CN115330985B - Data processing method and device for three-dimensional model optimization - Google Patents
Data processing method and device for three-dimensional model optimization Download PDFInfo
- Publication number
- CN115330985B CN115330985B CN202210880323.9A CN202210880323A CN115330985B CN 115330985 B CN115330985 B CN 115330985B CN 202210880323 A CN202210880323 A CN 202210880323A CN 115330985 B CN115330985 B CN 115330985B
- Authority
- CN
- China
- Prior art keywords
- vertex
- data
- model
- preset
- vertex data
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- 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
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/70—Labelling scene content, e.g. deriving syntactic or semantic representations
-
- 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
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Architecture (AREA)
- Computational Linguistics (AREA)
- Remote Sensing (AREA)
- Processing Or Creating Images (AREA)
Abstract
The application discloses a data processing method and device for three-dimensional model optimization. The method comprises the following steps: obtaining a model to be processed, and carrying out vertex extraction processing on the model to be processed based on a preset semantic label to obtain marked vertex data; performing suspension judgment processing on the marked vertex data based on a preset suspension rule, and taking the marked vertex data as suspension vertex data if the marked vertex data meets the preset suspension rule; and deleting the suspended matter vertex data in the model to be processed to obtain the target optimization model. The vertex extraction based on semantic segmentation is carried out on the model to be processed, and suspended matter judgment is carried out according to the extracted vertex, so that suspended matter vertices in the model to be processed are obtained, the suspended matter in the three-dimensional model is automatically judged and deleted, the problem that in the prior art, the suspended matter elimination efficiency of the three-dimensional model is low is solved, and the optimization efficiency of the three-dimensional model with suspended matters is improved.
Description
Technical Field
The application relates to the field of three-dimensional models, in particular to a data processing method and device for three-dimensional model optimization.
Background
With the continuous development of information technology, the space simulation display of the target environment and the product is carried out by constructing a three-dimensional model, more detailed and more visual display can be provided relative to a two-dimensional image, in the three-dimensional reconstruction model, the three-dimensional model obtained by reconstruction has suspension due to the problems of complex structure and motion characteristics of a simulation target, the effect of the three-dimensional model is influenced, for example, when the three-dimensional model of a city is constructed, trees on the roadside are often moved due to the characteristics of complex characteristics, and the like, and the model obtained by three-dimensional reconstruction has the suspension of the trees on the ground, so that the model effect is influenced. In the prior art, aiming at the problem that the model has suspension to influence the model effect due to complex simulation target structure and motion characteristics in a three-dimensional model, the model is optimized in a manual processing mode, and the problem of low working efficiency exists.
Therefore, the three-dimensional model suspended matter elimination in the prior art has the problem of low efficiency.
Disclosure of Invention
The application mainly aims to provide a data processing method and device for three-dimensional model optimization, so as to solve the problem of low efficiency of three-dimensional model suspended matter removal in the prior art, and realize the technical effect of improving the three-dimensional model suspended matter removal efficiency.
To achieve the above object, a first aspect of the present application proposes a data processing method for three-dimensional model optimization, comprising:
obtaining a to-be-processed model, wherein the to-be-processed model is a three-dimensional reconstruction model to be optimized and the to-be-processed model is a model subjected to semantic tag pretreatment;
performing vertex extraction processing on the to-be-processed model based on a preset semantic tag to obtain marked vertex data, wherein the marked vertex data are data extracted from the to-be-processed model, and the marked vertex data are vertex data with the preset semantic tag, and the preset semantic tag corresponds to suspended matters in the to-be-processed model;
performing suspension judgment processing on the marked vertex data based on a preset suspension rule, wherein if the marked vertex data meets the preset suspension rule, the marked vertex data is used as suspension vertex data; and
and deleting the suspended matter vertex data in the model to be processed to obtain a target optimization model.
Optionally, performing a suspension judgment process on the marked vertex data based on a preset suspension rule, and if the marked vertex data meets the preset suspension rule, taking the marked vertex data as suspension vertex data includes:
Clustering the marked vertex data to obtain first vertex data, wherein the first vertex data is data for representing clustered first vertices;
performing vertex growth processing on the first vertex based on a preset topological structure to obtain second vertex data, wherein the second vertex data are data used for representing the second vertex obtained by the first vertex through multiple topological growth; and
traversing the second vertex data in a preset traversal vertex database to judge whether the second vertex data exists in the preset traversal vertex database,
if the second vertex data exists in the preset traversal vertex database, the marked vertex data meets the preset suspended matter rule, and the suspended matter vertex data is obtained; and
and if the second vertex data does not exist in the preset traversal vertex database, performing vertex growth processing on the first vertex based on a topological structure in an iterative manner until the second vertex data exists in the traversal vertex database, so as to obtain the suspended matter vertex data.
Optionally, iteratively performing topology-based vertex growth processing on the first vertex until the second vertex data exists in the traversal vertex database, and obtaining the suspended object vertex data includes:
Performing vertex growth processing on the first vertex based on a topological structure in an iteration manner to obtain updated second vertex data;
identifying the updated second vertex data to obtain updated growth times;
comparing the updated growth times with a preset updated growth threshold,
if the update growth times are equal to the preset update growth threshold, terminating the vertex growth processing to obtain judgment result data without suspended matters;
and if the update growth times are smaller than the preset update growth threshold, performing vertex growth processing on the first vertex based on a topological structure in an iteration mode until the second vertex data exist in the traversal vertex database, and obtaining the suspended object vertex data.
Optionally, clustering the marked vertex data to obtain first vertex data includes:
identifying the marked vertex data to obtain a plurality of marked vertices, wherein the marked vertices are a plurality of model vertices corresponding to preset semantic tags in a model to be processed;
performing mean value clustering processing on the plurality of marked vertexes to obtain a plurality of clustered data, wherein the clustered data are data for the plurality of marked vertexes of the same class;
And carrying out vertex extraction processing on the first clustering data to obtain the first vertex data, wherein the first clustering data is any one of the plurality of clustering data, and the first vertex data is data of a plurality of marked vertices meeting a preset extraction rule in the first clustering.
