Steelmaking slab intelligent control method and system based on visual three-dimensional imaging

Abstract

The invention discloses a steel-making slab intelligent control method and system based on visual three-dimensional imaging. The method comprises the step of scanning a hot delivery area from different angles through a plurality of sets of laser radars. According to the method, the point cloud data is extracted from the three-dimensional imaging data, the point cloud model is formed through registration splicing and fine registration, automatic modeling of a scene is achieved through the process, manual intervention is reduced, noise is reduced through a bilateral filtering algorithm, triangular meshes are formed through a subdivision algorithm, and the NURBS curved surface modeling technology is adopted; the method further improves the fineness and fidelity of the model, and records the data of the plate blank on the roller way in real time, including the current position and the final position, so that the management personnel can master the transportation and processing states of the plate blank at any time, analyze according to the plate blank roller way data, and send corresponding instructions to the pusher equipment and the turntable equipment. The automatic control that the plate blanks are sequentially machined according to the set plate stacking table sequence is achieved, and the production efficiency and the intelligent management level are improved.

Description

Technical Field

[0001] The present invention relates to a steelmaking slab intelligent control method and system based on visual three-dimensional imaging, and relates to the technical field of three-dimensional imaging. Background Art

[0002] Steelmaking slabs are transported to the hot delivery area of the plate mill via rollers. Currently, the slab conveying, rotation, and depalletizing processes are all performed manually. Operators visually estimate the relative position of the slabs using images from on-site surveillance cameras. They then manually manipulate the rollers, turntable, and pallet stacking platform to change the slab's position and transport it. This method requires operators to simultaneously operate the pallet stacking platform, turntable, and rollers, resulting in high labor intensity and low production efficiency. This can lead to errors during busy production periods, impacting production schedules. Summary of the Invention

[0003] The purpose of the present invention is to address the defects or shortcomings in the existing technology and provide a steelmaking slab intelligent control method and system based on visual three-dimensional imaging. By scanning the hot delivery area from different angles through laser radar, high-precision three-dimensional imaging data can be obtained, point cloud data is extracted from the three-dimensional imaging data, and a point cloud model is formed through alignment, splicing and fine alignment. This process realizes the automatic modeling of the scene, so that management personnel can grasp the transportation and processing status of the slab at any time, analyze the slab roller data, and send corresponding instructions to the pusher equipment and turntable equipment, thereby realizing the automatic control of the slab processing in the set stacking order, thereby improving production efficiency and the intelligent level of management.

[0004] To achieve the above object, the present invention provides a steelmaking slab intelligent control method based on visual three-dimensional imaging, comprising the following steps:

[0005] S101, scanning the hot delivery area from different angles using several sets of laser radars to obtain three-dimensional imaging data of the slab, stacking platform, turntable, and roller conveyor;

[0006] S102, preprocessing the three-dimensional imaging data, where the preprocessing includes filling in missing imaging data to obtain complete three-dimensional imaging data;

[0007] S103, based on the complete 3D imaging data, extracting point cloud data from it, adjusting the initial position of the point cloud, and using a preset algorithm to align and stitch the point cloud blocks, and then using an iterative closest point method to accurately align the point cloud data to form a point cloud model;

[0008] S104, using a bilateral filtering algorithm to perform noise reduction on the point cloud model, using a subdivision algorithm to form a triangular mesh, and using NURBS surface modeling technology to generate a heat delivery area scene model;

[0009] S105. In the hot delivery area scene modeling, the data of the slabs on the roller conveyor are recorded in real time, including the current position and the end position of each slab on the roller conveyor;

[0010] S106: Analyze the slab roller table data and send corresponding instructions to the pusher equipment and the turntable equipment to process the slabs in the set stacking order.

[0011] Furthermore, filling in the missing imaging data specifically includes the following steps:

[0012] S201, arranging all imaging data into a column in the form of coordinates (x, y);

[0013] S202, retrieve the coordinates (x, y) of all imaging data;

[0014] S203: If there is a gap in x or y, fill in the gap with data.

