CN117058330B - Three-dimensional reconstruction method, reconstruction model and related equipment for electric power corridor - Google Patents
Three-dimensional reconstruction method, reconstruction model and related equipment for electric power corridor Download PDFInfo
- Publication number
- CN117058330B CN117058330B CN202311314243.8A CN202311314243A CN117058330B CN 117058330 B CN117058330 B CN 117058330B CN 202311314243 A CN202311314243 A CN 202311314243A CN 117058330 B CN117058330 B CN 117058330B
- Authority
- CN
- China
- Prior art keywords
- point cloud
- cloud data
- noise reduction
- reconstruction
- corridor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title abstract description 29
- 230000009467 reduction Effects 0.000 claims abstract description 50
- 238000012545 processing Methods 0.000 claims abstract description 42
- 239000012212 insulator Substances 0.000 claims description 7
- 238000012876 topography Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 description 23
- 230000008569 process Effects 0.000 description 8
- 238000004891 communication Methods 0.000 description 6
- 238000007689 inspection Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004590 computer program Methods 0.000 description 2
- 238000009795 derivation Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/005—General purpose rendering architectures
-
- 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
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Graphics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Geometry (AREA)
- Architecture (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention provides a three-dimensional reconstruction method, a reconstruction model and related equipment for an electric power corridor, and relates to the technical field of three-dimensional reconstruction. The three-dimensional reconstruction method of the electric power corridor comprises the following steps: acquiring first point cloud data of a target area; performing point cloud sparsification processing on the first point cloud data to obtain processed second point cloud data; encoding the second point cloud data to obtain encoded third point cloud data; carrying out noise reduction treatment on the third point cloud data for multiple times to obtain fourth point cloud data after noise removal; decoding the fourth point cloud data to obtain decoded fifth point cloud data; and reconstructing a three-dimensional model of the power corridor according to the fifth point cloud data. The three-dimensional reconstruction method of the electric power corridor effectively shortens the reconstruction period under the condition of ensuring the reconstruction efficiency and accuracy.
Description
Technical Field
The invention relates to the technical field of three-dimensional reconstruction, in particular to a three-dimensional reconstruction method, a reconstruction model and related equipment of an electric power corridor.
Background
In order to ensure the normal operation of the power transmission system, the power corridor needs to be regularly inspected, and because the power corridor is generally erected in mountain forests with large relief and complex structures and areas with complex structures, the manual inspection in the areas is low in efficiency and high in danger, and therefore, unmanned aerial vehicles are generally adopted for inspection.
In contrast, the current unmanned aerial vehicle inspection generally adopts a laser radar, a depth camera and an oblique photography mode to obtain images, and the oblique photography mode performs photogrammetry through a plurality of angles, so that relatively clear images can be obtained, and the unmanned aerial vehicle inspection is more used for electric inspection.
The three-dimensional reconstruction of the power corridor is an important foundation for inspection of the unmanned aerial vehicle power corridor and mainly comprises reconstruction of a power tower pole, a power line, an insulator and corridor topography, high requirements are provided for reconstruction efficiency and accuracy, in the prior art, the unmanned aerial vehicle oblique photography three-dimensional reconstruction is generally generated by adopting commercial visual three-dimensional modeling software, and the problem of long modeling period exists.
In view of the above problems, no effective technical solution is currently available.
Disclosure of Invention
The invention aims to provide a three-dimensional reconstruction method, a reconstruction model and related equipment for an electric power corridor, which can effectively shorten the reconstruction period under the condition of ensuring the reconstruction efficiency and accuracy.
In a first aspect, the present invention provides a three-dimensional reconstruction method for an electric power corridor, which is applied to an image processing system, and includes the following steps:
s1, acquiring first point cloud data of a target area;
s2, performing point cloud sparsification processing on the first point cloud data to obtain processed second point cloud data;
s3, encoding the second point cloud data to obtain encoded third point cloud data;
s4, carrying out noise reduction treatment on the third point cloud data for multiple times to obtain fourth point cloud data after noise removal;
s5, decoding the fourth point cloud data to obtain decoded fifth point cloud data;
s6, reconstructing a three-dimensional model of the electric power corridor according to the fifth point cloud data.
