CN111192321B - Target three-dimensional positioning method and device - Google Patents
Target three-dimensional positioning method and device Download PDFInfo
- Publication number
- CN111192321B CN111192321B CN201911404383.8A CN201911404383A CN111192321B CN 111192321 B CN111192321 B CN 111192321B CN 201911404383 A CN201911404383 A CN 201911404383A CN 111192321 B CN111192321 B CN 111192321B
- Authority
- CN
- China
- Prior art keywords
- camera
- dimensional
- position point
- target object
- coordinate system
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/10—Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
Abstract
The invention provides a three-dimensional positioning method and device for a target object, wherein the method comprises the steps of carrying out BIM modeling on a construction site to obtain a construction site BIM three-dimensional model, carrying out target object identification in an image or video shot by a camera, and obtaining a two-dimensional coordinate of the target object in the image or video to serve as a first position point; according to the first position point mapping calculation, obtaining a two-dimensional coordinate of the projection of the target object on the near projection surface of the camera in the camera coordinate system as a second position point; calculating three-dimensional coordinates in a camera coordinate system taking a camera as an origin as a third position point according to the second position point; and performing product transformation processing on the third position point and the world coordinate matrix to obtain a position point of the target object in the three-dimensional world coordinate system of the BIM three-dimensional model as a fourth position point, and displaying the three-dimensional positioning of the target object in the BIM three-dimensional model according to the fourth position point. The positioning effect of the invention is not influenced by the behavior of the target object and the existence of the positioning base station, and is more robust and reliable.
Description
Technical Field
The invention relates to the technical field of digital management of construction engineering construction, in particular to a three-dimensional positioning method and device for a target object.
Background
The three-dimensional visualization of the target object is realized based on a building information model (BIM three-dimensional model), which is a widely adopted technology in the current construction site management, and coordinate information of the target object in the BIM three-dimensional model is obtained through positioning technologies such as GIS, beidou GNSS, bluetooth, wifi ranging and the like or a video intelligent analysis technology, and three-dimensional visual display is performed.
The existing constructor positioning technology needs to wear a positioning tag, an electronic tag, a positioning device with a positioning chip, a mobile terminal or other small electronic positioning equipment by personnel, and realizes the positioning of the personnel position through the data exchange with a positioning base station.
In the prior art, no matter outdoor or indoor positioning is carried out, each constructor is provided with a portable electronic tag or equipment, and the indoor positioning is also provided with a plurality of positioning base stations, so that the positioning cost is increased, on the other hand, the positioning base stations are deployed with larger workload, and for a construction site, new base station deployment is required to be continuously carried out along with the advancement of construction, so that the implementation is complex, and the problem of difficult implementation exists. In addition, the positioning is effective, if the constructor does not wear the related equipment intentionally or unintentionally, the positioning is completely disabled, so that the reliability is difficult to ensure.
Disclosure of Invention
Aiming at the technical problems of high positioning cost, difficult implementation, difficult guarantee of reliability and the like of constructors in the prior art, the invention provides a three-dimensional positioning method for a target object. The invention also provides a three-dimensional positioning device for the target object.
The technical scheme of the invention is as follows:
a method for three-dimensional localization of objects for localization of objects at a construction site, the method comprising the steps of:
deploying a monitoring camera on a construction site or utilizing an existing monitoring camera on the construction site and carrying out BIM modeling on the construction site to obtain a BIM three-dimensional model of the construction site, obtaining position information and view angle directions of the camera in a three-dimensional world coordinate system of the BIM three-dimensional model, and calculating a world coordinate matrix of the camera according to the position information and the view angle directions;
Carrying out object identification in an image or video shot by a camera, and acquiring a two-dimensional coordinate of an object in the image or video to serve as a first position point;
according to the first position point mapping calculation, a two-dimensional coordinate of the target object projected on a near projection plane of a camera in a camera coordinate system is obtained to serve as a second position point;
calculating three-dimensional coordinates in a camera coordinate system taking a camera as an origin as a third position point according to the second position point;
and performing product transformation processing on the third position point and the world coordinate matrix to obtain a position point of the target object in a three-dimensional world coordinate system of the BIM three-dimensional model as a fourth position point, and displaying three-dimensional positioning of the target object in the BIM three-dimensional model according to the fourth position point.
Further, identifying a target object in an image or video shot by a camera, and acquiring a two-dimensional coordinate of a head center point of the target object in the image or video as a first position point; and carrying out mapping calculation according to the first position point and combining with the horizontal and vertical pixel values of the camera to obtain the two-dimensional coordinates of the head center point of the target object projected on the near projection plane of the camera in the camera coordinate system as the second position point.
Further, the calculating the three-dimensional coordinate in the camera coordinate system with the camera as the origin according to the second position point to be used as the third position point comprises:
making a straight line from the camera position to the second position point by taking a camera as an origin and calculating the straight line direction, wherein the straight line direction comprises an X-axis coordinate of the straight line calculated according to the tangent value of the second position point and a half view angle of the camera, a Y-axis coordinate of the straight line calculated according to the second position point and a horizontal and vertical pixel value and a Z-axis coordinate determined by the default of the camera looking to the Z-axis negative direction in a camera coordinate system; calculating the distance between the head of the target object and the camera according to the aspect ratio of the head area of the target object in the image or the video image; and scaling the straight line according to the distance between the head of the target object and the camera obtained by calculation according to the principle of imaging near-far and far-far of the camera, so as to determine the third position point.
Further, the calculating the world coordinate matrix of the camera according to the position information and the viewing angle direction includes: determining Euler angles of the camera, which need to rotate around three coordinate axes respectively, according to the visual angle direction of the camera in a three-dimensional world coordinate system, and calculating to obtain a first two-three column of a world coordinate matrix of the camera according to the Euler angles; and obtaining a fourth column of the world coordinate matrix of the camera according to the position information of the camera in the three-dimensional world coordinate system.