Optionally, based on a preset semantic label, vertex extraction processing is performed on the model to be processed, and obtaining marked vertex data includes:
carrying out orthographic projection processing on the model to be processed to obtain a model projection image;
carrying out recognition processing based on the road semantic tags on the model projection image to obtain road interval data;
constructing a suspended matter identification interval according to the road interval data; and
and carrying out vertex extraction processing based on the preset semantic label in the suspended matter identification interval to obtain the marked vertex data.
Optionally, based on a preset semantic label, vertex extraction processing is performed on the model to be processed, and obtaining marked vertex data includes:
identifying the model to be processed to obtain suspended matter characteristic data, wherein the suspended matter characteristic data is used for representing suspended matter characteristics in the model to be processed;
Matching semantic tags corresponding to the feature data of the suspended matters in a preset semantic tag database to obtain the preset semantic tags;
carrying out orthographic projection processing on the model to be processed to obtain a model projection image;
and carrying out vertex extraction processing based on the preset semantic label on the model projection image to obtain marked vertex data, wherein the marked vertex data are data used for representing suspended matter vertices in the model to be processed.
According to a second aspect of the present application, there is provided a data processing apparatus for three-dimensional model optimization, comprising:
the model acquisition module is used for acquiring a to-be-processed model, wherein the to-be-processed model is a three-dimensional reconstruction model to be optimized and the to-be-processed model is a model subjected to semantic tag pretreatment;
the vertex extraction module is used for carrying out vertex extraction processing on the to-be-processed model based on a preset semantic tag to obtain marked vertex data, wherein the marked vertex data are data extracted from the to-be-processed model, the marked vertex data are vertex data with the preset semantic tag, and the preset semantic tag corresponds to suspended matters in the to-be-processed model;
The suspended matter judging module is used for carrying out suspended matter judging processing on the marked vertex data based on a preset suspended matter rule, wherein if the marked vertex data meets the preset suspended matter rule, the marked vertex data is used as suspended matter vertex data; and
and the result module is used for deleting the suspended matter vertex data in the model to be processed to obtain a target optimization model.
Optionally, the suspended matter judging module includes:
the clustering module is used for carrying out clustering processing on the marked vertex data to obtain first vertex data, wherein the first vertex data is data for representing the clustered first vertices;
the vertex growth module is used for carrying out vertex growth processing on the first vertex based on a preset topological structure to obtain second vertex data, wherein the second vertex data are data used for representing the second vertex obtained by the first vertex through multiple topological growth; and
a traversal judging module for traversing the second vertex data in a preset traversal vertex database to judge whether the second vertex data exists in the preset traversal vertex database,
If the second vertex data exists in the preset traversal vertex database, the marked vertex data meets the preset suspended matter rule, and the suspended matter vertex data is obtained; and
and if the second vertex data does not exist in the preset traversal vertex database, performing vertex growth processing on the first vertex based on a topological structure in an iterative manner until the second vertex data exists in the traversal vertex database, so as to obtain the suspended matter vertex data.
According to a third aspect of the present application, a computer-readable storage medium storing computer instructions for causing the computer to execute the above-described data processing method for three-dimensional model optimization is provided.
According to a fourth aspect of the present application, there is provided an electronic device comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executable by the at least one processor to cause the at least one processor to perform the data processing method for three-dimensional model optimization described above.
The technical scheme provided by the embodiment of the application can comprise the following beneficial effects:
in the application, a model to be processed is obtained, the model to be processed is a three-dimensional reconstruction model to be optimized, and the model to be processed is a model subjected to semantic tag pretreatment; performing vertex extraction processing on the model to be processed based on a preset semantic tag to obtain marked vertex data, wherein the marked vertex data is data extracted from the model to be processed and the marked vertex data is vertex data with the preset semantic tag, and the preset semantic tag corresponds to suspended matters in the model to be processed; performing suspension judgment processing on the marked vertex data based on a preset suspension rule, and taking the marked vertex data as suspension vertex data if the marked vertex data meets the preset suspension rule; and deleting the suspended matter vertex data in the model to be processed to obtain the target optimization model. The vertex extraction based on semantic segmentation is carried out on the model to be processed, and suspended matter judgment is carried out according to the extracted vertex, so that suspended matter vertices in the model to be processed are obtained, the suspended matter in the three-dimensional model is automatically judged and deleted, the problem that in the prior art, the suspended matter elimination efficiency of the three-dimensional model is low is solved, and the optimization efficiency of the three-dimensional model with suspended matters is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application, are incorporated in and constitute a part of this specification. The drawings and their description are illustrative of the application and are not to be construed as unduly limiting the application. In the drawings:
FIG. 1 is a flow chart of a data processing method for three-dimensional model optimization provided by the application;
FIG. 2 is a flow chart of a data processing method for three-dimensional model optimization provided by the application;
FIG. 3 is a schematic diagram of a construction suspension identification interval according to the present application;
FIG. 4 is a flow chart of a data processing method for three-dimensional model optimization provided by the application;
FIG. 5 is a schematic diagram of vertex growth based on a preset topology according to the present application;
FIG. 6 is a schematic diagram of a data process for three-dimensional model optimization according to the present application;
FIG. 7 is a schematic diagram of another data processing apparatus for three-dimensional model optimization according to the present application.
Detailed Description
In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate in order to describe the embodiments of the application herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
In the present application, the terms "upper", "lower", "left", "right", "front", "rear", "top", "bottom", "inner", "outer", "middle", "vertical", "horizontal", "lateral", "longitudinal" and the like indicate an azimuth or a positional relationship based on that shown in the drawings. These terms are only used to better describe the present application and its embodiments and are not intended to limit the scope of the indicated devices, elements or components to the particular orientations or to configure and operate in the particular orientations.