According to the three-dimensional reconstruction method for the electric power corridor, commercial vision three-dimensional modeling software is not required, only the acquired image is needed to be analyzed and processed during implementation, point cloud data related to the electric power corridor are obtained, the point cloud data are further encoded and decoded, and the effects of high reconstruction efficiency and high reconstruction precision are achieved by matching with point cloud sparsification processing and noise reduction processing.
Further, the specific steps in step S2 include:
s21, according to the first point cloud data, obtaining characteristic points of the power tower pole, the power line, the insulator and corridor topography as key characteristic points;
s22, performing point cloud sparsification processing on the characteristic points outside the key characteristic points to obtain the second point cloud data.
The key feature points are highlighted, and the interference of most of irrelevant point clouds is eliminated, so that the three-dimensional model of the electric power corridor is easier to reconstruct, and the three-dimensional reconstruction efficiency is improved.
Further, the specific steps in step S3 include:
s31, encoding the second point cloud data according to the following formula to obtain the third point cloud data:
；
；
wherein,for the third point cloud data, +.>For coding function +.>For the second point cloud data, +.>For the first parameter set, ++>For the first multi-headed attention parameter, +.>Is the first full connection parameter.
The data is encoded, so that the data processing speed can be improved, the processing time of the subsequent noise reduction processing is shortened, and the noise of the data is reduced after the noise reduction processing, so that the accuracy of three-dimensional reconstruction is improved.
Further, the specific steps in step S4 include:
s41, carrying out noise reduction processing on the third point cloud data for a plurality of times according to the following formula to obtain fourth point cloud data:
；
；
；
wherein,for the fourth point cloud data after the first noise reduction, < > is>For the noise reduction function->For the second parameter set, +.>Is fourth point cloud data after i+1st noise reduction,/for the fourth point cloud data>For the fourth point cloud data after the ith noise reduction,>for the total noise reduction times, < >>For the second multi-headed attention parameter, +.>Is the second full connection parameter.
And through repeated noise reduction treatment, the three-dimensional model of the electric power corridor reconstructed subsequently is ensured to have higher precision.
Further, the specific steps in step S5 include:
s51, decoding the fourth point cloud data to obtain fifth point cloud data according to the following formula:
；
；
wherein,for the fifth point cloud data, +.>For the decoding function +.>For the third parameter set, +.>For the third multi-head attention parameter, +.>For the third full connection parameter, < >>And the fourth point cloud data after the n+1st noise reduction is obtained.
In a second aspect, the present invention provides a reconstruction model for use in an image processing system, the reconstruction model being operable to perform the steps of the power corridor three-dimensional reconstruction method as described above.
Further, the reconstruction model is provided with a loss function, and the loss function is used for updating a first parameter set, a second parameter set and a third parameter set when the reconstruction model is subjected to iterative training;
the specific expression of the loss function is as follows:
；
wherein,for the third point cloud data, +.>For the fourth point cloud data after the N-th noise reduction,>for the second point cloud data, +.>For the fifth point cloud data, +.>For the loss function, +.>Indicating desire->Representing the binary norm.
In a third aspect, the present invention provides a three-dimensional reconstruction device for an electric power corridor, which is applied to an image processing system, and includes:
the first acquisition module is used for acquiring first point cloud data of the target area;
the second acquisition module is used for carrying out point cloud sparsification processing on the first point cloud data to obtain processed second point cloud data;
the third acquisition module is used for encoding the second point cloud data to obtain encoded third point cloud data;
the fourth acquisition module is used for carrying out noise reduction processing on the third point cloud data for a plurality of times to obtain denoised fourth point cloud data;
a fifth obtaining module, configured to decode the fourth point cloud data, and obtain decoded fifth point cloud data;
and the reconstruction module is used for reconstructing a three-dimensional model of the electric power corridor according to the fifth point cloud data.
The three-dimensional reconstruction device for the electric power corridor provided by the invention has the advantages of both reconstruction efficiency and reconstruction precision, and realizes the reconstruction of a high-precision three-dimensional model of the electric power corridor in a short period.
In a fourth aspect, the invention provides an electronic device comprising a processor and a memory storing computer readable instructions which, when executed by the processor, perform the steps of the method as provided in the first aspect above.
