Proximity photogrammetry device and method based on visual reference positioning

Abstract

The invention discloses a proximity photogrammetry device and method based on visual reference positioning, and relates to the technical field of photogrammetry, and the device comprises a first device used for photographing, a second device used for driving the first device to rotate in multiple dimensions, a third device used for cooperating with power supply communication, and a fourth device used for main power supply communication. The first device comprises a regular polyhedron, one surface of the regular polyhedron is reserved for mounting the second device, a camera is fixed on each surface of the rest regular polyhedron, two non-adjacent surfaces on the regular polyhedron are communicated, and the communicated space is a cylinder; the first battery occupies the through space, and the connecting parts of the second device and the fourth device are connected; according to the method, close photogrammetry of the ground or the object surface can be effectively carried out in an environment without RTK positioning signals or with weak RTK positioning signals and weak textures, and fine geographic information models such as digital orthoimages and three-dimensional models are generated.

Description

Technical Field

[0001] This invention belongs to the field of photogrammetry, specifically a close-up photogrammetry device and method based on visual reference positioning. Background Technology

[0002] Close-up photogrammetry is an object-oriented form of photogrammetry. It uses the "surface" of an object as the photographic object, employing rotary-wing drones to acquire ultra-high-resolution images for detailed geographic information extraction. Therefore, it can highly reproduce the fine structure of the earth's surface and objects. Currently, close-up photogrammetry technology is supported by two main components: first, the drone's gimbal attitude control capability, which controls the camera's pitch and yaw angles; and second, high-precision drone positioning technology, achieving centimeter-level positioning accuracy through integrated RTK (Real-time Kinematic) technology. RTK is a method for real-time processing of the carrier phase observations from two measurement stations. It sends the carrier phase data acquired by the base station to the user receiver for differential calculation of coordinates. This is a new and commonly used satellite positioning measurement method. Therefore, conventional close-up photogrammetry techniques rely on the high-precision RTK positioning technology of rotary-wing drones and are not suitable for drone photogrammetry in environments with no or weak RTK positioning signals.

[0003] AprilTag is a visual benchmark library widely used in AR, robotics, and camera calibration. It enables rapid marker detection and calculation of relative positions using specific markers (similar to QR codes, but with reduced complexity to meet real-time requirements).

[0004] AprilTag content mainly consists of three parts:

[0005] The first part describes how to detect various edges in an image based on gradients;

[0006] The second part is how to find and filter the required quadrilateral pattern in the edge image. AprilTag detects the detected edges as much as possible, first eliminating non-straight edges, then searching for adjacent edges on the straight edges, and finally forming a closed loop to detect a quadrilateral.

[0007] The third part is about how to encode and decode QR codes. There are usually three encoding methods: Tag36h11, Tag25h9, and Tag16h5. One code element corresponds to one color block. The length of the outermost black border color block for the three encoding methods is 8, 7, and 6 color block lengths, respectively. The total number of color blocks corresponding to all code elements for the three encoding methods is 6×6, 5×5, and 4×4, respectively. For decoding, a dot array is generated within the detected quadrilateral to calculate the value of each color block. Then, a simple classifier is constructed based on Local Binary Patterns to classify the color blocks within the quadrilateral. Positive color blocks are encoded as 1, and negative color blocks are encoded as 0. This gives the QR code encoding. After obtaining the encoding, it is matched with the encoding in the known database to determine whether the decoded QR code is correct. Summary of the Invention

[0008] To address the challenge of effectively performing close-up photogrammetry of ground or object surfaces in environments with no or weak RTK positioning signals and weak texture, this invention aims to provide the following technical solution:

[0009] A close-up photogrammetry device based on visual reference positioning includes a first device 1 for taking pictures, a second device for driving the first device to rotate in multiple dimensions, a third device for cooperating with power supply and communication, and a fourth device for main power supply and communication.

[0010] The first device includes a regular polyhedron, a camera unit consisting of multiple cameras, a first battery and a control and communication module, fixing components and connecting components. One face of the regular polyhedron is reserved for installation with the second device. Each of the remaining faces of the regular polyhedron is fixed with a camera by a camera nut, a camera washer and a camera bolt. Two non-adjacent faces of the regular polyhedron are connected to form a through space.

[0011] The second device includes a three-axis rotating and fixing component and multiple connecting rods;

[0012] The third device includes a second battery and a second control and communication module;

[0013] The fourth device includes a first battery, a first control and communication module, a fixing component, and a connecting component. The fourth device is installed on one of two non-adjacent surfaces at the through space location via a camera nut, a module washer, and a module bolt. The first battery occupies the through space, and the second device is connected to the connecting component of the fourth device.

[0014] As a further aspect of the present invention, the regular polyhedron is a regular icosahedron. Three camera nuts are fixed to each of the nineteen faces of the regular icosahedron of the first device. The nineteen cameras of the first device are respectively fixed to the three camera nuts by a camera gasket and three camera bolts. Each of the nineteen faces of the regular icosahedron has a number or letter marking to represent the camera code. The camera gasket has three through holes and a direction marking. The diameter of the through holes is larger than the diameter of the camera bolts. Three adjacent vertices of the regular icosahedron form a triangular plane. The camera is fixed on the triangular plane of the regular icosahedron, and the camera's image plane is parallel to the triangular plane. The line connecting the center of the camera's image to the center of the triangle is perpendicular to the triangular plane. The center of the camera's image is the camera center or the camera image center. The center of the regular icosahedron is taken as the origin of the local coordinate system of the regular icosahedron. A vertex of the regular icosahedron is taken as a point on the Z-axis of the local coordinate system to determine the Z-axis direction. A plane passing through the origin and perpendicular to the Z-axis intersects the edge of the regular icosahedron to obtain corresponding intersection points. One of these intersection points is selected as the Y-axis or X-axis. The points on the icosahedron determine the Y-axis or X-axis direction of the local coordinate system. Another axis direction, the X-axis or Y-axis direction of the local coordinate system, is determined using a left-handed or right-handed coordinate system. This establishes the origin and axis vectors of the local coordinate system. Within this system, the local coordinates of all vertices of the icosahedron, the local coordinates of the centers of all triangles formed by adjacent vertices, and the local coordinates of the nineteen camera image centers are determined. The origin is the center of a triangle on the icosahedron, and three vectors are formed through three vertices. (The last two lines appear to be unrelated and possibly machine-translated gibberish.) The fifth device, consisting of a shim and a direction marker, has a vector corresponding to the vertex closest to the direction marker, which is the Y-axis or X-axis vector of the camera plane. The vector formed by the two vertices farther from the direction marker is the other axis vector of the camera plane, namely the X-axis or Y-axis direction of the camera plane. The camera Z-axis vector is obtained by calculating the vector product of the Y and X-axis vectors of the camera plane, which is the camera direction vector. In summary, the center coordinates, camera plane axis vectors, and direction vectors of the nineteen camera images in the local coordinate system of the icosahedron are calculated.