A three-dimensional object positioning device for positioning an object at a construction site, comprising:
the information acquisition module is used for acquiring position information and view angle directions of a camera deployed on a construction site in a three-dimensional world coordinate system of a construction site BIM three-dimensional model obtained by BIM modeling the construction site; carrying out object identification in an image or video shot by a camera, and acquiring a two-dimensional coordinate of an object in the image or video to serve as a first position point;
the calculation module is used for calculating a world coordinate matrix of the camera according to the position information and the view angle direction; according to the first position point mapping calculation, a two-dimensional coordinate of the target object projected on a near projection plane of a camera in a camera coordinate system is obtained to serve as a second position point; calculating three-dimensional coordinates in a camera coordinate system taking a camera as an origin as a third position point according to the second position point; performing product transformation processing on the third position point and the world coordinate matrix to obtain a position point of the target object in a three-dimensional world coordinate system of the BIM three-dimensional model as a fourth position point;
and the positioning display module is used for displaying the three-dimensional positioning of the target object in the BIM three-dimensional model according to the fourth position point.
Further, the information acquisition module acquires two-dimensional coordinates of a target object in the image or the video by using an image or video intelligent analysis technology as a first position point.
Further, the information acquisition module performs object identification in an image or video shot by a camera, and acquires a two-dimensional coordinate of a head center point of an object in the image or video as a first position point by utilizing an image or video intelligent analysis technology.
Further, the calculation module performs mapping calculation according to the first position point and by combining with the horizontal pixel value and the vertical pixel value of the camera to obtain the two-dimensional coordinate of the projection of the head center point of the target object on the near projection plane of the camera in the camera coordinate system as the second position point.
Further, the calculating module calculates, according to the second location point, a three-dimensional coordinate in a camera coordinate system with a camera as an origin as a third location point, specifically:
making a straight line from the camera position to the second position point by taking a camera as an origin, and calculating the straight line direction, wherein the straight line direction comprises an X-axis coordinate of the straight line calculated according to the tangent value of the second position point and a half view angle of the camera, a Y-axis coordinate of the straight line calculated according to the second position point and a transverse-longitudinal pixel ratio and a Z-axis coordinate determined by the default of the camera looking to the Z-axis negative direction in a camera coordinate system; calculating the distance between the head of the target object and the camera according to the aspect ratio of the head area of the target object in the image or the video image; and scaling the straight line according to the distance between the head of the target object and the camera obtained by calculation according to the principle of imaging near-far and far-far of the camera, so as to determine the third position point.
Further, the calculating module is configured to calculate a world coordinate matrix of the camera according to the position information and the view direction, and includes: determining Euler angles of the camera, which need to rotate around three coordinate axes respectively, according to the visual angle direction of the camera in a three-dimensional world coordinate system, and calculating to obtain a first two-three column of a world coordinate matrix of the camera according to the Euler angles; and obtaining a fourth column of the world coordinate matrix of the camera according to the position information of the camera in the three-dimensional world coordinate system.
The invention has the following technical effects:
the invention provides a three-dimensional positioning method of a target object, which is based on intelligent analysis of a monitoring video or an image, acquires two-dimensional position information of the target object in the image or the video, then converts the two-dimensional position information of the target object into a position point of a projection of a camera near projection plane of the target object in a camera coordinate system through the position and a visual angle of a camera corresponding to the monitoring video in a BIM three-dimensional model, acquires the position of the target object in the camera coordinate system according to the position point and the distance between the target object and the camera position, and acquires the position of the target object in the three-dimensional world coordinate system according to the position of the target object in the camera coordinate system and a world coordinate matrix, thereby realizing the positioning of the target object in the BIM three-dimensional model. The method is based on intelligent analysis of the monitoring video to realize the positioning of the target object such as constructors, and because a building site is usually provided with a plurality of video monitoring devices-cameras, the positioning of the target object is based on the existing devices, thereby remarkably reducing the cost and the deployment workload. Of course, the camera may be redeployed when the camera at the construction site is insufficient. The two-dimensional position information of the target object in the image or video in the image plane coordinate system is converted into the three-dimensional position information in the camera coordinate system, and finally converted into the three-dimensional position information in the three-dimensional world coordinate system, so that the positioning of the target object in the BIM three-dimensional model is realized, the accurate positioning can be realized without wearing a positioning tag, an electronic tag, a positioning device containing a positioning chip, a mobile terminal or other small electronic positioning equipment on the target object, the problems that the positioning cost is high, the implementation is difficult, the reliability is difficult to guarantee and the like caused by the fact that a constructor needs to wear special equipment in the prior art are completely solved, and no matter whether the positioning base station is arranged indoors or outdoors, the positioning cost is saved, and the easy accurate positioning is realized more easily. The positioning effect of the invention is not influenced by the action of a target object, namely constructors, and is more robust and reliable.
The invention also provides a three-dimensional target positioning device which corresponds to the three-dimensional target positioning method, can be also understood as a device for realizing the three-dimensional target positioning method, can acquire the position information and the view angle direction of a camera with a target object in a view cone range in a world coordinate system through the mutual cooperation of an information acquisition module, a calculation module and a positioning display module, and acquire the two-dimensional position information of the target object in an image plane coordinate system of an image or video shot by the camera, converts the two-dimensional position information into the three-dimensional position information in a camera coordinate system, and converts the three-dimensional position information in the camera coordinate system into the three-dimensional position information in a three-dimensional world coordinate system through a world coordinate matrix, so that the positioning of the target object in a BIM three-dimensional model is realized, the accurate positioning can be realized without wearing a positioning tag, an electronic tag, a positioning device with a positioning chip, a mobile terminal or other small electronic positioning equipment, and the accurate positioning cost is saved, the accurate positioning is easier to realize no matter whether the positioning is indoor or outdoor, and the three-dimensional positioning precision is improved.
Drawings
FIG. 1 is a flow chart of a method for three-dimensional positioning of an object according to the present invention.
FIG. 2 is a flow chart of a preferred embodiment of the method for three-dimensional localization of objects of the present invention.
Fig. 3 is a block diagram of a three-dimensional positioning device for an object according to the present invention.
Fig. 4 is a diagram showing the positional relationship between a camera and a near projection surface of the three-dimensional positioning device for an object of the present invention.
Fig. 5 is a diagram showing the positional relationship between the camera position and the near projection surface and the far projection surface, respectively, of the three-dimensional positioning device for an object of the present invention.