In a fifth aspect, the present invention provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method as provided in the first aspect above.
According to the three-dimensional reconstruction method of the electric power corridor, the point cloud data are subjected to sparsification processing, so that other characteristics irrelevant to the electric power corridor are weakened, the three-dimensional reconstruction efficiency is improved, noise is estimated and removed through multiple noise reduction processing, the three-dimensional reconstruction precision is improved, and finally, under the condition that the reconstruction efficiency and precision are guaranteed, the point cloud data are encoded and decoded in a short time, and the three-dimensional model of the electric power corridor is obtained through reconstruction.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the embodiments of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and drawings.
Drawings
Fig. 1 is a flowchart of a three-dimensional reconstruction method for an electric power corridor according to an embodiment of the present invention.
Fig. 2 is a schematic structural diagram of a three-dimensional reconstruction device for an electric power corridor according to an embodiment of the present invention.
Fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Description of the reference numerals:
100. a first acquisition module; 200. a second acquisition module; 300. a third acquisition module; 400. a fourth acquisition module; 500. a fifth acquisition module; 600. a reconstruction module; 13. an electronic device; 1301. a processor; 1302. a memory; 1303. a communication bus.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be made by a person skilled in the art without making any inventive effort, are intended to be within the scope of the present invention.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only to distinguish the description, and are not to be construed as indicating or implying relative importance.
Referring to fig. 1, fig. 1 is a flowchart of a three-dimensional reconstruction method of an electric power corridor. The three-dimensional reconstruction method of the electric power corridor is applied to an image processing system and comprises the following steps of:
s1, acquiring first point cloud data of a target area;
s2, performing point cloud sparsification processing on the first point cloud data to obtain processed second point cloud data;
s3, encoding the second point cloud data to obtain encoded third point cloud data;
s4, carrying out noise reduction treatment on the third point cloud data for multiple times to obtain fourth point cloud data after noise removal;
s5, decoding the fourth point cloud data to obtain decoded fifth point cloud data;
s6, reconstructing a three-dimensional model of the electric power corridor according to the fifth point cloud data.
In this embodiment, in practical application, by mounting an oblique photography camera on an unmanned plane, the oblique photography camera obtains stereoscopic image data from a vertical direction and four oblique directions of specific angles, and then the stereoscopic image data is processed by a preprocessing algorithm, a shake blur algorithm, an image distortion correction algorithm and an adaptive feature matching algorithm (all of which belong to the prior art and are not described in detail herein), so as to finally obtain first point cloud data.
The first point cloud data is recorded as，/>For three-dimensional point cloud data, <' > a->(position of the dot in space), -A-T>Denoted as->Is a point cloud matrix of:
；
wherein,is->The X-axis coordinate values of the points in the spatial coordinate system, and (2)>Is->Y-axis coordinate value of point in space coordinate system, < >>Is->Z-axis coordinate value of point in space coordinate system, < >>Is->The X-axis coordinate values of the points in the spatial coordinate system, and (2)>Is->Y-axis coordinate value of point in space coordinate system, < >>Is->The Z-axis coordinate value of the point in the spatial coordinate system.
And because the obtained point cloud data in the whole target area always exists characteristic points irrelevant to the power corridor, the irrelevant characteristic points can be weakened by carrying out point cloud sparsification processing on the first point cloud data, and key characteristic points highly associated with the power corridor are highlighted in the second point cloud data, so that the three-dimensional reconstruction efficiency is improved.
Further, the data are encoded, so that the data processing speed can be increased, the processing time of the subsequent noise reduction processing is shortened, the noise of the data is reduced after the noise reduction processing, the accuracy of three-dimensional reconstruction is improved, the fourth point cloud data are obtained after the noise reduction processing, and the fifth point cloud data (equivalent to obtaining the complete three-dimensional information of a scene) are obtained by decoding the fourth point cloud data and fusing the fourth point cloud data in the decoding process; and finally, reconstructing a three-dimensional model of the power corridor according to the decoded fifth point cloud data.
The whole reconstruction process gives consideration to efficiency and precision, and realizes reconstructing the high-precision three-dimensional model of the electric power corridor in a short period.