Also, some of the terms described above may be used to indicate other meanings in addition to orientation or positional relationships, for example, the term "upper" may also be used to indicate some sort of attachment or connection in some cases. The specific meaning of these terms in the present application will be understood by those of ordinary skill in the art according to the specific circumstances.
Furthermore, the terms "mounted," "configured," "provided," "connected," "coupled," and "sleeved" are to be construed broadly. For example, "connected" may be in a fixed connection, a removable connection, or a unitary construction; may be a mechanical connection, or an electrical connection; may be directly connected, or indirectly connected through intervening media, or may be in internal communication between two devices, elements, or components. The specific meaning of the above terms in the present application can be understood by those of ordinary skill in the art according to the specific circumstances.
FIG. 1 is a flow chart of a data processing method for three-dimensional model optimization according to the present application, as shown in FIG. 1, the method includes the following steps:
s101: obtaining a model to be processed;
the model to be processed is a three-dimensional reconstruction model to be optimized, the model to be processed is a model subjected to semantic tag pretreatment, and the obtaining of the model to be processed comprises the following steps: obtaining an initial three-dimensional model, wherein the initial three-dimensional model is a model obtained after the three-dimensional model is constructed, such as an urban three-dimensional model, and performing semantic tag pretreatment on the initial three-dimensional model to obtain the model to be processed. The semantic label preprocessing of the initial three-dimensional model comprises the following steps: acquiring initial three-dimensional model data, wherein the initial three-dimensional model data is data for representing an initial three-dimensional model and comprises vertex data, image data and the like in the initial three-dimensional model; identifying initial three-dimensional model data to obtain image data, wherein the image data is data of model images in the initial three-dimensional model, carrying out semantic label distribution processing on each pixel in the model images based on a preset convolutional neural network model to obtain preprocessed model images, and carrying out semantic label distribution processing on model vertexes and model images in the initial three-dimensional model based on the preset convolutional neural network model to obtain a model to be processed. For example, if the initial three-dimensional image is a three-dimensional model of a city, the three-dimensional model of the city includes markers such as roads, trees, houses, etc. for representing different marks in the city, semantic label distribution processing based on a preset convolutional neural network model is performed on the three-dimensional model of the initial city, semantic label distribution is performed on model vertices in the three-dimensional model of the initial city for representing various city markers, road vertex marking is performed on pixels in the three-dimensional model of the city for representing model vertices of the roads, tree vertex marking is performed on pixels in the three-dimensional model of the city for representing model vertices of the trees, etc., so as to obtain the city model to be processed.
The label with semantic categories marked on the vertex of the three-dimensional model is obtained by preprocessing the three-dimensional model based on semantic labels, so that in the process of model optimization, labels corresponding to the requirements in the three-dimensional model are conveniently optimized based on the model optimization requirements of users, and the optimization efficiency in the three-dimensional model is improved.
S102: performing vertex extraction processing on the model to be processed based on a preset semantic label to obtain marked vertex data;
the marking vertex data is data extracted from a model to be processed, the marking vertex data is vertex data with preset semantic labels, the preset semantic labels correspond to suspended matters in the model to be processed, various types of suspended matters possibly exist in the model to be processed, the corresponding model labels are different, the suspended matters corresponding to the optimizing requirements of a user model are obtained according to the optimizing requirements of the user model, the preset semantic labels are determined according to the suspended matters, for example, in the optimizing of a city model, suspended trees exist in the city model due to the structures of the trees and the movement of the trees, the suspended trees in the city model are optimized, and the preset semantic labels are set as the semantic labels of the trees.
FIG. 2 is a flow chart of a data processing method for three-dimensional model optimization according to the present application, as shown in FIG. 2, the method includes the following steps:
s201: carrying out orthographic projection treatment on the model to be treated to obtain a model projection image;
the model to be processed is a three-dimensional reconstruction model, when the vertexes in the three-dimensional reconstruction model are extracted, a top view of the three-dimensional reconstruction model is obtained by projecting the three-dimensional reconstruction model from top to bottom, and suspended matter distribution areas in the three-dimensional reconstruction model are conveniently identified through the top view. For example, when the model optimization is performed on the three-dimensional city model, the three-dimensional city model is subjected to front projection from top to bottom to obtain a top view of the three-dimensional city model, and the top view of the three-dimensional city model includes spatial distribution information in the city model.
S202: carrying out recognition processing based on the road semantic tags on the model projection image to obtain road interval data;
the method comprises the steps of carrying out recognition processing based on a road semantic label on a model projection image to obtain a road vertex, carrying out road plane construction processing on the road vertex to obtain a road section, wherein in a three-dimensional reconstruction model, the road has a positioning reference function on marker distribution in the model, and a suspended matter detection area is constructed based on the road section by extracting the road section.
S203: constructing a suspended matter identification interval according to the road interval data;
fig. 3 is a schematic diagram of a construction of a suspended matter identification section according to the present application, where as shown in fig. 3, road section data is identified to obtain a road width W, a road first side R1 and a road second side R2, the road first side R1 is moved to the outside by a distance W1, and the road second side is moved to the outside by a distance W2, so as to obtain a suspended matter identification section, where the suspended matter identification section includes: the interval formed by the first side R1 and the distance W1, the interval formed by the second side R2 and the distance W2, wherein the threshold range of W1 and W2 is determined by the road width W, e.g. the threshold range of W1 and W2 is set to [5%,15% ] of the road width W.
