Intelligent scanning method and device for storage yard materials

Abstract

The invention provides a storage yard material intelligent scanning method and device, and belongs to the technical field of automatic scanning, and the method comprises the steps: responding to the received scanning information of a plurality of laser scanners, and determining the point cloud data of a target region of a storage yard and the point cloud data of a ground unloading region of the storage yard based on the scanning information; determining three-dimensional coordinates of the storage yard based on the point cloud data of the target area of the storage yard and the point cloud data of the ground unloading area of the storage yard; determining a point cloud model of the storage yard based on the three-dimensional coordinates of the storage yard; and determining an operation strategy of the storage yard according to the point cloud model. According to the intelligent storage yard material scanning method and device provided by the invention, the accuracy of storage yard material scanning can be improved, and the requirement of modern storage yard material scanning can be met.

Description

Technical Field

[0001] The present invention belongs to the field of automated scanning technology, and more specifically, relates to a method and device for intelligently scanning materials in a storage yard. Background Art

[0002] Accurate measurement and monitoring of materials are essential in various storage yards, such as ore storage yards, coal storage yards, and construction material storage yards. Since the unloading and grabbing points of bulk cargo terminal machinery are not fixed, and the materials in the storage yard are in real-time dynamic changes as the operation proceeds, the existing scanning methods are not adaptable to complex storage yards. Therefore, traditional scanning methods will produce large errors and are difficult to meet the needs of modern storage yard material scanning. Summary of the invention

[0003] The purpose of the present invention is to provide a method and device for intelligent scanning of storage yard materials, so as to improve the accuracy of storage yard material scanning.

[0004] A first aspect of an embodiment of the present disclosure provides a method for intelligently scanning materials in a storage yard, comprising: Determine the target three-dimensional coordinates based on the first point cloud data and the second point cloud data, the first point cloud data is the point cloud data of the target area of ​​the stockyard, the second point cloud data is the point cloud data of the ground unloading area of ​​the stockyard, and the target three-dimensional coordinates are the three-dimensional coordinates of the stockyard; Determine the point cloud model of the storage yard based on the target three-dimensional coordinates; Determine the yard's operating strategy based on the point cloud model.

[0005] A second aspect of the embodiments of the present disclosure provides an intelligent scanning device for materials in a storage yard, comprising: A coordinate acquisition module, used to determine the target three-dimensional coordinates based on the first point cloud data and the second point cloud data, wherein the first point cloud data is the point cloud data of the target area of ​​the stockyard, the second point cloud data is the point cloud data of the ground unloading area of ​​the stockyard, and the target three-dimensional coordinates are the three-dimensional coordinates of the stockyard; A model building module, used to determine the point cloud model of the storage yard based on the target three-dimensional coordinates; The strategy selection module is used to determine the operation strategy of the yard based on the point cloud model.

[0006] According to a third aspect of an embodiment of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of the above-mentioned method for intelligent scanning of materials in a yard when executing the computer program.

[0007] According to a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the steps of the above-mentioned method for intelligent scanning of materials in a yard are implemented.

[0008] The beneficial effects of the intelligent scanning method and device for storage yard materials provided by the embodiments of the present disclosure are: By accurately integrating the first point cloud data and the second point cloud data to determine the target three-dimensional coordinates, and performing data preprocessing and other operations, an accurate data basis is provided for subsequent processing, ensuring the reliability of subsequent analysis. The point cloud model constructed based on the target three-dimensional coordinates can fully and accurately reflect the real-time status of the yard, which helps the yard manager to intuitively understand the material distribution and space utilization, so that every decision can be based on real and accurate data. The operation strategy formulated according to the point cloud model can optimize the material loading and unloading process, significantly improve the yard operation efficiency, and enhance the automation and intelligence level of yard management. This design not only greatly improves the accuracy of the yard material scanning results, but also meets the needs of modern yard material scanning. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic diagram of a flow chart of a method for intelligently scanning materials in a storage yard provided by an embodiment of the present disclosure; Figure 2 A structural block diagram of a storage yard material intelligent scanning device provided by an embodiment of the present disclosure; Figure 3 A schematic block diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0011] In the following description, specific details such as specific system structures and technologies are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, it should be clear to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present disclosure with unnecessary details.