[0015] As a further embodiment of the present invention, the connecting rod of the second device is composed of a fixed rod and a tightening sleeve. The fixed rod is composed of a central cylindrical rod and two fixed terminals. The fixed terminals have three mating columns. The inner diameter of the tightening sleeve is larger than the diameter of the fixed terminals. The diameter of the central through hole of the cylindrical rod fixedly connected to the tightening sleeve is smaller than the diameter of the fixed terminals. The inner wall of the tightening sleeve is threaded. The tightening sleeve is used to lock the connecting component of the fourth device or the first or second component of the three-axis rotating fixing component.

[0016] As a further embodiment of the present invention, the connecting component of the fourth device has three or more mating column grooves on one bottom surface. The diameter of the mating column grooves is larger than the diameter of the mating column of the fixing terminal of the fixing rod of the connecting rod of the second device. The mating column can be inserted into the mating column groove. The camera nut of the first device is fixedly connected to the fourth device. In the axial direction of the fixing rod, the rotation angle between the mating column and the mating column groove determines the rotation angle between the first device and the second device. The connecting component of the fourth device has threads on its side. The threads match the threads on the inner wall of the tightening sleeve of the connecting rod of the second device. The tightening sleeve is tightened and fixed to the connecting component of the fourth device through the threads.

[0017] As a further embodiment of the present invention, the three-axis rotating fixing component of the second device is composed of a first component, a second component, and a third component. The second component consists of a connecting column and a connecting body with four or more mating holes. The first component consists of a connecting column and a connecting body with four or more mating columns. The diameter of the mating holes of the connecting body with mating holes is larger than the diameter of the mating columns of the connecting body with mating columns. The height of the middle mating column of the connecting body is greater than the height of the surrounding mating columns. The connecting bodies with mating holes and the connecting bodies with mating columns overlap. The middle mating column of the connecting body protrudes above the connecting body with mating holes. The third component is a tightening column with a column groove. The middle column groove of the tightening column has threads that match the threads of the middle mating column. Tightening is achieved through the two threads. The first component, the second component, and the third component are connected by the middle mating column. The first component is fixed along its centerline. The mating column of the first component rotates in the centerline and inserts into the mating hole of the second component. The bottom surface of the connecting column of the first component and the second component has three or more mating column grooves. The diameter of the mating column groove is larger than the diameter of the mating column of the fixing terminal of the fixing rod of the connecting rod of the second device. The mating column can be inserted into the mating column groove. In the axial direction of the first or second component of the three-axis rotating fixing part of the second device, the rotation angle between the mating column and the mating column groove determines the rotation angle between the first or second component of the three-axis rotating fixing part and the fixing rod of the connecting rod of the second device. The side of the connecting column has threads. The threads match the threads on the inner wall of the tightening sleeve of the connecting rod. By tightening through the two threads, in the centerline direction of the connecting column, the tightening sleeve of the connecting rod is fixed to the connecting column of the first or second component of the three-axis rotating fixing part.

[0018] As a further embodiment of the present invention, the fourth device includes a first battery and a first control and communication module. The first battery supplies power to the nineteen cameras of the icosahedron and the first control and communication module. The first control and communication module includes a control processing module and a wireless data transmission or image transmission communication module. The control processing module includes functions for controlling the shooting function of the nineteen cameras and acquiring the camera images. The wireless data transmission or image transmission communication module includes functions for sending the camera images to the second control and communication module of the third device and receiving control commands from the second control and communication module.

[0019] As a further embodiment of the present invention, the third device includes a second battery and a second control and communication module. The second battery supplies power to the second control and communication module. The second control and communication module includes a human-computer interaction module, a wireless data transmission or image transmission module, and a control processing module. The human-computer interaction module includes functions for user input parameters and displaying nineteen camera images of the icosahedron. The wireless data transmission or image transmission module includes functions for receiving camera images from the first device and sending control commands to the first device. The control processing module includes functions for calculating and saving the center coordinates of the camera images in the measurement space coordinate system, as well as the camera plane axis vector and direction vector.

[0020] Another object of the present invention is to provide:

[0021] A close-up photogrammetry method based on visual reference positioning specifically includes the following steps:

[0022] S1: Establish a measurement space coordinate system, dividing the measurement space into one or more measurement subspaces; under the icosahedral local coordinates, calculate the center coordinates, camera plane axis vectors, and direction vectors of the nineteen camera images of the first device; using the center coordinates of each of the nineteen camera images as the origin, the camera plane axis vector as the X-axis or Y-axis, and the camera direction vector as the Z-axis, establish corresponding camera local coordinate systems, i.e., establish a total of nineteen camera local coordinate systems; calculate the three rotation angles and origin offsets between any two camera local coordinate systems in the nineteen camera local coordinate systems, i.e., calculate a total of 19×18 sets of the three rotation angles and origin offsets; fix the connection between the first device, the fourth device, and the second device.

[0023] S2: Perform close-up photogrammetry on each measurement subspace, and arrange one or more AprilTag codes in the space outside the current measurement subspace; measure the center position and orientation of the AprilTag code in the measurement space coordinate system; the second device utilizes the rotation and fixing function of the three-axis rotation fixing component to assemble a second device including one or more connecting rods, wherein the rotation angles between the connecting rods are the same or different, forming second devices of different shapes to adapt to measurement subspaces with different spatial characteristics; perform close-up photogrammetry by holding one end of the second device, or fix one end of the second device to a drone and perform close-up photogrammetry by controlling the drone;

[0024] S3: The first device is in different positions or different postures in the current measurement subspace. Each camera on the first device at the current position and posture captures and generates a camera image. The nineteen cameras of the first device capture and generate nineteen camera images. The first control communication module of the fourth device sends the camera images to the second control communication module of the third device.