The plane construction based on the model vertexes consumes a great amount of computational resources, and for the suspended matters widely distributed in the three-dimensional model, the construction of the plane area corresponding to the suspended matters needs to consume a great amount of computational resources, the road has the function of positioning reference on the distribution of the markers in the three-dimensional model, the range of the suspended matter identification interval is determined based on the extracted road interval, the construction of the suspended matter identification interval based on the suspended matter vertexes is not needed, the computational power consumption of constructing the detection interval through the suspended matter vertexes is reduced, the construction efficiency of the suspended matter identification interval is improved, and the technical effect of improving the model optimization efficiency is realized.
S204: and carrying out vertex extraction processing based on a preset semantic label in the suspended matter identification interval to obtain marked vertex data.
The method comprises the steps of constructing a suspended matter identification section based on a road, extracting vertexes of preset semantic labels corresponding to suspended matters in the suspended matter identification section to obtain marked vertex data, constructing a suspended matter detection section through the road, extracting vertexes in the detection section, and extracting vertexes in different areas, so that the vertex extraction efficiency is improved, and the technical effect of improving the model optimization efficiency is realized.
In another alternative embodiment of the present application, there is provided a method for extracting vertices of a model to be processed to obtain labeled vertex data, the method including: identifying a model to be processed to obtain feature data of suspended matters, wherein the feature data of the suspended matters are used for representing the feature of the suspended matters in the model to be processed; matching semantic tags corresponding to the feature data of the suspended matters in a preset semantic tag database to obtain preset semantic tags; carrying out orthographic projection treatment on the model to be treated to obtain a model projection image; projecting an image on a model
S102: performing vertex extraction processing on the model to be processed based on a preset semantic label to obtain marked vertex data;
The marking vertex data is data extracted from a model to be processed, the marking vertex data is vertex data with preset semantic labels, the preset semantic labels correspond to suspended matters in the model to be processed, various types of suspended matters possibly exist in the model to be processed, the corresponding model labels are different, the suspended matters corresponding to the optimizing requirements of a user model are obtained according to the optimizing requirements of the user model, the preset semantic labels are determined according to the suspended matters, for example, in the optimizing of a city model, suspended trees exist in the city model due to the structures of the trees and the movement of the trees, the suspended trees in the city model are optimized, and the preset semantic labels are set as the semantic labels of the trees.
FIG. 2 is a flow chart of a data processing method for three-dimensional model optimization according to the present application, as shown in FIG. 2, the method includes the following steps:
s201: carrying out orthographic projection treatment on the model to be treated to obtain a model projection image;
the model to be processed is a three-dimensional reconstruction model, when the vertexes in the three-dimensional reconstruction model are extracted, a top view of the three-dimensional reconstruction model is obtained by projecting the three-dimensional reconstruction model from top to bottom, and suspended matter distribution areas in the three-dimensional reconstruction model are conveniently identified through the top view. For example, when the model optimization is performed on the three-dimensional city model, the three-dimensional city model is subjected to front projection from top to bottom to obtain a top view of the three-dimensional city model, and the top view of the three-dimensional city model includes spatial distribution information in the city model.
S202: carrying out recognition processing based on the road semantic tags on the model projection image to obtain road interval data;
the method comprises the steps of carrying out recognition processing based on a road semantic label on a model projection image to obtain a road vertex, carrying out road plane construction processing on the road vertex to obtain a road section, wherein in a three-dimensional reconstruction model, the road has a positioning reference function on marker distribution in the model, and a suspended matter detection area is constructed based on the road section by extracting the road section.
S203: constructing a suspended matter identification interval according to the road interval data;
fig. 3 is a schematic diagram of a construction of a suspended matter identification section according to the present application, where as shown in fig. 3, road section data is identified to obtain a road width W, a road first side R1 and a road second side R2, the road first side R1 is moved to the outside by a distance W1, and the road second side is moved to the outside by a distance W2, so as to obtain a suspended matter identification section, where the suspended matter identification section includes: the interval formed by the first side R1 and the distance W1, the interval formed by the second side R2 and the distance W2, wherein the threshold range of W1 and W2 is determined by the road width W, e.g. the threshold range of W1 and W2 is set to [5%,15% ] of the road width W.
The plane construction based on the model vertexes consumes a great amount of computational resources, and for the suspended matters widely distributed in the three-dimensional model, the construction of the plane area corresponding to the suspended matters needs to consume a great amount of computational resources, the road has the function of positioning reference on the distribution of the markers in the three-dimensional model, the range of the suspended matter identification interval is determined based on the extracted road interval, the construction of the suspended matter identification interval based on the suspended matter vertexes is not needed, the computational power consumption of constructing the detection interval through the suspended matter vertexes is reduced, the construction efficiency of the suspended matter identification interval is improved, and the technical effect of improving the model optimization efficiency is realized.
S204: and carrying out vertex extraction processing based on a preset semantic label in the suspended matter identification interval to obtain marked vertex data.
The method comprises the steps of constructing a suspended matter identification section based on a road, extracting vertexes of preset semantic labels corresponding to suspended matters in the suspended matter identification section to obtain marked vertex data, constructing a suspended matter detection section through the road, extracting vertexes in the detection section, and extracting vertexes in different areas, so that the vertex extraction efficiency is improved, and the technical effect of improving the model optimization efficiency is realized.
In another alternative embodiment of the present application, there is provided a method for extracting vertices of a model to be processed to obtain labeled vertex data, the method including: identifying a model to be processed to obtain feature data of suspended matters, wherein the feature data of the suspended matters are used for representing the feature of the suspended matters in the model to be processed; matching semantic tags corresponding to the feature data of the suspended matters in a preset semantic tag database to obtain preset semantic tags; carrying out orthographic projection treatment on the model to be treated to obtain a model projection image; and carrying out vertex extraction processing based on a preset semantic label on the model projection image to obtain marked vertex data, wherein the marked vertex data is data for representing suspended matter vertices in the model to be processed.