[0012] In order to make the purpose, technical solutions and advantages of the present disclosure more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0013] Please refer to Figure 1 , Figure 1 A schematic diagram of a flow chart of a method for intelligently scanning materials in a storage yard provided in an embodiment of the present disclosure, the method comprising: S101: Determine the target three-dimensional coordinates based on the first point cloud data and the second point cloud data, the first point cloud data is the point cloud data of the target area of ​​the yard, the second point cloud data is the point cloud data of the ground unloading area of ​​the yard, and the target three-dimensional coordinates are the three-dimensional coordinates of the yard.

[0014] In this embodiment, point cloud data refers to a collection of a large number of discrete points in a three-dimensional space. These points usually contain location information (usually expressed in the form of X, Y, and Z coordinates), and in some cases may also contain other attribute information such as color, reflection intensity, and normal. Point cloud data can be obtained in a variety of ways, such as three-dimensional laser scanning, photogrammetry, depth cameras and other technologies. The yard includes a target area and a ground unloading area. The target area of ​​the yard may include areas such as the position of the ship's hatch, the height of the ship's side, the height of the hatch cover, the material position, and the loading and unloading position. The ground unloading area of ​​the yard may include areas such as the ground material position.

[0015] The first point cloud data provides point cloud data of the target area of ​​the storage yard, and the point cloud data includes the three-dimensional coordinates of each point on the surface of the object in the target area. The second point cloud data provides point cloud data of the ground unloading area of ​​the storage yard, and the point cloud data includes the three-dimensional coordinates of each point on the surface of the object in the ground unloading area.

[0016] Finally, by integrating the first point cloud data and the second point cloud data, the target three-dimensional coordinates, that is, the three-dimensional coordinates of the yard, are obtained.

[0017] Exemplarily, the first point cloud data and the second point cloud data may be subjected to data preprocessing. The data preprocessing process may include denoising, downsampling, and coordinate normalization. Denoising refers to removing noise points in the first point cloud data or the second point cloud data by filtering and other methods, which can improve the accuracy of the point cloud model. During the acquisition process of the first point cloud data or the second point cloud data, noise may be introduced due to factors such as measurement errors or environmental interference. Downsampling refers to reducing the amount of the first point cloud data and the second point cloud data, thereby improving computing efficiency while retaining key information features. If the amount of the first point cloud data or the second point cloud data is too large, downsampling is required. Coordinate normalization refers to normalizing the coordinate range of the first point cloud data or the second point cloud data to a standard range for subsequent processing.

[0018] S102: Determine a point cloud model of the storage yard based on the target three-dimensional coordinates.

[0019] In this embodiment, first, the predicted area volume of the target area and the three-dimensional coordinates of the predicted area can be calculated according to the target three-dimensional coordinates. The actual area volume of the target area can be obtained according to the real design drawings (such as the design drawings of the ship), and the three-dimensional coordinates of the actual area can also be obtained.

[0020] Secondly, the overlap between the predicted area volume and the actual area volume is calculated. If the overlap is higher, the accuracy of the target three-dimensional coordinates of this embodiment is higher; the target distance between the three-dimensional coordinates of the predicted area and the three-dimensional coordinates of the actual area is calculated. If the target distance is smaller, the positioning of the target three-dimensional coordinates of this embodiment is more accurate. Afterwards, a comprehensive loss function can be generated based on the weighted calculation of the overlap and the target distance to measure the performance of the point cloud model.

[0021] Then, choose a suitable model architecture, for example, deep learning-based methods such as PointNet and its derivative models.