[0025] S4: The second control communication module receives nineteen camera images of the current position and attitude in the current measurement subspace, calculates and saves the center coordinates, camera plane axis vector and direction vector of the nineteen camera images in the measurement space coordinate system;

[0026] S5: For all positions or all postures in the current measurement subspace, if the device completes the calculation and saving of the center coordinates, camera plane axis vectors and direction vectors of the close-up photography and the nineteen camera images in the measurement space coordinate system, then proceed to step S6; otherwise, the first device moves to the next different position or is in the next different posture in the current measurement subspace, and proceeds to step S3.

[0027] S6: If the measurement subspace has not completed the close-up photography and the calculation of camera image parameters in the measurement space coordinate system, then proceed to step S2. Otherwise, use the camera images in all measurement subspaces that do not contain the AprilTag code and the center coordinates, camera plane axis vector and direction vector of the camera images in the measurement space coordinate system to perform close-up photography measurement calculations, and generate a detailed digital orthophoto map and a geographic information model of the three-dimensional model.

[0028] Step S4 specifically includes the following steps:

[0029] Step 1: If no one or more images containing the AprilTag code are calculated and identified from the camera images, the first device moves to the next different position or is in the next different pose in the current measurement subspace and proceeds to step S3; otherwise, proceed to step 2.

[0030] Step 2: For one or more camera images containing the AprilTag code, perform the following calculations: If one or two AprilTag codes are identified in the camera image, based on the position and orientation of the AprilTag code in the measurement space, obtain the measurement space positions of the four vertices of the AprilTag code. Based on the image coordinates of the four vertices and the measurement space positions, use spatial resection to calculate the center coordinates and orientation of the camera image in the measurement space coordinate system. If three or more AprilTag codes are identified in the camera image, obtain the measurement space positions of the centers of the three or more AprilTag codes. Combine this with the image coordinates of the AprilTag code centers, use spatial resection to calculate the center coordinates and orientation of the camera image in the measurement space coordinate system. Through these calculations, the center coordinates and orientation of one or more camera images in the measurement space coordinate system are obtained.

[0031] Step 3: There are two methods to calculate the center coordinates, camera plane axis vectors, and direction vectors of the nineteen camera images in the measurement space coordinate system. You can choose one of these methods for calculation. The two calculation methods are as follows:

[0032] The first calculation method is as follows:

[0033] Based on the center coordinates and orientation of one or more camera images in the measurement space coordinate system, select the center coordinates and orientation of one camera image in the measurement space coordinate system, calculate the origin, X-axis vector, Y-axis vector, and Z-axis vector of one of the camera local coordinate systems in the measurement space coordinate system, and based on the three rotation angles and origin offset between any two camera local coordinate systems calculated in step S, use the absolute orientation relationship between the two coordinate systems to calculate the origin, X-axis vector, Y-axis vector, and Z-axis vector of eighteen camera local coordinate systems (excluding one of the cameras) in the measurement space coordinate system, and obtain and save the center coordinates, camera plane axis vectors, and direction vectors of the nineteen camera images in the measurement space coordinate system.

[0034] The second calculation method is as follows:

[0035] Based on the center coordinates and pose of the images from one or more cameras in the measurement space coordinate system, the origin, X-axis vector, Y-axis vector, and Z-axis vector of the local coordinate system of one or more cameras in the measurement space coordinate system are calculated. Based on the geometric relationship between the origin of the icosahedral local coordinate system and the local coordinate systems of one or more cameras, the origin coordinates of the one or more icosahedral local coordinate systems in the measurement space coordinate system are calculated. The average of these origin coordinates is the origin coordinate of the icosahedral local coordinate system in the measurement space coordinate system. Based on the icosahedral local coordinate system and the measurement space coordinate system, the scaling factor of the model scale of the two coordinate systems is calculated, and the origin of the two coordinate systems is calculated. The translation of coordinates; based on the center coordinates of one or more camera images in the measurement space coordinate system and their center coordinates in the icosahedral local coordinate system, the three rotation angles between the two coordinate systems are calculated using the analytical absolute orientation method; based on the scaling factor of the model scale of the two coordinate systems, the translation of the origin coordinates of the two coordinate systems, and the three rotation angles between the two coordinate systems, the center coordinates of the nineteen camera images in the icosahedral local coordinate system, and the camera plane axis vector and direction vector are substituted into the absolute orientation relationship of the two coordinate systems to calculate and save the center coordinates, camera plane axis vector, and direction vector of the nineteen camera images in the measurement space coordinate system.

[0036] Compared with the prior art, the beneficial effects of the present invention are:

[0037] By employing the above methods, the present invention has the following advantages compared with the prior art: The present invention provides a close-up photogrammetry device and method based on visual reference positioning, which can effectively perform close-up photogrammetry of the ground or object surface in environments with no RTK positioning signal or weak RTK positioning signal and weak texture, and generate detailed digital orthophoto maps and three-dimensional models and other geographic information models. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of a close-up photogrammetry device based on visual reference positioning.

[0039] Figure 2 This is a schematic diagram of the structure of the first device;

[0040] Figure 3 This is a schematic diagram of the structure of a regular icosahedron;

[0041] Figure 4 for Figure 3 Schematic diagram of the structure after adding the nut;

[0042] Figure 5 This is a schematic diagram of the camera unit structure;

[0043] Figure 6 This is a schematic diagram of the fifth device;

[0044] Figure 7 This is a schematic diagram of the fourth device;

[0045] Figure 8 This is a schematic diagram of the connecting rod structure;

[0046] Figure 9 for Figure 8 Cross-sectional view along CC;

[0047] Figure 10 This is a structural diagram of the rotating fixed component;

[0048] Figure 11 for Figure 10 A partial exploded view.