S103: performing suspended matter judgment processing on the marked vertex data based on a preset suspended matter rule to obtain suspended matter vertex data;
and if the marked vertex data meets the preset suspension rule, taking the marked vertex data as suspension vertex data.
FIG. 4 is a flowchart of a data processing method for three-dimensional model optimization according to the present application, as shown in FIG. 4, the method includes the following steps:
S301: clustering the marked vertex data to obtain first vertex data;
the first vertex data is data for representing the clustered first vertices; identifying the marked vertex data to obtain a plurality of marked vertices, wherein the marked vertices are a plurality of model vertices corresponding to preset semantic tags in the model to be processed; carrying out mean value clustering processing on a plurality of marked vertexes to obtain a plurality of clustered data, wherein the clustered data are data for the plurality of marked vertexes of the same class; and carrying out vertex extraction processing on the first clustering data to obtain first vertex data, wherein the first clustering data is any one of a plurality of clustering data, and the first vertex data is data of a plurality of marked vertices meeting a preset extraction rule in the first clustering.
The suspended matter identification interval comprises a plurality of suspended matters distributed at different positions, the marked vertexes at the same position are clustered into the same class by clustering the marked vertexes, the marked vertexes are clustered by a K-means (K-means clustering algorithm) to obtain a plurality of the same class, and each same class comprises a plurality of marked vertexes. And extracting the seed vertexes from the plurality of marked vertexes in the same class based on a preset seed vertex extraction rule to obtain a seed vertex set.
For example, when optimizing the three-dimensional city model, eliminating the suspended tree in the three-dimensional city model, extracting the tree semantic label-based vertexes in the suspended tree detection interval to obtain a plurality of tree vertexes, clustering the plurality of tree vertexes by a K-means method, clustering the vertexes of the same tree into the same class, for example, clustering the plurality of tree vertexes in the suspended tree detection interval to obtain a class P corresponding to the tree,P={,/>,/>.../>...}，/>for tree vertexes in the current tree, extracting seed vertexes of a class P corresponding to the current tree, and calculating the mass center of the three-dimensional point of the class P>Calculate to satisfy->All +.>Point (/ -)>Representing the minimum threshold distance between the three-dimensional point and the centroid, wherein the threshold range is [0.05,1 ]]Between, in meters), a seed set SP = { = {>,/>,/>.../>...}。
S302: performing vertex growth processing on the first vertex based on a preset topological structure to obtain second vertex data;
the second vertex data are data used for representing a second vertex obtained by multiple times of topological growth of the first vertex, the preset topological structure is a topological structure in the model to be processed, the vertex growth is carried out according to the preset topological structure based on the first vertex, the second vertex connected with the first vertex in the model to be processed is obtained, the second vertex is a vertex in the model to be processed, and the vertex in the model connected with the first vertex is searched through the topological structure growth, so that the second vertex is obtained.
S303: traversing the second vertex data in the preset traversal vertex database to judge whether the second vertex data exist in the preset traversal vertex database or not, and obtaining suspended matter vertex data.
If the second vertex data exists in the preset traversal vertex database, marking the vertex data to meet a preset suspended matter rule, and obtaining suspended matter vertex data; and if the second vertex data does not exist in the preset traversal vertex database, performing vertex growth processing on the first vertex based on the topological structure in an iterative manner until the second vertex data exists in the traversal vertex database, so as to obtain suspended matter vertex data.
The first preset traversal vertex database is a database comprising a plurality of first vertex data, the second traversal vertex database is empty, vertex growth processing based on a topological structure is carried out on any first vertex to obtain a second vertex, whether the second vertex exists in the first preset traversal vertex initial database or not is judged, if yes, the second vertex data is deleted from the first preset traversal vertex initial database, the second vertex data is added to the second preset traversal vertex database, the second vertex data obtained after the vertex growth processing of the topological structure is carried out on the second preset traversal vertex database, the second preset traversal vertex database is searched, the first vertex database comprises the second vertex data obtained after the vertex growth processing of the topological structure is carried out on the first vertex, the second vertex data is judged to be found, and the second vertex data is not updated again to obtain the second vertex data of the new traversal vertex database.
In another alternative embodiment of the present application, there is provided a data processing method for three-dimensional model optimization, the method comprising the steps of:
performing vertex growth processing based on a topological structure on the first vertex iteration to obtain updated second vertex data; identifying the updated second vertex data to obtain updated growth times; comparing the updated growth times with a preset updated growth threshold value, and if the updated growth times are equal to the preset updated growth threshold value, terminating the vertex growth processing to obtain the judgment result data without suspended matters; and if the update growth times are smaller than a preset update growth threshold, performing vertex growth processing on the first vertex based on the topological structure in an iterative manner until second vertex data exist in the traversal vertex database, and obtaining suspended matter vertex data. When the vertex growth processing based on the topological structure is carried out, the threshold value of the growth times is set, so that when suspended matters are not existed in the model to be processed in the process of suspended matters judgment, the suspended matters judgment is controlled to stop the suspended matters searching based on the vertex growth processing based on the topological structure, and the consumption of calculation force is reduced.
For example, fig. 5 is a schematic diagram of vertex growth based on a preset topology structure, where a suspension tree vertex set TP is constructed, and a seed set sp= { based on the suspension tree vertex set TP ,/>,/>.../>...}, initial TP, tp=sp= { = { }>,/>,.../>.. for each three-dimensional point, a set tp_label= { } is marked by the floating tree vertices, which is used to mark the set of three-dimensional points that have been traversed, the set of points to be traversed initially tp_new=sp= { = {>,/>,/>.../>.., growth calculation by topology of three-dimensional model, as shown by dark dots in fig. 5 (i.e. black solid dots in fig. 5), dark dots represent three-dimensional dots +_ in SP seed set>I.e. a set of three-dimensional points in the initial TP, traversing the three-dimensional points in TP_New, e.g. traversing to the three-dimensional points +.>Then it is deleted from tp_new and inserted into tp_label set, and vertex set directly connected with it is obtained by using topology structure of three-dimensional modelIf the light color point is in tp_label, if the light color point is already in tp_label, if the light color point is not already in tp_label, inserting a New light color point into TP and tp_new, repeating the above process until triggering the termination condition, wherein the termination condition is: if no New node can be inserted into TP_New in the growth process, namely TP_New is empty, determining that a suspended tree is found; if the number of vertices in the TP set is greater than NUM_P during growth (NUM_P threshold range is [500, 50000) ]Between) and there are New three-dimensional points in TP New, then it is determined that no floating tree is found.