[0022] For example, the input layer of the PointNet model receives the target 3D coordinates; the feature extraction layer uses a multi-layer perceptron to perform independent feature extraction for each target 3D coordinate. The multi-layer perceptron can learn the local features of each target 3D coordinate; the symmetric function aggregation layer uses a symmetric function (such as maximum pooling) to aggregate the features of all target 3D coordinates in order to obtain global features that are independent of the order of the target 3D coordinates, since the target 3D coordinates are unordered. The maximum pooling operation can capture the most significant features in the target 3D coordinates, thereby obtaining a global feature vector that is invariant to the order of the input target 3D coordinates; the fully connected layer inputs the aggregated global feature vector into the fully connected layer for further feature learning and prediction of tasks such as classification or segmentation. For example, in the classification task, the probability distribution of the category to which the object belongs is finally given through the output layer. In the segmentation task, the corresponding category label will be output for each target 3D coordinate.

[0023] Finally, after the model architecture is determined, the model is trained based on the historical target three-dimensional coordinates and the comprehensive loss function. During the training process, the parameters and model are continuously optimized according to the comprehensive loss function to construct a point cloud model.

[0024] S103: Determine the operation strategy of the storage yard according to the point cloud model.

[0025] In this embodiment, the operation strategy of the preset yard includes multiple operation strategies, such as the gantry crane loading and unloading operation process, path planning, and material stacking analysis. The actual situation of the ship, materials, etc. in the yard is analyzed by the point cloud model, and the type, location, and height of the material can be considered based on this to determine the operation target point of the gantry crane (including the grabbing point and the unloading point), and the grabbing and unloading can be carried out in sequence, and the driving path of the gantry crane can also be planned.

[0026] For example, the free space distribution and size of the yard in the point cloud model are analyzed to determine the best storage location for newly arrived goods, so as to maximize the use of the yard space and avoid congestion and chaos, thereby achieving space planning. Special storage areas can also be divided according to the characteristics of different materials (such as weight, size, etc.) and storage requirements to improve storage efficiency and safety.

[0027] For example, based on the stacking height, shape and position of the materials in the point cloud model, the operation path of the gantry crane is planned to reduce the operation time and energy consumption. The optimal working position and angle of the gantry crane is determined to avoid collision with other materials or yard facilities.

[0028] From the above, it can be concluded that by accurately integrating the first point cloud data and the second point cloud data to determine the target three-dimensional coordinates and performing data preprocessing and other operations, an accurate data basis is provided for subsequent processing, ensuring the reliability of subsequent analysis. The point cloud model constructed based on the target three-dimensional coordinates can fully and accurately reflect the real-time status of the yard, which helps the yard manager to intuitively understand the material distribution and space utilization, so that every decision can be based on real and accurate data. The operation strategy formulated according to the point cloud model can optimize the material loading and unloading process, significantly improve the yard operation efficiency, and enhance the automation and intelligence level of yard management. This design not only greatly improves the accuracy of the yard material scanning results, but also meets the needs of modern yard material scanning.

[0029] In one embodiment of the present disclosure, a method for intelligently scanning materials in a storage yard further includes: In response to receiving first scanning information sent by the first laser scanner, determining first point cloud data based on the first scanning information; The first laser scanner is arranged in a target area, and the target area includes at least one area among a ship hatch area, a ship side area, a cabin cover area and a loading and unloading area.

[0030] In this embodiment, a laser scanner is used to emit a laser beam to a target object, and then the time from the laser emission to the laser being reflected by the object is measured, thereby calculating the distance between each point on the surface of the object and the scanner. By continuously changing the laser emission angle, a large amount of point data can be obtained to form a point cloud.

[0031] After receiving the first scanning information sent by the first laser scanner set in the specific target area, this embodiment will perform corresponding processing operations to determine the first point cloud data.

[0032] The first laser scanner is set in the target area, which covers at least one of the ship hatch area, the ship side area, the cabin cover area and the loading and unloading area. When the first laser scanner is working, it will continuously emit lasers to these areas and receive reflected signals to form first scanning information.