It should be noted that, the point cloud data fusion related to the decoding process belongs to the prior art, and is not described herein.
In certain embodiments, the specific steps in step S2 include:
s21, acquiring characteristic points of the power tower pole, the power line, the insulator and the corridor topography as key characteristic points according to the first point cloud data;
s22, performing point cloud sparsification processing on the characteristic points outside the key characteristic points to obtain second point cloud data.
In this embodiment, for the power corridor, the power tower, the power line, the insulator and the corridor topography belong to key feature points highly associated with the power corridor, so that in addition to the feature points, it is necessary to perform a point cloud thinning process on other feature points, so that the point cloud of the power tower, the power line, the insulator and the corridor topography is highlighted, most of the interference of irrelevant point clouds is eliminated, and the three-dimensional model of the power corridor is easier to reconstruct, thereby improving the efficiency of three-dimensional reconstruction.
In certain embodiments, the specific steps in step S3 include:
s31, encoding the second point cloud data according to the following formula to obtain third point cloud data:
；
；
wherein,for the third point cloud data, +.>For coding function +.>For the second point cloud data, +.>For the first parameter set, ++>For the first multi-headed attention parameter, +.>Is the first full connection parameter.
In this embodiment, the second point cloud data is also a point cloud matrix, and the dimension is，/>In order to measure the length of time,is the number of each point in the point cloud.
It should be noted that, encoding the point cloud data belongs to the prior art, and specific encoding functions thereof are not described herein.
In certain embodiments, the specific steps in step S4 include:
s41, performing noise reduction processing on the third point cloud data for a plurality of times according to the following formula to obtain fourth point cloud data:
；
；
；
wherein,for the fourth point cloud data after the first noise reduction, < > is>For the noise reduction function->For the second parameter set, +.>Is fourth point cloud data after i+1st noise reduction,/for the fourth point cloud data>For the fourth point cloud data after the ith noise reduction,>for the total noise reduction times, < >>For the second multi-headed attention parameter, +.>Is the second full connection parameter.
The denoising function is constructed by full convolution, the input point cloud is represented in a low dimension, and then the input point cloud is restored to a high dimension through an up-sampling method, so that the denoising effect is achieved, in the embodiment, the point cloud data after the last denoising is input into the denoising function again for repeated denoising, so that the denoising effect is gradually improved, the noise in the point cloud data is reduced as much as possible, and the fourth point cloud data (namely the fourth point cloud data after the denoising) after the last denoising is used for decoding to obtain fifth point cloud data, so that the electric corridor three-dimensional model reconstructed later is ensured to have higher precision.
It should be noted that, the noise reduction processing of the point cloud data belongs to the prior art, and the specific noise reduction function is not described herein.
In certain embodiments, the specific steps in step S5 comprise:
s51, decoding the fourth point cloud data to obtain fifth point cloud data according to the following formula:
；
；
wherein,for the fifth point cloud data, +.>For the decoding function +.>For the third parameter set, +.>For the third multi-head attention parameter, +.>For the third full connection parameter, < >>For the (n+1) -th time (i.e. last time)And the fourth point cloud data after noise reduction.
It should be noted that, decoding the point cloud data belongs to the prior art, and specific decoding functions thereof are not described herein.
The invention also provides a reconstruction model which is applied to an image processing system and executes the steps in the three-dimensional reconstruction method of the electric power corridor in the embodiment.
In some embodiments, the reconstruction model is provided with a loss function for updating the first, second and third parameter sets as the reconstruction model is iteratively trained;
the specific expression of the loss function is:
；
wherein,for the third point cloud data, +.>For the fourth point cloud data after the N-th noise reduction,>for the second point cloud data, +.>For the fifth point cloud data, +.>For loss function->Indicating desire->Representing the binary norm.
In this embodiment, since the objective is to obtain the three-dimensional model of the power corridor, training of the modeling model is required before practical application, so as to ensure that the three-dimensional model of the power corridor can be successfully reconstructed and has higher accuracy.
In the iterative training process, the loss function is used to determine whether the reconstructed model is trained, e.g. whenTime (e.g.)>，/>A preset reference value) to consider the reconstructed model as having completed training.
