A container model construction method based on three-dimensional software

Abstract

The present application relates to the technical field of container model, and particularly relates to a container model construction method based on three-dimensional software. The method comprises the following steps: obtaining point cloud data of a container through three-dimensional laser scanning, and performing feature profiling to generate a container feature set, performing spatial positioning on the features to obtain a vertex coordinate set, and performing geometric constraint modeling based on the vertex coordinate set to form a container skeleton model, obtaining a target cargo image, extracting a basic contour, performing three-dimensional reconstruction to generate a cargo model, performing spatial planning on the container skeleton based on the cargo model, and calculating a gravity center distribution to generate cargo balance parameters, thereafter, performing interference simulation and conflict detection to obtain interference conflict data, positioning a conflict focus area, and finally performing variant design on the container skeleton according to the conflict area to generate an adjusted container model. The present application realizes a more comprehensive and more accurate container model construction method.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of container modeling, and particularly relates to a container model construction method based on three-dimensional software. BACKGROUND

[0002] Traditional container loading methods often rely on manual experience, lack scientific planning and design, resulting in low space utilization and increased risk of damage during loading. In container loading, due to the diversity of cargo shape, size and weight, how to effectively plan the space to ensure the stability and safety of the cargo becomes a problem to be solved. The main problem in container loading design is the lack of efficient three-dimensional modeling and simulation tools, which makes it impossible to evaluate the rationality of the loading scheme in real time. Many existing methods rely on two-dimensional drawings or simple three-dimensional models, which cannot fully consider the actual shape of the goods and the complexity of the loading environment. In addition, traditional methods often rely on manual inspection or simple collision detection algorithms when performing interference detection and conflict analysis, which makes it difficult to comprehensively and accurately identify potential interference areas, further increasing the risk in the cargo loading process. SUMMARY

[0003] Therefore, it is necessary to provide a container model construction method based on three-dimensional software to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a container model construction method based on three-dimensional software includes the following steps:

[0005] Step S1: performing three-dimensional laser scanning on the container to obtain container point cloud data; performing feature segmentation on the container point cloud data to generate a container feature set;

[0006] Step S2: performing feature point space positioning on the container feature set to obtain a vertex coordinate set; performing geometric constraint modeling based on the vertex coordinate set to generate a container skeleton model;

[0007] Step S3: obtaining a target cargo image; performing basic contour extraction on the target cargo image to generate a cargo basic contour image; performing three-dimensional reconstruction on the cargo basic contour image to obtain a reconstructed cargo model;

[0008] Step S4: performing space planning processing on the container skeleton model based on the reconstructed cargo model to obtain an initial cargo planning model; performing barycenter distribution calculation on the initial cargo planning model to generate cargo balance parameters;

[0009] Step S5: performing interference simulation on the initial cargo planning model based on the cargo balance parameters to obtain simulated cargo interference data; performing conflict detection on the simulated cargo interference data to generate interference conflict data;

[0010] Step S6: conflict focus positioning is performed on the interference conflict data to obtain a conflict focus region; and a variant design is performed on the container skeleton model according to the conflict focus region to generate an adjusted container model.

[0011] The present application ensures high-precision point cloud data acquisition through the implementation of three-dimensional laser scanning of the container, and the container feature set generated through feature segmentation of the point cloud data can comprehensively reflect the geometric shape and structural features of the container. The implementation of feature point spatial positioning on the container feature set ensures accurate positioning of each feature point, and the vertex coordinate set generated provides necessary data support for three-dimensional modeling. The operation of geometric constraint modeling based on the vertex coordinate set can generate a container skeleton model. The implementation of obtaining a target cargo image and extracting a basic contour ensures accurate capture of the shape features of the cargo, and the basic contour image of the cargo generated provides necessary visual information for subsequent three-dimensional reconstruction. Through three-dimensional reconstruction of the contour image, the reconstructed cargo model can truly reflect the spatial occupation of the cargo, providing accurate data support for container loading. The implementation of spatial planning processing of the container skeleton model based on the reconstructed cargo model ensures efficient arrangement of the layout of the cargo in the container. The initial cargo planning model generated provides a basis for subsequent balance analysis. The result of the center of gravity distribution calculation can quantify the balance parameters of the cargo. The implementation of interference simulation based on the cargo balance parameters can identify potential conflicts between the cargo and between the cargo and the container. The simulation cargo interference data generated provides detailed information support for subsequent conflict detection. The interference conflict data generated through conflict detection can efficiently identify problems in the loading process, ensuring that the scheme can be adjusted in time in actual operation, optimizing the loading layout of the cargo. The implementation of conflict focus positioning on the interference conflict data ensures that the specific conflict region can be determined, providing an important basis for subsequent variant design of the container skeleton model. The model adjustment according to the conflict focus region can effectively avoid interference problems in the loading process. The adjusted container model generated can better adapt to actual loading requirements, ultimately improving the use efficiency of the container and the safety of cargo transportation.

[0012] Preferably, step S1 comprises the following steps:

[0013] Step S11: The container is scanned from multiple angles to obtain original container point cloud data; and noise filtering is performed on the original container point cloud data to generate filtered point cloud data.

[0014] Step S12: Coordinate conversion processing is performed on the filtered point cloud data to generate container point cloud data.

[0015] Step S13: geometric feature point recognition is performed on the container point cloud data to obtain a container feature point set; topological relationship analysis is performed on the container feature point set to obtain feature point topological relationship;

[0016] Step S14: feature point connection is performed on the container feature point set based on the feature point topological relationship to obtain a feature point connection graph; geometric feature analysis is performed on the feature point connection graph to generate a container feature set.

[0017] The present application can comprehensively capture the three-dimensional shape of the container through multi-angle scanning, remove invalid point data through noise filtering, improve the precision and quality of the point cloud data, and through coordinate conversion processing, the container point cloud data can be unified in a coordinate system, eliminate coordinate deviation caused by different scanning devices or environmental changes, geometric feature point recognition can accurately extract the key geometric features of the container, topological relationship analysis reveals the spatial correlation between feature points, through feature point connection and geometric feature analysis, a detailed feature map of the container can be formed, the geometric structure of the container is accurately described, the accuracy and operability of the three-dimensional model are ensured, and the efficiency and quality of model construction are improved.

[0018] Preferably, step S2 comprises the following steps:

[0019] Step S21: spatial coordinate mapping is performed on the container feature set to obtain a three-dimensional coordinate point set; vertex clustering analysis is performed on the three-dimensional coordinate point set to generate a candidate vertex set;

[0020] Step S22: accurate positioning calculation is performed on the candidate vertex set according to the three-dimensional coordinate point set to obtain a vertex coordinate set;

[0021] Step S23: geometric relationship extension is performed on the vertex coordinate set to generate surface domain relationship data; geometric constraint analysis is performed on the surface domain relationship data to generate geometric constraint data;

[0022] Step S24: a framework structure is established based on the geometric constraint data to obtain an initial skeleton model; node connection filling is performed on the initial skeleton model according to the surface domain relationship data to generate a container skeleton model.

