CN111460735B - Camera layout function optimization method based on genetic inheritance and related equipment - Google Patents
Camera layout function optimization method based on genetic inheritance and related equipment Download PDFInfo
- Publication number
- CN111460735B CN111460735B CN202010268735.8A CN202010268735A CN111460735B CN 111460735 B CN111460735 B CN 111460735B CN 202010268735 A CN202010268735 A CN 202010268735A CN 111460735 B CN111460735 B CN 111460735B
- Authority
- CN
- China
- Prior art keywords
- camera
- score
- observability
- point
- target
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- 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)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Genetics & Genomics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Physiology (AREA)
- Studio Devices (AREA)
Abstract
The invention relates to the field of computer vision, and discloses a genetic-based camera layout function optimization method and related equipment, which are used for improving the camera layout efficiency of an optical motion capture system. The camera layout function optimization method based on genetic inheritance comprises the following steps: carrying out parameter configuration on a preset type camera to obtain camera parameters and scene parameters; determining an objective function in a preset genetic algorithm according to the camera parameters and the scene parameters to obtain an initial camera layout function; optimizing actual installation position parameters in the initial camera layout function to obtain candidate camera layout functions; and carrying out target parameter configuration on the candidate camera layout function to obtain a target camera layout function, wherein the target parameters comprise the number of genes, the number of populations, the maximum iteration number, the minimum iteration number, the stable iteration number, the cross probability and the variation probability.
Description
Technical Field
The invention relates to the field of computer vision, in particular to a camera layout function optimization method based on genetic inheritance and related equipment.
Background
The optical motion capturing system is a high-precision camera installed in a dynamic capturing space to position and track a marker point on a moving target in a scene so as to realize the task of capturing motion. The quality of the final capture effect is affected by many steps, an important one of which includes an excellent camera mounting location. The current camera installation method in the market mainly determines the installation position according to engineering experience, and adopts an equidistant distribution mode to determine the installation position, namely, one camera is installed at a certain distance.
In the prior art, the method for determining the installation position according to engineering experience has a plurality of disadvantages, firstly, the method excessively depends on the engineering experience of staff, and the effect is difficult to control; secondly, the installation method is extremely time-consuming and labor-consuming, needs to be continuously disassembled and reinstalled, and is subjected to a large number of test adjustment, so that the workload is obviously not easy; thirdly, the final effect is not particularly good due to rough installation, and the later motion capturing effect is seriously affected. The mounting positions are determined in an equidistant distribution mode, so that the problems of redundancy of cameras and increased number of required cameras are solved, and the cost is increased. Thus, the camera layout of the optical motion capture system is rendered inefficient.
Disclosure of Invention
The invention mainly aims to solve the problem of low camera layout efficiency of an optical motion capture system.
The first aspect of the invention provides a camera layout function optimization method based on genetic inheritance, which comprises the following steps:
carrying out parameter configuration on a preset type camera to obtain camera parameters and scene parameters;
determining an objective function in a preset genetic algorithm according to the camera parameters and the scene parameters to obtain an initial camera layout function;
optimizing actual installation position parameters in the initial camera layout function to obtain candidate camera layout functions;
and carrying out target parameter configuration on the candidate camera layout function to obtain a target camera layout function, wherein the target parameters comprise the number of genes, the number of populations, the maximum iteration number, the minimum iteration number, the stable iteration number, the cross probability and the variation probability.
Optionally, in a first implementation manner of the first aspect of the present invention, the performing parameter configuration on the preset type camera to obtain a camera parameter and a scene parameter includes:
setting the expected installation quantity of the preset type cameras, the arrangement range information corresponding to the preset type cameras and the installable area information of the preset type cameras to obtain camera parameters;
And discretizing the scene capturing area of the preset type camera and setting the size of the area to obtain scene parameters.
Optionally, in a second implementation manner of the first aspect of the present invention, the performing discretization processing and region size setting on the scene capturing region of the preset type camera to obtain a scene parameter includes:
performing discretization processing on a scene capturing area of the preset type camera according to preset discrete precision to obtain a discrete point group;
the range of the scene capturing area is retracted inwards relative to the position of the row frame in the mountable area information, so that the size of the scene capturing area is obtained;
and generating a point cloud according to the discrete point group and the size of the scene capturing area, and determining the point cloud as a scene parameter.
Optionally, in a third implementation manner of the first aspect of the present invention, the determining, according to the camera parameter and the scene parameter, an objective function in a preset genetic algorithm to obtain an initial camera layout function includes:
calculating a bi-quad observability score for each point in the point cloud;
determining the maximum value of the double four-azimuth observability score of each point as the initial observability score of each point;
Acquiring the total number of points in the point cloud, and carrying out normalization processing on the initial observability score of each point according to the total number of points in the point cloud to obtain a target observability score;
maximizing a score function in a preset genetic algorithm according to the target observability score to obtain a target function;
and generating an initial camera layout function according to the objective function and the genetic algorithm.