FIG. 6 is an overall flow chart of a method for three-dimensional positioning of an object according to the present invention.
FIG. 7 is a drawing of the position of a target in a BIM three-dimensional model according to the three-dimensional positioning method of the target of the present invention.
Detailed Description
The technical scheme of the invention is described in detail below with reference to the accompanying drawings.
The invention provides a three-dimensional positioning method of a target object, which is used for positioning the target object on a construction site, as shown in fig. 1, and comprises the following steps: deploying a monitoring camera on a construction site or utilizing an existing monitoring camera on the construction site and carrying out BIM modeling on the construction site to obtain a BIM three-dimensional model of the construction site, obtaining the position information and the view angle direction of the camera in a three-dimensional world coordinate system of the BIM three-dimensional model, and calculating a world coordinate matrix of the camera according to the position information and the view angle direction; carrying out object identification in an image or video shot by a camera, and acquiring a two-dimensional coordinate of an object in the image or video to serve as a first position point; according to the first position point mapping calculation, obtaining a two-dimensional coordinate of the projection of the target object on the near projection surface of the camera in the camera coordinate system as a second position point; calculating three-dimensional coordinates in a camera coordinate system taking a camera as an origin as a third position point according to the second position point; and performing product transformation processing on the third position point and the world coordinate matrix to obtain a position point of the target object in the three-dimensional world coordinate system of the BIM three-dimensional model as a fourth position point, and displaying the three-dimensional positioning of the target object in the BIM three-dimensional model according to the fourth position point.
Specifically, the present invention relates to conversion between three coordinate systems, which are an image plane coordinate system, a camera coordinate system and a three-dimensional world coordinate system (abbreviated as world coordinate system), wherein the image plane coordinate system and the camera coordinate system are in an imaging projection relationship, it can be understood that the image plane coordinate system is a coordinate system located on a near projection plane of a camera, two-dimensional coordinates of a target object in an image or a video are obtained through the image plane coordinate system, the world coordinate system and the camera coordinate system can be converted into another coordinate system through a rigid body conversion mode, and the camera coordinate system can be converted into coordinates in the world coordinate system through a world coordinate matrix, so that the coordinates of the target object in the world coordinate system can be determined, as shown in fig. 5, a range which can be shot by the camera is between the near projection plane and the far projection plane of a camera position Z, and the target object (constructor) moving in the range can be positioned by the camera.
Based on the embodiment of the invention, the three-dimensional positioning method of the target object is based on intelligent analysis of a monitoring video or an image, two-dimensional position information of the target object in the image or the video is obtained, then the two-dimensional position information of the target object is converted into a position point of the target object projected by a camera near projection plane in a camera coordinate system through the position and a visual angle of a camera corresponding to the monitoring video in a BIM three-dimensional model, the position of the target object in the camera coordinate system is obtained according to the position point and the distance between the target object and the camera position, and the position of the target object in the camera coordinate system is obtained according to the position of the target object in the camera coordinate system and a world coordinate matrix, so that the positioning of the target object in the BIM three-dimensional model is realized. The method is based on intelligent analysis of the monitoring video to realize the positioning of the target object such as constructors, and because a building site is usually provided with a plurality of video monitoring devices-cameras, the positioning of the target object is based on the existing devices, thereby remarkably reducing the cost and the deployment workload. Of course, the camera may be redeployed when the camera at the construction site is insufficient. The positioning of the position of the target object in the BIM three-dimensional model is realized by converting the two-dimensional position information of the target object in the image or video in the image plane coordinate system into the three-dimensional position information in the camera coordinate system and finally into the three-dimensional position information in the three-dimensional world coordinate system, and the accurate positioning can be realized without wearing a positioning tag, an electronic tag, a positioning device containing a positioning chip, a mobile terminal or other small electronic positioning equipment on the target object, namely, the positioning of constructors is realized by adopting a monitoring camera instead of a special positioning device or equipment; the three-dimensional coordinates of the personnel in the BIM three-dimensional model are obtained through the mapping of the two-dimensional view corresponding to the camera in the BIM three-dimensional model and the monitoring video of the camera, so that the problems of high positioning cost, difficult implementation, difficult guarantee of reliability and the like caused by the fact that the constructor needs to wear special equipment in the prior art are completely solved, and no matter in a room or in an outdoor place, a plurality of positioning base stations are not needed to be equipped, so that the positioning cost is saved, and the easy and accurate positioning is easy to realize. The positioning effect of the invention is not influenced by the action of a target object, namely constructors, and is more robust and reliable.
As a preferred embodiment of the three-dimensional positioning method of the object of the present invention, as shown in a preferred flow chart of FIG. 2, the method specifically comprises the following steps:
a first step of: the method comprises the steps of deploying a monitoring camera on a construction site or utilizing an existing monitoring camera on the construction site, carrying out BIM modeling on the construction site to obtain a construction site BIM three-dimensional model, obtaining detailed construction site three-dimensional model information, obtaining position information and view angle directions of the camera in a three-dimensional world coordinate system of the BIM three-dimensional model, namely position points Z (xz, yz, zz) in the three-dimensional world coordinate system, and calculating a world coordinate matrix M of the camera according to the position points Z and the view angle directions R.
Specifically, the world coordinate matrix of the camera is composed of four columns, the first three columns representing three directions, the fourth column representing the positional information of the camera in the world coordinate system, and the last row of the world coordinate matrix is denoted as [0 0 0 1], wherein 0 of the last row represents the direction, 1 of the last row represents one positional point, that is, represents the positional point of the camera in the world coordinate system, and the world coordinate matrix of the camera is obtained by: determining Euler angles of the camera, which need to rotate around three coordinate axes respectively, according to the visual angle direction of the camera in a three-dimensional world coordinate system, and calculating according to the Euler angles to obtain a first column and a second column of a world coordinate matrix of the camera except for the last row; and obtaining the fourth column of the last row of the world coordinate matrix of the camera according to the position information of the camera in the three-dimensional world coordinate system. That is, according to the view angle direction of the camera in the world coordinate system, obtaining euler angles (α, β, γ) that the camera needs to rotate around xyz axes respectively, and calculating the euler angles (α, β, γ) according to a preset matrix formula to obtain a first two-three column of the world coordinate matrix of the camera except for the last row, wherein the preset matrix formula (1) is as follows:
The fourth column of the world coordinate matrix excluding the last row is the position information of the camera in the world coordinate system, expressed as
The last row of the world coordinate matrix, denoted [0 0 0 1];
the world coordinate matrix M of the camera is obtained from the above analysis.