Specifically, the training process adopts a gradient descent method:
；
；
；
wherein,for the set of parameters after the t-th iteration, < >>For the set of parameters after the t-1 th iteration,representation pair->Derivation and->Representation pair->Derivation and->For the first parameter set after the t-1 th iteration,/a first parameter set after the t-1 th iteration>For the second parameter set after the t-1 th iteration,/a second parameter set after the t-1 th iteration>Is the third parameter set after the t-1 th iteration.
In each iteration process, the first parameter set, the second parameter set and the third parameter set are updated, so that the value of the loss function is changed, and when the loss function is converged to be smaller than or equal to a reference value, the reconstruction model reaches the required error range, and the high-precision three-dimensional model of the electric power corridor can be reconstructed.
Referring to fig. 2, fig. 2 is a schematic diagram of a three-dimensional reconstruction device of a power corridor, which is integrated in a back-end control apparatus in the form of a computer program and applied to an image processing system, according to some embodiments of the present invention, and includes:
a first acquiring module 100, configured to acquire first point cloud data of a target area;
the second obtaining module 200 is configured to perform a point cloud sparsification process on the first point cloud data, and obtain processed second point cloud data;
the third obtaining module 300 is configured to encode the second point cloud data to obtain encoded third point cloud data;
a fourth obtaining module 400, configured to perform noise reduction processing on the third point cloud data for multiple times, to obtain denoised fourth point cloud data;
a fifth obtaining module 500, configured to decode the fourth point cloud data to obtain decoded fifth point cloud data;
and the reconstruction module 600 is configured to reconstruct a three-dimensional model of the power corridor according to the fifth point cloud data.
In some embodiments, the second obtaining module 200 performs, when performing the point cloud sparsification processing on the first point cloud data to obtain the processed second point cloud data:
s21, acquiring characteristic points of the power tower pole, the power line, the insulator and the corridor topography as key characteristic points according to the first point cloud data;
s22, performing point cloud sparsification processing on the characteristic points outside the key characteristic points to obtain second point cloud data.
In some embodiments, the third obtaining module 300 performs, when configured to encode the second point cloud data and obtain the encoded third point cloud data:
s31, encoding the second point cloud data according to the following formula to obtain third point cloud data:
；
；
wherein,for the third point cloud data, +.>For coding function +.>For the second point cloud data, +.>For the first parameter set, ++>For the first multi-headed attention parameter, +.>Is the first full connection parameter.
In some embodiments, the fourth obtaining module 400 performs, when performing noise reduction processing on the third point cloud data for a plurality of times to obtain denoised fourth point cloud data:
s41, performing noise reduction processing on the third point cloud data for a plurality of times according to the following formula to obtain fourth point cloud data:
；
；
；
wherein,for the fourth point cloud data after the first noise reduction, < > is>For the noise reduction function->For the second parameter set, +.>Is fourth point cloud data after i+1st noise reduction,/for the fourth point cloud data>For the fourth point cloud data after the ith noise reduction,>for the total noise reduction times, < >>For the second multi-headed attention parameter, +.>Is the second full connection parameter.
In some embodiments, the fifth obtaining module 500 performs, when configured to decode the fourth point cloud data to obtain decoded fifth point cloud data:
s51, decoding the fourth point cloud data to obtain fifth point cloud data according to the following formula:
；
；
wherein,for the fifth point cloud data, +.>For the decoding function +.>For the third parameter set, +.>For the third multi-head attention parameter, +.>For the third full connection parameter, < >>And the fourth point cloud data after the n+1st noise reduction is obtained.
Referring to fig. 3, fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, and the present invention provides an electronic device 13, including: processor 1301 and memory 1302, processor 1301 and memory 1302 interconnected and in communication with each other by a communication bus 1303 and/or other form of connection mechanism (not shown), memory 1302 storing computer readable instructions executable by processor 1301, which when the electronic device is running, processor 1301 executes the computer readable instructions to perform the method of three-dimensional reconstruction of a power corridor in any of the alternative implementations of the above embodiments, when executed, to perform the functions of: acquiring first point cloud data of a target area; performing point cloud sparsification processing on the first point cloud data to obtain processed second point cloud data; encoding the second point cloud data to obtain encoded third point cloud data; carrying out noise reduction treatment on the third point cloud data for multiple times to obtain fourth point cloud data after noise removal; decoding the fourth point cloud data to obtain decoded fifth point cloud data; and reconstructing a three-dimensional model of the power corridor according to the fifth point cloud data.