[0023] The application ensures effective conversion of feature points into a three-dimensional coordinate point set by implementing spatial coordinate mapping of a feature set of a container, and the generated three-dimensional coordinate point set provides basic data for subsequent analysis and modeling. Through vertex clustering analysis of the three-dimensional coordinate point set, the obtained candidate vertex set can efficiently identify important geometric features. According to the implementation of accurate positioning calculation of the candidate vertex set based on the three-dimensional coordinate point set, the accurate coordinates of each candidate vertex can be ensured, and the obtained vertex coordinate set provides accurate data support for geometric modeling. The accurate positioning result can reduce errors in the subsequent modeling process and improve the overall accuracy of the model. The implementation of geometric relationship extension of the vertex coordinate set ensures comprehensive analysis of the spatial relationship between vertices, and the generated surface domain relationship data provides necessary information support for subsequent geometric constraint analysis. Through geometric constraint analysis of the surface domain relationship data, the generated geometric constraint data can clearly define the constraint relationship between each feature point. The implementation of establishing a framework structure based on the geometric constraint data can effectively construct an initial skeleton model of the container, ensuring that the model can truly reflect the geometric features of the container. Through the operation of connecting and filling nodes of the initial skeleton model based on the surface domain relationship data, the generated container skeleton model is more complete and has practical application value, improving the structural strength and stability of the model.

[0024] Preferably, step S23 comprises the following steps:

[0025] Adjacent relationship analysis is performed on the vertex coordinate point set to obtain a point pair relationship set, and edge line connection processing is performed on the point pair relationship set to generate edge line data;

[0026] Parallel degree detection is performed on the edge line data to obtain parallel edge data, and perpendicular degree detection is performed on the parallel edge data to generate an orthogonal edge data set;

[0027] Surface element relationship construction is performed based on the orthogonal edge data set to generate surface domain relationship data;

[0028] Dimension parameter extraction is performed on the surface domain relationship data to obtain dimension constraint data;

[0029] Angle relationship analysis is performed on the dimension constraint data to generate angle constraint data, and symmetry calculation is performed on the angle constraint data to obtain symmetry constraint data;

[0030] Constraint integration is performed on the dimension constraint data, the angle constraint data and the symmetry constraint data to generate geometric constraint data.

[0031] The application ensures that the correlation between each vertex can be identified through the implementation of the adjacency relationship analysis on the vertex coordinate point set, and the generated point pair relationship set provides a basis for subsequent edge line generation, and the implementation of the edge line connection processing can efficiently construct the edge line data connecting each vertex, ensures the geometric structure integrity of the model, through the parallel degree detection on the edge line data, the parallel edge data obtained provides an important basis for subsequent geometric analysis, the implementation of the perpendicularity detection generates the orthogonal edge data set, which lays a foundation for the construction of the surface element relationship, the implementation of the surface element relationship construction based on the orthogonal edge data set can accurately form the surface domain relationship data, ensures that the geometric features of the model can be clearly reflected, the implementation of the size parameter extraction ensures that the key size information of the surface domain can be obtained, the generated size constraint data provides important support for subsequent geometric constraint analysis, the implementation of the angle relationship analysis generates the angle constraint data which can be quantified, and the implementation of the constraint integration on the size constraint data, the angle constraint data and the symmetry constraint data can integrate multiple constraint conditions, and the generated geometric constraint data provides a comprehensive constraint basis for the construction of the container model, ensures the accuracy and rationality of the model in the subsequent modeling process, and improves the structure quality and application effect of the container model.

[0032] Preferably, step S3 comprises the following steps:

[0033] Step S31: Multi-angle shooting is performed on the target cargo to obtain a target cargo image; and edge detection is performed on the target cargo image to obtain an edge data set;

[0034] Step S32: Contour line segmentation processing is performed on the edge data set to obtain a line segment data set; and basic contour cutting is performed on the target cargo image based on the line segment data set to generate a cargo basic contour image;

[0035] Step S33: Stereoscopic matching is performed on the cargo basic contour image to obtain a cargo disparity map set; and depth calculation is performed on the cargo disparity map set to generate a depth data set;

[0036] Step S34: Point cloud conversion is performed on the depth data set to generate a space point set; and surface fitting is performed on the space point set to generate a complete surface set;

[0037] Step S35: Three-dimensional model reconstruction is performed on the complete surface set to obtain a reconstructed cargo model.

[0038] The present application ensures that multi-dimensional information of the target goods can be captured through the implementation of multi-angle shooting of the target goods, and the generated target goods image provides rich visual data for subsequent processing. Through goods edge detection on the target goods image, the obtained edge data set can clearly reflect the contour of the goods. The implementation of contour line segmentation processing on the edge data set ensures that the complex edge information can be simplified into a manageable line segment data set. The generated line segment data set provides a clear reference for the basic contour cutting. The implementation of basic contour cutting on the target goods image based on the line segment data set generates a goods basic contour image that can effectively extract the shape features of the goods. The implementation of stereo matching on the goods basic contour image can obtain the parallax information of the goods. The generated goods parallax map set provides necessary data support for subsequent depth calculation. Through depth calculation on the goods parallax map set, the obtained depth data set can reflect the real position of the goods in space. The implementation of point cloud conversion on the depth data set ensures that the depth information can be converted into a spatial point set. The generated spatial point set provides basic data for subsequent surface fitting. The implementation of surface fitting on the spatial point set generates a complete surface set that can effectively describe the three-dimensional shape of the goods, improving the smoothness and realism of the model. The implementation of three-dimensional model reconstruction on the complete surface set can integrate all processing results into a complete three-dimensional goods model. The generated reconstructed goods model realistically reproduces the appearance and structure of the goods, providing an accurate three-dimensional view for subsequent container loading planning, ensuring the reasonable layout and safety of the goods in the container.

[0039] Preferably, step S4 comprises the following steps:

[0040] Step S41: voxelizing the container skeleton model to obtain a spatial grid set; marking the available space of the spatial grid set to generate loading partition data;

[0041] Step S42: calculating the size of the reconstructed goods model to generate goods parameters; sorting the loading sequence based on the goods parameters to generate a goods loading sequence;

[0042] Step S43: spatial matching planning of the goods loading sequence based on the loading partition data to generate an initial goods planning model;

[0043] Step S44: centroid positioning of the initial goods planning model to obtain goods centroid data; moment calculation of the goods centroid data to generate a moment data set;

[0044] Step S45: stability analysis of the moment data set to generate goods balance parameters.

[0045] The application ensures that the model can be converted into a spatial grid set through the implementation of voxelizing the container skeleton model, the generated spatial grid set provides a basic unit for subsequent loading analysis, the generated loading partition data can clearly divide the available space in the container through the available space marking of the spatial grid set, optimize the loading layout of the goods, the implementation of size calculation of the reconstructed goods model ensures that the detailed parameters of the goods can be obtained, the generated goods parameters provide necessary information for subsequent loading sequence sorting, the implementation of loading sequence sorting based on the goods parameters can effectively optimize the loading sequence of the goods, so that the loading sequence is rationalized, the implementation of spatial matching planning of the goods loading sequence based on the loading partition data ensures that the position of the goods in the container can be reasonably arranged, the generated initial goods planning model reflects the best adaptation of the goods and the available space, the implementation of centroid positioning of the initial goods planning model can accurately identify the barycenter position of each goods, and the obtained goods centroid data provides a basis for subsequent moment calculation, the implementation of moment calculation can quantify the stress condition of the goods in the container, the implementation of stability analysis of the moment data set can comprehensively evaluate the balance state of the goods in the container, and the generated goods balance parameter provides a scientific basis for optimizing the loading scheme, through the stability analysis, the safety and stability of the goods in the transportation process are ensured, the risk of goods damage is reduced, and the safety and reliability of the entire transportation process are improved.