And a second step of: carrying out object identification in an image or video shot by a camera, and acquiring a two-dimensional coordinate of a head center point of an object in the image or video to serve as a first position point; and carrying out mapping calculation according to the first position point and combining with the horizontal pixel value and the vertical pixel value of the camera to obtain the two-dimensional coordinate of the head center point of the target object projected on the near projection plane of the camera in the camera coordinate system as the second position point.
Specifically, by utilizing the existing mature image or video intelligent analysis technology such as machine learning, the target objects such as constructors and the like in the images or videos in the monitoring cameras deployed in the building site are identified, and the position information of the constructors in the monitoring images or videos is obtained; whenever a constructor is detected in the video, the constructor's head is obtainedTwo-dimensional coordinates of the central position in the image, i.e. the first position point coordinates P 0 (x 0 ,y 0 ) And the aspect ratio W of the constructor head area in the image or video 0 ,H 0 ；
The image in the surveillance video can be regarded as the near projection plane of the camera in the three-dimensional world coordinate system. Let the horizontal and vertical pixel values of the camera be W 1 ，H 1 The center point P of the head of the person is calculated according to the following method 0 Second position point P projected by camera near projection surface in three-dimensional world 1 (x 1 ,y 1 )：
x 1 ＝(x 0 /W 1 )*2-1
y 1 ＝-(y 0 /H 1 )*2+1
According to the embodiment of the invention, the second position point can be positioned only by combining the inherent camera parameter value with the initial plane coordinate of the target object in the image or video, so that the positioning operation step is simplified, and when the target object is a constructor, as the head and the body of the constructor are integrated in the BIM three-dimensional model (BIM model for short), the head part is placed at the calculated head position of the constructor in the BIM three-dimensional model by only positioning the head center point, namely the positioning of the constructor is realized, and the positioning operation is simplified and the positioning precision is improved by only positioning the head center point of the constructor.
And a third step of: and calculating three-dimensional coordinates in a camera coordinate system taking a video camera as an origin according to the second position point to obtain a third position point. Preferably, a straight line is made from the camera position to the second position point by taking the camera as an origin and a straight line direction is calculated, wherein the straight line direction comprises an X-axis coordinate of the straight line calculated according to the tangent value of the second position point and the half view angle of the camera, a Y-axis coordinate of the straight line calculated according to the second position point and the horizontal and vertical pixel value and a Z-axis coordinate which is determined in a camera coordinate system when the camera defaults to a Z-axis negative direction; calculating the distance between the head of the target object and the camera according to the aspect ratio of the head area of the target object in the image or the video image; and scaling the straight line according to the principle of the near-far-small imaging of the camera and the distance between the head of the target object and the camera obtained through calculation to determine a third position point.
Specifically, from camera position Z to point P 1 A straight line L (x) L ,y L ,z L ) The direction of L is calculated:
x L ＝x 1 * halfTan 1, wherein halfTan is the tan value of the real camera half view angle;
y L ＝y 1 l is H1/W1, and a Y-axis coordinate is obtained according to the horizontal-vertical pixel ratio;
z L -1, depending on the default view of the camera in the three-dimensional world to the negative Z-axis direction.
According to the aspect ratio W of the head of the worker in the image 0 ,H 0 Calculating the distance D (the principle of the imaging of the camera) between the head of the worker and the position of the camera:
D＝λ/(W 0 +H 0 ) Where λ is an empirical constant based on the monitor video picture size statistics.
The straight line L is scaled by D times to obtain a third position point P of the head of the constructor in a camera coordinate system taking a camera as an origin 2 (x 2 ,y 2 ,z 2 ):
x 2 ＝x L *D
y 2 ＝y L *D
z 2 ＝z L *D
According to the embodiment of the invention, the second position point is the projection position point of the real object on the near projection surface of the camera, the third position point is the position point of the real object between the near projection surface and the far projection surface of the camera, namely the position point of the real object in the camera coordinate system, the second position point is the image (projection) of the third position point, the image (projection) of the third position point is connected with the position point of the image (projection) of the object on the camera by taking the camera as the origin, and the third position point of the real object is found and determined by scaling the straight line, so that the position of the constructor in the camera coordinate system can be positioned without wearing a positioning tag, a moving end or other small electronic positioning equipment by the constructor, the positioning cost is saved, and the easy and accurate positioning is realized. The positioning effect of the invention is not influenced by the action of a target object, namely constructors, and is more robust and reliable.
Fourth step: and performing product transformation processing on the third position point and the world coordinate matrix to obtain a position point of the target object in the three-dimensional world coordinate system of the BIM three-dimensional model as a fourth position point, and displaying the three-dimensional positioning of the target object in the BIM three-dimensional model according to the fourth position point. Specifically, the third position point P 2 Multiplying the world coordinate matrix M of the camera to obtain a fourth position point P of the head of the constructor in the three-dimensional world coordinate system 3 (x 3 ,y 3 ,z 3 )＝M*P 2 . That is, the third position point is converted into a fourth position point through the world coordinate matrix, and the fourth position point is the coordinate of the target object in the world coordinate system. Finally according to four position points P 3 The coordinates of a target (e.g., a constructor) are plotted in a BIM three-dimensional model.
According to the embodiment of the invention, the second position point (actual position point) of the target object in the camera coordinate system is converted into the position point of the target object in the three-dimensional world coordinate system by utilizing the Euler angle determined by the visual angle direction of the camera in the three-dimensional world coordinate system and the world coordinate matrix obtained by combining the position information of the camera in the three-dimensional world coordinate system, and the position of the target object (constructor) is drawn in the BIM three-dimensional model, so that the positioning process of the target object is simplified, a plurality of positioning base stations are not required, and the target object can be simply, conveniently and accurately positioned without wearing a positioning device by only collecting video images and identifying two-dimensional or three-dimensional position coordinates of the target object in different coordinate systems and simple coordinate conversion. The positioning effect of the invention is not influenced by the action of a target object, namely constructors, and is more robust and reliable.