An embodiment of the present invention provides a computer readable storage medium, which when executed by a processor, performs the method for three-dimensional reconstruction of a power corridor in any optional implementation manner of the foregoing embodiment, so as to implement the following functions: acquiring first point cloud data of a target area; performing point cloud sparsification processing on the first point cloud data to obtain processed second point cloud data; encoding the second point cloud data to obtain encoded third point cloud data; carrying out noise reduction treatment on the third point cloud data for multiple times to obtain fourth point cloud data after noise removal; decoding the fourth point cloud data to obtain decoded fifth point cloud data; and reconstructing a three-dimensional model of the power corridor according to the fifth point cloud data.
The computer readable storage medium may be implemented by any type or combination of volatile or non-volatile Memory devices, such as static random access Memory (Static Random Access Memory, SRAM), electrically erasable Programmable Read-Only Memory (EEPROM), erasable Programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), programmable Read-Only Memory (PROM), read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk, or optical disk.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus and method may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
Further, the units described as separate units may or may not be physically separate, and units displayed as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
Furthermore, functional modules in various embodiments of the present invention may be integrated together to form a single portion, or each module may exist alone, or two or more modules may be integrated to form a single portion.
In this document, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.
The above description is only an example of the present invention and is not intended to limit the scope of the present invention, and various modifications and variations will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (2)
1. An image processing system for three-dimensional reconstruction of a power corridor, comprising a reconstruction model, said reconstruction model executing the following steps at run-time:
s1, acquiring first point cloud data of a target area;
s2, performing point cloud sparsification processing on the first point cloud data to obtain processed second point cloud data;
s3, encoding the second point cloud data to obtain encoded third point cloud data;
s4, carrying out noise reduction treatment on the third point cloud data for multiple times to obtain fourth point cloud data after noise removal;
s5, decoding the fourth point cloud data to obtain decoded fifth point cloud data;
s6, reconstructing a three-dimensional model of the electric power corridor according to the fifth point cloud data;
the specific steps in the step S3 include:
s31, encoding the second point cloud data according to the following formula to obtain the third point cloud data:
；
；
wherein,for the third point cloud data, +.>For coding function +.>For the second point cloud data, +.>For the first parameter set, ++>For the first multi-headed attention parameter, +.>Is a first full connection parameter;
the specific steps in the step S4 include:
s41, carrying out noise reduction processing on the third point cloud data for a plurality of times according to the following formula to obtain fourth point cloud data:
；
；
；
wherein,for the fourth point cloud data after the first noise reduction, < > is>For the noise reduction function->As a second set of parameters,is fourth point cloud data after i+1st noise reduction,/for the fourth point cloud data>For the fourth point cloud data after the ith noise reduction,>for the total noise reduction times, < >>For the second multi-headed attention parameter, +.>Is a second full connection parameter;
the specific steps in the step S5 include:
s51, decoding the fourth point cloud data to obtain fifth point cloud data according to the following formula:
；
；
wherein,for the fifth point cloud data, +.>For the decoding function +.>For the third parameter set, +.>For the third multi-head attention parameter, +.>For the third full connection parameter, < >>The fourth point cloud data after the n+1st noise reduction is obtained;
the reconstruction model is provided with a loss function, and the loss function is used for updating a first parameter set, a second parameter set and a third parameter set when the reconstruction model is subjected to iterative training;
the specific expression of the loss function is as follows:
；
wherein,for the third point cloud data, +.>For the fourth point cloud data after the N-th noise reduction,>for the second point cloud data, +.>For the fifth point cloud data, +.>For the loss function, +.>Indicating desire->Representing the binary norm.