Optionally, in a fourth implementation manner of the first aspect of the present invention, the calculating a bi-quad-azimuth observability score of each point in the point cloud includes:
acquiring a 360-degree coordinate system corresponding to each point in the point cloud;
dividing the 360-degree coordinate system into four equal-partition sections and performing preset angle rotation to obtain a target four equal-partition section;
acquiring the number of observable target orientations of each point in the point cloud in the target quarter section;
and scoring each point in the point cloud according to the number of the target orientations to obtain a double four-orientation observability score of each point.
Optionally, in a fifth implementation manner of the first aspect of the present invention, the obtaining the total number of points in the point cloud, and performing normalization processing on an initial observability score of each point according to the total number of points in the point cloud, to obtain a target observability score includes:
Acquiring the total number of points in the point cloud, and determining an observability score and a value according to the total number and the initial observability score of each point;
dividing the observability score sum value by the total number to obtain an initial observability score;
dividing the initial observability score by 4 results in a target observability score.
Optionally, in a sixth implementation manner of the first aspect of the present invention, the optimizing the actual installation location parameter in the initial camera layout function to obtain a candidate camera layout function includes:
normalizing the actual installation position parameters in the initial camera layout function to obtain optimized parameter values;
and determining the optimized parameter value as a numerical range of the individual genes in the initial camera layout function to obtain a candidate camera layout function.
The second aspect of the present invention provides a genetic-based camera layout function optimizing apparatus, comprising:
the first configuration module is used for carrying out parameter configuration on a preset type camera to obtain camera parameters and scene parameters;
the determining module is used for determining an objective function in a preset genetic algorithm according to the camera parameters and the scene parameters to obtain an initial camera layout function;
The optimization module is used for optimizing the actual installation position parameters in the initial camera layout function to obtain candidate camera layout functions;
the second configuration module is used for carrying out target parameter configuration on the candidate camera layout function to obtain a target camera layout function, wherein the target parameters comprise the number of genes, the number of populations, the maximum iteration number, the minimum iteration number, the stable iteration number, the cross probability and the variation probability.
A third aspect of the present invention provides a genetic inheritance-based camera layout function optimizing apparatus, comprising: a memory and at least one processor, the memory having instructions stored therein, the memory and the at least one processor being interconnected by a line; the at least one processor invokes the instructions in the memory to cause the genetic based camera layout function optimization device to perform the genetic based camera layout function optimization method described above.
A fourth aspect of the present invention provides a computer-readable storage medium having instructions stored therein, which when run on a computer, cause the computer to perform the above-described genetic-based camera layout function optimization method.
In the technical scheme provided by the invention, parameter configuration is carried out on a preset type camera to obtain camera parameters and scene parameters; determining an objective function in a preset genetic algorithm according to the camera parameters and the scene parameters to obtain an initial camera layout function; optimizing actual installation position parameters in the initial camera layout function to obtain candidate camera layout functions; and carrying out target parameter configuration on the candidate camera layout function to obtain a target camera layout function, wherein the target parameters comprise the number of genes, the number of populations, the maximum iteration number, the minimum iteration number, the stable iteration number, the cross probability and the variation probability. According to the invention, the genetic algorithm is optimized to obtain the genetic-based camera layout optimization function (target camera layout function), so that the target camera layout function is free from continuous and micro constraint and can be set in a global optimization mode with a definition domain arbitrarily, the target camera layout function is combined with global preference of the genetic algorithm, the risk of falling into a local optimal solution is reduced, parallelization is easy to realize, probability transition rules are adopted, and the characteristics of self-organization, self-adaption and self-learning are adopted, so that the method is suitable for the situation that the score function (target function) is discrete, the installation constraint condition is multiple and discontinuous (such as a preset type camera is required to be installed on a specified row frame), thereby greatly saving unnecessary manpower resources, continuously adjusting the installation position information of the target camera without repeated disassembly and reinstallation, and further improving the camera layout efficiency of the optical motion capture system.
Drawings
FIG. 1 is a diagram showing an embodiment of a genetic-based camera layout function optimization method according to an embodiment of the present invention;
FIG. 2 is a diagram showing another embodiment of a genetic-based camera layout function optimization method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an embodiment of a genetic-based camera layout function optimizing apparatus according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of another embodiment of a genetic-based camera layout function optimization device according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of an embodiment of a genetic-based camera layout function optimization apparatus according to an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a camera layout function optimization method based on genetic inheritance and related equipment, which can improve the camera layout efficiency of an optical motion capture system.
The terms "first," "second," "third," "fourth" and the like in the description and in the claims and in the above drawings, if any, 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 such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," or any other variation 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 or inherent to such process, method, article, or apparatus.
For ease of understanding, a specific flow of an embodiment of the present invention is described below with reference to fig. 1, and one embodiment of a method for optimizing a camera layout function based on genetic inheritance in the embodiment of the present invention includes:
101. carrying out parameter configuration on a preset type camera to obtain camera parameters and scene parameters;
it will be appreciated that the execution subject of the present invention may be a genetic-based camera layout function optimizing device, or may be a terminal or a server, and is not limited herein. The embodiment of the invention is described by taking a server as an execution main body as an example.