[0033] After receiving the first scan information, this embodiment uses downsampling to process the first scan information, including reducing the amount of data and extracting main features, because the target area covers a large range. Then, the first scan information is converted into first point cloud data that can intuitively present the spatial form and material distribution of the target area.

[0034] Through the first point cloud data, we can clearly understand the corresponding three-dimensional coordinates of the size and position of the ship's hatch, the shape of the ship's side, the status of the cabin cover, and the accumulation of materials in the loading and unloading area.

[0035] Exemplarily, the first laser scanner collects data from the target area, performs noise reduction on the point cloud data, matches feature points in the noise-reduced data in combination with the characteristic structure of the target to be identified, and defines the position of the target object in the space in the laser viewing angle through the feature points. Combined with spatial coordinate conversion, the spatial coordinate system under the laser viewing angle is converted into the actual earth material coordinate system to accurately locate the actual physical coordinate position of feature targets including hatches, ship decks, material areas, etc.

[0036] From the above, it can be concluded that by setting up the first laser scanner in the key target area, efficient and accurate capture of material distribution is achieved. The first point cloud data constructed based on the first scanning information not only improves the accuracy of material identification, but also significantly improves the level of intelligence in logistics operations. This design enables accurate and real-time monitoring of these key areas, providing an important data foundation for material management in the yard, planning and scheduling of loading and unloading operations, and safe and efficient execution of ship-related operations.

[0037] In one embodiment of the present disclosure, a method for intelligently scanning materials in a storage yard further includes: In response to receiving second scanning information sent by a second laser scanner, determining second point cloud data based on the second scanning information; Wherein, the second laser scanner is arranged in the ground unloading area.

[0038] In this embodiment, the second laser scanner is set in the ground unloading area to continuously scan the objects in the ground unloading area. The second laser scanner is used to emit a laser beam to the object, and then the time from the laser emission to the laser being reflected by the object is measured, so as to calculate the distance between each point on the surface of the object and the scanner. By continuously changing the emission angle of the laser, a large amount of point data can be obtained to form a point cloud.

[0039] When the second scanning information sent by the second laser scanner is received, this embodiment analyzes and processes the second scanning information to determine the second point cloud data.

[0040] Exemplarily, when the second laser scanner is working, it continuously emits laser into the unloading area on the ground and receives reflected signals to form second scanning information.

[0041] After receiving the second scanning information, this embodiment may use a noise reduction technology to remove noise or interference information in the second scanning information, thereby determining the second point cloud data.

[0042] Through the second point cloud data, the three-dimensional coordinates corresponding to the accumulation of materials in the ground unloading area can be clearly understood.

[0043] For example, the second point cloud data can display three-dimensional coordinate information such as the terrain undulations of the ground unloading area, the distribution of materials, and the location of staff.

[0044] It can be concluded from the above that by introducing a second laser scanner and setting it up in the ground unloading area, this method realizes real-time and accurate monitoring of the ground unloading area. The second point cloud data constructed based on the second scanning information supplements the spatial distribution information of the material during the unloading stage. This collaborative working mode of multiple laser scanners significantly improves the real-time and accuracy of material management, and provides yard managers with more comprehensive material status information, which helps to optimize resource allocation, improve work efficiency, and reduce potential risks caused by information asymmetry.

[0045] In one embodiment of the present disclosure, determining a point cloud model of a storage yard based on target three-dimensional coordinates includes: Determine the comprehensive loss function of the storage yard based on the target three-dimensional coordinates; Construct a point cloud model of the storage yard based on the comprehensive loss function.

[0046] In this embodiment, the determination of the comprehensive loss function usually takes multiple factors into consideration. For example, the difference between the predicted three-dimensional coordinates and the target three-dimensional coordinates will be considered, including position deviation, shape deviation, density deviation, etc. This embodiment can calculate the overlap between the predicted area volume and the actual area volume, and calculate the target distance between the target point of the predicted area and the target point of the actual area, thereby obtaining a comprehensive loss function based on the weighted calculation of the overlap and the target distance. Then, a point cloud model of the yard is constructed according to the comprehensive loss function. During the construction process, the value of the loss function is minimized by continuously adjusting the parameters of the model. This process indicates that the model is constantly being optimized and can accurately match the actual situation of the yard represented by the target three-dimensional coordinates.

