A planar steel gate installation method based on computer vision positioning

Abstract

The present invention provides a planar steel gate installation method based on computer vision positioning, comprising measuring and marking the center line of the gate, arranging targets on the lower leaf gate and the upper leaf gate; performing preliminary installation and positioning of the lower leaf gate; vertically aligning the lower leaf gate and the gate slot using computer vision technology; performing preliminary positioning and alignment of the upper leaf gate; vertically aligning the upper leaf gate and the lower leaf gate using computer vision technology; setting up scaffolding and welding the gate. The invented gate is directly installed in the gate slot without the need to set up an assembly platform, has low reliance on site area, and can save gate installation costs. During installation, the gate is hoisted and installed in sections, which can reduce the hoisting weight and reduce the hoisting safety risk compared to the overall hoisting method. The present invention uses computer vision technology to control the installation and positioning of the gate. Compared with traditional measurement and control technology, the gate installation accuracy can be improved and the monitoring workload can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of gate installation in water conservancy projects, and in particular to a planar steel gate installation method based on computer vision positioning. Background Art

[0002] The gate is an important component of water conservancy projects. Its function is to close the openings of hydraulic structures and to open these openings completely or partially as needed to regulate the upstream and downstream water levels and discharge flow, so as to play the engineering role of the sluice in flood control and drainage, navigation irrigation, alkali diversion, water source protection and ecological environment improvement. The gate is an extremely important engineering facility of the sluice project. Its normal and safe operation directly affects the effectiveness of the sluice project and the safety of the sluice operation.

[0003] Large flat steel gates are large in size and heavy in weight. They are too wide and too heavy to be transported. The gates need to be manufactured in sections in the factory and then transported to the construction site for assembly. The existing traditional assembly method requires building an assembly platform with steel sections in an open space near the gate, using a spirit level to control its flatness, and ensuring that it has a certain rigidity so that it will not deform during assembly. The gates are then laid flat on the platform for welding and installation.

[0004] Therefore, the existing assembly method relies more on a sufficiently large and relatively flat site to build an assembly platform, and setting up the platform requires a large amount of steel sections and steel plates. At the same time, after the gate is assembled into a whole, large lifting equipment is required to lift it, and the overall lifting is relatively dangerous. Therefore, when restricted by production, site, transportation and other conditions, it is more difficult to use the traditional method of setting up an assembly platform to install large flat steel gates, and it is impossible to assemble them directly in the gate slot. In addition, traditional measuring devices such as spirit levels are used to control its flatness. Due to the low measurement accuracy, the overall flatness and other installation accuracy of the gate after splicing are low. Summary of the Invention

[0005] The purpose of the present invention is to provide a planar steel gate installation method based on computer vision positioning to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for installing a planar steel gate based on computer vision positioning, the specific steps include:

[0008] S1: Measure and mark the center lines of the gate leaf and the gate slot on the gate sill, and arrange targets A, B, C and D on the lower leaf gate, and targets A', B', C' and D' on the upper leaf gate;

[0009] S2: Perform preliminary installation and positioning of the lower gate, and align the lower gate by the centerline mark of the gate groove on the gate sill and the centerline mark of the lower gate leaf;

[0010] S3: Using computer vision technology, a sub-pixel edge detection algorithm based on the Sobel operator and orthogonal polynomial fitting principle is used to obtain the coordinates of targets A, B, C, and D in the pixel coordinate system. By calculating the deviation information of the target in the three-dimensional space coordinates, the lower gate is adjusted and positioned using a jack so that the lower gate and the gate slot are vertically aligned;

[0011] S4: Lift the upper gate into the slot and perform preliminary positioning and alignment of the upper gate using the centerline mark of the lower gate and the centerline mark of the upper gate leaf;

[0012] S5: Obtain the coordinates of targets A', B', C', and D' in the pixel coordinate system using a sub-pixel edge detection algorithm based on the Sobel operator and orthogonal polynomial fitting principle. Calculate the deviation information of the targets in the three-dimensional space coordinates and use the jack to adjust the position of the upper gate so that the upper gate and the lower gate are vertically aligned.

[0013] S6: Set up scaffolding and weld the gate in the order of beams first and panels later to control the deformation of the gate during welding.