The camera parameters are expected installation number of the preset type cameras, range information corresponding to the preset type cameras and line frame information in a preset field, the range information corresponding to the preset type cameras can be shooting range information of the cameras, dynamic range information of the cameras and visible range information of the cameras which are vertebral bodies, the scene parameters are scene capturing area sizes of the preset type cameras, and the scene capturing area sizes are point cloud areas. The camera parameters and scene parameters of the preset type camera are configured so as to facilitate the subsequent judgment of the capturing effect of the points in the scene capturing area, and the camera layout information is merged into a preset genetic algorithm in the subsequent design of the objective function.
102. Determining an objective function in a preset genetic algorithm according to the camera parameters and the scene parameters to obtain an initial camera layout function;
the server ensures that each point of the point cloud area in the camera parameters and the scene parameters is captured and tracked by at least two cameras of preset types simultaneously by a simple, visual and effective method, namely a double four-azimuth observability method, so that the triangulation of the target point can be realized. For example: the coordinate system corresponding to each point of the point cloud area is divided into four subareas in average, each subarea has two directions, four directions of the four subareas form a four-direction, the four-direction score of each point is calculated, the server can take the average value of all four-direction scores of each point as the final score of the point, and the weighted average value of all four-direction scores of each point can also be taken as the final score of the point. The final score is calculated as the objective function in the preset genetic algorithm.
The objective function is set by adopting the function of the double four-azimuth observability method, and the score obtained by the objective function is mapped to the adaptation value in the preset genetic algorithm so as to realize the setting of the objective function in the preset genetic algorithm.
103. Optimizing actual installation position parameters in the initial camera layout function to obtain candidate camera layout functions;
the server may perform a corresponding screening reduction on the number of actual installation position parameters in the initial camera layout function according to a limited condition of the camera installation position, where the limited condition of the camera installation position may be a frame structure of the camera installation, an angle of rotation around the X-axis, a distance requirement between the camera and the camera, or a distance requirement between the camera and the object, and the actual installation position parameters include a camera position (spatial three-dimensional coordinate), a camera horizontal angle, and a camera pitch angle, for example: if the frame structure of the camera is not changed in the longitudinal direction, the parameters of the longitudinal axis in the spatial three-dimensional coordinate position of the camera are not required to be added into the genotype for optimization so as to reduce the calculation amount. By optimizing the actual installation position parameters in the initial camera layout function, the calculated amount of the actual installation position parameters is reduced, and the efficiency and accuracy of the method are improved.
104. And carrying out target parameter configuration on the candidate camera layout function to obtain a target camera layout function, wherein the target parameters comprise the number of genes, the number of populations, the maximum iteration number, the minimum iteration number, the stable iteration number, the cross probability and the variation probability.
The server sets the actual problems to be solved in the target camera layout function and the solutions corresponding to the actual problems by configuring the number of genes and the number of populations. The server sets the iteration times for obtaining the number of excellent populations by configuring the maximum iteration times, the minimum iteration times, the stable iteration times, the crossover probability and the variation probability. Before the population number is configured, a required uniform design table is created through a lattice point method, camera parameters, scene parameters, gene numbers, preset type camera installation positions, maximum iteration times, minimum iteration times, stable iteration times, cross probability and variation probability range values and camera installation positions (initial population) of the preset type cameras under different scenes are determined, experimental calculation is performed according to the uniform design table, and the optimal population number set value is selected according to experimental calculation results.
A plurality of new population individuals (camera mounting position information) are generated by the set crossover probability and mutation probability. And obtaining final required target population individuals (target camera mounting position information) through the cross variation of the maximum iteration number, the minimum iteration number and the stable iteration number and roulette selection. Thus, the target camera layout function which can greatly save unnecessary human resources and can obtain the installation position information of the target camera without repeated disassembly and reinstallation and continuous adjustment is obtained.
In the embodiment of the invention, the genetic algorithm is optimized to obtain the genetic-based camera layout optimization function (target camera layout function) so as to be free from continuous and slightly constraint and have a global optimization mode with a arbitrarily set definition domain, so that the target camera layout function is combined with global preference of the genetic algorithm, reduces the risk of falling into a locally optimal solution, is easy to realize parallelization, adopts a probability transition rule and has the characteristics of self-organization, self-adaption and self-learning, is suitable for the characteristics that the score function (target function) is discrete, has a plurality of installation constraint conditions and is discontinuous (such as a preset type camera is required to be installed on a specified row frame), thereby greatly saving unnecessary human resources, avoiding repeated disassembly and installation and continuously adjusting the effect of obtaining the installation position information of the target camera, and further improving the camera layout efficiency of the optical motion capture system.
Referring to fig. 2, another embodiment of the method for optimizing a camera layout function based on genetic inheritance in an embodiment of the present invention includes:
201. setting the expected installation quantity of the preset type cameras, the arrangement range information corresponding to the preset type cameras and the installable area information of the preset type cameras to obtain camera parameters;
The arrangement range information corresponding to the preset type camera comprises a farthest shooting distance, a horizontal range angle and a vertical range angle, the mountable area information of the preset type camera is row frame information of the preset type camera in the field, namely the total length, the total width and the total height of the row frame or the length, the width and the height of each layer of the row frame.