[0015] Furthermore, the adjusting the initial position of the point cloud specifically includes the following method:

[0016] Build the target model:

[0017]

[0018] Where q i and P i Represents the matching point pair, R and T represent the rotation and translation matrices respectively, and m represents the number of point clouds;

[0019] Calculate the initial point cloud P i (i=1,2,3,...,m) in the target point cloud q i The closest point of (i=1,2,3,...,m);

[0020] Calculate the rotation and translation matrices R and T so that the target model smallest;

[0021] By rotating and translating the matrices R and T, we can exchange the overlapping point cloud P to obtain a new point cloud P. ， ;

[0022] Calculate P ， Distance to the target point cloud:

[0023]

[0024] When the distance D is less than the preset error and exceeds the target number of iterations, the iteration is stopped. Otherwise, the calculation returns to the first step and continues until the convergence condition is met.

[0025] Furthermore, the preset algorithm is:

[0026]

[0027] Where m i is the default value, l i is the distance difference after point cloud position transformation.

[0028] Furthermore, the bilateral filtering algorithm is:

[0029]

[0030] Where n is the normal vector of data point O, and α represent the data points and bilateral filter factors after filtering, respectively. The representative formula of α is:

[0031]

[0032] Where x = PP i Represents the distance between data points, m represents the M nearest neighbors Nm(P i ), the number of data points in p and θ c is the M nearest neighbors Nm(P i )Normal height length and internal tangent plane Gaussian filtering.

[0033] Furthermore, the NURBS surface modeling technology is used to generate the heat delivery area scene modeling, which specifically includes the following steps:

[0034] S301. In the modeling software, select NURBS surface modeling tool, such as Maya, 3ds Max or UG;

[0035] S302, creating a surface of the hot delivery area using a selected surface generation tool according to the point cloud model curve;

[0036] S303, edit the curve, including modifying the curvature, direction, length, etc., to ensure that the curve meets the shape requirements of the hot delivery area;

[0037] S304: Output the generated hot delivery area scene model to a file format, such as OBJ, STL, etc., for subsequent application.

[0038] Furthermore, the analysis based on the slab roller table data specifically includes the following steps:

[0039] S401, starting from the slab corresponding to the first processing step number on the stacking platform, check each slab in turn;

[0040] S402: Determine whether the currently checked slab meets the loading conditions. If so, analyze the target position of the current slab, update the target position of the next slab as the final target position, and send instructions to the pusher and turntable to control them to suspend operation in preparation for loading the current slab.

[0041] S403, determining the span and corresponding loading roller conveyor according to the order of the current slab on the stacking platform, the span in which it is located, and the corresponding loading roller conveyor, and sending a command to the turntable device in the span in which the current slab is located to control the turntable device to transfer the current slab to its corresponding loading roller conveyor;

[0042] S404: After the loading preparation of the current slab is completed, the next slab that has not yet been loaded is checked in sequence, and the above steps are repeated until all slabs are checked.

[0043] Furthermore, the determination of whether the loading conditions are met specifically includes the following steps:

[0044] If the slab before the current slab needs to pass through the loading roller corresponding to the current slab but has not passed through the loading roller, it is determined that it does not meet the loading conditions; otherwise, it is determined that it meets the loading conditions.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] Scanning the hot delivery area from different angles using LiDAR can acquire high-precision 3D imaging data, providing a solid foundation for subsequent processing and analysis. Point cloud data is extracted from the 3D imaging data and formed into a point cloud model through registration, splicing, and fine registration. This process enables automated scene modeling, reducing manual intervention. The use of bilateral filtering algorithms for noise reduction, subdivision algorithms for triangular mesh formation, and NURBS surface modeling technology further enhances the model's precision and realism, providing a more reliable basis for subsequent real-time data analysis. The system also records slab data on the roller conveyor, including its current and final positions, in real time, allowing managers to monitor the transport and processing status of the slabs at all times. Based on the slab roller conveyor data, corresponding instructions are sent to the pusher and turntable equipment, enabling automated control of the slabs as they are processed in the set stacking sequence, improving production efficiency and intelligent management. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0048] Figure 1 Schematic diagram of the method for intelligent control of steelmaking slabs based on visual three-dimensional imaging in the present invention;

[0049] Figure 2 Schematic diagram of a method for filling missing imaging data in the present invention;

[0050] Figure 3 This is a schematic diagram of a method for generating a heat delivery area scene model using NURBS surface modeling technology in the present invention;

[0051] Figure 4 Schematic diagram of the method for analyzing slab roller table data in the present invention. DETAILED DESCRIPTION

[0052] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are merely examples, and those skilled in the art may conceive of other obvious variations.