S104: and deleting the suspended matter vertex data in the model to be processed to obtain a target optimization model.
FIG. 6 is a schematic structural diagram of a data processing apparatus for three-dimensional model optimization according to the present application, as shown in FIG. 6, the apparatus includes:
the model obtaining module 51 is configured to obtain a to-be-processed model, where the to-be-processed model is a three-dimensional reconstruction model to be optimized and the to-be-processed model is a model that is subjected to semantic tag preprocessing;
the vertex extraction module 52 performs vertex extraction processing on the model to be processed based on a preset semantic tag to obtain marked vertex data, wherein the marked vertex data is data extracted from the model to be processed and the marked vertex data is vertex data with the preset semantic tag, and the preset semantic tag corresponds to suspended matters in the model to be processed;
the suspended matter judging module 53 performs suspended matter judging processing on the marked vertex data based on a preset suspended matter rule, wherein if the marked vertex data meets the preset suspended matter rule, the marked vertex data is used as suspended matter vertex data; and
And a result module 54, configured to delete the suspended matter vertex data in the model to be processed, so as to obtain a target optimization model.
FIG. 7 is a schematic structural diagram of another data processing apparatus for three-dimensional model optimization according to the present application, as shown in FIG. 7, the apparatus includes:
the clustering module 61 is configured to perform clustering processing on the marked vertex data to obtain first vertex data, where the first vertex data is data for representing the clustered first vertices;
the vertex growth module 62 is configured to perform vertex growth processing on the first vertex based on a preset topology structure to obtain second vertex data, where the second vertex data is data for representing a second vertex obtained by performing multiple topological growth on the first vertex; and
a traversal judging module 63 for traversing the second vertex data in the preset traversal vertex database to judge whether the second vertex data exists in the preset traversal vertex database,
if second vertex data exists in the preset traversal vertex database, marking the vertex data to meet a preset suspended matter rule, and obtaining suspended matter vertex data; and
and if the second vertex data does not exist in the preset traversal vertex database, performing vertex growth processing on the first vertex based on the topological structure in an iterative manner until the second vertex data exists in the traversal vertex database, so as to obtain suspended matter vertex data.
The specific manner in which the operations of the units in the above embodiments are performed has been described in detail in the embodiments related to the method, and will not be described in detail here.
In summary, in the present application, by acquiring the model to be processed, the model to be processed is a three-dimensional reconstruction model to be optimized, and the model to be processed is a model subjected to semantic tag preprocessing; performing vertex extraction processing on the model to be processed based on a preset semantic tag to obtain marked vertex data, wherein the marked vertex data is data extracted from the model to be processed and the marked vertex data is vertex data with the preset semantic tag, and the preset semantic tag corresponds to suspended matters in the model to be processed; performing suspension judgment processing on the marked vertex data based on a preset suspension rule, and taking the marked vertex data as suspension vertex data if the marked vertex data meets the preset suspension rule; and deleting the suspended matter vertex data in the model to be processed to obtain the target optimization model. The vertex extraction based on semantic segmentation is carried out on the model to be processed, and suspended matter judgment is carried out according to the extracted vertex, so that suspended matter vertices in the model to be processed are obtained, the suspended matter in the three-dimensional model is automatically judged and deleted, the problem that in the prior art, the suspended matter elimination efficiency of the three-dimensional model is low is solved, and the optimization efficiency of the three-dimensional model with suspended matters is improved.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
It will be apparent to those skilled in the art that the elements or steps of the application described above may be implemented in a general purpose computing device, they may be concentrated on a single computing device, or distributed across a network of computing devices, or they may alternatively be implemented in program code executable by computing devices, so that they may be stored in a memory device for execution by the computing devices, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps within them may be fabricated into a single integrated circuit module. Thus, the present application is not limited to any specific combination of hardware and software.
The above description is only of the preferred embodiments of the present application and is not intended to limit the present application, but various modifications and variations can be made to the present application by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.
Claims (8)
1. A data processing method for three-dimensional model optimization, comprising:
obtaining a to-be-processed model, wherein the to-be-processed model is a three-dimensional reconstruction model to be optimized and the to-be-processed model is a model subjected to semantic tag pretreatment;
performing vertex extraction processing on the to-be-processed model based on a preset semantic tag to obtain marked vertex data, wherein the marked vertex data are data extracted from the to-be-processed model, and the marked vertex data are vertex data with the preset semantic tag, and the preset semantic tag corresponds to suspended matters in the to-be-processed model;
performing suspension judgment processing on the marked vertex data based on a preset suspension rule, wherein if the marked vertex data meets the preset suspension rule, taking the marked vertex data as suspension vertex data comprises the following steps:
clustering the marked vertex data to obtain first vertex data, wherein the first vertex data is data for representing clustered first vertices;
performing vertex growth processing on the first vertex based on a preset topological structure to obtain second vertex data, wherein the second vertex data are data used for representing the second vertex obtained by the first vertex through multiple topological growth;
Traversing the second vertex data in a preset traversal vertex database to judge whether the second vertex data exists in the preset traversal vertex database,
if the second vertex data exists in the preset traversal vertex database, the marked vertex data meets the preset suspended matter rule, and the suspended matter vertex data is obtained; and
if the second vertex data does not exist in the preset traversal vertex database, performing vertex growth processing on the first vertex based on a topological structure in an iterative manner until the second vertex data exists in the traversal vertex database, so as to obtain the suspended matter vertex data;
and deleting the suspended matter vertex data in the model to be processed to obtain a target optimization model.