202. Discretizing a scene capturing area of a preset type camera and setting the size of the area to obtain scene parameters;
specifically, the server performs discretization processing on a scene capturing area of a preset type camera according to preset discrete precision to obtain a discrete point group; the range of the scene capturing area is retracted inwards relative to the position of the row frame in the mountable area information, so that the size of the scene capturing area is obtained; and generating a point cloud according to the discrete point group and the size of the scene capturing area, and determining the point cloud as a scene parameter.
The server facilitates the determination of the capturing effect in the vicinity of a preset type of camera by discretizing the scene capture area of the point into a series of points (discrete point groups) with a preset discretization accuracy so as to calculate the observability score of the points. The capture area is sized to be set in an inward indentation in position relative to the row rack, for example: the positions of the row frames are (A, B and C), the position of the row frames is taken as a center point, the row frames are retracted inwards by 5, and the scene capturing area sizes are (A-5, B-5 and C-5). The size of the scene capturing area is a space coordinate, and the discrete point group and the space coordinate (the size of the scene capturing area) corresponding to the discrete point group are point clouds.
203. Determining an objective function in a preset genetic algorithm according to the camera parameters and the scene parameters to obtain an initial camera layout function;
specifically, the server calculates a bi-quad observability score for each point in the point cloud; determining the maximum value of the double four-azimuth observability score of each point as the initial observability score of each point; acquiring the total number of points in the point cloud, and carrying out normalization processing on the initial observability score of each point according to the total number of points in the point cloud to obtain a target observability score; maximizing a score function in a preset genetic algorithm according to the target observability score to obtain a target function; and generating an initial camera layout function according to the objective function and the genetic algorithm.
For example: taking the point B in the point cloud as an example, the point B is divided into 3 points and 4 points in double four directions, and taking the 4 points as initial observability points of the point B, and carrying out normalization processing on the 4 points according to the total number of points in the point cloud to obtain target observability points. Also for example: and if the total number of points in the point cloud is 2, the initial observability score of the point is 4, and the normalization processing is carried out on the 2 points and the 4 points according to the total number of points in the point cloud to obtain the target observability score. The objective observability score is the score obtained by the maximum score function, the function process of the objective observability score is the function process of the objective function, and the initial camera layout function is obtained by setting the objective function in the preset genetic algorithm, so that the preliminary optimization of the camera layout function based on genetic inheritance is realized.
Specifically, a server acquires a 360-degree coordinate system corresponding to each point in a point cloud; dividing a 360-degree coordinate system into four equal-partition sections and performing preset angle rotation to obtain a target four equal-partition section; acquiring the number of observable target orientations of each point in the point cloud in a target quarter; and scoring each point in the point cloud according to the number of the target orientations to obtain a double four-orientation observability score of each point.
For example: the 360-degree coordinate system corresponding to the propyl point in the point cloud is divided into four subareas in average, each subarea corresponds to two directions, namely an upper direction and a lower direction of the subarea, the four directions of the four subareas form a four direction, each subarea is rotated by 45 degrees to obtain a new four-way subarea (target four-way subarea), the number of directions observable by the propyl point in the first four-way subarea of the target four-way subarea is 4, the number of directions observable by the second four-way subarea is 3, the observability score of each direction is 1 minute, the observability score of the propyl point corresponding to the first four-way subarea is 4 minutes, the observability score of the propyl point corresponding to the second four-way subarea is 3 minutes, and the two four-way observability score of the propyl point is 4 minutes and 3 minutes.
When the server rotates the quarter section by a preset angle, the quarter section can be rotated by a preset angle anticlockwise or clockwise at the same time, or one part of the quarter section can be rotated by a preset angle clockwise, and the other part of the quarter section can be rotated by a preset angle anticlockwise. The server may obtain the number of observable orientations of each point in the point cloud in the target quadrant by mapping the quadrant of each point on the visual range of the view cone of the preset type camera according to the number of observable orientations in the visual range of the view cone.
Specifically, the server acquires the total number of points in the point cloud, and determines an observability score and a value according to the total number and the initial observability score of each point; dividing the observability score sum by the total number to obtain an initial observability score; dividing the initial observability score by 4 yields the target observability score.
For example: the total number of points in the point cloud is 3, namely a point D, a point E and a point F, and the initial observability scores of the point D, the point E and the point F are 4, 2 and 3 respectively, so that the initial observability score sum is 4+2+3=9, the candidate observability score (3) is obtained by dividing 9 by 3, and the candidate observability score (3) is divided by 4, namely, the initial observability score of each point is normalized according to the total number of points in the point cloud, so that the target observability score 3/4 is obtained.
204. Optimizing actual installation position parameters in the initial camera layout function to obtain candidate camera layout functions;
specifically, the server normalizes actual installation position parameters in the initial camera layout function to obtain optimized parameter values; and determining the optimized parameter value as the numerical range of the individual genes in the initial camera layout function to obtain the candidate camera layout function.