[0046] Preferably, step S43 comprises the following steps:

[0047] The available volume of the loading partition data is calculated to generate partition available volume data; boundary limitation mapping is performed based on the partition available volume data to generate partition boundary limitation data;

[0048] The loading partition data is subjected to connectivity analysis to obtain partition connectivity data; and the partition loading capacity is obtained through partition bearing capacity evaluation of the loading partition data according to the partition connectivity data;

[0049] The adaptation capability value is generated through adaptation capability judgment based on the partition bearing capacity and the partition boundary limitation data;

[0050] The goods loading sequence is subjected to bearing demand calculation to obtain goods bearing demand data;

[0051] The initial goods planning model is generated through spatial matching planning according to the region adaptation capability value and the goods bearing demand data.

[0052] The application ensures that the available space in each partition can be accurately identified by implementing the available volume calculation of the loading partition data, the generated partition available volume data provides key data for subsequent space planning, the generated partition boundary limit data can clearly define the boundary of the loading area by implementing the boundary limit mapping based on the partition available volume data, the implementation of the connectivity analysis of the loading partition data ensures that the connection state between each partition can be evaluated, and the obtained partition connectivity data provides an important basis for subsequent bearing capacity evaluation, by implementing the partition bearing capacity evaluation of the loading partition data, the amount of goods that each partition can bear can be quantified, and it is ensured that overloading does not occur in the actual loading process, the implementation of the adaptive capacity determination based on the partition bearing capacity and the partition boundary limit data can evaluate the adaptive degree of each partition to different goods, and the generated region adaptive capacity value provides a scientific basis for subsequent loading decision, by implementing the bearing demand calculation of the goods loading sequence, the obtained goods bearing demand data can clearly indicate the space required by each piece of goods, and the implementation of the space matching planning according to the region adaptive capacity value and the goods bearing demand data ensures that the position of the goods in the container can be reasonably arranged, and the generated initial goods planning model reflects the best adaptation of the goods and the available space, improves the loading efficiency, reduces the potential loading problem, and provides a scientific and reasonable basis for subsequent transportation.

[0053] Preferably, step S5 comprises the following steps:

[0054] Step S51: performing adjacent goods distance calculation on the initial goods planning model to obtain adjacent goods distance parameters; performing translation contact simulation according to the adjacent goods distance parameters to generate a simulated contact area;

[0055] Step S52: performing contact pressure distribution analysis based on the goods balance parameters to generate pressure distribution data;

[0056] Step S53: performing goods interference simulation on the simulated contact area according to the pressure distribution data to obtain simulated goods interference data;

[0057] Step S54: performing pressure threshold mapping based on the simulated contact area and the pressure distribution data to generate goods interference pressure thresholds;

[0058] Step S55: performing conflict point detection on the simulated goods interference data according to the goods interference pressure thresholds to generate interference conflict data.

[0059] The application ensures that the relative positions between the goods can be quantified by implementing the adjacent goods distance calculation of the initial goods planning model, the obtained adjacent goods distance parameters provide necessary data support for subsequent contact simulation, the simulation contact area generated by implementing the translation contact simulation according to the adjacent goods distance parameters can reflect the contact condition between the goods, the implementation of the contact pressure distribution analysis based on the goods balance parameters can deeply evaluate the pressure transmission condition between the goods, the generated pressure distribution data provides a basis for the identification of contact defects, through detailed analysis of the pressure distribution data, potential compression areas can be accurately identified, the effectiveness of the model is enhanced, the implementation of the goods interference simulation on the simulation contact area according to the pressure distribution data can simulate the interaction between the goods, the obtained simulation goods interference data reflects the interference condition of the goods in actual loading, the implementation of the pressure threshold mapping based on the simulation contact area and the pressure distribution data can quantize the pressure limit of the contact area, the generated goods interference pressure threshold provides an important basis for subsequent conflict detection, the implementation of the conflict point detection on the simulation goods interference data according to the goods interference pressure threshold can accurately identify the interference points between the goods, the generated interference conflict data provides detailed information for subsequent loading adjustment, ensures that the goods arrangement can be optimized in time, potential loading problems are prevented, and the overall transportation efficiency and safety are improved.

[0060] Preferably, step S53 comprises the following steps:

[0061] According to the pressure distribution data, the simulation contact area is subjected to pressure concentration region positioning, and a key contact point is obtained;

[0062] The key contact point is subjected to force direction identification, and pressure transmission direction data is generated; the key contact point is subjected to pressure value detection, and contact point pressure data is obtained;

[0063] The key contact point is subjected to goods contact simulation based on the pressure transmission direction data and the contact point pressure data, and simulation goods contact data is generated;

[0064] The simulation goods contact data is subjected to local deformation analysis, and a local deformation data set is obtained; the local deformation data set is subjected to deformation propagation range definition, and a deformation propagation range is generated;

[0065] According to the local deformation data set and the deformation propagation range, relative displacement analysis is performed, and a goods displacement amount is generated;

[0066] Based on the goods displacement amount, goods interference mapping is performed to obtain simulation goods interference data.

[0067] The application ensures that the key contact points can be identified through the implementation of the pressure concentration area positioning of the simulated contact area according to the pressure distribution data, and the key contact points obtained provide a basis for subsequent pressure transmission analysis; the pressure transmission direction data generated through the implementation of the force direction identification of the key contact points can clearly reflect the transmission path of the pressure; the actual pressure of each contact point can be quantified through the implementation of the pressure value detection of the key contact points, and the contact point pressure data obtained provides necessary information for subsequent contact simulation; the simulated cargo contact data generated through the implementation of the cargo contact simulation of the key contact points based on the pressure transmission direction data and the contact point pressure data can reflect the actual contact situation; the local deformation data set obtained through the implementation of the local deformation analysis of the simulated cargo contact data ensures that the deformation of the contact area can be identified, and the local deformation data set provides a necessary basis for subsequent deformation propagation analysis; the deformation propagation range generated through the implementation of the deformation propagation range definition of the local deformation data set can accurately reflect the influence area of the deformation; the relative displacement of the cargos can be quantified through the implementation of the relative displacement analysis according to the local deformation data set and the deformation propagation range, and the cargo displacement amount generated provides an important basis for subsequent interference mapping; the simulated cargo interference data obtained through the implementation of the cargo interference mapping based on the cargo displacement amount can reflect the interference between the cargos in detail, provides scientific guidance for actual loading optimization, and improves the safety and efficiency of the transportation process.

[0068] Preferably, step S6 comprises the following steps:

[0069] Step S61: locating the conflict point positions according to the interference conflict data, generating conflict point position coordinates; calculating the point group concentration degree of the conflict point position coordinates, and generating the conflict point position concentration degree;

[0070] Step S62: positioning the conflict focus based on the conflict point position concentration degree, and obtaining a conflict focus area;

[0071] Step S63: mapping the container skeleton model according to the conflict focus area, and generating a container model conflict area;

[0072] Step S64: analyzing the container restriction based on the interference conflict data, and obtaining a container restriction factor;

[0073] Step S65: optimizing the container skeleton model based on the container restriction factor, and generating an adjusted container model.