[0014] In one embodiment, the targets A', B' and targets A, B are located above and to the left of the upper and lower leaf gates, respectively, and the targets C', D' and targets C, D are located below and to the left of the upper and lower leaf gates, respectively.

[0015] In one embodiment, the specific steps of installing and positioning the lower leaf gate in step S2 include:

[0016] S21: hoisting the lower gate so that the bottom elevation of the lower gate is higher than the top elevation of the gate, and then rotating the crane to insert the lower gate into the gate slot;

[0017] S22: Align the center line of the sill and the center line of the lower gate, and position the lower gate symmetrically in the gate slot. At the same time, place wooden plugs of the same thickness on the water-facing side of the bottom of the gate slot on both sides of the lock chamber, so that the bottom of the lower gate is close to the wooden plugs and lowered, ensuring that the bottom of the lower gate is parallel to the axis of the gate slot of the lock chamber;

[0018] S23: A jack is set on the top of the lower gate. Use the jack to adjust the distance between the top and the water-facing side of the gate slot to make it equal to the thickness of the wooden plug at the bottom of the gate slot, and perform preliminary positioning of the lower gate.

[0019] In one embodiment, the specific steps of using computer vision technology in conjunction with the jack to adjust and position the lower leaf gate in step 3 include:

[0020] S31: Using a jack to adjust the verticality of the lower gate, during the adjustment process, a high-speed camera is used to track the motion of targets A, B, C, and D of the lower gate, and the captured video images are transmitted to a computer;

[0021] S32: According to the video image, the coordinates (u, v) of targets A, B, C, and D in the pixel coordinate system are obtained using a sub-pixel edge detection algorithm based on the Sobel operator and orthogonal polynomial fitting principle. T ;

[0022] S33: Convert the pixel coordinates of targets A, B, C, and D into coordinates (X, Y, Z) in a three-dimensional space coordinate system T ;

[0023] S34: Calculate the difference between the three-dimensional space coordinates of target A and target C, and the difference between the three-dimensional space coordinates of target B and target D, and obtain the verticality deviation of the lower leaf gate;

[0024] S35: According to the position deviations of targets A, C, B, and D, the verticality of the lower gate is adjusted using a jack;

[0025] S36: Repeat steps S31 to S35 until the position deviations of targets A, B, C and D meet the accuracy requirements, and the vertical alignment installation of the lower leaf gate is achieved.

[0026] In one embodiment, the specific steps of installing and positioning the upper leaf gate in step S4 include:

[0027] S41: hoisting the upper gate so that the bottom elevation of the upper gate is higher than the top elevation of the gate, and then rotating the crane to insert the upper gate into the gate slot;

[0028] S42: Positioning the upper gate symmetrically in the gate slot by aligning the center line of the upper gate with the center line of the lower gate;

[0029] S43: The flange plate extending from the top of the lower gate is engaged with the groove at the bottom of the upper gate to form a positioning slot.

[0030] In one embodiment, the specific steps of using computer vision technology in conjunction with the jack to adjust and position the upper leaf gate in step 5 include:

[0031] S51: Using a jack to adjust the verticality and alignment of the upper gate, during the adjustment process, a high-speed camera is used to track the motion of the target of the upper gate, and the captured video image is transmitted to the computer;

[0032] S52: According to the video image, the coordinates (u, v) of the targets A', B', C' and D' in the pixel coordinate system are obtained using a sub-pixel edge detection algorithm based on the Sobel operator and orthogonal polynomial fitting principle. T ;

[0033] S53: Convert the pixel coordinates of targets A', B', C' and D' into coordinates (X, Y, Z) in a three-dimensional space coordinate system T ;

[0034] S54: Calculate the difference between the three-dimensional coordinates of target A' and target C', and the difference between the three-dimensional coordinates of target B' and target D', and obtain the verticality deviation of the upper gate. Simultaneously, calculate the difference between the three-dimensional coordinates of target A' and target A, and the difference between the three-dimensional coordinates of target B' and target B, and obtain the alignment deviation of the upper and lower gates.