2. The data processing method according to claim 1, wherein iteratively performing topology-based vertex growth processing on the first vertex until the second vertex data exists in the traversal vertex database, obtaining the suspended object vertex data includes:
performing vertex growth processing on the first vertex based on a topological structure in an iteration manner to obtain updated second vertex data;
Identifying the updated second vertex data to obtain updated growth times;
comparing the updated growth times with a preset updated growth threshold,
if the update growth times are equal to the preset update growth threshold, terminating the vertex growth processing to obtain judgment result data without suspended matters;
and if the update growth times are smaller than the preset update growth threshold, performing vertex growth processing on the first vertex based on a topological structure in an iteration mode until the second vertex data exist in the traversal vertex database, and obtaining the suspended object vertex data.
3. The data processing method according to claim 1, wherein clustering the labeled vertex data to obtain first vertex data includes:
identifying the marked vertex data to obtain a plurality of marked vertices, wherein the marked vertices are a plurality of model vertices corresponding to preset semantic tags in a model to be processed;
performing mean value clustering processing on the plurality of marked vertexes to obtain a plurality of clustered data, wherein the clustered data are data for the plurality of marked vertexes of the same class;
and carrying out vertex extraction processing on the first clustering data to obtain the first vertex data, wherein the first clustering data is any one of the plurality of clustering data, and the first vertex data is data of a plurality of marked vertices meeting a preset extraction rule in the first clustering.
4. The data processing method according to claim 1, wherein performing vertex extraction processing on the model to be processed based on a preset semantic tag, obtaining labeled vertex data includes:
carrying out orthographic projection processing on the model to be processed to obtain a model projection image;
carrying out recognition processing based on the road semantic tags on the model projection image to obtain road interval data;
constructing a suspended matter identification interval according to the road interval data; and
and carrying out vertex extraction processing based on the preset semantic label in the suspended matter identification interval to obtain the marked vertex data.
5. The data processing method according to claim 1, wherein performing vertex extraction processing on the model to be processed based on a preset semantic tag, obtaining labeled vertex data includes:
identifying the model to be processed to obtain suspended matter characteristic data, wherein the suspended matter characteristic data is used for representing suspended matter characteristics in the model to be processed;
matching semantic tags corresponding to the feature data of the suspended matters in a preset semantic tag database to obtain the preset semantic tags;
Carrying out orthographic projection processing on the model to be processed to obtain a model projection image;
and carrying out vertex extraction processing based on the preset semantic label on the model projection image to obtain marked vertex data, wherein the marked vertex data are data used for representing suspended matter vertices in the model to be processed.
6. A data processing apparatus for three-dimensional model optimization, comprising:
the model acquisition module is used for acquiring a to-be-processed model, wherein the to-be-processed model is a three-dimensional reconstruction model to be optimized and the to-be-processed model is a model subjected to semantic tag pretreatment;
the vertex extraction module is used for carrying out vertex extraction processing on the to-be-processed model based on a preset semantic tag to obtain marked vertex data, wherein the marked vertex data are data extracted from the to-be-processed model, the marked vertex data are vertex data with the preset semantic tag, and the preset semantic tag corresponds to suspended matters in the to-be-processed model;
the suspended matter judging module is used for carrying out suspended matter judging processing on the marked vertex data based on a preset suspended matter rule, wherein if the marked vertex data meets the preset suspended matter rule, the marked vertex data is used as suspended matter vertex data;
Wherein, the suspended matter judging module includes:
the clustering module is used for carrying out clustering processing on the marked vertex data to obtain first vertex data, wherein the first vertex data is data for representing the clustered first vertices;
the vertex growth module is used for carrying out vertex growth processing on the first vertex based on a preset topological structure to obtain second vertex data, wherein the second vertex data are data used for representing the second vertex obtained by the first vertex through multiple topological growth;
a traversal judging module for traversing the second vertex data in a preset traversal vertex database to judge whether the second vertex data exists in the preset traversal vertex database,
if the second vertex data exists in the preset traversal vertex database, the marked vertex data meets the preset suspended matter rule, and the suspended matter vertex data is obtained; and
if the second vertex data does not exist in the preset traversal vertex database, performing vertex growth processing on the first vertex based on a topological structure in an iterative manner until the second vertex data exists in the traversal vertex database, so as to obtain the suspended matter vertex data;
And the result module is used for deleting the suspended matter vertex data in the model to be processed to obtain a target optimization model.
7. A computer-readable storage medium storing computer instructions for causing the computer to execute the data processing method for three-dimensional model optimization according to any one of claims 1 to 5.