[0074] The application ensures that all conflict points can be accurately identified by implementing the conflict point positioning of the initial cargo planning model based on interference conflict data, and the generated conflict point coordinates provide an important basis for subsequent analysis, the conflict point concentration degree generated by implementing the point group concentration degree calculation of the conflict point coordinates can quantify the severity of the conflict, the conflict focus positioning based on the conflict point concentration degree can effectively focus on the most critical conflict area, and the obtained conflict focus area provides a clear target for subsequent optimization design, the implementation of the region mapping of the container skeleton model according to the conflict focus area can directly reflect the identified conflict area to the container model, and the generated container model conflict area provides intuitive data support for subsequent analysis, the implementation of the container restriction analysis based on the container model conflict area based on the interference conflict data can comprehensively evaluate various limiting factors affecting loading, and the obtained container restriction factor provides an important basis for subsequent optimization, and the implementation of the variant optimization of the container skeleton model based on the container restriction factor ensures that the identified restrictions can be effectively adjusted, and the generated adjusted container model can better adapt to the actual loading demand, improves the overall design efficiency of the container, reduces potential loading problems, and ensures the safety and efficiency of the transportation process. BRIEF DESCRIPTION OF DRAWINGS

[0075] Figure 1 It is a step flowchart of a container model construction method based on three-dimensional software;

[0076] Figure 2 It is Figure 1 It is a detailed implementation step flowchart of step S2 in the embodiment;

[0077] Figure 3 It is Figure 1 It is a detailed implementation step flowchart of step S3 in the embodiment;

[0078] The implementation of the application, functional characteristics and advantages will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0079] The technical method of the application will be described clearly and completely in combination with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor belong to the protection scope of the application.

[0080] Moreover, the attached drawings are only schematic and are non-limiting. Each block denoted by a reference numeral makes up one unit among others of the same type. Identical references thus designate identical or similar parts that are detailed only once; the repeated description thereof will be omitted for the sake of simplicity. Some of the blocks shown in the attached drawings are functional blocks that do not necessarily correspond to physically or logically independent entities. Functional blocks can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0081] It is to be understood that, although terms such as "first", "second", and so on can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of the exemplary embodiments, a first element can be called a second element, and similarly a second element can be called a first element. The term "and / or" as used herein includes any and all combinations of one or more of the associated associated items.

[0082] To achieve the above object, there is provided Figures 1 to 3 A container model construction method based on three-dimensional software, comprising the following steps:

[0083] Step S1: performing three-dimensional laser scanning on the container to obtain container point cloud data; performing feature segmentation on the container point cloud data to generate a container feature set;

[0084] Step S2: performing feature point spatial positioning on the container feature set to obtain a vertex coordinate set; performing geometric constraint modeling based on the vertex coordinate set to generate a container skeleton model;

[0085] Step S3: obtaining a target cargo image; performing basic contour extraction on the target cargo image to generate a cargo basic contour image; performing three-dimensional reconstruction on the cargo basic contour image to obtain a reconstructed cargo model;

[0086] Step S4: performing spatial planning processing on the container skeleton model based on the reconstructed cargo model to obtain an initial cargo planning model; performing barycenter distribution calculation on the initial cargo planning model to generate cargo balance parameters;

[0087] Step S5: performing interference simulation on the initial cargo planning model based on the cargo balance parameters to obtain simulated cargo interference data; performing conflict detection on the simulated cargo interference data to generate interference conflict data;

[0088] Step S6: performing conflict focus positioning on the interference conflict data to obtain a conflict focus area; performing variant design on the container skeleton model according to the conflict focus area to generate an adjusted container model.

[0089] The application ensures high-precision point cloud data acquisition through the implementation of three-dimensional laser scanning of the container, the generated container feature set can comprehensively reflect the geometric shape and structural features of the container through feature segmentation of the point cloud data, the implementation of feature point spatial positioning on the container feature set ensures accurate positioning of each feature point, the generated vertex coordinate set provides necessary data support for three-dimensional modeling, the operation of geometric constraint modeling based on the vertex coordinate set can generate a container skeleton model, the implementation of obtaining target cargo images and extracting basic contours ensures that the shape features of the cargo can be accurately captured, the generated cargo basic contour image provides necessary visual information for subsequent three-dimensional reconstruction, through three-dimensional reconstruction of the contour image, the reconstructed cargo model can truly reflect the space occupation of the cargo, providing accurate data support for container loading, the implementation of spatial planning processing of the container skeleton model based on the reconstructed cargo model ensures efficient arrangement of the layout of the cargo in the container, the generated initial cargo planning model provides a basis for subsequent balance analysis, the result of the center of gravity distribution calculation can quantify the balance parameters of the cargo, the implementation of interference simulation based on the cargo balance parameters can identify potential conflicts between the cargo and between the cargo and the container, the generated simulation cargo interference data provides detailed information support for subsequent conflict detection, the interference conflict data generated through conflict detection can efficiently identify problems in the loading process, ensuring that the scheme can be adjusted in time in actual operation, optimizing the loading layout of the cargo, the implementation of conflict focus positioning on the interference conflict data ensures that the specific conflict area can be determined, providing an important basis for subsequent container skeleton model variant design, model adjustment according to the conflict focus area can effectively avoid interference problems in the loading process, and the generated adjusted container model can better adapt to actual loading requirements, finally improving the use efficiency of the container and the safety of cargo transportation.

[0090] In the embodiment of the application, refer to Figure 1 The application is a kind of container model construction method based on three-dimensional software, and the steps of the method are as follows:

[0091] Step S1: three-dimensional laser scanning of the container is performed to obtain container point cloud data; feature segmentation is performed on the container point cloud data to generate a container feature set;

[0092] In this embodiment, the exterior of the container is scanned in all directions using a three-dimensional laser scanner (such as Leica RTC360), covering all surfaces of the container to ensure the completeness of the scanning data. The scanning result generates a high-density point cloud data, each point in the point cloud data contains the coordinate information of the container surface. Then, the point cloud data is preprocessed, including removing noise points, filling blank areas, and smoothing. A point cloud filtering algorithm (such as Voxel Grid filter) is used to reduce redundant points and obtain more accurate container point cloud data. Then, a feature segmentation algorithm (such as a plane fitting method based on RANSAC algorithm) is applied to the container point cloud data to identify geometric features related to the container shape, such as faces, edges, and corners, forming a feature set of the container. These feature sets contain the geometric information of the container shape, providing basic data for subsequent model construction.

[0093] Step S2: Feature point space positioning is performed on the feature set of the container to obtain a vertex coordinate set; geometric constraint modeling is performed based on the vertex coordinate set to generate a container skeleton model;

[0094] In this embodiment, the feature set of the container is input into a geometric modeling tool (such as Autodesk Revit). By spatially positioning the feature points, the key vertices in the container structure are identified to obtain a vertex coordinate set. On this basis, geometric constraint modeling techniques are applied to the vertex coordinate set, such as defining boundary conditions and constraint relationships, using least squares method to optimize the vertex coordinates to eliminate errors and ensure the accuracy of the coordinate data. A container skeleton model is generated, which provides an accurate geometric shape basis for subsequent cargo arrangement and interference simulation, and ensures that there is no deformation and geometric error in the modeling process.

[0095] Step S3: Obtain the target cargo image; perform basic contour extraction on the target cargo image to generate a cargo basic contour image; perform three-dimensional reconstruction on the cargo basic contour image to obtain a reconstructed cargo model;

[0096] In this embodiment, a high-resolution digital camera (such as Canon EOS 5D Mark IV) is used to take multiple-angle images of the target cargo. The images should include the front, side, and top of the cargo to ensure sufficient image data. Then, an edge detection algorithm (such as Canny edge detection) is used to perform basic contour extraction on the target cargo image to obtain a cargo basic contour image. The contour image includes the shape information of the cargo. Subsequently, three-dimensional reconstruction is performed on the cargo basic contour image using structured light scanning technology or stereo vision technology (such as using Intel RealSense depth camera) to convert two-dimensional image data into a three-dimensional model. After data fusion and optimization, an accurate reconstructed cargo model is obtained.