[0035] S55: According to the target position deviation information, the position of the upper gate is adjusted using the jack to change the verticality of the upper gate and the forward and backward movement of the docking position;

[0036] S56: Repeat S51 to S55 until the deviation of the target position meets the accuracy requirement, and the upper and lower leaf gates are installed vertically and aligned.

[0037] In one embodiment, the pixel coordinates of the target are converted into coordinates in a three-dimensional space coordinate system (X, Y, Z) T The specific steps include:

[0038] According to the camera calibration, the camera's intrinsic parameter coefficient K and extrinsic parameter coefficient T are obtained as follows:

[0039]

[0040]

[0041] Where: f x =f / dx;f y =f / dy, f is the focal length of the camera, dx and dy are the physical sizes of the unit pixel in the x and y directions respectively, u0 and v0 are the optical centers, R 3×3 、T 3×1 The rotation matrix and translation matrix from the three-dimensional coordinate system to the camera coordinate system;

[0042] The coordinates (u, v) in the pixel coordinate system T Convert the coordinates to the image coordinate system (x, y, 1) T , the conversion relationship is:

[0043]

[0044] The coordinates (x, y, 1) in the image coordinate system T Convert to the camera coordinate system (X', Y', Z') T , the conversion relationship is:

[0045]

[0046] Where: s is the proportional coefficient, s = Z c , Z c is the point in the camera coordinate system (X', Y', Z') T The distance from the camera's optical center in the Z' axis direction;

[0047] The coordinates (X', Y', Z') in the camera coordinate system T , converted to coordinates in the three-dimensional space coordinate system (X, Y, Z) T , the conversion relationship is:

[0048]

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1. The gate of the present invention is installed directly in the gate groove, without the need to set up an assembly platform, which has a small dependence on the site area and can save the gate installation cost. During installation, the gate is hoisted and installed in sections, which can reduce the hoisting weight and reduce the hoisting safety risk compared to the overall hoisting method.

[0051] 2. The present invention uses computer vision technology to control the installation and positioning of the gate. Compared with traditional measurement and control technology, it can improve the gate installation accuracy and reduce the monitoring workload. It can also be used for continuous deformation monitoring of the gate during operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 Schematic diagram of the overall method flow of the present invention;

[0053] Figure 2 This is a schematic diagram of the overall structure of the plane steel gate in the present invention;

[0054] Figure 3 Schematic diagram of the structure of the target in the present invention;

[0055] Figure 4 It is a structural schematic diagram of the jack in the present invention;

[0056] Figure 5 It is a schematic diagram of the overall cross-sectional structure of the present invention;

[0057] Figure 6 It is a schematic diagram of the installation structure of the overall jack in the present invention.

[0058] In the figure: 10 upper leaf gate, 11 groove, 20 lower leaf gate, 21 flange plate, 30 gate sill, 31 door groove, 40 wooden plug, 50 jack, 60 mark. DETAILED DESCRIPTION

[0059] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0060] Example:

[0061] See also Figures 1 to 5 , the present invention provides a technical solution:

[0062] A method for installing a planar steel gate based on computer vision positioning. Before using this method, the gate needs to be manufactured in sections and hoisted to the construction site in sections. The gate consists of an upper leaf gate 10 and a lower leaf gate 20. The specific steps of the installation method include:

[0063] S1: Measure the center lines of the gate leaf and the gate slot 31 on the gate sill 30 and mark them 60, and arrange targets A, B, C and D on the lower leaf gate 20, and arrange targets A', B', C' and D' on the upper leaf gate 10.

[0064] Furthermore, the targets A', B' and targets A, B are located above and to the left of the upper and lower gates 10 and 20 respectively, and the targets C', D' and targets C, D are located below and to the left of the upper and lower gates 10 and 20 respectively.

[0065] After the gate is hoisted into the gate slot 31, the gate slot 31 will block a part of the gate. Therefore, the setting of the target position must ensure that it is not within the blocking range of the gate slot 31. Therefore, the setting of the targets A, B, C and D and A', B', C' and D' must ensure that they are not within the blocking range of the gate slot 31. The mark 60 is released by the measuring instrument, and it is necessary to ensure that the mark is relatively eye-catching.