In order to ensure the consistency of the parameters of each actual installation position in dimension and facilitate the cross mutation operation in the genetic algorithm, the server performs normalization operation on the parameters of each actual installation position, takes the optimized parameter values obtained after the normalization operation as the numerical range of the individual genes in the genetic algorithm, wherein each optimized parameter value belongs to the [0,1] interval, each optimized parameter obtained after the normalization operation corresponds to one individual gene, and each three individual genes corresponds to the installation position of the camera in one scene.
In the process of optimizing the actual installation position parameters in the initial camera layout function, the number of the actual installation position parameters in the initial camera layout function is correspondingly reduced according to the particularity of the camera installation position, namely the frame structure of the row frame on which the camera is installed, wherein the frame structure of the row frame can be one layer or multiple layers, for example: taking a layer of row frame as an example, the preset type cameras are arranged on the row frame, if the row frames are all one layer, the longitudinal coordinates CameraZ in the camera position (space three-dimensional coordinates) are fixed, so that the optimization is not needed to be carried out, and the horizontal coordinates CameraX and CameraY in the camera position (space three-dimensional coordinates) have an equality relation, so that the number of the optimization parameters can be expressed by only one parameter, the number of the parameters to be optimized finally or the number of genes can be further reduced, the expected installation number of the preset type cameras is 3, namely each preset type camera has 3 optimization parameters, and the value of each optimization parameter belongs to the section [0,1 ].
205. And carrying out target parameter configuration on the candidate camera layout function to obtain a target camera layout function, wherein the target parameters comprise the number of genes, the number of populations, the maximum iteration number, the minimum iteration number, the stable iteration number, the cross probability and the variation probability.
The server sets the actual problems to be solved in the target camera layout function and the solutions corresponding to the actual problems by configuring the number of genes and the number of populations. The server sets the iteration times for obtaining the number of excellent populations by configuring the maximum iteration times, the minimum iteration times, the stable iteration times, the crossover probability and the variation probability. A plurality of new population individuals (camera mounting position information) are generated by the set crossover probability and mutation probability. And obtaining final required target population individuals (target camera mounting position information) through the cross variation of the maximum iteration number, the minimum iteration number and the stable iteration number and roulette selection. Therefore, unnecessary human resources can be greatly saved, the installation position information of the target camera can be obtained without repeated disassembly and installation and continuous adjustment, the performance of each camera can be fully and efficiently utilized, and the target camera layout function of the motion capturing effect is greatly improved.
According to the embodiment of the invention, unnecessary human resources can be greatly saved, the effect of continuously adjusting the installation position information of the target camera can be obtained without repeated disassembly and installation, and further, the camera layout efficiency of the optical motion capture system is improved, the number of parameters and the corresponding calculated amount are reduced in a simple, visual and effective manner, the dimensional consistency of the parameters of each actual installation position is ensured, and the cross variation operation in the genetic algorithm is facilitated, so that the camera layout efficiency of the optical motion capture system is improved.
The method for optimizing a camera layout function based on genetic inheritance in the embodiment of the present invention is described above, and the device for optimizing a camera layout function based on genetic inheritance in the embodiment of the present invention is described below, referring to fig. 3, one embodiment of the device for optimizing a camera layout function based on genetic inheritance in the embodiment of the present invention includes:
the first configuration module 301 is configured to perform parameter configuration on a preset type camera to obtain a camera parameter and a scene parameter;
the determining module 302 is configured to determine an objective function in a preset genetic algorithm according to the camera parameter and the scene parameter, so as to obtain an initial camera layout function;
The optimizing module 303 is configured to optimize an actual installation position parameter in the initial camera layout function to obtain a candidate camera layout function;
the second configuration module 304 is configured to perform target parameter configuration on the candidate camera layout function to obtain a target camera layout function, where the target parameter includes a number of genes, a number of populations, a maximum iteration number, a minimum iteration number, a stable iteration number, a crossover probability, and a variation probability.
The function implementation of each module in the genetic-based camera layout function optimization device corresponds to each step in the genetic-based camera layout function optimization method embodiment, and the functions and implementation processes of the genetic-based camera layout function optimization device are not described in detail herein.
In the embodiment of the invention, the genetic algorithm is optimized to obtain the genetic-based camera layout optimization function (target camera layout function) so as to be free from continuous and slightly constraint and have a global optimization mode with a arbitrarily set definition domain, so that the target camera layout function is combined with global preference of the genetic algorithm, reduces the risk of falling into a locally optimal solution, is easy to realize parallelization, adopts a probability transition rule and has the characteristics of self-organization, self-adaption and self-learning, is suitable for the characteristics that the score function (target function) is discrete, has a plurality of installation constraint conditions and is discontinuous (such as a preset type camera is required to be installed on a specified row frame), thereby greatly saving unnecessary human resources, avoiding repeated disassembly and installation and continuously adjusting the effect of obtaining the installation position information of the target camera, and further improving the camera layout efficiency of the optical motion capture system.