[0097] Step S4: based on the reconstructed cargo model, the container skeleton model is subjected to spatial planning processing to obtain an initial cargo planning model; the initial cargo planning model is subjected to center of gravity distribution calculation to generate cargo balance parameters;

[0098] In this embodiment, based on the reconstructed cargo model, a spatial planning algorithm (such as a spatial search method based on A* algorithm) is used to perform spatial planning processing on the container skeleton model, to ensure the matching of the cargo and the internal structure of the container, to generate an initial cargo planning model, and then a mass distribution calculation method (such as the center force method) is used to perform center of gravity distribution calculation on the cargo planning model to obtain the center of gravity position of the cargo, and further calculate the balance parameters of the cargo, including the center of gravity offset and stability index, which provide necessary basis for subsequent cargo interference simulation and adjustment.

[0099] Step S5: based on the cargo balance parameters, interference simulation is performed on the initial cargo planning model to obtain simulated cargo interference data; conflict detection is performed on the simulated cargo interference data to generate interference conflict data;

[0100] In this embodiment, based on the cargo balance parameters, interference simulation is performed on the initial cargo planning model, and an interference detection module in computer-aided design (CAD) software (such as SolidWorks) is used to simulate the interference between the cargo and the inner wall of the container and other cargos to obtain simulated cargo interference data, which records the spatial relationship and interference area between the cargo and the container structure, and then a conflict detection algorithm (such as a collision detection algorithm based on volume calculation) is used to analyze the simulated cargo interference data to identify and generate interference conflict data, which provides specific optimization direction for adjusting the container model in the next step.

[0101] Step S6: conflict focus positioning is performed on the interference conflict data to obtain a conflict focus area; and a variant design is performed on the container skeleton model according to the conflict focus area to generate an adjusted container model.

[0102] In this embodiment, according to the interference conflict data, the spatial distribution of the conflict focus area is analyzed, and a minimum conflict optimization algorithm (such as an optimization method based on simulated annealing) is used for focus positioning to accurately identify the specific area of interference in the container, and then a variant design is performed on the container skeleton model based on the conflict focus area, in which a parametric modeling method (such as parametric design through Grasshopper plug-in) is used to adjust the skeleton structure of the container to generate an adjusted container model, and the size or shape of the container is adjusted to ensure that the cargo can be reasonably placed in the container to avoid interference or damage, and finally an optimized container model that meets the cargo balance and space utilization requirements is obtained.

[0103] Preferably, step S1 comprises the following steps:

[0104] Step S11: Multi-angle scanning of the container to obtain original container point cloud data; noise filtering of the original container point cloud data to generate filtered point cloud data;

[0105] Step S12: Coordinate conversion processing of the filtered point cloud data to generate container point cloud data;

[0106] Step S13: Geometric feature point recognition of the container point cloud data to obtain a container feature point set; topological relationship analysis of the container feature point set to obtain a feature point topological relationship;

[0107] Step S14: Feature point connection of the container feature point set based on the feature point topological relationship to obtain a feature point connection graph; geometric feature analysis of the feature point connection graph to generate a container feature set.

[0108] In this embodiment, the container is scanned by a three-dimensional laser scanner (such as Leica RTC360) at multiple angles, ensuring that data is collected from different angles of the container. The scanning data contains all point cloud data of the container surface, with a point cloud density of about 2,000 points per square meter to ensure data detail. After the original scanning data is generated, noise filtering is performed using point cloud processing software (such as CloudCompare). The point cloud is denoised using a statistical outlier removal algorithm (Statistical Outlier Removal, SOR) to remove abnormal points away from the main point cloud cluster. The filtered point cloud data will contain the main geometric information of the container surface. The number of generated filtered point cloud data points is reduced by about 10% to 15%. The filtered point cloud data is input into a coordinate conversion tool (such as AutoCAD or PDMS) to perform coordinate system conversion. The container original point cloud data is converted from the local coordinate system of the scanning device to the world coordinate system. The conversion uses a least squares-based coordinate matching algorithm to ensure that the converted point cloud data is consistent with the actual coordinates by comparing known geographic coordinate points. After the coordinate system of the container is converted, the container point cloud data is generated, with a conversion accuracy of 0.5 mm. Using the container point cloud data, geometric feature point recognition is performed using a geometric feature recognition algorithm (such as the RANSAC algorithm). First, the parameters are set to extract the plane, straight line, and circular arc features of the container surface. Special attention is paid to the edges, corner points, and face points of the container. By setting a point cloud density threshold, the container surface feature point set is identified, such as the four corner points on the top, the midpoint of each side, and the bottom intersection key points. Next, topological relationship analysis is performed on the identified feature point set using a connectivity analysis method (such as a topological analysis algorithm based on neighborhood relationships). The relative position and connection relationship between points are analyzed to obtain the topological relationship of the feature points. Based on the container feature point set and the topological relationship, the feature points are connected using a graphical modeling tool (such as the Grasshopper plug-in of Rhino). Based on the topological relationship, the feature points are connected in order of spatial position and geometric relationship to generate a feature point connection graph of the container. Each node in the graph represents a feature point, and each edge represents the connection relationship between adjacent feature points. After generation, the feature point connection graph is further analyzed using a geometric analysis method (such as least squares fitting-based geometric analysis) to analyze the geometric shapes and symmetry in the feature point connection graph. Finally, a container feature set is formed, which contains all the key geometric features of the container, including the outline, face, edge, and spatial topological structure of the container, providing detailed geometric descriptions for container modeling and subsequent analysis.

[0109] Preferably, step S2 comprises the following steps:

[0110] Step S21: mapping the container feature set to a spatial coordinate to obtain a three-dimensional coordinate point set; performing vertex clustering analysis on the three-dimensional coordinate point set to generate a candidate vertex set;

[0111] Step S22: performing accurate positioning calculation on the candidate vertex set according to the three-dimensional coordinate point set to obtain a vertex coordinate set;

[0112] Step S23: performing geometric relationship extension on the vertex coordinate set to generate surface domain relationship data; performing geometric constraint analysis on the surface domain relationship data to generate geometric constraint data;

[0113] Step S24: establishing a framework structure based on the geometric constraint data to obtain an initial skeleton model; performing node connection filling on the initial skeleton model according to the surface domain relationship data to generate a container skeleton model.