[0053] Example 1

[0054] Please refer to Figure 1 As shown, the technical solution adopted in this specific embodiment is: providing a steelmaking slab intelligent control method based on visual three-dimensional imaging, comprising the following steps:

[0055] S101. Scan the hot delivery area from different angles using several sets of laser radars to obtain three-dimensional imaging data of the slab, stacking platform, turntable, and roller conveyor, providing a solid foundation for subsequent processing and analysis.

[0056] S102, preprocessing the 3D imaging data. The preprocessing includes filling in missing imaging data to obtain complete 3D imaging data. This step pre-processes the required data so that data information can be obtained from the complete database as quickly as possible later.

[0057] S103. Based on the complete 3D imaging data, point cloud data is extracted from it, the initial position of the point cloud is adjusted, and the point cloud blocks are aligned and spliced using a preset algorithm. The point cloud data is then precisely aligned using an iterative closest point method to form a point cloud model. This process realizes automated modeling of the scene, reduces manual intervention, and improves efficiency.

[0058] S104: A bilateral filtering algorithm is used to reduce noise on the point cloud model, a subdivision algorithm is used to form a triangular mesh, and NURBS surface modeling technology is used to generate a scene model of the heat transfer area. This further improves the model's precision and realism, providing a more reliable basis for subsequent real-time data analysis.

[0059] S105. In the hot delivery area scenario modeling, the data of the slabs on the roller conveyor is recorded in real time, including the current position and the terminal position of each slab on the roller conveyor, so that the management personnel can grasp the transportation and processing status of the slabs at any time;

[0060] S106. Analyze the slab roller data and send corresponding instructions to the pusher equipment and turntable equipment to process the slabs in the set stacking order. Automated control improves production efficiency and management intelligence.

[0061] Through the above steps, those skilled in the art can understand that the present invention uses laser radar to scan the hot delivery area from different angles to obtain high-precision three-dimensional imaging data, which provides a solid foundation for subsequent processing and analysis. Point cloud data is extracted from the three-dimensional imaging data, and a point cloud model is formed through alignment, splicing and fine alignment. This process realizes the automatic modeling of the scene and reduces manual intervention. The use of bilateral filtering algorithm for noise reduction, subdivision algorithm to form a triangular mesh and NURBS surface modeling technology further improves the fineness and realism of the model, providing a more reliable basis for subsequent real-time data analysis, and records the data of the slab on the roller in real time, including the current position and the end position, so that managers can grasp the transportation and processing status of the slab at any time, analyze the slab roller data, and send corresponding instructions to the pusher equipment and turntable equipment, thereby realizing the automatic control of the slab processing in the set stacking order, improving production efficiency and the intelligent level of management.

[0062] For more specific instructions, please refer to Figure 2 As shown, filling the missing imaging data specifically includes the following steps:

[0063] S201, arranging all imaging data into a column in the form of coordinates (x, y);

[0064] S202, retrieve the coordinates (x, y) of all imaging data;

[0065] S203: If there is a gap in x or y, fill in the gap with data.

[0066] Adjusting the initial position of the point cloud includes the following methods:

[0067] Build the target model:

[0068]

[0069] Where q i and P i Represents the matching point pair, R and T represent the rotation and translation matrices respectively, and m represents the number of point clouds;

[0070] Calculate the initial point cloud P i (i=1,2,3,...,m) in the target point cloud q i The closest point of (i=1,2,3,...,m);

[0071] Calculate the rotation and translation matrices R and T so that the target model smallest;

[0072] By rotating and translating the matrices R and T, we can exchange the overlapping point cloud P to obtain a new point cloud P. ， ;

[0073] Calculate P ， Distance to the target point cloud:

[0074]

[0075] When the distance D is less than the preset error and exceeds the target number of iterations, the iteration is stopped. Otherwise, the calculation returns to the first step and continues until the convergence condition is met.

[0076] It will be understood by those skilled in the art that, since there is data overlap in the point cloud acquisition frame of the hot delivery area equipment, the point cloud data needs to be registered and spliced. The point cloud data precise registration algorithm adopted is the iterative nearest point method, which minimizes the spatial distance between the two groups of point clouds by spatially transforming the overlapping area point cloud and the target point cloud of the overlapping point cloud.

[0077] The default algorithm is:

[0078]

[0079] Where m i is the default value, l i is the distance difference after point cloud position transformation.