The invention also provides a three-dimensional object positioning device, which corresponds to the three-dimensional object positioning method, and can be understood as a device for realizing the three-dimensional object positioning method, wherein the embodiment of the three-dimensional object positioning device, the achieved effect and the solved technical problem are the same as those achieved by the three-dimensional object positioning method, only a brief description is provided herein, and repeated parts are not repeated. As shown in the block diagram of fig. 3, the apparatus is used for target positioning at a construction site, and includes: the information acquisition module is used for acquiring position information and view angle directions of a camera deployed on a construction site in a three-dimensional world coordinate system of a construction site BIM three-dimensional model obtained by BIM modeling the construction site; carrying out object identification in an image or video shot by a camera, and acquiring a two-dimensional coordinate of an object in the image or video to serve as a first position point; the calculation module is used for calculating a world coordinate matrix of the camera according to the position information and the view angle direction; according to the first position point mapping calculation, obtaining a two-dimensional coordinate of the projection of the target object on the near projection surface of the camera in the camera coordinate system as a second position point; calculating three-dimensional coordinates in a camera coordinate system taking a camera as an origin as a third position point according to the second position point; performing product transformation processing on the third position point and the world coordinate matrix to obtain a position point of the target object in a three-dimensional world coordinate system of the BIM three-dimensional model as a fourth position point; and the positioning display module is used for displaying the three-dimensional positioning of the target object in the BIM three-dimensional model according to the fourth position point.
According to the three-dimensional positioning device for the target object, the information acquisition module, the calculation module and the positioning display module which are sequentially connected are used for mutually and cooperatively working, the position information and the visual angle direction of the camera with the target object in the world coordinate system in the view cone range are acquired, the two-dimensional position information of the target object in the image or video shot by the camera in the image plane coordinate system is acquired, the two-dimensional position information is converted into the three-dimensional position information in the camera coordinate system, the three-dimensional position information in the camera coordinate system is converted into the three-dimensional position information in the three-dimensional world coordinate system through the world coordinate matrix, the positioning of the position of the target object in the BIM three-dimensional model is realized, the positioning of the target object can be realized without wearing a positioning tag, an electronic tag, a positioning device with a positioning chip, a mobile terminal or other small electronic positioning equipment, and a plurality of positioning base stations are not needed indoors or outdoors, so that the positioning cost is saved, the easy and the accurate positioning is realized, and the three-dimensional positioning precision is improved.
Preferably, in this embodiment, the information acquisition module acquires two-dimensional coordinates of the object in the image or video as the first location point using an image or video intelligent analysis technique.
Based on the embodiment of the application, the video image is analyzed in real time by a computer by utilizing an image or intelligent video analysis technology, and the event violating the set rule is alarmed by rule filtering, so that the method is applied to automatically positioning and displaying the target object with the set characteristics, such as preset shape, preset contour, preset height and the like, with high accuracy and high speed. The image or intelligent video analysis technology is based on a deep learning algorithm such as Convolutional Neural Network (CNN) and has a plurality of different algorithms, and is roughly classified into a 2-stage algorithm represented by R-CNN and a 1-stage algorithm represented by YOLO. At present, a system provided by a third party is adopted, a CNN network is used for directly predicting the types and positions of different targets based on a YOLO algorithm, and finally, the identification of human targets is realized. Compared with R-CNN algorithm, the method has lower accuracy but higher speed.
Preferably, the information acquisition module performs object identification in an image or video shot by the camera, and acquires two-dimensional coordinates of a head center point of an object in the image or video as a first position point by utilizing an image or video intelligent analysis technology.
According to the embodiment of the invention, the head center point of the target object is utilized to replace the whole target object for positioning, and as the head and the body of the constructor are integrated in the BIM three-dimensional model, the head position of the constructor calculated by the BIM three-dimensional model is placed at the head position by only positioning the head center point, namely the positioning of the constructor is realized, and the positioning operation is simplified and the positioning precision is improved by only positioning the head center point of the constructor.
Preferably, the calculation module performs mapping calculation according to the first position point and combined with the horizontal and vertical pixel values of the camera to obtain a two-dimensional coordinate of the head center point of the target object projected on the near projection plane of the camera in the camera coordinate system as the second position point.
According to the embodiment of the invention, the second position point can be positioned only by combining the inherent camera parameter value with the initial plane coordinate of the target object in the image or video, so that the positioning operation step is simplified, and when the target object is a constructor, as the head and the body of the constructor are integrated in the BIM three-dimensional model, the head position of the constructor calculated by the BIM three-dimensional model is placed by only positioning the head center point, namely the positioning of the constructor is realized, and the positioning operation is simplified and the positioning precision is improved by only positioning the head center point of the constructor.
Preferably, the calculating module calculates, as the third location point, a three-dimensional coordinate in a camera coordinate system with the camera as an origin, according to the second location point, specifically: taking a camera as an origin, making a straight line from the position of the camera to a second position point, and calculating a straight line direction, wherein the straight line direction comprises an X-axis coordinate of the straight line calculated according to the second position point and the tangent value of a half visual angle of the camera, a Y-axis coordinate of the straight line calculated according to the second position point and a horizontal-vertical pixel ratio and a Z-axis coordinate determined by the default of the camera looking in a Z-axis negative direction in a camera coordinate system; calculating the distance between the head of the target object and the camera according to the aspect ratio of the head area of the target object in the image or the video image; and scaling the straight line according to the principle of the near-far-small imaging of the camera and the distance between the head of the target object and the camera obtained through calculation to determine a third position point.
According to the embodiment of the invention, the second position point is the projection position point of the real object on the near projection surface of the camera, the third position point is the position point of the real object between the near projection surface and the far projection surface of the camera, namely the position point of the real object in the camera coordinate system, the second position point is the image (projection) of the third position point, the image (projection) of the third position point is connected with the position point of the image (projection) of the object on the camera by taking the camera as the origin, and the third position point of the real object is found and determined by scaling the straight line, so that the position of the constructor in the camera coordinate system can be positioned without wearing a positioning tag, a moving end or other small electronic positioning equipment by the constructor, the positioning cost is saved, and the easy and accurate positioning is realized. The positioning effect of the invention is not influenced by the action of a target object, namely constructors, and is more robust and reliable.