Referring to fig. 4, another embodiment of the genetic-based camera layout function optimizing apparatus according to the embodiment of the present invention includes:
the first configuration module 301 is configured to perform parameter configuration on a preset type camera to obtain a camera parameter and a scene parameter;
the first configuration module 301 specifically includes:
a setting unit 3011, configured to set an expected installation number of preset type cameras, arrangement range information corresponding to the preset type cameras, and installable area information of the preset type cameras, to obtain camera parameters;
a first processing unit 3012, configured to perform discretization processing and region size setting on a scene capturing region of a preset type camera, so as to obtain scene parameters;
the determining module 302 is configured to determine an objective function in a preset genetic algorithm according to the camera parameter and the scene parameter, so as to obtain an initial camera layout function;
the optimizing module 303 is configured to optimize an actual installation position parameter in the initial camera layout function to obtain a candidate camera layout function;
the second configuration module 304 is configured to perform target parameter configuration on the candidate camera layout function to obtain a target camera layout function, where the target parameter includes a number of genes, a number of populations, a maximum iteration number, a minimum iteration number, a stable iteration number, a crossover probability, and a variation probability.
Optionally, the first processing unit 3012 may be further specifically configured to:
performing discretization processing on a scene capturing area of a preset type camera according to preset discrete precision to obtain a discrete point group;
the range of the scene capturing area is retracted inwards relative to the position of the row frame in the mountable area information, so that the size of the scene capturing area is obtained;
and generating a point cloud according to the discrete point group and the size of the scene capturing area, and determining the point cloud as a scene parameter.
Optionally, the determining module 302 includes:
a computing unit 3021 for computing a bi-quad observability score for each point in the point cloud;
a determining unit 3022 for determining, as an initial observability score of each point, a score having a largest value among the two-four-way observability scores of each point;
a second processing unit 3023, configured to obtain a total number of points in the point cloud, normalize the initial observability score of each point according to the total number of points in the point cloud, and obtain a target observability score;
a third processing unit 3024, configured to maximize a score function in a preset genetic algorithm according to the target observability score, so as to obtain a target function;
a generating unit 3025 for generating an initial camera layout function based on the objective function and the genetic algorithm.
Optionally, the computing unit 3021 may be further specifically configured to:
acquiring a 360-degree coordinate system corresponding to each point in the point cloud;
dividing a 360-degree coordinate system into four equal-partition sections and performing preset angle rotation to obtain a target four equal-partition section;
acquiring the number of observable target orientations of each point in the point cloud in a target quarter;
and scoring each point in the point cloud according to the number of the target orientations to obtain a double four-orientation observability score of each point.
Optionally, the second processing unit 3023 may be further specifically configured to:
acquiring the total number of points in the point cloud, and determining an observability score and a value according to the total number and the initial observability score of each point;
dividing the observability score sum by the total number to obtain an initial observability score;
dividing the initial observability score by 4 yields the target observability score.
Optionally, the optimization module 303 may be further specifically configured to:
normalizing the actual installation position parameters in the initial camera layout function to obtain optimized parameter values;
and determining the optimized parameter value as the numerical range of the individual genes in the initial camera layout function to obtain the candidate camera layout function.
The function implementation of each module and each unit in the genetic-based camera layout function optimization device corresponds to each step in the genetic-based camera layout function optimization method embodiment, and the functions and implementation processes of the genetic-based camera layout function optimization device are not described in detail herein.
According to the embodiment of the invention, unnecessary human resources can be greatly saved, the effect of continuously adjusting the installation position information of the target camera can be obtained without repeated disassembly and installation, and further, the camera layout efficiency of the optical motion capture system is improved, the number of parameters and the corresponding calculated amount are reduced in a simple, visual and effective manner, the dimensional consistency of the parameters of each actual installation position is ensured, and the cross variation operation in the genetic algorithm is facilitated, so that the camera layout efficiency of the optical motion capture system is improved.
The genetic-based camera layout function optimizing apparatus in the embodiment of the present invention is described in detail above in fig. 3 and 4 from the point of view of the modularized functional entity, and the genetic-based camera layout function optimizing device in the embodiment of the present invention is described in detail below from the point of view of hardware processing.
Fig. 5 is a schematic structural diagram of a genetic-based camera layout function optimization device 500 according to an embodiment of the present invention, where the genetic-based camera layout function optimization device 500 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 510 (e.g., one or more processors) and a memory 520, and one or more storage media 530 (e.g., one or more mass storage devices) storing application programs 533 or data 532. Wherein memory 520 and storage medium 530 may be transitory or persistent storage. The program stored in the storage medium 530 may include one or more modules (not shown), each of which may include a series of instruction operations in the genetic based camera layout function optimization device 500. Still further, the processor 510 may be configured to communicate with the storage medium 530 to execute a series of instruction operations in the storage medium 530 on the genetic based camera layout function optimization device 500.
The genetic based camera layout function optimization device 500 may also include one or more power supplies 540, one or more wired or wireless network interfaces 550, one or more input/output interfaces 560, and/or one or more operating systems 531, such as Windows Serve, mac OS X, unix, linux, freeBSD, and the like. It will be appreciated by those skilled in the art that the genetic based camera layout function optimization device structure shown in fig. 5 does not constitute a limitation of the genetic based camera layout function optimization device, and may include more or fewer components than shown, or may combine certain components, or may be a different arrangement of components.