[0049] In the diagram: First device 1, Second device 2, Third device 3, Camera unit 5, Fourth device 6, Module gasket 60, Connecting component 61, Module bolt 62, First control and communication module 63, First battery 64, Icosahedron 7, Nut 70, Fifth device 8, Camera gasket 80, Direction mark 81, Camera 82, Camera bolt 83, Connecting rod 9, Fixing rod 90, Tightening sleeve 91, Rotating fixing component 10, First component 11, Second component 12, Third component 13. Detailed Implementation

[0050] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Preferred embodiments of the invention are shown in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a thorough and complete understanding of the disclosure of the invention.

[0051] Furthermore, elements in this invention are referred to as being "fixed to" or "set on" another element, which may be directly on the other element or may also include an intervening element. When an element is considered to be "connected" to another element, it may be directly connected to the other element or may also include an intervening element. The terms "vertical," "horizontal," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only possible implementations.

[0052] Please see the appendix Figures 1 to 11 In this embodiment of the invention, a close-up photogrammetry device based on visual reference positioning includes a first device 1 for taking pictures, a second device 2 for driving the first device to rotate in multiple dimensions, a third device 3 for cooperating with power supply and communication, and a fourth device 6 for main power supply and communication.

[0053] The first device 1 includes a regular polyhedron, a camera unit 5 consisting of multiple cameras 82, a first battery 64, a control and communication module 63, fixing components, and connecting components. One face of the regular polyhedron is reserved for installation with the second device 2. Each of the remaining faces of the regular polyhedron is fixed with a camera 82 by a camera nut 70, a camera washer 80, and a camera bolt 83. Two non-adjacent faces of the regular polyhedron are connected to form a through space.

[0054] The second device 2 includes a three-axis rotating and fixing component 10 and multiple connecting rods 9;

[0055] The third device 3 includes a second battery and a second control and communication module;

[0056] The fourth device 6 includes a first battery 64, a first control and communication module 63, a fixing component, and a connecting component 61. The fourth device 6 is installed on one of two non-adjacent surfaces at the through space position via a camera nut 70, a module washer 60, and a module bolt 62. The first battery 64 occupies the through space, and the second device 2 is connected to the connecting component 61 of the fourth device 6.

[0057] As a further embodiment of the present invention, the regular polyhedron is a regular icosahedron 7. Three camera nuts 70 are fixed to each of the nineteen faces of the regular icosahedron 7 of the first device 1. The nineteen cameras 82 of the first device 1 are respectively fixed to the three camera nuts 70 via a camera washer 80 and three camera bolts 83. Each of the nineteen faces of the regular icosahedron 7 has a number or letter marking to represent the camera 82 code. The camera washer 80 has three through holes and a direction mark 81. The diameter of the through holes is larger than the diameter of the camera bolts 83. Three adjacent vertices of the regular icosahedron 7 form a... A triangular plane is formed, with the camera 82 fixed on the triangular plane of the regular icosahedron 7 and its photographic plane parallel to the triangular plane. The line connecting the center of the photographic plane of the camera 82 to the center of the triangle is perpendicular to the triangular plane. The center of the photographic plane of the camera 82 is the camera center or the camera image center. The center of the regular icosahedron 7 is taken as the origin of the local coordinate system of the regular icosahedron 7, and a vertex of the regular icosahedron 7 is taken as a point on the Z-axis of the local coordinate system to determine the direction of the Z-axis. The intersection point is obtained by the plane passing through the origin and perpendicular to the Z-axis intersecting the edge of the regular icosahedron 7. One of the intersection points is selected. A point is used as a point on the Y-axis or X-axis to determine the Y-axis or X-axis direction of the local coordinate system of the icosahedron. Another axis direction, namely the X-axis or Y-axis direction of the local coordinate system of the icosahedron, is determined through a left-handed or right-handed coordinate system. Thus, the origin and axis vectors of the local coordinate system of the icosahedron are determined. In this local coordinate system, the local coordinates of all vertices of the icosahedron 7, the local coordinates of the centers of all triangles formed by adjacent vertices, and the local coordinates of the image centers of the nineteen cameras 82 are determined. Taking the center of the triangle of the icosahedron 7 as the origin and forming three vectors through three vertices, the camera 82 and camera bolt 83 are... The fifth device 8, consisting of a camera pad 80 and a direction marker 81, has a vector corresponding to the vertex closest to the direction marker 81, which is the Y-axis or X-axis vector of the camera plane. The vector formed by the two vertices farther from the direction marker 81 is the other axis vector of the camera plane, namely the X-axis or Y-axis direction of the camera plane. The camera Z-axis vector is obtained by calculating the vector product of the Y and X-axis vectors of the camera plane, which is the camera direction vector. In summary, the center coordinates, camera plane axis vectors, and direction vectors of the nineteen camera images in the local coordinate system of the icosahedron are calculated.

[0058] As a further embodiment of the present invention, the connecting rod 9 of the second device 2 is composed of a fixed rod 90 and a tightening sleeve 91. The fixed rod 90 is composed of a central cylindrical rod and two fixed terminals. The fixed terminals have three mating columns. The inner diameter of the tightening sleeve 91 is larger than the diameter of the fixed terminals. The diameter of the central through hole of the cylindrical rod fixedly connected to the tightening sleeve 91 is smaller than the diameter of the fixed terminals. The inner wall of the tightening sleeve 91 is threaded. The tightening sleeve 91 is used to lock the connecting component 61 of the fourth device 6 or the first component 11 or the second component 12 of the three-axis rotating fixing component 10.

[0059] As a further embodiment of the present invention, the connecting component 61 of the fourth device 6 has three or more mating column grooves on one bottom surface. The diameter of the mating column grooves is larger than the diameter of the mating column of the fixing terminal of the fixing rod 90 of the connecting rod 9 of the second device 2. The mating column can be inserted into the mating column groove. The camera nut 70 of the first device 1 is fixedly connected to the fourth device 6. In the axial direction of the fixing rod, the rotation angle between the mating column and the mating column groove determines the rotation angle between the first device 1 and the second device 2. The connecting component 61 of the fourth device 6 has threads on its side. The threads match the threads on the inner wall of the tightening sleeve 91 of the connecting rod 9 of the second device 2. The tightening sleeve 91 is tightened and fixed to the connecting component 61 of the fourth device 6 through the threads.