[0114] In this embodiment, the container is scanned by a three-dimensional laser scanning device (such as Leica RTC360) at multiple angles to obtain point cloud data of the container, which contains all the geometric feature information of the container outside. Then, the original point cloud data is preprocessed using point cloud processing software (such as CloudCompare) to remove noise points. A filtering algorithm (such as Voxel Grid filtering) is applied to the data to reduce redundant points and enhance the clarity of the data. The processed point cloud data is used as input for spatial coordinate mapping. The point cloud data is converted into a unified three-dimensional coordinate system using coordinate transformation methods to generate a set of three-dimensional coordinate points. Then, a K-means clustering algorithm is used to analyze the clustering of the three-dimensional coordinate point set to identify the main feature points related to the geometric shape of the container to obtain a candidate vertex set. A least squares-based precise positioning algorithm is used to calculate the positioning of the candidate vertex set to ensure the accuracy of the vertex position. First, the relative position and distance in the three-dimensional coordinate point set are calculated to optimize the preliminary positioning of the candidate vertex set and reduce the positioning error to a minimum to ensure that the position of each vertex is highly matched with the true form of the container. The spatial position of the vertex is further corrected using geometric constraint methods (such as rigid body transformation method) to obtain an accurate vertex coordinate set. Topological methods are used to build geometric relationships for the precisely positioned vertex coordinate set based on the relative positions between the vertices to identify each face domain that constitutes the surface and skeleton of the container. Connectivity analysis and topological structure construction techniques (such as using Graph Theory methods) are used to establish the relationship between the face domains to generate face domain relationship data. Geometric constraint analysis is then applied to the face domain relationship data to identify the geometric constraint conditions that affect the stability and variability of the container skeleton structure by analyzing the angle, distance, and other constraint relationships between different face domains to generate geometric constraint data. Based on the geometric constraint data, a parametric modeling method (such as the parametric modeling tool in Grasshopper) is used to establish the container frame structure. The frame structure first generates an initial skeleton model based on the geometric constraint data, which includes the basic structural skeleton of the container, the nodes and boundaries of the connecting framework, and ensures that the geometric properties of the container skeleton meet the design requirements of the container by calculating the relative positions and shape constraints between the nodes. Then, the nodes of the initial skeleton model are connected and filled according to the face domain relationship data. The grid division technique is used to ensure accurate connection between each node. The generated container skeleton model has a complete geometric structure that is suitable for subsequent cargo layout and spatial optimization design.

[0115] Preferably, step S23 comprises the following steps:

[0116] The vertex coordinate point set is analyzed for adjacency relationship to obtain a point pair relationship set. The point pair relationship set is processed for edge line connection to generate edge line data.

[0117] Parallelism detection is performed on the edge line data to obtain parallel edge data; perpendicularity detection is performed on the parallel edge data to generate an orthogonal edge data set;

[0118] Based on the orthogonal edge data set, a face element relationship is constructed to generate face domain relationship data;

[0119] Size parameter extraction is performed on the face domain relationship data to obtain size constraint data;

[0120] Angle relationship analysis is performed on the size constraint data to generate angle constraint data; symmetry calculation is performed on the angle constraint data to obtain symmetry constraint data;

[0121] Constraint integration is performed on the size constraint data, the angle constraint data and the symmetry constraint data to generate geometric constraint data.

[0122] In this embodiment, the input precise positioning of the vertex coordinate set, by calculating the Euclidean distance of each pair of vertices, using the Adjacency Matrix method to analyze which vertices exist between the direct geometric relationship, according to the distance threshold to determine whether the point pair is adjacent, the threshold is set to 5mm, if the distance between the vertices is less than the threshold, they are considered to be adjacent in space, the point pair relationship set is generated, which contains the information of all adjacent vertex pairs, indicating the connection relationship between the container surface and the skeleton structure, by traversing the point pair relationship set, using the straight line fitting algorithm (such as least squares method) to connect adjacent vertex pairs, generate edge data, each edge is defined by two vertices, edge data contains the boundary information of the container skeleton, further analyze the length of each edge and its direction in space, construct the spatial coordinates of the edge, and combine these edges with the vertex data to ensure the rationality of each edge in the physical space, by calculating the angle between the edges, using the method of vector operation, calculate the angle value of each pair of edges, if the angle is less than 5°, the two edges are considered to be parallel edges, using the parallel degree detection algorithm to compare all edges, extract the edges that meet the parallel degree condition, generate parallel edge data, which represents the spatial relationship of parallel edges in the container structure, by calculating the angle of each pair of edges in the parallel edge data set, if the angle is 90°, the two edges are considered to be perpendicular, using the orthogonal detection algorithm to calculate the perpendicularity of all edge pairs, extract all perpendicular edge pairs, generate orthogonal edge data set, by analyzing the orthogonal edge data set, determine which orthogonal edges combine to form a face domain, based on the connection relationship of the edges, using the polygon construction algorithm (such as quadrilateral or triangular mesh generation algorithm) to construct the face element (face domain) of the container surface, in the construction process of the face element, the edge is used as the boundary condition to ensure the planarity and accuracy of the angle of the face element in space, the generated face domain relationship data records the geometric relationship between each face in the container structure, extract the size parameters (such as length, width, height, diagonal length, etc.) of each face domain from the face domain relationship data, apply the geometric size calculation method to measure the boundary of each face domain to get accurate size data, the size parameters include the edge length and angle of each face domain, generate size constraint data, by further analyzing the edge length and angle relationship extracted from the size constraint data, calculate the angle between each two face domains, apply the geometric calculation method (such as cosine theorem) to analyze the angle between adjacent face domains, get the angle constraint data, the angle constraint data ensures the consistency of the angle between each face in the container model, further analyze the angle constraint data, using the symmetry analysis algorithm (such as reflection transformation detection method) to identify the symmetry structure in the model, calculate whether there is symmetry between the faces in the angle constraint data, if there is symmetry, generate symmetry constraint data, by integrating the size constraint data, angle constraint data and symmetry constraint data, using the constraint optimization algorithm (such as linear programming method) to unify all constraint conditions,The complete geometric constraint data is generated, which ensures that the container model meets all geometric constraint conditions in space and guarantees the accuracy and stability of the model.

[0123] Preferably, the step S3 comprises the following steps:

[0124] Step S31: Multi-angle shooting of the target cargo to obtain target cargo images; cargo edge detection on the target cargo images to obtain an edge data set;

[0125] Step S32: Contour line segmentation processing on the edge data set to obtain a line segment data set; basic contour cutting on the target cargo images based on the line segment data set to generate a cargo basic contour image;

[0126] Step S33: Stereo matching on the cargo basic contour image to obtain a cargo disparity map set; depth calculation on the cargo disparity map set to generate a depth data set;

[0127] Step S34: Point cloud conversion on the depth data set to generate a space point set; surface fitting on the space point set to generate a complete surface set;

[0128] Step S35: Three-dimensional model reconstruction on the complete surface set to obtain a reconstructed cargo model.