Preferably, the calculating module is configured to calculate a world coordinate matrix of the camera according to the position information and the view direction, including: determining Euler angles of the camera, which need to rotate around three coordinate axes respectively, according to the visual angle direction of the camera in a three-dimensional world coordinate system, and calculating to obtain a first two-three column of a world coordinate matrix of the camera according to the Euler angles; and obtaining the fourth column of the world coordinate matrix of the camera according to the position information of the camera in the three-dimensional world coordinate system.
According to the embodiment of the invention, the second position point (actual position point) of the target object in the camera coordinate system is converted into the position point of the target object in the three-dimensional world coordinate system by utilizing the Euler angle determined by the visual angle direction of the camera in the three-dimensional world coordinate system and the world coordinate matrix obtained by combining the position information of the camera in the three-dimensional world coordinate system, and the position of the target object (constructor) is drawn in the BIM three-dimensional model, so that the positioning process of the target object is simplified, a plurality of positioning base stations are not required, and the target object can be simply, conveniently and accurately positioned without wearing a positioning device by only collecting video images and identifying two-dimensional or three-dimensional position coordinates of the target object in different coordinate systems and simple coordinate conversion. The positioning effect of the invention is not influenced by the action of a target object, namely constructors, and is more robust and reliable.
The following three-dimensional positioning method/device for the target object is specifically described by combining data:
the three-dimensional positioning method and the three-dimensional positioning device of the target object acquire two-dimensional position information of personnel in an image or a video based on intelligent analysis of a monitoring video of a construction site; then converting the two-dimensional position information of the personnel into three-dimensional position information through the position and the visual angle of the monitoring video corresponding to the camera in the BIM three-dimensional model; finally, the positioning of the constructor in the world coordinate system in the BIM three-dimensional model is realized, the main steps of which are shown in fig. 6 (fig. 6 can be understood as a working flow chart of the three-dimensional positioning method of the target object of the invention, and can also be understood as a working schematic diagram of the three-dimensional positioning device of the target object of the invention), taking the target object as an example of the constructor:
(1) Deploying a monitoring camera at a construction site;
(2) Building site BIM three-dimensional modeling is carried out, and a BIM three-dimensional model is built so as to obtain detailed building site three-dimensional model information; identifying three-dimensional world coordinates Z (xz, yz, zz) of the camera in the BIM three-dimensional model;
(3) Identifying constructors in images or videos in monitoring cameras deployed in a building site by utilizing existing mature image or video intelligent analysis technologies such as machine learning and the like, and obtaining two-dimensional image coordinates of two-dimensional images of constructors, namely first position point coordinates P 0 ；
(4) Mapping calculation to obtain BIM three-dimensional world coordinate P of constructor 3 Specifically, the method can be divided into three steps, namely, the two-dimensional coordinate of the constructor on the near projection surface of the camera, namely, the second position point coordinate P, is calculated firstly 1 Then, a third position point P which is a three-dimensional coordinate of the constructor in a camera coordinate system taking the camera position as an origin coordinate is calculated 2 Then calculate the position of constructor in three-dimensional world coordinate system, namely the fourth position point coordinate P 3 ；
(5) According to P 3 (x 3 ,y 3 ,z 3 ) The three-dimensional positioning of the constructor is displayed in the BIM three-dimensional model as shown in fig. 7.
According to the embodiment, the specific calculation process of three-dimensional positioning of constructors in the BIM three-dimensional model is described by the enumerated data:
specifically, as shown in FIG. 5, in a computer three-dimensional world, the camera has a cone range with a near projection surface and a far projection surface, and only the object between the surfaces can be movedSeen by the camera, then from the camera and P on its near projection surface 1 The point is connected with a straight line L, the straight line L is prolonged, and the information between the near projection plane and the far projection plane on the straight line L is projected to P on the near projection plane 1 In the point, all the points projected on the near projection surface are combined into one image (negative film), which is the content shot by the camera.
Assuming that the position point z= (643.07, 21.54, 196.36) and the viewing angle direction r= (-1, 0, -0.5) of the camera in the three-dimensional world coordinates are further calculated to have euler angles {2.356, -0.615,2.618}, the matrix of the camera coordinate system after rotational translation is derived via the preset matrix formula (1) as described above to beThus, the fourth column and the last row are added, thus deriving the final world matrix M as +.>
Assume that a first position point P is obtained after AI intelligent analysis is performed on video images acquired by a camera 0 (1473.5，561)，W 0 ＝1494-1453＝41，H 0 The first position point P is calculated by calculating the video camera frame size of 1920 x 1080 with the range of 586-536=50 0 The position point projected by the camera near the projection plane in the three-dimensional world is the second position point P 1 (x 1 ,y 1 )，
x 1 ＝1473.5/1920*2-1＝0.534896，
y 1 ＝-1*561/1080*2+1＝-0.038889。
A straight line L (x L ,y L ,z L )：
x L ＝0.534896*tan(45*Math.PI/180/2)*1＝0.221561，
y L ＝-0.038889*tan(45*Math.PI/180/2)*1080/1920*1＝-0.009061，
z L ＝-1。
As shown in fig. 4, assuming that the distance from the camera position to the center point of the near projection plane is 1, the length of half of the width of the near projection plane on the X-axis is tan (45 °/2) ×1, and 45 °/2 is converted into radian, that is, tan (45×math.pi/180/2) ×1, the field angle of the human eye is usually 124 degrees, so that the human eye cannot see the object behind the brain, and the camera also has the field angle, that is, 45 degrees, that is, the range that can be displayed in the lateral direction of the image.
Pi is pi in mathematics, i.e., 3.1415926,360 degrees is 2 pi, and math pi/180 converts angle into radian.
According to the aspect ratio W of the head of the worker in the image 0 ，H 0 The distance D of the head of the worker from the camera position is calculated, d=1200/(41+50) = 13.186813, and the constant λ is 1200, and the value of λ is not particularly limited in the present invention.