[0066] S2: Perform preliminary installation and positioning of the lower leaf gate 20 by aligning the lower leaf gate 20 with the center line mark 60 of the gate groove 31 on the gate bottom sill 30 and the center line mark 60 of the gate leaf of the lower leaf gate 20.

[0067] Furthermore, the specific steps of installing and positioning the lower leaf gate 20 include:

[0068] S21: hoisting the lower gate 20 so that the bottom elevation of the lower gate 20 is higher than the gate top elevation, and then rotating the crane so that the lower gate 20 is inserted into the gate slot 31;

[0069] S22: By aligning the center line of the sill and the center line of the lower gate 20, the lower gate 20 is positioned symmetrically in the gate slot 31. At the same time, wooden plugs 40 of the same thickness are placed on the water-facing side of the bottom of the gate slot 31 on both sides of the lock chamber. The bottom of the lower gate 20 is placed close to the wooden plugs 40 to ensure that the bottom of the lower gate 20 is parallel to the axis of the lock chamber gate slot 31.

[0070] S23: A jack 50 is set on the top of the lower leaf gate 20, and the jack 50 is used to adjust the distance between the top and the water-facing side of the gate groove 31 to make it equal to the thickness of the wooden plug 40 at the bottom of the gate groove 31, thereby preliminarily positioning the lower leaf gate 20.

[0071] Among them, when using the jack 50 to position the lower leaf gate 20, one end of the jack 50 is against the side of the lower leaf gate 20, and the other end is against the inner side of the gate slot 31, and the gate slot 31 is used as a reaction point to push the lower leaf gate 20 to adjust its posture. In this embodiment, this method is used when the jack 50 is used to adjust the lower leaf gate 20 and the upper leaf gate 10.

[0072] S3: Using computer vision technology, the coordinates of targets A, B, C, and D in the pixel coordinate system are obtained through the sub-pixel edge detection algorithm based on the Sobel operator and orthogonal polynomial fitting principle. By calculating the deviation information of the target in the three-dimensional space coordinates, the lower leaf gate 20 is adjusted and positioned using the jack 50 so that the lower leaf gate 20 and the gate slot 31 are vertically aligned.

[0073] Furthermore, the specific steps of using computer vision technology in conjunction with the jack 50 to adjust and position the lower leaf gate 20 include:

[0074] S31: Using the jack 50 to adjust the verticality of the lower gate 20, during the adjustment process, a high-speed camera is used to track the motion of the targets A, B, C, and D of the lower gate 20, and the captured video images are transmitted to the computer;

[0075] S32: According to the video image, the sub-pixel edge detection algorithm based on the Sobel operator and orthogonal polynomial fitting principle is used to obtain the coordinates of targets A, B, C, and D in the pixel coordinate system (u A , v A ) T 、(u B , v B ) T 、(u C , v C ) T and(u D, v D ) T ;

[0076] S33: The pixel coordinates of targets A, B, C and D are converted into coordinates in a three-dimensional space coordinate system (X A , Y A , Z A ) T 、(X B , Y B , Z B ) T 、(X C , Y C , Z C ) T and (X D , Y D , Z D ) T ;

[0077] S34: Calculate the difference (ΔX AC , ΔY AC , ΔZ AC ), and the difference between the three-dimensional coordinates of target B and target D (ΔX BD , ΔY BD , ΔZ BD ), the verticality deviation of the lower leaf gate 20 is obtained;

[0078] S35: According to the position deviations of targets A, C, B, and D, the verticality of the lower leaf gate 20 is adjusted using the jack 50;

[0079] S36: Repeat steps S31 to S35 until the position deviations of targets A, B, C and D meet the accuracy requirements, and the vertical alignment installation of the lower leaf gate 20 is achieved.

[0080] S4: The upper gate 10 is hoisted into the gate slot 31, and the upper gate 10 is preliminarily positioned and aligned by the center line mark 60 of the lower gate 20 and the center line mark 60 of the gate leaf of the upper gate 10.

