) * | 2023-03-31 | 2023-07-28 | 云南电网有限责任公司曲靖供电局 | Line point cloud tree segmentation method based on ground point removal and density guide filtering |
CN116612132A (en) * | 2023-05-25 | 2023-08-18 | 重庆茂侨科技有限公司 | 3D point cloud target segmentation method based on aggregate characteristics |
CN116738339A (en) * | 2023-06-09 | 2023-09-12 | 北京航空航天大学 | Multi-classification deep learning recognition detection method for small-sample electric signals |
CN116883754A (en) * | 2023-07-20 | 2023-10-13 | 成都市勘察测绘研究院(成都市基础地理信息中心) | Building information extraction method for ground LiDAR point cloud |
CN117115012A (en) * | 2023-07-25 | 2023-11-24 | 山东科技大学 | Road surface point cloud marking segmentation denoising method |
CN116977679A (en) * | 2023-08-09 | 2023-10-31 | 王子静 | Image acquisition method and system based on image recognition |
CN117218343A (en) * | 2023-09-11 | 2023-12-12 | 电子科技大学 | Semantic component attitude estimation method based on deep learning |
CN117292181A (en) * | 2023-09-20 | 2023-12-26 | 吉林大学 | Sheet metal part hole group classification and full-size measurement method based on 3D point cloud processing |
CN117408913A (en) * | 2023-12-11 | 2024-01-16 | 浙江托普云农科技股份有限公司 | Method, system and device for denoising point cloud of object to be measured |
Non-Patent Citations (4)
Title |
---|
REN, Y等: "Overall filtering algorithm for multiscale noise removal from point cloud data", 《IEEE ACCESS》, vol. 9, 14 July 2021 (2021-07-14), pages 110723 - 110734, XP011871838, DOI: 10.1109/ACCESS.2021.3097185 * |
周钦坤等: "一种改进的形态学LiDAR点云滤波算法", 《甘肃科学学报》, vol. 32, no. 5, 31 October 2020 (2020-10-31), pages 1 - 5 * |
李红卫等: "基于聚类的航空导管三维重建方法", 《计算机工程与设计》, vol. 44, no. 4, 30 April 2023 (2023-04-30), pages 1206 - 1212 * |
王月阳: "基于激光雷达的障碍物三维检测与跟踪方法研究", 《中国优秀硕士学位论文全文数据库 工程科技II辑》, no. 12, 15 December 2023 (2023-12-15), pages 035 - 22 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118392054A (en) * | 2024-06-27 | 2024-07-26 | 南京航空航天大学 | Method for measuring thickness of aerospace wallboard based on point cloud data and Haosdorf measurement |
Also Published As
Publication number | Publication date |
---|---|
CN117808703B (en) | 2024-05-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2018107939A1 (en) | Edge completeness-based optimal identification method for image segmentation | |
CN117808703B (en) | Multi-scale large-scale component assembly gap point cloud filtering method | |
CN110222642B (en) | Plane building component point cloud contour extraction method based on global graph clustering | |
WO2022141145A1 (en) | Object-oriented high-resolution remote sensing image multi-scale segmentation method and system | |
CN115222625A (en) | Laser radar point cloud denoising method based on multi-scale noise | |
CN114749342B (en) | Lithium battery pole piece coating defect identification method, device and medium | |
CN109102004A (en) | Cotton-plant pest-insects method for identifying and classifying and device | |
CN108596920A (en) | A kind of Target Segmentation method and device based on coloured image | |
CN109741358B (en) | Superpixel segmentation method based on adaptive hypergraph learning | |
CN114862861B (en) | Lung lobe segmentation method and device based on few-sample learning | |
CN117710603B (en) | Unmanned aerial vehicle image three-dimensional building modeling method under constraint of linear geometry | |
CN106815851A (en) | A kind of grid circle oil level indicator automatic reading method of view-based access control model measurement | |
WO2020164042A1 (en) | Region merging image segmentation algorithm based on boundary extraction | |
CN117788735A (en) | Dynamic point cloud removing method based on grid division | |
CN115147433A (en) | Point cloud registration method | |
CN116452604B (en) | Complex substation scene segmentation method, device and storage medium | |
Guo et al. | Image guided fuzzy clustering for image segmentation | |
CN107492101B (en) | Multi-modal nasopharyngeal tumor segmentation algorithm based on self-adaptive constructed optimal graph | |
Liu et al. | Multi-scale selective image texture smoothing via intuitive single clicks | |
CN111311586B (en) | Nonlinear health analysis system-based data multi-index dynamic integration algorithm and system | |
CN114066923A (en) | 3D Otsu threshold segmentation method based on iteration and dimension decomposition | |
Hyunki et al. | A noble Image segmentation using local area splitting and merging method based on intensity change | |
CN113570630A (en) | Road remote sensing image segmentation method and device, electronic equipment and storage medium | |
CN117274294B (en) | Homologous chromosome segmentation method | |
CN118115767B (en) | Image data sampling method based on second-order adjacent guidance |
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 |