The present invention also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, and which may also be a volatile computer readable storage medium, in which instructions are stored which, when executed on a computer, cause the computer to perform the steps of a genetic-based camera layout function optimization method.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (7)
1. The genetic-based camera layout function optimization method is characterized by comprising the following steps of:
carrying out parameter configuration on a preset type camera to obtain camera parameters and scene parameters, wherein the scene parameters are used for indicating point clouds;
determining an objective function in a preset genetic algorithm according to the camera parameters and the scene parameters to obtain an initial camera layout function;
optimizing actual installation position parameters in the initial camera layout function to obtain candidate camera layout functions;
performing target parameter configuration on the candidate camera layout function to obtain a target camera layout function, wherein the target parameters comprise the number of genes, the number of populations, the maximum iteration number, the minimum iteration number, the stable iteration number, the cross probability and the variation probability;
determining an objective function in a preset genetic algorithm according to the camera parameters and the scene parameters to obtain an initial camera layout function, wherein the method comprises the following steps: calculating a bi-quad observability score for each point in the point cloud; determining the maximum value of the double four-azimuth observability score of each point as the initial observability score of each point; acquiring the total number of points in the point cloud, and carrying out normalization processing on the initial observability score of each point according to the total number of points in the point cloud to obtain a target observability score; maximizing a score function in a preset genetic algorithm according to the target observability score to obtain a target function; generating an initial camera layout function according to the objective function and the genetic algorithm;
The calculating a bi-quad observability score for each point in the point cloud includes: acquiring a 360-degree coordinate system corresponding to each point in the point cloud; dividing the 360-degree coordinate system into four equal-partition sections and performing preset angle rotation to obtain a target four equal-partition section; acquiring the number of observable target orientations of each point in the point cloud in the target quarter section; scoring each point in the point cloud according to the number of the target orientations to obtain a double four-orientation observability score of each point;
the obtaining the total number of points in the point cloud, and normalizing the initial observability score of each point according to the total number of points in the point cloud to obtain a target observability score, including: acquiring the total number of points in the point cloud, and determining an observability score and a value according to the total number and the initial observability score of each point; dividing the observability score sum value by the total number to obtain an initial observability score; dividing the initial observability score by 4 results in a target observability score.
2. The genetic-based camera layout function optimization method according to claim 1, wherein the performing parameter configuration on the preset type camera to obtain camera parameters and scene parameters comprises:
Setting the expected installation quantity of the preset type cameras, the arrangement range information corresponding to the preset type cameras and the installable area information of the preset type cameras to obtain camera parameters;
and discretizing the scene capturing area of the preset type camera and setting the size of the area to obtain scene parameters.
3. The genetic-based camera layout function optimization method according to claim 2, wherein the discretizing the scene capturing area of the preset type camera and setting the area size to obtain scene parameters includes:
performing discretization processing on a scene capturing area of the preset type camera according to preset discrete precision to obtain a discrete point group;
the range of the scene capturing area is retracted inwards relative to the position of the row frame in the mountable area information, so that the size of the scene capturing area is obtained;
and generating a point cloud according to the discrete point group and the size of the scene capturing area, and determining the point cloud as a scene parameter.
4. A method of optimizing a genetic based camera layout function according to any one of claims 1-3, wherein optimizing actual installation location parameters in the initial camera layout function to obtain candidate camera layout functions comprises:
Normalizing the actual installation position parameters in the initial camera layout function to obtain optimized parameter values;
and determining the optimized parameter value as a numerical range of the individual genes in the initial camera layout function to obtain a candidate camera layout function.
5. A genetic-based camera layout function optimizing apparatus, characterized in that the genetic-based camera layout function optimizing apparatus comprises:
the first configuration module is used for carrying out parameter configuration on a preset type camera to obtain camera parameters and scene parameters, wherein the scene parameters are used for indicating point clouds;
the determining module is used for determining an objective function in a preset genetic algorithm according to the camera parameters and the scene parameters to obtain an initial camera layout function;
the optimization module is used for optimizing the actual installation position parameters in the initial camera layout function to obtain candidate camera layout functions;
the second configuration module is used for carrying out target parameter configuration on the candidate camera layout function to obtain a target camera layout function, wherein the target parameters comprise the number of genes, the number of populations, the maximum iteration number, the minimum iteration number, the stable iteration number, the cross probability and the variation probability;
The determining module comprises: the computing unit is used for computing a double four-azimuth observability score of each point in the point cloud; a determining unit, configured to determine a score with a maximum median value of the bi-four-azimuth observability scores of each point as an initial observability score of each point; the second processing unit is used for obtaining the total number of points in the point cloud, and carrying out normalization processing on the initial observability score of each point according to the total number of points in the point cloud to obtain a target observability score; the third processing unit is used for maximizing a score function in a preset genetic algorithm according to the target observability score to obtain a target function; the generating unit is used for generating an initial camera layout function according to the objective function and the genetic algorithm;
the computing unit is specifically configured to: acquiring a 360-degree coordinate system corresponding to each point in the point cloud; dividing the 360-degree coordinate system into four equal-partition sections and performing preset angle rotation to obtain a target four equal-partition section; acquiring the number of observable target orientations of each point in the point cloud in the target quarter section; scoring each point in the point cloud according to the number of the target orientations to obtain a double four-orientation observability score of each point;
The second processing unit is specifically configured to: acquiring the total number of points in the point cloud, and determining an observability score and a value according to the total number and the initial observability score of each point; dividing the observability score sum value by the total number to obtain an initial observability score; dividing the initial observability score by 4 results in a target observability score.