[0060] As a further embodiment of the present invention, the three-axis rotating fixing component 10 of the second device 2 is composed of a first component 11, a second component 12, and a third component 13. The second component 12 is composed of a connecting column and a connecting body with four or more mating holes. The first component 11 is composed of a connecting column and a connecting body with four or more mating columns. The diameter of the mating holes of the connecting body with mating holes is larger than the diameter of the mating columns of the connecting body with mating columns. The height of the middle mating column of the connecting body is greater than the height of the surrounding mating columns. The connecting bodies with mating holes and the connecting bodies with mating columns overlap. The middle mating column of the connecting body protrudes above the connecting body with mating holes. The third component 13 is a tightening column with a column groove. The middle column groove of the tightening column has a thread. The thread matches the thread of the middle mating column, and tightening is achieved through the two threads. The first component 11, the second component 12, and the third component 13 are fixed along the centerline direction of the middle mating column. The first component 11's mating column rotates in the central direction and inserts into the mating hole of the second component 12. The bottom surface of the connecting column of the first component 11 and the second component 12 has three or more mating column grooves. The diameter of the mating column groove is larger than the diameter of the mating column of the fixing terminal of the fixing rod 90 of the connecting rod 9 of the second device 2. The mating column can be inserted into the mating column groove. In the axial direction of the first component 11 or the second component 12 of the three-axis rotating fixing part 10 of the second device 2, the rotation angle between the mating column and the mating column groove determines the rotation angle between the first component 11 or the second component 12 of the three-axis rotating fixing part 10 and the fixing rod of the connecting rod 9 of the second device 2. The side of the connecting column has threads, which match the threads on the inner wall of the tightening sleeve 91 of the connecting rod 9. By tightening through the two threads, in the central line direction of the connecting column, the tightening sleeve 91 of the connecting rod 9 is fixed to the connecting column of the first component 11 or the second component 12 of the three-axis rotating fixing part 10.

[0061] As a further embodiment of the present invention, the fourth device 6 includes a first battery 64 and a first control communication module 63. The first battery 63 supplies power to the nineteen cameras 82 of the icosahedron and the first control communication module 63. The first control communication module 63 includes a control processing module and a wireless data transmission or image transmission communication module. The control processing module includes functions for controlling the nineteen cameras 82 to take pictures and for acquiring the camera images. The wireless data transmission or image transmission communication module includes functions for sending the camera images to the second control communication module of the third device 3 and for receiving control commands from the second control communication module.

[0062] As a further embodiment of the present invention, the third device 3 includes a second battery and a second control and communication module. The second battery supplies power to the second control and communication module. The second control and communication module includes a human-computer interaction module, a wireless data transmission or image transmission module, and a control processing module. The human-computer interaction module includes functions for user input parameters and displaying images from the nineteen cameras 82 of the icosahedron. The wireless data transmission or image transmission module includes functions for receiving camera images from the first device 1 and sending control commands to the first device 1. The control processing module includes functions for calculating and saving the center coordinates of the camera images in the measurement space coordinate system and the camera plane axis vector and direction vector.

[0063] Another object of the present invention is to provide:

[0064] A close-up photogrammetry method based on visual reference positioning specifically includes the following steps:

[0065] S1: Establish a measurement space coordinate system, dividing the measurement space into one or more measurement subspaces; under the icosahedral local coordinates, calculate the center coordinates, camera plane axis vectors, and direction vectors of the nineteen camera images of the first device 1; using the center coordinates of each of the nineteen camera images as the origin, the camera plane axis vector as the X-axis or Y-axis, and the camera direction vector as the Z-axis, establish corresponding camera local coordinate systems, i.e., establish a total of nineteen camera local coordinate systems; calculate the three rotation angles and origin offsets between any two camera local coordinate systems in the nineteen camera local coordinate systems, i.e., calculate a total of 19×18 sets of the three rotation angles and origin offsets; fix the connection between the first device 1, the fourth device 6, and the second device 2.

[0066] S2: Perform close-up photogrammetry on each measurement subspace, and arrange one or more AprilTag codes in the space outside the current measurement subspace; measure the center position and orientation of the AprilTag code in the measurement space coordinate system; the second device 2 utilizes the rotation and fixing function of the three-axis rotation fixing component 10 to assemble a second device 2 including one or more connecting rods 9, wherein the rotation angles of the connecting rods 9 are the same or different, forming second devices 2 of different shapes to adapt to measurement subspaces with different spatial characteristics; perform close-up photogrammetry by holding one end of the second device 2, or fix one end of the second device 2 to a drone and perform close-up photogrammetry by controlling the drone;

[0067] S3: The first device 1 is in different positions or different postures in the current measurement subspace. Each camera on the first device 1 at the current position and posture captures and generates a camera image. The nineteen cameras 5 of the first device 1 capture and generate nineteen camera images. The first control communication module 63 of the fourth device 6 sends the camera images to the second control communication module of the third device 3.

[0068] S4: The second control communication module receives images from nineteen cameras 82 at the current position and orientation in the current measurement subspace, calculates and saves the center coordinates, camera plane axis vectors and direction vectors of the nineteen camera images in the measurement space coordinate system;

[0069] S5: For all positions or all postures in the current measurement subspace, if the device completes the calculation and saving of the center coordinates, camera plane axis vectors and direction vectors of the close-up photography and the images of the nineteen cameras 82 in the measurement space coordinate system, then proceed to step S6; otherwise, the first device moves to the next different position or is in the next different posture in the current measurement subspace, and proceeds to step S3.

[0070] S6: If the measurement subspace has not completed the close-up photography and the calculation of camera image parameters in the measurement space coordinate system, then proceed to step S2. Otherwise, use the camera images in all measurement subspaces that do not contain the AprilTag code and the center coordinates, camera plane axis vector and direction vector of the camera images in the measurement space coordinate system to perform close-up photography measurement calculations, and generate a detailed digital orthophoto map and a geographic information model of the three-dimensional model.