[0047] For example, if the comprehensive loss function mainly focuses on the position difference between the position of each point in the point cloud model and the target three-dimensional coordinates, then when building the model, the model will continuously adjust the position of the point to reduce the position difference, thereby achieving the purpose of reducing the loss function value.

[0048] From the above, it can be concluded that by constructing a comprehensive loss function and optimizing the point cloud model accordingly, the accuracy and applicability of the model are significantly improved. This method not only ensures that the point cloud data is highly consistent with the actual material distribution, but also reduces the model error through the optimization algorithm, providing more refined and reliable data support for yard material management.

[0049] In one embodiment of the present disclosure, determining a comprehensive loss function of a storage yard based on target three-dimensional coordinates includes: Determine the standard area volume of the target area based on the reference three-dimensional coordinates; the reference three-dimensional coordinates are the real three-dimensional coordinates of the target area; Determine a predicted region volume of a target region based on the target three-dimensional coordinates; Determine the regional loss function of the target region based on the standard region volume and the predicted region volume; The comprehensive loss function of the storage yard is determined based on the regional loss function.

[0050] In this embodiment, the target area includes at least one area including the ship hatch position, ship side height, hatch cover height, material position and loading and unloading position. The reference three-dimensional coordinates are the real three-dimensional coordinates of the target area. According to the reference three-dimensional coordinates, the standard area volume of the target area can be accurately calculated, and the volume reflects the actual size of the target area.

[0051] For example, the design of a ship has corresponding drawings and three-dimensional images, and based on this information, the real three-dimensional coordinates of the ship's hatch position, the ship's side height, and the hatch cover height can be obtained.

[0052] Secondly, the predicted region volume of the target region is calculated based on the given three-dimensional coordinates of the target, where the predicted region volume is an estimated value obtained according to the current model or algorithm.

[0053] Then, the regional loss function of the target region can be determined based on the standard region volume and the predicted region volume. The regional loss function is usually a mathematical expression that can quantify the degree of difference between the two. For example, the absolute value of the volume difference, the square of the difference, or a more complex mathematical relationship can be used to express the difference.

[0054] Finally, the comprehensive loss function of the entire yard is determined based on the regional loss function of the target area. In this case, not only the target area is considered, but also the loss of other related areas or factors in the yard. For example, the loss of different target areas and different types of errors (position error, shape error, etc.) are comprehensively considered, and a comprehensive loss function that can fully reflect the overall situation of the yard is formed through a certain weighting or combination method.

[0055] From the above, it can be concluded that by accurately calculating the difference between the standard area volume and the predicted area volume, and constructing the regional loss function based on this, a comprehensive loss assessment system for the yard is formed, which not only improves the estimation accuracy of the point cloud data of the yard, but also enhances the flexibility and response speed of yard management.

[0056] In one embodiment of the present disclosure, determining a region loss function of a target region based on a standard region volume and a predicted region volume includes: Calculate the overlap between the standard region volume and the predicted region volume as the first loss parameter of the target region; Calculate the target distance between the reference three-dimensional coordinates and the target three-dimensional coordinates as the second loss parameter of the target area; A region loss function of the target region is determined based on the first loss parameter and the second loss parameter.

[0057] In this embodiment, the overlap can reflect the degree of spatial overlap between the predicted area volume and the standard area volume. For example, if the overlap is high, it means that the prediction is more accurate; if the overlap is low, it means that there is a large deviation between the prediction and the actual. The target distance can measure the degree of deviation between the predicted position and the actual position.

[0058] The overlap is used as the first loss parameter of the target area, and the target distance is used as the second loss parameter of the target area, so as to determine the regional loss function of the target area. For example, the first loss parameter and the second loss parameter are combined by some mathematical combination, such as weighted summation, multiplication, or a more complex functional relationship to form a regional loss function that can comprehensively evaluate the difference between the prediction result and the actual situation.