[0080] The bilateral filtering algorithm is:

[0081]

[0082] Where n is the normal vector of data point O, and α represent the data points and bilateral filter factors after filtering, respectively. The representative formula of α is:

[0083]

[0084] Where x = PP i Represents the distance between data points, m represents the M nearest neighbors Nm(P i ), the number of data points in p and θ c is the M nearest neighbors Nm(P i )Normal height length and internal tangent plane Gaussian filtering.

[0085] For more specific instructions, please refer to Figure 3 As shown in the figure, the NURBS surface modeling technology is used to generate the heat transfer area scene modeling, which specifically includes the following steps:

[0086] S301. In the modeling software, select NURBS surface modeling tool, such as Maya, 3ds Max or UG;

[0087] S302, creating a surface of the hot delivery area using a selected surface generation tool according to the point cloud model curve;

[0088] S303, edit the curve, including modifying the curvature, direction, length, etc., to ensure that the curve meets the shape requirements of the hot delivery area;

[0089] S304: Output the generated hot delivery area scene model to a file format, such as OBJ, STL, etc., for subsequent application.

[0090] For more specific instructions, please refer to Figure 4 As shown in the figure, the analysis based on the slab roller table data includes the following steps:

[0091] S401, starting from the slab corresponding to the first processing step number on the stacking platform, check each slab in turn;

[0092] S402: Determine whether the currently checked slab meets the loading conditions. If so, analyze the target position of the current slab, update the target position of the next slab as the final target position, and send instructions to the pusher and turntable to control them to suspend operation in preparation for loading the current slab.

[0093] S403, determining the span and corresponding loading roller conveyor according to the order of the current slab on the stacking platform, the span in which it is located, and the corresponding loading roller conveyor, and sending a command to the turntable device in the span in which the current slab is located to control the turntable device to transfer the current slab to its corresponding loading roller conveyor;

[0094] S404: After the loading preparation of the current slab is completed, the next slab that has not yet been loaded is checked in sequence, and the above steps are repeated until all slabs are checked.

[0095] More specifically, determining whether the loading conditions are met includes the following steps:

[0096] If the slab before the current slab needs to pass through the loading roller corresponding to the current slab but has not passed through the loading roller, it is determined that it does not meet the loading conditions; otherwise, it is determined that it meets the loading conditions.

[0097] More specifically, in this embodiment, a steelmaking slab intelligent control system based on visual three-dimensional imaging is provided, including:

[0098] Scanning module: The scanning module is used to scan the hot delivery area from different angles through several sets of laser radars to obtain three-dimensional imaging data of slabs, stacking platforms, turntables and rollers;

[0099] A preprocessing module is used to preprocess the three-dimensional imaging data, including filling in missing imaging data to obtain complete three-dimensional imaging data;

[0100] The stitching module is used to extract point cloud data from the complete 3D imaging data, adjust the initial position of the point cloud, and use a preset algorithm to align and stitch the point cloud blocks. It then uses the iterative closest point method to precisely align the point cloud data to form a point cloud model.

[0101] Modeling module: The modeling module is used to reduce noise on the point cloud model using bilateral filtering algorithm, form a triangular mesh using subdivision algorithm, and generate heat delivery area scene modeling using NURBS surface modeling technology;

[0102] Recording module: The recording module is used to record the data of the slab on the roller in real time in the hot delivery area scene modeling, including the current position and end position of each slab on the roller;

[0103] The control module is used to analyze the slab roller data and send corresponding instructions to the pusher equipment and turntable equipment to enable the slabs to be processed in sequence according to the set stacking order.

[0104] To be more specific, in this embodiment, an electronic device is also provided.

[0105] Electronic device is intended to represent various forms of digital computers, such as laptops, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers; electronic device may also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices and other similar computing devices; the components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit implementations of the inventions described and / or claimed herein.

[0106] The electronic device includes a computing unit, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory ROM or a computer program loaded from a storage unit into a random access memory RAM; various programs and data required for the operation of the electronic device can also be stored in the RAM; the computing unit, ROM and RAM are connected to each other through a bus; and an input / output (I / O) interface is also connected to the bus.

[0107] Multiple components in an electronic device are connected to the I / O interface, including: input units, such as a keyboard, mouse, etc.; output units, such as various types of displays, speakers, etc.; storage units, such as magnetic disks, optical disks, etc.; and communication units, such as network cards, modems, wireless communication transceivers, etc.; communication units allow electronic devices to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunication networks.