[0071] Step S4 specifically includes the following steps:

[0072] Step 1: If no one or more images containing the AprilTag code are calculated and identified from the camera images, the first device moves to the next different position or is in the next different pose in the current measurement subspace and proceeds to step S3; otherwise, proceed to step 2.

[0073] Step 2: For one or more camera images containing the AprilTag code, perform the following calculations: If one or two AprilTag codes are identified in the camera image, based on the position and orientation of the AprilTag code in the measurement space, obtain the measurement space positions of the four vertices of the AprilTag code. Based on the image coordinates of the four vertices and the measurement space positions, use spatial resection to calculate the center coordinates and orientation of the camera image in the measurement space coordinate system. If three or more AprilTag codes are identified in the camera image, obtain the measurement space positions of the centers of the three or more AprilTag codes. Combine this with the image coordinates of the AprilTag code centers, use spatial resection to calculate the center coordinates and orientation of the camera image in the measurement space coordinate system. Through these calculations, the center coordinates and orientation of one or more camera images in the measurement space coordinate system are obtained.

[0074] Step 3: There are two methods to calculate the center coordinates, camera plane axis vectors, and direction vectors of the 82 images from the 19 cameras in the measurement space coordinate system. You can choose one of these methods to calculate them. The two calculation methods are as follows:

[0075] The first calculation method is as follows:

[0076] Based on the center coordinates and orientation of one or more camera images in the measurement space coordinate system, select the center coordinates and orientation of one camera image in the measurement space coordinate system, calculate the origin, X-axis vector, Y-axis vector, and Z-axis vector of one of the camera local coordinate systems in the measurement space coordinate system, and based on the three rotation angles and origin offset between any two camera local coordinate systems calculated in step S1, use the absolute orientation relationship between the two coordinate systems to calculate the origin, X-axis vector, Y-axis vector, and Z-axis vector of eighteen camera local coordinate systems other than one of the cameras in the measurement space coordinate system, and obtain and save the center coordinates, camera plane axis vectors, and direction vectors of the nineteen camera images in the measurement space coordinate system.

[0077] The second calculation method is as follows:

[0078] Based on the center coordinates and pose of the images from one or more cameras in the measurement space coordinate system, the origin, X-axis vector, Y-axis vector, and Z-axis vector of the local coordinate system of one or more cameras in the measurement space coordinate system are calculated. Based on the geometric relationship between the origin of the icosahedral local coordinate system and the local coordinate systems of one or more cameras, the origin coordinates of the one or more icosahedral local coordinate systems in the measurement space coordinate system are calculated. The average of these origin coordinates is the origin coordinate of the icosahedral local coordinate system in the measurement space coordinate system. Based on the icosahedral local coordinate system and the measurement space coordinate system, the scaling factor of the model scale of the two coordinate systems is calculated, and the origin coordinates of the two coordinate systems are calculated. The translation amount; based on the center coordinates of one or more camera images in the measurement space coordinate system and their center coordinates in the icosahedral local coordinate system, the three rotation angles between the two coordinate systems are calculated by the analytical absolute orientation method; based on the scaling factor of the model scale of the two coordinate systems, the translation amount of the origin coordinates of the two coordinate systems, and the three rotation angles between the two coordinate systems, the center coordinates of the nineteen camera images in the icosahedral local coordinate system and the camera plane axis vector and direction vector are substituted into the absolute orientation relationship of the two coordinate systems, and the center coordinates, camera plane axis vector and direction vector of the nineteen camera images in the measurement space coordinate system are calculated and saved.

[0079] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Although this specification describes embodiments, not every embodiment contains only one technical solution. This method of description is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A close-up photogrammetry device based on visual reference positioning, comprising a first device (1) for taking pictures, a second device (2) for driving the first device to rotate in multiple dimensions, a third device (3) for cooperating with power supply and communication, and a fourth device (6) for main power supply and communication; The first device (1) includes a regular polyhedron, a camera unit (5) consisting of multiple cameras (82), a first battery (64), a control and communication module (63), a fixing component, and a connecting component. The regular polyhedron has one face reserved for installation with the second device (2). Each of the remaining faces of the regular polyhedron is fixed with a camera (82) by a camera nut (70), a camera washer (80), and a camera bolt (83). The regular polyhedron has a through space between two non-adjacent faces. The second device (2) includes a three-axis rotating and fixing component (10) and multiple connecting rods (9); The third device (3) includes a second battery and a second control and communication module; The fourth device (6) includes a first battery (64), a first control and communication module (63), a fixing component, and a connecting component (61). The fourth device (6) is installed on one of two non-adjacent surfaces at the through space position by a camera nut (70), a module gasket (60), and a module bolt (62). The first battery (64) occupies the through space. The second device (2) is connected to the connecting component (61) of the fourth device (6).

2. The close-up photogrammetry device based on visual reference positioning according to claim 1, characterized in that, The regular polyhedron is a regular icosahedron (7). Three camera nuts (70) are fixed to the nineteen faces of the regular icosahedron (7) of the first device (1). The nineteen cameras (82) of the first device (1) are fixed to the three camera nuts (70) by a camera pad (80) and three camera bolts (83). Each of the nineteen faces of the regular icosahedron (7) has a number or letter mark to represent the camera (82) code. The camera pad (80) has three through holes and a direction mark (81). The diameter of the through hole is larger than the diameter of the camera bolt (83). The three adjacent faces of the regular icosahedron (7) are fixed to the camera nuts (70). The vertices form a triangular plane. The camera (82) is fixed on the triangular plane of the regular icosahedron (7), and the camera's image plane is parallel to the triangular plane. The line connecting the image center of the camera (82) and the center of the triangle is perpendicular to the triangular plane. The image center of the camera (82) is the camera center or the camera image center. The center of the regular icosahedron (7) is taken as the origin of the local coordinate system of the regular icosahedron. A vertex of the regular icosahedron (7) is taken as a point on the Z-axis of the local coordinate system to determine the direction of the Z-axis. The plane passing through the origin and perpendicular to the Z-axis intersects the edge of the regular icosahedron (7) to obtain the corresponding intersection point. Choose one intersection point as the point on the Y-axis or X-axis to determine the Y-axis or X-axis direction of the local coordinate system of the icosahedron. Determine the other axis direction through the left-hand or right-hand coordinate system, that is, the X-axis or Y-axis direction of the local coordinate system of the icosahedron. In this way, determine the origin and axis vector of the local coordinate system of the icosahedron. In the local coordinate system of the icosahedron, determine the local coordinates of all vertices of the icosahedron (7), the local coordinates of the centers of all triangles formed by adjacent vertices, and the local coordinates of the image centers of the nineteen cameras (82). Take the center of the triangle of the icosahedron (7) as the origin and form three vectors through three vertices. Camera (82), camera bolt The fifth device (8) consists of (83), camera pad (80) and direction marker (81). The vector corresponding to the vertex closest to the direction marker (81) is the Y-axis or X-axis vector of the camera plane. The vector formed by the two vertices far from the direction marker (81) is the other axis vector of the camera plane, that is, the X-axis or Y-axis direction of the camera plane. The camera Z-axis vector is obtained by calculating the vector product of the Y and X-axis vectors of the camera plane. The camera Z-axis vector is the camera direction vector. Thus, the center coordinates, camera plane axis vectors and direction vectors of the nineteen camera (5) images in the local coordinate system of the icosahedron are calculated.