[0129] In this embodiment, a high-resolution camera is used to take multiple-angle shots of the target goods. During the shooting process, a camera automatic calibration method is used to ensure the accuracy of the shooting angle. The included angle between each shooting angle is set to 15°, and 3 pictures are taken at each angle. The camera focal length is set to 50mm and the aperture is adjusted to F8 during shooting to ensure that clear images of the goods are obtained. The images from multiple angles can provide sufficient perspective differences. The Canny edge detection algorithm is used to process each target goods image. During the edge detection process, the high threshold is set to 200, the low threshold is set to 100, and the image size is set to 1024x1024 pixels. After edge detection, the edge data set of the target goods is obtained, which contains the surface contour information of the goods. The edge data set is subjected to contour line segmentation processing. The Hough Transform algorithm is used to detect straight lines in the edge data, with a threshold set to 50. All significant straight lines in the image are detected. Then, the detected straight lines are fitted using a line segment algorithm (such as the RANSAC algorithm) to generate a line segment data set containing all line segments. The line segment data set records the linear features of the target goods' outer contour. Based on the line segment data set, the basic contour is cut. Image segmentation techniques (such as the threshold-based region growing method) are used to cut the target goods image to obtain an accurate external contour of the goods. During the cutting process, the threshold is set to 0.3, and all pixel points in the image region are classified according to whether they belong to the goods boundary. Finally, a clear goods basic contour image is generated. A suitable stereo matching algorithm (such as Block Matching or Semi-Global Matching) is selected to perform stereo matching on the goods basic contour image. The left and right image pairs are input, the image window size is set to 9x9, and the disparity range is set to -64 to 64. The algorithm generates a disparity map by calculating the pixel difference between the left and right images. The disparity map set contains the depth information of different regions of the goods surface. The goods surface structure under different viewing angles can be obtained through the disparity map. The depth of each disparity map in the goods disparity map set is calculated using the depth calculation formula: Depth = Baseline x Focal Length / Disparity ("Depth" = ("Baseline" x "Focal Length") / "Disparity"). The baseline is set to the camera distance.1 meter, the focal length is set to 50mm, the calculated depth data set provides accurate depth information of each region of the goods, according to the depth data set, the depth value of each pixel is converted into a spatial point through a three-dimensional reconstruction algorithm, the position of the spatial point is converted through the formula: X=(x*"Depth") / "Focal Length", Y=(y*"Depth") / "Focal Length", Z="Depth", during the conversion process, x and y are pixel coordinates, "Depth" is the depth value corresponding to the pixel, "Focal Length" is the focal length of the camera, the finally generated spatial point set records the three-dimensional point cloud data of the surface of the goods, the least squares method is used for surface fitting of the spatial point set, the surface of the goods is fitted by selecting an appropriate fitting model (such as B-spline surface or NURBS surface), the number of control points is set to 100 during the fitting process, and the fitting accuracy is set to 0.5mm, the generated surface set contains a smooth three-dimensional surface model of the surface of the goods, the three-dimensional model is reconstructed by using the surface set, the surface is converted into a three-dimensional grid model by using a meshing algorithm (such as Delaunay triangulation algorithm), the mesh density is set to 5mm during the meshing process, after the three-dimensional grid model is generated, the grid is smoothed by using a surface refinement algorithm (such as Laplacian smoothing), and the reconstructed goods model has high-precision three-dimensional geometric shape.

[0130] Preferably, step S4 comprises the following steps:

[0131] Step S41: voxelizing the container frame model to obtain a spatial grid set; marking the available space of the spatial grid set to generate loading partition data;

[0132] Step S42: calculating the size of the reconstructed goods model to generate goods parameters; sorting the loading sequence based on the goods parameters to generate a goods loading sequence;

[0133] Step S43: spatially matching and planning the goods loading sequence based on the loading partition data to generate an initial goods planning model;

[0134] Step S44: positioning the centroid of the initial goods planning model to obtain goods centroid data; calculating the moment of the goods centroid data to generate a moment data set;

[0135] Step S45: analyzing the stability of the moment data set to generate goods balance parameters.

[0136] In this embodiment, the container skeleton model is converted into a three-dimensional grid format, and voxelization is performed based on the grid. In the voxelization process, the space of the container is first divided into multiple small cubic units, each unit representing a voxel. The size of the voxel is set according to the actual size of the container, for example, the length, width, and height of the container are divided into equal intervals. These voxels represent the space inside the container in a discrete manner. Through voxelization, the geometric position of each voxel can be clearly defined, and the number and shape of the voxels can accurately represent all the space inside the container. The result of voxelization generates a set of space grids. Then, the space grids are marked with available space. The state of each voxel is marked as available or unavailable. The specific marking method is to detect the existing objects in the skeleton model and identify the occupied space. The corresponding voxels of the occupied space are marked as unavailable, and the remaining voxels are marked as available. Finally, a set of space grids containing all available voxels is obtained, and the loading partition data is generated. The three-dimensional reconstruction model of the cargo is converted into a standard geometric format, such as STL or OBJ file. By analyzing the geometric data in the file, the basic dimensions of the cargo, such as length, width, and height, are obtained, and the volume and surface area of the cargo are calculated. In the size calculation, the geometric analysis method is used, and the bounding box method is used to calculate the maximum circumscribed cuboid size of the cargo in three-dimensional space. By scanning the surface of the cargo point by point and calculating the minimum and maximum values in each coordinate axis direction, the minimum length, width, and height of the cargo are determined. Then, based on these size data, the parameters of the cargo are further generated, including the volume, mass, center of gravity position, and shape characteristics of the cargo. The center of gravity position is calculated using the centroid formula based on the geometric shape and mass distribution of the cargo. All cargo parameters provide data support for subsequent loading sequence sorting, and generate the basis for cargo loading sequence. According to the previously generated cargo loading sequence, combined with the loading partition data, a three-dimensional space matching algorithm is used to adapt each cargo to the space. In the matching process, the loading partition is first divided and optimized to ensure that the cargo can be reasonably loaded into the appropriate space region. In the space matching, the size, shape, weight, and center of gravity position of the cargo are considered, and the heuristic search algorithm (such as A* algorithm) is used to distribute all the cargos according to the available space and carrying capacity of each partition to ensure the rationality of the loading. In each matching, the most suitable loading position for the current cargo is selected to ensure the optimization of the layout of the cargo in the container and the maximization of the space utilization rate. Through this matching planning, a preliminary cargo planning model is generated, which includes the specific position, direction, and loading sequence of each cargo. According to the position and size of each cargo in the initial cargo planning model, the centroid of each cargo is calculated. The centroid calculation is based on the geometric shape and mass distribution of the cargo, and the centroid formula is used to calculate the center of gravity coordinates of the cargo through integration.In actual operation, by scanning the three-dimensional model of the cargo, the geometric volume of the cargo is divided into a plurality of small volume units, the mass density of each unit is combined, the centroid coordinates of each volume unit are calculated, and then the overall centroid of the cargo is calculated according to the weighted average method. Then, by using these centroid data, the position of the cargo in the container is adjusted to ensure that the centroid of the cargo can be evenly distributed and the situation of the gravity center deviating too much is avoided, so as to provide accurate data support for subsequent torque calculation. The centroid data of the cargo is obtained, the centroid coordinates of each cargo are recorded, and the torque distribution of the cargo in the container is calculated according to the centroid data of each cargo and its position in the container through the torque calculation formula. The torque calculation is based on the centroid position of each cargo and the length of the force arm relative to the container, and the traditional torque formula is used to solve it. The specific method is to calculate the relative position of the centroid coordinates and the container coordinate system to obtain the torque value of each cargo. In the calculation process, the torque of the cargo is mapped into the coordinate system of the container through the three-dimensional coordinate system, and the torque data set is generated by combining the position and mass information of the cargo. The torque data set records the torque values of all cargos and the influence of these torque values on the overall stability of the container. The generated torque data is simulated by using a special mechanical analysis software (such as ANSYS or SolidWorks Simulation), the balance state of the cargo under different loading conditions is analyzed, and multiple scenes are set in the simulation process, including different cargo arrangement and container inclination angle. The cargo balance parameters are generated, and the analysis software determines the stability under each arrangement mode by calculating the centroid and stress of the cargo.