6. A genetic-based camera layout function optimizing apparatus, characterized in that the genetic-based camera layout function optimizing apparatus comprises: a memory and at least one processor, the memory having instructions stored therein, the memory and the at least one processor being interconnected by a line;
the at least one processor invokes the instructions in the memory to cause the genetic based camera layout function optimization device to perform the genetic based camera layout function optimization method of any of claims 1-4.
7. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the genetic based camera layout function optimization method according to any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010268735.8A CN111460735B (en) | 2020-04-08 | 2020-04-08 | Camera layout function optimization method based on genetic inheritance and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010268735.8A CN111460735B (en) | 2020-04-08 | 2020-04-08 | Camera layout function optimization method based on genetic inheritance and related equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111460735A CN111460735A (en) | 2020-07-28 |
| CN111460735B true CN111460735B (en) | 2023-09-01 |
Family
ID=71680490
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010268735.8A Active CN111460735B (en) | 2020-04-08 | 2020-04-08 | Camera layout function optimization method based on genetic inheritance and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111460735B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115529437A (en) * | 2021-06-25 | 2022-12-27 | 青岛海信智慧生活科技股份有限公司 | Method, device, equipment and medium for determining monitoring equipment arrangement information |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103630072A (en) * | 2013-10-25 | 2014-03-12 | 大连理工大学 | Layout optimization method for camera in binocular vision measuring system |
| CN109635332A (en) * | 2018-11-08 | 2019-04-16 | 北京航空航天大学 | A kind of variable step constellation orbital optimization method and device based on genetic algorithm |
| CN110930182A (en) * | 2019-11-08 | 2020-03-27 | 中国农业大学 | Improved particle swarm optimization algorithm-based client classification method and device |
-
2020
- 2020-04-08 CN CN202010268735.8A patent/CN111460735B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103630072A (en) * | 2013-10-25 | 2014-03-12 | 大连理工大学 | Layout optimization method for camera in binocular vision measuring system |
| CN109635332A (en) * | 2018-11-08 | 2019-04-16 | 北京航空航天大学 | A kind of variable step constellation orbital optimization method and device based on genetic algorithm |
| CN110930182A (en) * | 2019-11-08 | 2020-03-27 | 中国农业大学 | Improved particle swarm optimization algorithm-based client classification method and device |
Non-Patent Citations (1)
| Title |
|---|
| 运动捕获系统中多摄像机布局优化方法;袁金刚;张艳宁;郑江滨;郗润平;;微处理机(第03期) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111460735A (en) | 2020-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111460735B (en) | Camera layout function optimization method based on genetic inheritance and related equipment | |
| CN106600570A (en) | Massive point cloud filtering method based on cloud calculating | |
| CN111161331B (en) | Registration method of BIM model and GIS model | |
| CN111126459A (en) | Method and device for identifying fine granularity of vehicle | |
| CN107220647A (en) | Crop location of the core method and system under a kind of blade crossing condition | |
| CN116310849B (en) | Tree point cloud monomerization extraction method based on three-dimensional morphological characteristics | |
| CN113538264A (en) | Denoising method and device for point cloud data and storage medium | |
| CN116628903A (en) | Optimal arrangement method for urban wind field environment monitoring sensors | |
| CN112991300B (en) | Single wood skeleton extraction and visualization method based on neighborhood characteristics | |
| CN106910180A (en) | A kind of image quality measure method and device | |
| CN111445525A (en) | Camera installation position layout method, device, equipment and storage medium | |
| CN108510010A (en) | A kind of density peaks clustering method and system based on prescreening | |
| CN111383353B (en) | Fractured bone model registration method based on Gaussian mixture model and contour descriptor | |
| CN109741358A (en) | Superpixel segmentation method based on the study of adaptive hypergraph | |
| CN116862106A (en) | Urban construction land updating potential analysis method, device, terminal and medium | |
| CN111047684A (en) | Model simplification method based on three-dimensional model characteristics | |
| CN113377891B (en) | Space vector data pattern spot area-oriented adjustment method | |
| CN111540019B (en) | Method, device, equipment and storage medium for determining camera mounting position | |
| CN115588086A (en) | Map dividing method, map dividing device, computer readable storage medium and processor | |
| CN108596234A (en) | A kind of constraint Spectral Clustering based on shared nearest neighbor | |
| CN112435342A (en) | Region division method and system for special-shaped curved surface | |
| Athira et al. | An integrated method for identifying essential proteins from multiplex network model of protein–protein interactions | |
| CN111540018B (en) | Score calculation method of symmetrical layout mode of camera and related equipment | |
| CN111583278A (en) | Graph center axis extraction method based on double-normal tracking | |
| CN115388893B (en) | Mobile robot mapping method based on genetic filtering SLAM algorithm |
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 |