[0108] The computing unit can be various general and / or special processing components with processing and computing capabilities; some examples of computing units include but are not limited to central processing units CPU, graphics processing units GPU, various special artificial intelligence AI computing chips, various computing units running machine learning model algorithms, digital signal processors DSP, and any appropriate processors, controllers, microcontrollers, etc. The computing unit performs the various methods and processes described above, such as methods S101 to S106; for example, in some embodiments, methods S101 to S106 can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit; in some embodiments, part or all of the computer program can be loaded and / or installed on an electronic device via a ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the computing unit, one or more steps of methods S101 to S106 described above can be executed. Alternatively, in other embodiments, the computing unit can be configured to execute methods S101 to S106 in any other appropriate manner (for example, with the aid of firmware).

[0109] Various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays FPGAs, application specific integrated circuits ASICs, application specific standard products ASSPs, systems on chips SOCs, load programmable logic devices CPLDs, computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special-purpose or general-purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0110] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that when the program code is executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine as an independent software package and partially on a remote machine, or entirely on the remote machine or server.

[0111] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or apparatus, and the machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory RAM, a read-only memory ROM, an erasable programmable read-only memory EPROM, an optical fiber, a portable compact disk read-only memory CD-ROM, an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0112] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0113] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components, and the components of the system may be interconnected by any form or medium of digital data communication (e.g., a communications network), examples of which include a local area network (LAN), a wide area network (WAN), and the Internet.

[0114] A computer system may include a client and a server, which are generally remote from each other and typically interact through a communication network. The client and server relationship arises through computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server in a distributed system, or a server in conjunction with a blockchain.

[0115] In summary, the present invention uses a laser radar to scan the hot delivery area from different angles, and can obtain high-precision three-dimensional imaging data, which provides a solid foundation for subsequent processing and analysis. Point cloud data is extracted from the three-dimensional imaging data, and a point cloud model is formed through registration, splicing and fine registration. This process realizes the automatic modeling of the scene, reduces manual intervention, and uses a bilateral filtering algorithm for noise reduction, a subdivision algorithm to form a triangular mesh, and NURBS surface modeling technology to further improve the fineness and realism of the model, providing a more reliable basis for subsequent real-time data analysis. It also records the data of the slab on the roller in real time, including the current position and the end position, so that managers can grasp the transportation and processing status of the slab at any time, analyze the slab roller data, and send corresponding instructions to the pusher equipment and turntable equipment, thereby realizing the automatic control of the slab processing in the set stacking order, improving production efficiency and the intelligent level of management.

[0116] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions merely illustrate the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A steelmaking slab intelligent control method based on visual three-dimensional imaging, characterized in that: include: S101, scanning the hot delivery area from different angles using several sets of laser radars to obtain three-dimensional imaging data of the slab, stacking platform, turntable, and roller conveyor; S102, preprocessing the three-dimensional imaging data, wherein the preprocessing includes filling in missing imaging data to obtain complete three-dimensional imaging data; S103, based on the complete 3D imaging data, extracting point cloud data from it, adjusting the initial position of the point cloud, and using a preset algorithm to align and stitch the point cloud blocks, and then using an iterative closest point method to accurately align the point cloud data to form a point cloud model; S104, using a bilateral filtering algorithm to perform noise reduction on the point cloud model, using a subdivision algorithm to form a triangular mesh, and using NURBS surface modeling technology to generate a heat delivery area scene model; S105. In the hot delivery area scene modeling, the data of the slabs on the roller conveyor are recorded in real time, including the current position and the end position of each slab on the roller conveyor; S106: Analyze the slab roller table data and send corresponding instructions to the pusher equipment and the turntable equipment to process the slabs in the set stacking order.

2. The method for intelligent control of steelmaking slabs based on visual three-dimensional imaging according to claim 1, characterized in that: The filling of the missing imaging data specifically includes the following steps: S201, arranging all imaging data into a column in the form of coordinates (x, y); S202, retrieve the coordinates (x, y) of all imaging data; S203: If there is a gap in x or y, fill in the gap with data.

3. The method for intelligent control of steelmaking slabs based on visual three-dimensional imaging according to claim 2, characterized in that: The method of adjusting the initial position of the point cloud specifically includes the following steps: Build the target model: Where q i and P i Represents the matching point pair, R and T represent the rotation and translation matrices respectively, and m represents the number of point clouds; Calculate the initial point cloud P i (i=1,2,3,...,m) in the target point cloud q i The closest point of (i=1,2,3,...,m); Calculate the rotation and translation matrices R and T so that the target model smallest; By rotating and translating the matrices R and T, we can exchange the overlapping point cloud P to obtain a new point cloud P. ， ; Calculate P ， Distance to the target point cloud: When the distance D is less than the preset error and exceeds the target number of iterations, the iteration is stopped. Otherwise, the calculation returns to the first step and continues until the convergence condition is met.