[0081] Furthermore, the specific steps of installing and positioning the upper leaf gate 10 include:

[0082] S41: hoisting the upper gate 10 so that the bottom elevation of the upper gate 10 is higher than the gate top elevation, and then rotating the crane so that the upper gate 10 is inserted into the gate slot 31;

[0083] S42: By aligning the center line of the upper gate 10 and the center line of the lower gate 20, the upper gate 10 is positioned symmetrically in the gate slot 31;

[0084] S43: The flange plate 21 extending from the top of the lower gate 20 engages with the groove 11 at the bottom of the upper gate 10 to form a positioning slot.

[0085] S5: The coordinates of targets A', B', C' and D' in the pixel coordinate system are obtained by using a sub-pixel edge detection algorithm based on the Sobel operator and orthogonal polynomial fitting principle. The deviation information of the targets in the three-dimensional space coordinates is calculated, and the upper leaf gate 10 is adjusted and positioned using the jack 50 so that the upper leaf gate 10 and the lower leaf gate are vertically aligned.

[0086] Furthermore, in step 5, the specific steps of using computer vision technology in conjunction with the jack 50 to adjust and position the upper leaf gate 10 include:

[0087] S51: Using the jack 50 to adjust the verticality and alignment of the upper leaf gate 10, during the adjustment process, a high-speed camera is used to track the motion of the target of the upper leaf gate 10, and the captured video image is transmitted to the computer;

[0088] S52: According to the video image, the coordinates of the targets A', B', C' and D' in the pixel coordinate system are obtained using a sub-pixel edge detection algorithm based on the Sobel operator and orthogonal polynomial fitting principle (u A ',v A ') T 、(u B ',v B ') T 、(u C ',v C ') T and(u D ',v D ') T ;

[0089] S53: The pixel coordinates of targets A', B', C' and D' are converted into coordinates in a three-dimensional space coordinate system (X A ', Y A ', Z A ') T 、(X B ', Y B ', Z B ') T 、(X C ', Y C ', Z C ') T and (X D ', Y D ', Z D ') T ;

[0090] S54: Calculate the difference ΔX between the three-dimensional coordinates of target A' and target C' respectively A ' C ', ΔY A ' C ', ΔZ A ' C '), and the difference between the three-dimensional coordinates of target B' and target D' (ΔX B ' D ', ΔY B ' D ', ΔZ B ' D '), the vertical deviation of the upper leaf gate 10 is obtained, and the difference ΔX between the three-dimensional space coordinates of target A' and target A is calculated respectively A ' A , ΔY A ' A , ΔZ A ' A ), and the difference between the three-dimensional coordinates of target B' and target B (ΔX B ' B , ΔY B ' B , ΔZ B ' B ), and obtain the alignment deviation of the upper leaf gate 10 and the lower leaf gate 20;

[0091] S55: According to the target position deviation information, the position of the upper gate 10 is adjusted using the jack 50 to change the verticality of the upper gate 10 and the forward and backward movement of the docking position;

[0092] S56: Repeat S51 to S55 until the deviation of the target position meets the accuracy requirement, and the upper leaf gate 10 and the lower leaf gate 20 are installed vertically and aligned.

[0093] S6: Set up scaffolding and weld the gate in the order of beams first and panels later to control the deformation of the gate during welding.

[0094] Furthermore, to reduce the deformation of the gate during welding, the order of welding should follow the principle of beam first and then panel. First, weld the main beam and longitudinal beam, then weld the seams of the backwater panel, and finally weld the seams of the panel facing the water. Due to the obstruction of the gate groove 31, the entire gate needs to be hoisted after the above welding work is completed, and then weld the gate edge. Stop welding when the welding temperature is too high, and continue construction when the temperature drops to the natural temperature to avoid gate deformation due to excessive temperature. After the welding and assembly are completed, the gate body appearance, panel flatness, and welds are inspected and evaluated according to relevant specifications.