3. The close-up photogrammetry device based on visual reference positioning according to claim 2, characterized in that, The connecting rod (9) of the second device (2) is composed of a fixed rod (90) and a tightening sleeve (91). The fixed rod (90) is composed of a central column rod and two fixed terminals. The fixed terminals have three mating columns. The inner diameter of the tightening sleeve (91) is larger than the diameter of the fixed terminal. The diameter of the central through hole of the column rod fixedly connected to the tightening sleeve (91) is smaller than the diameter of the fixed terminal. The inner wall of the tightening sleeve (91) has threads. The tightening sleeve (91) is used to lock the connecting component (61) of the fourth device (6) or the first component (11) or the second component (12) of the three-axis rotating fixing component (10).

4. The close-up photogrammetry device based on visual reference positioning according to claim 3, characterized in that, The connecting component (61) of the fourth device (6) has three or more mating column grooves on one bottom surface. The diameter of the mating column groove is larger than the diameter of the mating column of the fixing terminal of the fixing rod (90) of the connecting rod (9) of the second device (2). The mating column can be inserted into the mating column groove. The camera nut (70) of the first device (1) is fixedly connected to the fourth device (6). In the axial direction of the fixing rod, the rotation angle between the mating column and the mating column groove determines the rotation angle between the first device (1) and the second device (2). The connecting component (61) of the fourth device (6) has threads on its side. The threads match the threads on the inner wall of the tightening sleeve (91) of the connecting rod (9) of the second device (2). The tightening sleeve (91) is tightened and fixed to the connecting component (61) of the fourth device (6) through the threads.

5. The close-up photogrammetry device based on visual reference positioning according to claim 4, characterized in that, The three-axis rotating fixing component (10) of the second device (2) is composed of a first component (11), a second component (12), and a third component (13). The second component (12) is composed of a connecting column and a connector with four or more mating holes. The first component (11) is composed of a connecting column and a connector with four or more mating columns. The diameter of the mating hole of the connector with mating holes is larger than the diameter of the mating column of the connector with mating columns. The height of the mating column in the middle of the connector is greater than the height of the surrounding mating columns. A connector with a mating hole and a connector with a mating column overlap. The mating column in the middle of the connector protrudes above the connector with the mating hole. The third component (13) is a screw-on column with a column groove. The column groove in the middle of the screw-on column has a thread. The thread matches the thread of the middle mating column. Tightening is achieved through the two threads. The first component (11), the second component (12), and the third component (13) are fixed in the centerline direction of the middle mating column. The mating column of the first component (11) is in the middle The first component (11) and the second component (12) are rotated in the center direction and inserted into the mating hole of the second component (12). The bottom surface of the connecting column of the first component (11) and the second component (12) has three or more mating column grooves. The diameter of the mating column groove is larger than the diameter of the mating column of the fixing terminal of the fixing rod (90) of the connecting rod (9) of the second device (2). The mating column can be inserted into the mating column groove. In the axial direction of the first component (11) or the second component (12) of the three-axis rotating fixing component (10) of the second device (2), the mating column and the mating column groove are... The rotation angle determines the rotation angle of the fixing rod of the first component (11) or the second component (12) of the three-axis rotation fixing part (10) and the connecting rod (9) of the second device (2); the side of the connecting column has threads, which match the threads on the inner wall of the tightening sleeve (91) of the connecting rod (9). By tightening the two threads, the tightening sleeve (91) of the connecting rod (9) is fixed to the connecting column of the first component (11) or the second component (12) of the three-axis rotation fixing part (10) in the direction of the center line of the connecting column.

6. The close-up photogrammetry device based on visual reference positioning according to claim 1, characterized in that, The fourth device (6) includes a first battery (64) and a first control communication module (63). The first battery (63) supplies power to the nineteen cameras (82) of the icosahedron and the first control communication module (63). The first control communication module (63) includes a control processing module and a wireless data transmission or image transmission communication module. The control processing module includes the function of controlling the shooting function of the nineteen cameras (82) and the function of acquiring the camera images. The wireless data transmission or image transmission communication module includes the function of sending the camera images to the second control communication module of the third device (3) and receiving control commands from the second control communication module.

7. The close-up photogrammetry device based on visual reference positioning according to claim 1, characterized in that, The third device (3) includes a second battery and a second control communication module. The second battery powers the second control communication module. The second control communication module includes a human-computer interaction module, a wireless data transmission or image transmission module, and a control processing module. The human-computer interaction module includes functions for user input parameters and displaying images from the nineteen cameras (82) of the icosahedron. The wireless data transmission or image transmission module includes functions for receiving camera images from the first device (1) and sending control commands to the first device (1). The control processing module includes functions for calculating and saving the center coordinates of the camera images in the measurement space coordinate system and the camera plane axis vector and direction vector.