4. The method for intelligent control of steelmaking slabs based on visual three-dimensional imaging according to claim 3, characterized in that: The preset algorithm is: Where m i is the default value, l i is the distance difference after point cloud position transformation.

5. The method for intelligent control of steelmaking slabs based on visual three-dimensional imaging according to claim 4, characterized in that: The bilateral filtering algorithm is: Where n is the normal vector of data point O, and α represent the data points and bilateral filter factors after filtering, respectively. The representative formula of α is: Where x = PP i Represents the distance between data points, m represents the M nearest neighbors Nm(P i ), the number of data points in p and θ c is the M nearest neighbors Nm(P i )Normal height length and internal tangent plane Gaussian filtering.

6. The method for intelligent control of steelmaking slabs based on visual three-dimensional imaging according to claim 5, characterized in that: The NURBS surface modeling technology is used to generate the heat delivery area scene modeling, which specifically includes the following steps: S301. In the modeling software, select NURBS surface modeling tool, such as Maya, 3ds Max or UG; S302, creating a surface of the hot delivery area using a selected surface generation tool according to the point cloud model curve; S303, edit the curve, including modifying the curvature, direction, length, etc., to ensure that the curve meets the shape requirements of the hot delivery area; S304: Output the generated hot delivery area scene model to a file format, such as OBJ, STL, etc., for subsequent application.

7. The method for intelligent control of steelmaking slabs based on visual three-dimensional imaging according to claim 6, characterized in that: The analysis based on the slab roller table data specifically includes the following steps: S401, starting from the slab corresponding to the first processing step number on the stacking platform, check each slab in turn; S402: Determine whether the currently checked slab meets the loading conditions. If so, analyze the target position of the current slab, update the target position of the next slab as the final target position, and send instructions to the pusher and turntable to control them to suspend operation in preparation for loading the current slab. S403, determining the span and corresponding loading roller conveyor according to the order of the current slab on the stacking platform, the span in which it is located, and the corresponding loading roller conveyor, and sending a command to the turntable device in the span in which the current slab is located to control the turntable device to transfer the current slab to its corresponding loading roller conveyor; S404: After the loading preparation of the current slab is completed, the next slab that has not yet been loaded is checked in sequence, and the above steps are repeated until all slabs are checked.

8. The method for intelligent control of steelmaking slabs based on visual three-dimensional imaging according to claim 7, characterized in that: Determining whether the loading conditions are met specifically includes the following steps: If the slab before the current slab needs to pass through the loading roller corresponding to the current slab but has not passed through the loading roller, it is determined that it does not meet the loading conditions; otherwise, it is determined that it meets the loading conditions.

9. An intelligent control system for steelmaking slabs based on visual three-dimensional imaging, used to implement the intelligent control method for steelmaking slabs based on visual three-dimensional imaging according to any one of claims 1 to 8, characterized in that: include: A scanning module is used to scan the hot delivery area from different angles using several sets of laser radars to obtain three-dimensional imaging data of the slab, stacking platform, turntable and roller conveyor; A preprocessing module, the preprocessing module is used to preprocess the three-dimensional imaging data, the preprocessing includes filling in missing imaging data to obtain complete three-dimensional imaging data; A stitching module is used to extract point cloud data from the complete three-dimensional imaging data, adjust the initial position of the point cloud, and use a preset algorithm to align and stitch the point cloud blocks. Then, an iterative closest point method is used to precisely align the point cloud data to form a point cloud model. A modeling module, which is used to perform noise reduction processing on the point cloud model using a bilateral filtering algorithm, form a triangular mesh using a subdivision algorithm, and generate a heat delivery area scene model using NURBS surface modeling technology; A recording module, which is used to record the data of the slabs on the roller conveyor in real time during the hot delivery area scene modeling, including the current position and the end position of each slab on the roller conveyor; The control module is used to analyze the slab roller data and send corresponding instructions to the pusher equipment and turntable equipment to realize the processing of the slabs in the set stacking order.

10. An electronic device comprising at least one processor; and a memory communicatively connected to the at least one processor; characterized in that: The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.

Images