8. An electronic device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores a computer program executable by the at least one processor to cause the at least one processor to perform the data processing method for three-dimensional model optimization of any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210880323.9A CN115330985B (en) | 2022-07-25 | 2022-07-25 | Data processing method and device for three-dimensional model optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210880323.9A CN115330985B (en) | 2022-07-25 | 2022-07-25 | Data processing method and device for three-dimensional model optimization |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115330985A CN115330985A (en) | 2022-11-11 |
| CN115330985B true CN115330985B (en) | 2023-09-08 |
Family
ID=83920102
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210880323.9A Active CN115330985B (en) | 2022-07-25 | 2022-07-25 | Data processing method and device for three-dimensional model optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115330985B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008259879A (en) * | 2008-07-17 | 2008-10-30 | Samii Kk | Image processor, image processing method and program |
| CN107247927A (en) * | 2017-05-23 | 2017-10-13 | 浙江海洋大学 | A kind of remote sensing images coastline information extracting method and system based on K-T Transformation |
| WO2020103110A1 (en) * | 2018-11-22 | 2020-05-28 | 深圳市大疆创新科技有限公司 | Image boundary acquisition method and device based on point cloud map and aircraft |
| CN111325848A (en) * | 2020-03-02 | 2020-06-23 | 中国电建集团昆明勘测设计研究院有限公司 | Automatic batch deletion method for suspended matters of OSGB inclined model |
| CN112085840A (en) * | 2020-09-17 | 2020-12-15 | 腾讯科技（深圳）有限公司 | Semantic segmentation method, device, equipment and computer readable storage medium |
| CN113658309A (en) * | 2021-08-25 | 2021-11-16 | 北京百度网讯科技有限公司 | Three-dimensional reconstruction method, device, equipment and storage medium |
| CN114332370A (en) * | 2021-12-28 | 2022-04-12 | 埃洛克航空科技（北京）有限公司 | Road image processing method, device, equipment and storage medium |
| CN114373056A (en) * | 2021-12-17 | 2022-04-19 | 云南联合视觉科技有限公司 | Three-dimensional reconstruction method and device, terminal equipment and storage medium |
| CN114519712A (en) * | 2022-02-23 | 2022-05-20 | 广州极飞科技股份有限公司 | Point cloud data processing method and device, terminal equipment and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9805058B2 (en) * | 2014-06-01 | 2017-10-31 | Microsoft Technology Licensing, Llc | Visibility of a point of interest based on environmental conditions |
| US20210297453A1 (en) * | 2015-10-28 | 2021-09-23 | Qomplx, Inc. | Pathfinding in two and three-dimensional spaces using an automated planning service |
-
2022
- 2022-07-25 CN CN202210880323.9A patent/CN115330985B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008259879A (en) * | 2008-07-17 | 2008-10-30 | Samii Kk | Image processor, image processing method and program |
| CN107247927A (en) * | 2017-05-23 | 2017-10-13 | 浙江海洋大学 | A kind of remote sensing images coastline information extracting method and system based on K-T Transformation |
| WO2020103110A1 (en) * | 2018-11-22 | 2020-05-28 | 深圳市大疆创新科技有限公司 | Image boundary acquisition method and device based on point cloud map and aircraft |
| CN111325848A (en) * | 2020-03-02 | 2020-06-23 | 中国电建集团昆明勘测设计研究院有限公司 | Automatic batch deletion method for suspended matters of OSGB inclined model |
| CN112085840A (en) * | 2020-09-17 | 2020-12-15 | 腾讯科技（深圳）有限公司 | Semantic segmentation method, device, equipment and computer readable storage medium |
| CN113658309A (en) * | 2021-08-25 | 2021-11-16 | 北京百度网讯科技有限公司 | Three-dimensional reconstruction method, device, equipment and storage medium |
| CN114373056A (en) * | 2021-12-17 | 2022-04-19 | 云南联合视觉科技有限公司 | Three-dimensional reconstruction method and device, terminal equipment and storage medium |
| CN114332370A (en) * | 2021-12-28 | 2022-04-12 | 埃洛克航空科技（北京）有限公司 | Road image processing method, device, equipment and storage medium |
| CN114519712A (en) * | 2022-02-23 | 2022-05-20 | 广州极飞科技股份有限公司 | Point cloud data processing method and device, terminal equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 舒振宇 等.基于分层学习的三维模型兴趣点提取算法.《计算机辅助设计与图形学学报》.2020,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115330985A (en) | 2022-11-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106133756B (en) | System, method and the non-transitory computer-readable medium of filtering, segmentation and identification object | |
| CN109145171A (en) | A kind of multiple dimensioned map data updating method | |
| CN112070769A (en) | Layered point cloud segmentation method based on DBSCAN | |
| CN109584302A (en) | Camera pose optimization method, device, electronic equipment and computer-readable medium | |
| CN113920134B (en) | Slope ground point cloud segmentation method and system based on multi-line laser radar | |
| Xu et al. | Plane segmentation based on the optimal-vector-field in LiDAR point clouds | |
| CN112084871B (en) | High-resolution remote sensing target boundary extraction method based on weak supervised learning | |
| Miao et al. | Label3DMaize: toolkit for 3D point cloud data annotation of maize shoots | |
| US20220004740A1 (en) | Apparatus and Method For Three-Dimensional Object Recognition | |
| Xu et al. | A new clustering-based framework to the stem estimation and growth fitting of street trees from mobile laser scanning data | |
| CN114758086B (en) | Method and device for constructing urban road information model | |
| Li et al. | Unsupervised road extraction via a Gaussian mixture model with object-based features | |
| CN115861247A (en) | High-resolution remote sensing image contour multistage regularization method, system and application | |
| Liu et al. | Image-translation-based road marking extraction from mobile laser point clouds | |
| US11861855B2 (en) | System and method for aerial to ground registration | |
| Yao et al. | Semantic labelling of ultra dense MLS point clouds in urban road corridors based on fusing CRF with shape priors | |
| CN101887522A (en) | Method for extracting remote sensing image road based on multipoint simultaneous and rapid advance | |
| CN113807309A (en) | Orchard machine walking route planning method based on deep learning | |
| CN115330985B (en) | Data processing method and device for three-dimensional model optimization | |
| CN110363847B (en) | Map model construction method and device based on point cloud data | |
| Aksoy et al. | Image mining using directional spatial constraints | |
| CN113487741B (en) | Dense three-dimensional map updating method and device | |
| CN115330984B (en) | Data processing method and device for suspended matter rejection | |
| CN113378800A (en) | Automatic classification and vectorization method for road sign lines based on vehicle-mounted three-dimensional point cloud | |
| CN114445644A (en) | Laser point cloud parameterized semantic feature matching method, device, equipment and storage medium |
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 |