[0059] Exemplarily, the calculation formula of overlap is:

[0060] in, represents the degree of overlap, P represents the predicted region volume, and Q represents the standard region volume.

[0061] Assume that the diagonal vertices of the prediction area are and The diagonal vertices of the standard area are and , first calculate , , Overlap length in three directions: exist In direction,

[0062] exist In direction,

[0063] exist In direction,

[0064] If the overlap lengths in all three directions are greater than 0, the intersection volume for:

[0065]

[0066]

[0067]

[0068] in, , , and express The coordinate value in the direction, , , and express The coordinate value in the direction, , , and express The coordinate value in the direction, express The overlap length in the direction, express The overlap length in the direction, express The overlap length in the direction, Indicates prediction area, Indicates Standard area, and One to one correspondence.

[0069] Exemplarily, the target distance is calculated as:

[0070] in, Represents the target distance, and the three-dimensional coordinate points of the prediction area are , the three-dimensional coordinate points of the standard area are , and express The coordinate value in the direction, and express The coordinate value in the direction, and express The coordinate value in the direction, Indicates prediction area, Indicates Standard area, and One to one correspondence, express Axis coordinates, express Axis coordinates, express Axis coordinates.

[0071] Exemplarily, the calculation formula of the regional loss function is:

[0072] in, represents the regional loss function, represents the weight of overlap, The weight representing the target distance.

[0073] From the above, it can be concluded that this embodiment significantly improves the accuracy of target detection and positioning by introducing dual loss parameters to construct the regional loss function of the target area. First, by calculating the overlap between the standard area volume and the predicted area volume as the first loss parameter, the degree of fit between the predicted area and the actual target area is effectively measured, which promotes the model's accurate estimation of the target shape and size. Secondly, using the target distance between the reference three-dimensional coordinates and the target three-dimensional coordinates as the second loss parameter, the positioning accuracy of the target in three-dimensional space is further refined to ensure the spatial consistency of the detection results. This dual loss mechanism not only enhances the robustness of the model for target recognition in complex scenes, but also promotes the rapid convergence of the model during training.

[0074] In one embodiment of the present disclosure, determining a comprehensive loss function of a storage yard based on a regional loss function includes: Calculate the proportion of the target area to the yard; The regional loss function is weighted based on the proportion to determine the comprehensive loss function of the yard.

[0075] In this embodiment, the proportion of the target area in the entire storage yard is calculated, that is, the proportion of the target area in the storage yard. This proportion can be obtained by comparing the area, volume or other related metrics of the target area with the corresponding metrics of the entire storage yard.

[0076] Then, based on the calculated proportion, the regional loss function of the target area is weighted. By determining the comprehensive loss function of the yard in this weighted manner, the importance and impact of different areas in the yard can be more reasonably considered.

[0077] For example, assume that the region loss function of the target region is , which accounts for , then the comprehensive loss function of the yard is It can be expressed as .

[0078] From the above, it can be concluded that by calculating the proportion of the target area in the yard, the impact of different areas on the overall performance of the yard can be more accurately evaluated. Then, weighting the regional loss function based on the proportion can ensure that the comprehensive loss function more comprehensively reflects the actual situation of the yard. Different areas will have different importance and complexity. Through weighted processing, the performance of the point cloud model can be more reasonably optimized. Finally, constructing a comprehensive loss function for the yard helps to comprehensively consider multiple objectives during the model training process and improve the generalization ability and adaptability of the model. By minimizing the comprehensive loss function, the overall performance of the yard can be optimized, the risk of overfitting can be reduced, and the practicality and reliability of the model can be improved.