[0095] In this embodiment, the pixel coordinates of the target are converted into coordinates (X, Y, Z) in a three-dimensional space coordinate system. T The specific steps include:

[0096] According to the camera calibration, the camera's intrinsic parameter coefficient K and extrinsic parameter coefficient T are obtained as follows:

[0097]

[0098]

[0099] Where: f x =f / dx;f y =f / dy, f is the focal length of the camera, dx and dy are the physical sizes of the unit pixel in the x and y directions respectively, u0 and v0 are the optical centers, R 3×3 、T 3×1 The rotation matrix and translation matrix from the three-dimensional coordinate system to the camera coordinate system;

[0100] The coordinates (u, v) in the pixel coordinate system T Convert the coordinates to the image coordinate system (x, y, 1) T , the conversion relationship is:

[0101]

[0102] The coordinates (x, y, 1) in the image coordinate system T Convert to the camera coordinate system (X', Y', Z') T , the conversion relationship is:

[0103]

[0104] Where: s is the proportional coefficient, s = Z c , Z c is the point in the camera coordinate system (X', Y', Z') T The distance from the camera's optical center in the Z' axis direction;

[0105] The coordinates (X', Y', Z') in the camera coordinate system T , converted to coordinates in the three-dimensional space coordinate system (X, Y, Z) T , the conversion relationship is:

[0106]

[0107] While the present invention is described with reference to the accompanying drawings, the embodiments disclosed in the drawings are intended to illustrate preferred embodiments of the invention and are not to be construed as limiting the invention. The purpose of the drawings is to supplement the textual description of the specification and to provide an intuitive and visual understanding of each technical feature and the overall technical solution of the invention. For the purpose of illustrating components required for clarity, the scales in the drawings do not represent the actual proportional relationships of the components.

[0108] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A planar steel gate installation method based on computer vision positioning, characterized in that: The specific steps include: S1: Measure and mark the center lines of the gate leaf and the gate slot on the gate sill, and arrange targets A, B, C and D on the lower leaf gate, and targets A', B', C' and D' on the upper leaf gate; S2: Perform preliminary installation and positioning of the lower gate, and align the lower gate by the centerline mark of the gate groove on the gate sill and the centerline mark of the lower gate leaf; S3: Using computer vision technology, a sub-pixel edge detection algorithm based on the Sobel operator and orthogonal polynomial fitting principle is used to obtain the coordinates of targets A, B, C, and D in the pixel coordinate system. By calculating the deviation information of the target in the three-dimensional space coordinates, the lower gate is adjusted and positioned using a jack so that the lower gate and the gate slot are vertically aligned; S4: Lift the upper gate into the slot and perform preliminary positioning and alignment of the upper gate using the centerline mark of the lower gate and the centerline mark of the upper gate leaf; S5: Obtain the coordinates of targets A', B', C', and D' in the pixel coordinate system using a sub-pixel edge detection algorithm based on the Sobel operator and orthogonal polynomial fitting principle. Calculate the deviation information of the targets in the three-dimensional space coordinates and use the jack to adjust the position of the upper gate so that the upper gate and the lower gate are vertically aligned. S6: Set up scaffolding and weld the gate in the order of beams first and panels later to control the deformation of the gate during welding.

2. A planar steel gate installation method based on computer vision positioning according to claim 1, characterized in that: The targets A', B' and targets A, B are located above and to the left of the upper and lower leaf gates respectively, and the targets C', D' and targets C, D are located below and to the left of the upper and lower leaf gates respectively.

3. The method for installing a planar steel gate based on computer vision positioning according to claim 1 is characterized in that: The specific steps of installing and positioning the lower leaf gate in step S2 include: S21: hoisting the lower gate so that the bottom elevation of the lower gate is higher than the top elevation of the gate, and then rotating the crane to insert the lower gate into the gate slot; S22: Align the center line of the sill and the center line of the lower gate, and position the lower gate symmetrically in the gate slot. At the same time, place wooden plugs of the same thickness on the water-facing side of the bottom of the gate slot on both sides of the lock chamber, so that the bottom of the lower gate is close to the wooden plugs and lowered, ensuring that the bottom of the lower gate is parallel to the axis of the gate slot of the lock chamber; S23: A jack is set on the top of the lower gate. Use the jack to adjust the distance between the top and the water-facing side of the gate slot to make it equal to the thickness of the wooden plug at the bottom of the gate slot, and perform preliminary positioning of the lower gate.