The straight line L is scaled by D times to obtain a third position point P which is a position point of the head of the worker in a coordinate system taking the camera as an origin 2 (x 2 ,y 2 , z2 )，
x 2 ＝0.221561*13.186813＝2.921685，
y 2 ＝-0.009061*13.186813＝-0.119485，
z 2 ＝-1*13.186813＝-13.186813，
World matrix M multiplied by point P 2 Obtaining the position of the head of the constructor in the three-dimensional world coordinate system, namely a fourth position point P 3 ：
Namely P 3 ＝(631.39,24.46,190.32)。
Personnel locations are plotted in a BIM three-dimensional model as shown in fig. 7.
It should be noted that the above-described embodiments will enable those skilled in the art to more fully understand the invention, but do not limit it in any way. Therefore, although the present invention has been described in detail with reference to the drawings and examples, it will be understood by those skilled in the art that the present invention may be modified or equivalent, and in all cases, all technical solutions and modifications which do not depart from the spirit and scope of the present invention are intended to be included in the scope of the present invention.
Claims (4)
1. A method for three-dimensional localization of objects for localization of objects at a construction site, the method comprising the steps of:
deploying a monitoring camera on a construction site or utilizing an existing monitoring camera on the construction site and carrying out BIM modeling on the construction site to obtain a BIM three-dimensional model of the construction site, obtaining position information and view angle directions of the camera in a three-dimensional world coordinate system of the BIM three-dimensional model, and calculating a world coordinate matrix of the camera according to the position information and the view angle directions;
carrying out object identification in an image or video shot by a camera, and acquiring a two-dimensional coordinate of a head center point of an object in the image or video to serve as a first position point;
mapping calculation is carried out according to the first position point and by combining with the horizontal pixel value and the vertical pixel value of the camera, so that the two-dimensional coordinates of the head center point of the target object projected on the near projection surface of the camera in the camera coordinate system are obtained to serve as second position points;
calculating three-dimensional coordinates in a camera coordinate system taking a camera as an origin as a third position point according to the second position point; the calculating the three-dimensional coordinate in the camera coordinate system with the camera as the origin according to the second position point to be used as a third position point comprises the following steps: making a straight line from the camera position to the second position point by taking a camera as an origin and calculating the straight line direction, wherein the straight line direction comprises an X-axis coordinate of the straight line calculated according to the tangent value of the second position point and a half view angle of the camera, a Y-axis coordinate of the straight line calculated according to the second position point and a horizontal and vertical pixel value and a Z-axis coordinate determined by the default of the camera looking to the Z-axis negative direction in a camera coordinate system; calculating the distance between the head of the target object and the camera according to the aspect ratio of the head area of the target object in the image or the video image; scaling the straight line according to the distance between the head of the target object and the camera obtained by calculation according to the principle of imaging near-far and far of the camera to determine the third position point;
And performing product transformation processing on the third position point and the world coordinate matrix to obtain a position point of the target object in a three-dimensional world coordinate system of the BIM three-dimensional model as a fourth position point, and displaying three-dimensional positioning of the target object in the BIM three-dimensional model according to the fourth position point.
2. The method of claim 1, wherein said calculating a world coordinate matrix of the camera from the location information and the view direction comprises: determining Euler angles of the camera, which need to rotate around three coordinate axes respectively, according to the visual angle direction of the camera in a three-dimensional world coordinate system, and calculating to obtain a first two-three column of a world coordinate matrix of the camera according to the Euler angles; and obtaining a fourth column of the world coordinate matrix of the camera according to the position information of the camera in the three-dimensional world coordinate system.
3. A three-dimensional object positioning device for positioning an object at a construction site, comprising:
the information acquisition module is used for acquiring position information and view angle directions of a camera deployed on a construction site in a three-dimensional world coordinate system of a construction site BIM three-dimensional model obtained by BIM modeling the construction site; target object identification is carried out in an image or video shot by a camera, and a two-dimensional coordinate of a head center point of a target object in the image or video is obtained by utilizing an image or video intelligent analysis technology to serve as a first position point;
The calculation module is used for calculating a world coordinate matrix of the camera according to the position information and the view angle direction; mapping calculation is carried out according to the first position point and by combining with the horizontal pixel value and the vertical pixel value of the camera, so that the two-dimensional coordinates of the head center point of the target object projected on the near projection surface of the camera in the camera coordinate system are obtained to serve as second position points; calculating three-dimensional coordinates in a camera coordinate system taking a video camera as an origin according to the second position point to serve as a third position point, wherein the three-dimensional coordinates specifically comprise: making a straight line from the camera position to the second position point by taking a camera as an origin, and calculating the straight line direction, wherein the straight line direction comprises an X-axis coordinate of the straight line calculated according to the tangent value of the second position point and a half view angle of the camera, a Y-axis coordinate of the straight line calculated according to the second position point and a transverse-longitudinal pixel ratio and a Z-axis coordinate determined by the default of the camera looking to the Z-axis negative direction in a camera coordinate system; calculating the distance between the head of the target object and the camera according to the aspect ratio of the head area of the target object in the image or the video image; scaling the straight line according to the distance between the head of the target object and the camera obtained by calculation according to the principle of imaging near-far and far of the camera to determine the third position point; performing product transformation processing on the third position point and the world coordinate matrix to obtain a position point of the target object in a three-dimensional world coordinate system of the BIM three-dimensional model as a fourth position point;
And the positioning display module is used for displaying the three-dimensional positioning of the target object in the BIM three-dimensional model according to the fourth position point.