[0079] Corresponding to the intelligent scanning method of yard materials in the above embodiment, Figure 2 This is a structural block diagram of a storage yard material intelligent scanning device provided by an embodiment of the present disclosure. For the sake of convenience, only the parts related to the embodiment of the present disclosure are shown. Figure 2 The intelligent scanning device 20 for stockyard materials includes: a coordinate acquisition module 21, a model building module 22 and a strategy selection module 23. The coordinate acquisition module 21 is used to determine the target three-dimensional coordinates based on the first point cloud data and the second point cloud data, the first point cloud data is the point cloud data of the target area of ​​the stockyard, the second point cloud data is the point cloud data of the ground unloading area of ​​the stockyard, and the target three-dimensional coordinates are the three-dimensional coordinates of the stockyard; A model building module 22, for determining a point cloud model of the storage yard based on the target three-dimensional coordinates; The strategy selection module 23 is used to determine the operation strategy of the storage yard according to the point cloud model.

[0080] In one embodiment of the present disclosure, the model building module 22 is specifically used to: Determine the comprehensive loss function of the storage yard based on the target three-dimensional coordinates; Construct a point cloud model of the storage yard based on the comprehensive loss function.

[0081] In one embodiment of the present disclosure, the model building module 22 is further used to: Determine the standard area volume of the target area based on the reference three-dimensional coordinates; the reference three-dimensional coordinates are the real three-dimensional coordinates of the target area; Determine a predicted region volume of a target region based on the target three-dimensional coordinates; Determine the regional loss function of the target region based on the standard region volume and the predicted region volume; The comprehensive loss function of the storage yard is determined based on the regional loss function.

[0082] In one embodiment of the present disclosure, the model building module 22 is further configured to: Calculate the overlap between the standard region volume and the predicted region volume as the first loss parameter of the target region; Calculate the target distance between the reference three-dimensional coordinates and the target three-dimensional coordinates as the second loss parameter of the target area; A region loss function of the target region is determined based on the first loss parameter and the second loss parameter.

[0083] In one embodiment of the present disclosure, the model building module 22 is further used to: Calculate the proportion of the target area to the yard; The regional loss function is weighted based on the proportion to determine the comprehensive loss function of the yard.

[0084] In one embodiment of the present disclosure, a yard material intelligent scanning device 20 further includes: a first point cloud data acquisition module and a second point cloud data acquisition module; A first point cloud data acquisition module, configured to determine first point cloud data based on the first scanning information in response to receiving first scanning information sent by the first laser scanner; The first laser scanner is arranged in a target area, and the target area includes at least one area among a ship hatch area, a ship side area, a cabin cover area and a loading and unloading area.

[0085] A second point cloud data acquisition module, configured to determine second point cloud data based on the second scanning information in response to receiving second scanning information sent by the second laser scanner; Wherein, the second laser scanner is arranged in the ground unloading area.

[0086] See also Figure 3 , Figure 3 A schematic block diagram of an electronic device provided by an embodiment of the present disclosure. Figure 3 The electronic device 300 in the embodiment shown may include: one or more processors 301, one or more input devices 302, one or more output devices 303 and one or more memories 304. The processors 301, input devices 302, output devices 303 and memories 304 communicate with each other via a communication bus 305. The memory 304 is used to store computer programs, which include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to call the program instructions to execute the functions of each module / unit in the above-mentioned device embodiments, such as Figure 2 The functions of modules 21 to 23 are shown.

[0087] It should be understood that in the embodiment of the present disclosure, the processor 301 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0088] The input device 302 may include a touch panel, a fingerprint collection sensor (for collecting the user's fingerprint information and fingerprint direction information), a microphone, etc., and the output device 303 may include a display (LCD, etc.), a speaker, etc.

[0089] The memory 304 may include a read-only memory and a random access memory, and provide instructions and data to the processor 301. A portion of the memory 304 may also include a non-volatile random access memory. For example, the memory 304 may also store information about the device type.

[0090] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiments of the present disclosure can execute the implementation methods described in the first and second embodiments of a method for intelligent scanning of yard materials provided in the embodiments of the present disclosure, and can also execute the implementation methods of the electronic device described in the embodiments of the present disclosure, which will not be repeated here.