4. The method for installing a planar steel gate based on computer vision positioning according to claim 2 is characterized in that: The specific steps of using computer vision technology in conjunction with the jack to adjust and position the lower leaf gate in step 3 include: S31: Using a jack to adjust the verticality of the lower gate, during the adjustment process, a high-speed camera is used to track the motion of targets A, B, C, and D of the lower gate, and the captured video images are transmitted to a computer; S32: According to the video image, the coordinates (u, v) of targets A, B, C, and D in the pixel coordinate system are obtained using a sub-pixel edge detection algorithm based on the Sobel operator and orthogonal polynomial fitting principle. T ; S33: Convert the pixel coordinates of targets A, B, C, and D into coordinates (X, Y, Z) in a three-dimensional space coordinate system T ; S34: Calculate the difference between the three-dimensional space coordinates of target A and target C, and the difference between the three-dimensional space coordinates of target B and target D, and obtain the verticality deviation of the lower leaf gate; S35: According to the position deviations of targets A, C, B, and D, the verticality of the lower gate is adjusted using a jack; S36: Repeat steps S31 to S35 until the position deviations of targets A, B, C and D meet the accuracy requirements, and the vertical alignment installation of the lower leaf gate is achieved.

5. The method for installing a planar steel gate based on computer vision positioning according to claim 1 is characterized in that: The specific steps of installing and positioning the upper leaf gate in step S4 include: S41: hoisting the upper gate so that the bottom elevation of the upper gate is higher than the top elevation of the gate, and then rotating the crane to insert the upper gate into the gate slot; S42: Positioning the upper gate symmetrically in the gate slot by aligning the center line of the upper gate with the center line of the lower gate; S43: The flange plate extending from the top of the lower gate is engaged with the groove at the bottom of the upper gate to form a positioning slot.

6. The method for installing a planar steel gate based on computer vision positioning according to claim 2, characterized in that: The specific steps of using computer vision technology in conjunction with the jack to adjust and position the upper leaf gate in step 5 include: S51: Using a jack to adjust the verticality and alignment of the upper gate, during the adjustment process, a high-speed camera is used to track the motion of the target of the upper gate, and the captured video image is transmitted to the computer; S52: According to the video image, the coordinates (u, v) of the targets A', B', C' and D' in the pixel coordinate system are obtained using a sub-pixel edge detection algorithm based on the Sobel operator and orthogonal polynomial fitting principle. T ; S53: Convert the pixel coordinates of targets A', B', C' and D' into coordinates (X, Y, Z) in a three-dimensional space coordinate system T ; S54: Calculate the difference between the three-dimensional coordinates of target A' and target C', and the difference between the three-dimensional coordinates of target B' and target D', and obtain the verticality deviation of the upper gate. Simultaneously, calculate the difference between the three-dimensional coordinates of target A' and target A, and the difference between the three-dimensional coordinates of target B' and target B, and obtain the alignment deviation of the upper and lower gates. S55: According to the target position deviation information, the position of the upper gate is adjusted using the jack to change the verticality of the upper gate and the forward and backward movement of the docking position; S56: Repeat S51 to S55 until the deviation of the target position meets the accuracy requirement, and the upper and lower leaf gates are installed vertically and aligned.

7. A planar steel gate installation method based on computer vision positioning according to any one of claims 4 or 6, characterized in that: Convert the pixel coordinates of the target into coordinates in a three-dimensional space coordinate system (X, Y, Z) T The specific steps include: According to the camera calibration, the camera's intrinsic parameter coefficient K and extrinsic parameter coefficient T are obtained as follows: Where: f x =f / dx;f y =f / dy, f is the focal length of the camera, dx and dy are the physical sizes of the unit pixel in the x and y directions respectively, u0 and v0 are the optical centers, R 3×3 、T 3×1 The rotation matrix and translation matrix from the three-dimensional coordinate system to the camera coordinate system; The coordinates (u, v) in the pixel coordinate system T Convert the coordinates to the image coordinate system (x, y, 1) T , the conversion relationship is: The coordinates (x, y, 1) in the image coordinate system T Convert to the camera coordinate system (X', Y', Z') T , the conversion relationship is: Where: s is the proportional coefficient, s = Z c , Z c is the point in the camera coordinate system (X', Y', Z') T The distance from the camera's optical center in the Z' axis direction; The coordinates (X', Y', Z') in the camera coordinate system T , converted to coordinates in the three-dimensional space coordinate system (X, Y, Z) T , the conversion relationship is:

Images