8. A close-up photogrammetry method based on visual reference positioning, and a close-up photogrammetry device based on visual reference positioning according to any one of claims 2 to 7, characterized in that, Specifically, the steps include the following: S1: Establish a measurement space coordinate system, and divide the measurement space into one or more measurement subspaces; under the local coordinates of the regular icosahedron, calculate the center coordinates, camera plane axis vectors, and direction vectors of the nineteen camera (82) images of the first device (1); take the center coordinates of each camera image in the nineteen camera (82) images as the origin, take the camera plane axis vector as the X-axis or Y-axis, and take the camera direction vector as the Z-axis to establish the corresponding camera local coordinate system, that is, establish a total of nineteen camera local coordinate systems, calculate the three rotation angles and origin offsets between any two camera local coordinate systems in the nineteen camera local coordinate systems, that is, calculate a total of 19×18 sets of the three rotation angles and origin offsets; fix the connection part between the first device (1), the fourth device (6), and the second device (2); S2: Perform close-up photogrammetry on each measurement subspace, and arrange one or more AprilTag codes in the space outside the current measurement subspace; measure the center position and direction of the AprilTag code in the measurement space coordinate system; the second device (2) uses the rotation and fixing function of the three axes of the three-axis rotation fixing component (10) to assemble a second device (2) containing one or more connecting rods (9), the rotation angles between the connecting rods (9) are the same or different, forming a second device (2) of different shapes to adapt to measurement subspaces with different spatial characteristics; perform close-up photogrammetry by holding one end of the second device (2), or fix one end of the second device (2) on the drone and perform close-up photogrammetry by controlling the drone; S3: The first device (1) is in different positions or different postures in the current measurement subspace. Each camera on the first device (1) at the current position and posture captures and generates a camera image. The nineteen cameras (5) of the first device (1) capture and generate nineteen camera images. The first control communication module (63) of the fourth device (6) sends the camera images to the second control communication module of the third device (3). S4: The second control communication module receives images of the nineteen cameras (82) at the current position and attitude in the current measurement subspace, calculates and saves the center coordinates, camera plane axis vector and direction vector of the nineteen camera images in the measurement space coordinate system; S5: For all positions or all postures in the current measurement subspace, if the device completes the calculation and saving of the center coordinates, camera plane axis vectors and direction vectors of the close-up photography and the images of the nineteen cameras (82) in the measurement space coordinate system, then proceed to step S6; otherwise, the first device moves to the next different position or is in the next different posture in the current measurement subspace and proceeds to step S3. S6: If the measurement subspace has not completed the close-up photography and the calculation of camera image parameters in the measurement space coordinate system, then proceed to step S2. Otherwise, use the camera images in all measurement subspaces that do not contain the AprilTag code and the center coordinates, camera plane axis vector and direction vector of the camera images in the measurement space coordinate system to perform close-up photography measurement calculations, and generate a detailed digital orthophoto map and a geographic information model of the three-dimensional model.

9. The close-up photogrammetry device and method based on visual reference positioning according to claim 8, characterized in that, Step S4 specifically includes the following steps: Step 1: If no one or more images containing the AprilTag code are calculated and identified from the camera images, the first device moves to the next different position or is in the next different pose in the current measurement subspace and proceeds to step S3; otherwise, proceed to step 2. Step 2: For one or more camera images containing the AprilTag code, perform the following calculations: If one or two AprilTag codes are identified in the camera image, based on the position and orientation of the AprilTag code in the measurement space, obtain the measurement space positions of the four vertices of the AprilTag code. Based on the image coordinates of the four vertices and the measurement space positions, use spatial resection to calculate the center coordinates and orientation of the camera image in the measurement space coordinate system. If three or more AprilTag codes are identified in the camera image, obtain the measurement space positions of the centers of the three or more AprilTag codes. Combine this with the image coordinates of the AprilTag code centers, use spatial resection to calculate the center coordinates and orientation of the camera image in the measurement space coordinate system. Through these calculations, the center coordinates and orientation of one or more camera images in the measurement space coordinate system are obtained. Step 3: There are two methods to calculate the center coordinates, camera plane axis vectors, and direction vectors of the images from the nineteen cameras (82) in the measurement space coordinate system. You can choose one of these methods to calculate them. The two calculation methods are as follows: The first calculation method is as follows: Based on the center coordinates and orientation of one or more camera images in the measurement space coordinate system, select the center coordinates and orientation of one camera image in the measurement space coordinate system, calculate the origin, X-axis vector, Y-axis vector, and Z-axis vector of one camera local coordinate system in the measurement space coordinate system, based on the three rotation angles and origin offset between any two camera local coordinate systems calculated in step S1, and using the absolute orientation relationship between the two coordinate systems, calculate the origin, X-axis vector, Y-axis vector, and Z-axis vector of eighteen camera local coordinate systems other than one camera in the measurement space coordinate system, and obtain and save the center coordinates, camera plane axis vectors, and direction vectors of the nineteen camera (5) images in the measurement space coordinate system; The second calculation method is as follows: Based on the center coordinates and pose of the images from one or more cameras in the measurement space coordinate system, the origin, X-axis vector, Y-axis vector, and Z-axis vector of the local coordinate system of one or more cameras in the measurement space coordinate system are calculated. Based on the geometric relationship between the origin of the icosahedral local coordinate system and the local coordinate systems of one or more cameras, the origin coordinates of the one or more icosahedral local coordinate systems in the measurement space coordinate system are calculated. The average of these origin coordinates is the origin coordinate of the icosahedral local coordinate system in the measurement space coordinate system. Based on the icosahedral local coordinate system and the measurement space coordinate system, the scaling factor of the model scale of the two coordinate systems is calculated, and the average of the origin coordinates of the two coordinate systems is calculated. The displacement is calculated based on the center coordinates of one or more camera images in the measurement space coordinate system and their center coordinates in the local coordinate system of the icosahedron. The three rotation angles between the two coordinate systems are calculated by the analytical absolute orientation method. Based on the scaling factor of the model scale of the two coordinate systems, the translation of the origin coordinates of the two coordinate systems, and the three rotation angles between the two coordinate systems, the center coordinates of the nineteen camera (82) images in the local coordinate system of the icosahedron, the camera plane axis vector and the direction vector are substituted into the absolute orientation relationship of the two coordinate systems. The center coordinates, camera plane axis vector and direction vector of the nineteen camera (82) images in the measurement space coordinate system are calculated and saved.

Images