2. The image processing system for three-dimensional reconstruction of an electrical corridor according to claim 1, characterized in that the specific steps in step S2 comprise:
s21, according to the first point cloud data, obtaining characteristic points of the power tower pole, the power line, the insulator and corridor topography as key characteristic points;
s22, performing point cloud sparsification processing on the characteristic points outside the key characteristic points to obtain the second point cloud data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311314243.8A CN117058330B (en) | 2023-10-11 | 2023-10-11 | Three-dimensional reconstruction method, reconstruction model and related equipment for electric power corridor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311314243.8A CN117058330B (en) | 2023-10-11 | 2023-10-11 | Three-dimensional reconstruction method, reconstruction model and related equipment for electric power corridor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117058330A CN117058330A (en) | 2023-11-14 |
| CN117058330B true CN117058330B (en) | 2024-02-13 |
Family
ID=88653922
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311314243.8A Active CN117058330B (en) | 2023-10-11 | 2023-10-11 | Three-dimensional reconstruction method, reconstruction model and related equipment for electric power corridor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117058330B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022226850A1 (en) * | 2021-04-28 | 2022-11-03 | Oppo广东移动通信有限公司 | Point cloud quality enhancement method, encoding and decoding methods, apparatuses, and storage medium |
| CN116486020A (en) * | 2023-06-21 | 2023-07-25 | 季华实验室 | Three-dimensional reconstruction method and related equipment |
-
2023
- 2023-10-11 CN CN202311314243.8A patent/CN117058330B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022226850A1 (en) * | 2021-04-28 | 2022-11-03 | Oppo广东移动通信有限公司 | Point cloud quality enhancement method, encoding and decoding methods, apparatuses, and storage medium |
| CN116486020A (en) * | 2023-06-21 | 2023-07-25 | 季华实验室 | Three-dimensional reconstruction method and related equipment |
Non-Patent Citations (1)
| Title |
|---|
| Sparse-to-Dense Multi-Encoder Shape Completion of Unstructured Point Cloud;YANJUN PENG 等;IEEE Access;第8卷;第1-10页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117058330A (en) | 2023-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10984556B2 (en) | Method and apparatus for calibrating relative parameters of collector, device and storage medium | |
| US9767598B2 (en) | Smoothing and robust normal estimation for 3D point clouds | |
| KR101933860B1 (en) | Apparatus and method for radiance transfer sampling for augmented reality | |
| US8331615B2 (en) | Match, expand, and filter technique for multi-view stereopsis | |
| CN108230235B (en) | Disparity map generation system, method and storage medium | |
| CN116310076A (en) | Three-dimensional reconstruction method, device, equipment and storage medium based on nerve radiation field | |
| CN114073071B (en) | Video frame inserting method and device and computer readable storage medium | |
| CN111160298A (en) | Robot and pose estimation method and device thereof | |
| CN116958492B (en) | VR editing method for reconstructing three-dimensional base scene rendering based on NeRf | |
| CN110599586A (en) | Semi-dense scene reconstruction method and device, electronic equipment and storage medium | |
| CN110827341A (en) | Picture depth estimation method and device and storage medium | |
| CN111680573A (en) | Face recognition method and device, electronic equipment and storage medium | |
| CN115293992A (en) | Polarization image defogging method and device based on unsupervised weight depth model | |
| CN117274072A (en) | Point cloud denoising method and device based on two-dimensional multi-modal range image | |
| CN112150470A (en) | Image segmentation method, image segmentation device, image segmentation medium, and electronic device | |
| CN113421217A (en) | Method and device for detecting travelable area | |
| CN117058330B (en) | Three-dimensional reconstruction method, reconstruction model and related equipment for electric power corridor | |
| CN117635444A (en) | Depth completion method, device and equipment based on radiation difference and space distance | |
| CN115345990A (en) | Oblique photography three-dimensional reconstruction method and device for weak texture scene | |
| CN113496468B (en) | Depth image restoration method, device and storage medium | |
| CN110119721B (en) | Method and apparatus for processing information | |
| CN113537491A (en) | Neural network training method, device and computer readable medium | |
| CN116051427B (en) | Point cloud denoising model acquisition method, point cloud fusion method and related equipment thereof | |
| CN112085850B (en) | Face reconstruction method and related equipment | |
| CN115239915B (en) | VR scene real-time reconstruction method and device, electronic 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 |