[0091] In another embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by the processor, all or part of the processes in the above-mentioned embodiment method are implemented, and the computer program can also be completed by instructing the relevant hardware through the computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0092] The computer-readable storage medium may be an internal storage unit of the electronic device of any of the aforementioned embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium may also be an external storage device of the electronic device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Furthermore, the computer-readable storage medium may also include both an internal storage unit of the electronic device and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0093] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0094] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the electronic devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0095] In the several embodiments provided in the present application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or it can be an electrical, mechanical or other form of connection.

[0096] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present disclosure.

[0097] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0098] The above are only specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present disclosure, and these modifications or replacements should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A method for intelligent scanning of materials in a storage yard, characterized in that: include: Determine the target three-dimensional coordinates based on the first point cloud data and the second point cloud data, wherein the first point cloud data is the point cloud data of the target area of ​​the stockyard, the second point cloud data is the point cloud data of the ground unloading area of ​​the stockyard, and the target three-dimensional coordinates are the three-dimensional coordinates of the stockyard; Determine a point cloud model of the storage yard based on the target three-dimensional coordinates; An operation strategy of the storage yard is determined according to the point cloud model.

2. The intelligent scanning method for storage yard materials according to claim 1, characterized in that: Also includes: In response to receiving first scanning information sent by a first laser scanner, determining first point cloud data based on the first scanning information; Wherein, the first laser scanner is arranged in the target area, and the target area includes at least one area of ​​a ship hatch area, a ship side area, a cabin cover area and a loading and unloading area.

3. The intelligent scanning method for storage yard materials according to claim 1, characterized in that: Also includes: In response to receiving second scanning information sent by a second laser scanner, determining second point cloud data based on the second scanning information; Wherein, the second laser scanner is arranged in the ground unloading area.

4. The intelligent scanning method for storage yard materials according to claim 1, characterized in that: Determining a point cloud model of the storage yard based on the target three-dimensional coordinates includes: Determining a comprehensive loss function of the storage yard based on the target three-dimensional coordinates; A point cloud model of the storage yard is constructed based on the comprehensive loss function.

5. The intelligent scanning method for storage yard materials according to claim 4, characterized in that: The comprehensive loss function of the storage yard is determined based on the target three-dimensional coordinates, including: Determine the standard area volume of the target area based on reference three-dimensional coordinates; the reference three-dimensional coordinates are the real three-dimensional coordinates of the target area; Determining a predicted area volume of the target area based on the target three-dimensional coordinates; Determine a region loss function of the target region based on the standard region volume and the predicted region volume; A comprehensive loss function of the storage yard is determined based on the regional loss function.

6. The intelligent scanning method for storage yard materials according to claim 5, characterized in that: The determining of the region loss function of the target region based on the standard region volume and the predicted region volume includes: Calculating the overlap between the standard region volume and the predicted region volume as a first loss parameter of the target region; Calculating a target distance between the reference three-dimensional coordinates and the target three-dimensional coordinates as a second loss parameter of the target area; A regional loss function of the target area is determined based on the first loss parameter and the second loss parameter.

7. The intelligent scanning method for storage yard materials according to claim 5, characterized in that: The determining of the comprehensive loss function of the storage yard based on the regional loss function comprises: Calculate the proportion of the target area to the yard; The regional loss function is weighted based on the proportion to determine a comprehensive loss function of the storage yard.

8. An intelligent scanning device for yard materials, characterized in that: include: A coordinate acquisition module, used to determine a target three-dimensional coordinate based on first point cloud data and second point cloud data, wherein the first point cloud data is point cloud data of a target area of ​​the stockyard, the second point cloud data is point cloud data of a ground unloading area of ​​the stockyard, and the target three-dimensional coordinate is the three-dimensional coordinate of the stockyard; A model building module, used for determining a point cloud model of the storage yard based on the target three-dimensional coordinates; A strategy selection module is used to determine the operation strategy of the storage yard according to the point cloud model.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images