4. The apparatus of claim 3, wherein the computing module to compute the world coordinate matrix of the camera based on the location information and the view direction comprises: determining Euler angles of the camera, which need to rotate around three coordinate axes respectively, according to the visual angle direction of the camera in a three-dimensional world coordinate system, and calculating to obtain a first two-three column of a world coordinate matrix of the camera according to the Euler angles; and obtaining a fourth column of the world coordinate matrix of the camera according to the position information of the camera in the three-dimensional world coordinate system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911404383.8A CN111192321B (en) | 2019-12-31 | 2019-12-31 | Target three-dimensional positioning method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911404383.8A CN111192321B (en) | 2019-12-31 | 2019-12-31 | Target three-dimensional positioning method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111192321A CN111192321A (en) | 2020-05-22 |
| CN111192321B true CN111192321B (en) | 2023-09-22 |
Family
ID=70709704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911404383.8A Active CN111192321B (en) | 2019-12-31 | 2019-12-31 | Target three-dimensional positioning method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111192321B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112199754B (en) * | 2020-10-30 | 2023-05-09 | 久瓴(江苏)数字智能科技有限公司 | Coordinate positioning method and device, storage medium and electronic equipment |
| CN112652015B (en) * | 2020-11-30 | 2023-05-09 | 中国公路工程咨询集团有限公司 | BIM-based pavement disease marking method and device |
| CN112685812A (en) * | 2020-12-16 | 2021-04-20 | 中冶建筑研究总院有限公司 | Dynamic supervision method, device, equipment and storage medium |
| CN113329181B (en) * | 2021-06-08 | 2022-06-14 | 厦门四信通信科技有限公司 | Angle switching method, device, equipment and storage medium of camera |
| CN114820924A (en) * | 2022-03-24 | 2022-07-29 | 香港大学深圳研究院 | Method and system for analyzing museum visit based on BIM and video monitoring |
| CN114998768B (en) * | 2022-06-02 | 2023-04-04 | 广州市港航工程研究所 | Intelligent construction site management system and method based on unmanned aerial vehicle |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107248193A (en) * | 2017-05-22 | 2017-10-13 | 北京红马传媒文化发展有限公司 | The method, system and device that two dimensional surface is switched over virtual reality scenario |
| CN109272570A (en) * | 2018-08-16 | 2019-01-25 | 合肥工业大学 | A kind of spatial point three-dimensional coordinate method for solving based on stereoscopic vision mathematical model |
| CN109801301A (en) * | 2018-12-15 | 2019-05-24 | 华南理工大学 | A kind of automatic collection method of tile work progress msg based on BIM and computer vision |
| DE102018010252A1 (en) * | 2018-02-14 | 2019-08-14 | Gesellschaft zur Förderung angewandter Informatik eV | Method and system for dynamic structural analysis |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2542762B (en) * | 2015-09-21 | 2018-11-21 | Imetrum Ltd | Measuring device and method |
| JP6236600B1 (en) * | 2017-06-02 | 2017-11-29 | 株式会社Ｇｐｒｏ | Flight parameter measuring apparatus and flight parameter measuring method |
| US10282913B2 (en) * | 2017-07-24 | 2019-05-07 | Visom Technology, Inc. | Markerless augmented reality (AR) system |
| CN108648237B (en) * | 2018-03-16 | 2022-05-03 | 中国科学院信息工程研究所 | Space positioning method based on vision |
-
2019
- 2019-12-31 CN CN201911404383.8A patent/CN111192321B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107248193A (en) * | 2017-05-22 | 2017-10-13 | 北京红马传媒文化发展有限公司 | The method, system and device that two dimensional surface is switched over virtual reality scenario |
| DE102018010252A1 (en) * | 2018-02-14 | 2019-08-14 | Gesellschaft zur Förderung angewandter Informatik eV | Method and system for dynamic structural analysis |
| CN109272570A (en) * | 2018-08-16 | 2019-01-25 | 合肥工业大学 | A kind of spatial point three-dimensional coordinate method for solving based on stereoscopic vision mathematical model |
| CN109801301A (en) * | 2018-12-15 | 2019-05-24 | 华南理工大学 | A kind of automatic collection method of tile work progress msg based on BIM and computer vision |
Non-Patent Citations (2)
| Title |
|---|
| 朱志鹏.基于结构光三维空间测量方法的研究.《中国优秀硕士学位论文全文数据库信息科技辑》.2018,第1-62页. * |
| 洪灏 等.基于BIM的单目相机室内定位技术.《第三届全国 BIM 学术会议论文集》.2017,第140-145页. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111192321A (en) | 2020-05-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111192321B (en) | Target three-dimensional positioning method and device | |
| CN111031897B (en) | System and method for analyzing skin condition | |
| US10068344B2 (en) | Method and system for 3D capture based on structure from motion with simplified pose detection | |
| CN110599540B (en) | Real-time three-dimensional human body shape and posture reconstruction method and device under multi-viewpoint camera | |
| Zollmann et al. | Augmented reality for construction site monitoring and documentation | |
| Golparvar-Fard et al. | Evaluation of image-based modeling and laser scanning accuracy for emerging automated performance monitoring techniques | |
| US8848035B2 (en) | Device for generating three dimensional surface models of moving objects | |
| CN108810473B (en) | Method and system for realizing GPS mapping camera picture coordinate on mobile platform | |
| WO2023093217A1 (en) | Data labeling method and apparatus, and computer device, storage medium and program | |
| CN112686877B (en) | Binocular camera-based three-dimensional house damage model construction and measurement method and system | |
| CN115597659B (en) | Intelligent safety management and control method for transformer substation | |
| Hübner et al. | Marker-based localization of the microsoft hololens in building models | |
| IE86364B1 (en) | Closed loop 3D video scanner for generation of textured 3D point cloud | |
| JP6610640B2 (en) | Position recognition method and system, and abnormality determination method and system | |
| JP5079547B2 (en) | Camera calibration apparatus and camera calibration method | |
| CN102970514A (en) | Apparatus, method, and program for video surveillance system | |
| US20180020203A1 (en) | Information processing apparatus, method for panoramic image display, and non-transitory computer-readable storage medium | |
| CN112802208B (en) | Three-dimensional visualization method and device in terminal building | |
| Muffert et al. | The estimation of spatial positions by using an omnidirectional camera system | |
| CN108234932B (en) | Method and device for extracting personnel form in video monitoring image | |
| US8294814B2 (en) | System for accurately repositioning imaging devices | |
| CN111105467A (en) | Image calibration method and device and electronic equipment | |
| CN107274449B (en) | Space positioning system and method for object by optical photo | |
| JP2020135737A (en) | Gaze action search system, and control program | |
| US20240005556A1 (en) | System and method for assessing imaged object location |
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 |