[0137] Preferably, the step S43 comprises the following steps:

[0138] The available volume calculation is performed on the loading partition data to generate partition available volume data; the boundary restriction mapping is performed based on the partition available volume data to generate partition boundary restriction data;

[0139] The connectivity analysis is performed on the loading partition data to obtain partition connectivity data; the partition bearing capacity evaluation is performed on the loading partition data according to the partition connectivity data to obtain the partition bearing capacity;

[0140] The adaptive capacity determination is performed based on the partition bearing capacity and the partition boundary restriction data to generate a region adaptive capacity value;

[0141] The bearing demand calculation is performed on the cargo loading sequence to obtain cargo bearing demand data;

[0142] The space matching planning is performed according to the region adaptive capacity value and the cargo bearing demand data to generate an initial cargo planning model.

[0143] In this embodiment, based on the loading partition data, the spatial size of each partition is determined, and the volume of the loading partition is solved by using a three-dimensional geometric calculation method. In specific operation, each partition is divided into a plurality of small volume units by meshing the partition boundary, the volume of these small units is weighted and summed by using a volume integration algorithm, and the available volume of each partition is obtained. Through the accurate available volume calculation of the meshed partition, the idle space size of each partition can be accurately obtained, and this data becomes the basis for subsequent loading planning. The finally generated partition available volume data includes the volume size of each partition and the specific position of the available space region and the occupied region in the partition. According to the partition available volume data, the spatial boundary restriction mapping is carried out. The specific steps are to determine the boundary condition and spatial constraint of each partition by geometric analysis on the spatial range of each loading partition, and to use the constraint condition to carry out boundary restriction mapping. The mapping process of boundary restriction is to set the maximum and minimum size limits of the partition, to clearly define the shape and size of each partition, and to generate the limit data based on the available volume data of the partition, to combine the volume and boundary constraint, and to accurately describe the boundary of each partition. Through the boundary restriction mapping, the spatial range of each partition can be clearly defined, the connectivity analysis is carried out on the loading partition data, and the partition connectivity data is obtained. First, the connectivity analysis is carried out on the loading partition, the connection relationship between each partition and its adjacent partition is analyzed, in specific operation, the adjacency matrix between partitions is constructed to represent whether there is a direct path between partitions, based on the geometric calculation method, the spatial distance and relative position between each partition are calculated to determine whether the partitions are connected. The connectivity analysis adopts a depth-first search algorithm (Depth-First Search, DFS) or a breadth-first search algorithm (Breadth-First Search, BFS) to traverse all partitions, generates the connectivity data between partitions, records whether each partition is connected with other partitions, and the spatial range and position of the connected region, and according to the partition connectivity data, the partition carrying capacity is evaluated, and the partition carrying capacity is obtained. Based on the partition connectivity data, the carrying capacity of each partition is analyzed, and the specific operation is to evaluate according to the available volume data of each partition and the connection relationship with other partitions, and to combine the physical carrying model. In the evaluation, volume ratio, gravity distribution and other physical parameters are used to calculate the carrying capacity of each partition under a given weight, a mechanical analysis tool is used to simulate and calculate the carrying capacity of each partition, and the maximum load and carrying capacity of the partition are calculated.By considering the connectivity between the partitions, the carrying capacity data of each partition is obtained, and the carrying performance of each partition is further evaluated. According to the partition carrying capacity data and the partition boundary restriction data, the adaptive capacity of each loading partition is determined. In specific implementation, according to the carrying capacity of the partition and the space limitation condition of the boundary, the adaptability is evaluated, and it is judged whether each partition can meet the carrying demand and space requirement of the goods. In the evaluation process, the matching degree algorithm is used to determine the adaptive capacity value of each region in combination with the loading requirements and space restrictions. The adaptive capacity value reflects the adaptive degree of each partition under the requirements of weight, volume, shape, etc. Finally, the adaptive capacity data of each partition is obtained. The carrying demand of the goods loading sequence is calculated to obtain the goods carrying demand data. First, based on the goods loading sequence data, the carrying demand of each goods is calculated. In specific operation, first, according to the geometric size, weight and shape of the goods, the load distribution is analyzed by using a physical model (such as a physical mechanics model) to obtain the carrying demand of each goods to the partition. In the carrying demand calculation, the mechanical model (such as the center of mass moment balance) is used to analyze the weight distribution and moment distribution of the goods in the loading process, to ensure the matching of the center of gravity of each goods and the loading area, and to obtain the carrying demand data of the goods, which includes the weight demand, space demand and distribution of the goods. According to the region adaptive capacity value and the goods carrying demand data, the space matching planning is carried out to generate the initial goods planning model. Based on the region adaptive capacity value and the goods carrying demand data, the space matching planning is carried out. First, according to the carrying demand data of each goods and the adaptive capacity value of each partition, the space optimization algorithm (such as the minimum distance method or the greedy algorithm) is used to allocate the space of each goods, to determine the optimal position and loading sequence of each goods in the loading partition. In planning, the weight, volume, shape and space demand of the goods are considered, and the adaptive capacity of the partition is combined to finally generate the initial goods planning model, which includes the position, direction and loading sequence of each goods in the container.

[0144] Preferably, step S5 comprises the following steps:

[0145] Step S51: Calculate the distance between adjacent goods based on the initial goods planning model to obtain the adjacent goods distance parameter; and perform translation contact simulation according to the adjacent goods distance parameter to generate a simulated contact area;

[0146] Step S52: Perform contact pressure distribution analysis based on the goods balance parameter to generate pressure distribution data;

[0147] Step S53: Perform goods interference simulation on the simulated contact area according to the pressure distribution data to obtain simulated goods interference data;

[0148] Step S54: Perform pressure threshold mapping based on the simulated contact area and pressure distribution data to generate a cargo interference pressure threshold;

[0149] Step S55: Perform conflict point detection on the simulated cargo interference data according to the cargo interference pressure threshold to generate interference conflict data.

[0150] In this embodiment, based on the cargo position data in the initial cargo planning model, the distance between each pair of adjacent cargos is calculated to obtain the distance parameters of adjacent cargos. This calculation is performed by traversing the centroid positions of each pair of cargos and using the Euclidean distance formula to calculate the straight-line distance between each pair of cargos. Through these adjacent cargo distance parameters, the spatial relationship between the cargos is evaluated, and further translation contact simulation is performed. In the translation contact simulation process, the cargos are simulated by translation according to the adjacent cargo distance parameters, simulating the contact area of the cargos in the contact state. The translation simulation is calculated by setting the bounding box of the cargo, gradually simulating the translation process of the cargo until the cargo is in contact, and then determining the contact area, simulating the contact area as the geometric area of the intersection of the surfaces of the two cargos on the contact surface, forming the simulation contact data, determining the mass distribution and moment data of the cargo in the contact area according to the cargo balance parameters, and analyzing the distribution of contact pressure in combination with the surface characteristics and material properties of the cargo. The contact pressure distribution analysis obtains the distribution of pressure on the contact area by calculating the size and distribution of the force on the cargo surface. The specific method is to perform pressure simulation by combining the Finite Element Analysis (FEA) model to evaluate the pressure size at the contact point. According to the cargo balance parameters, the pressure value at each contact point is determined and spatially mapped to generate pressure distribution data, based on which cargo interference simulation is performed. The interference simulation is completed by shape overlap detection of the cargos within the simulated contact area to evaluate whether the cargos will interfere in the contact area. Using the generated contact pressure distribution, in combination with the geometry of the cargo, it is simulated whether the contact points between the two cargos will lead to physical overlap, and it is judged whether interference will occur according to the pressure value. When the pressure value exceeds the set threshold, it is considered that interference has occurred. Through this method, simulation cargo interference data is obtained, recording which cargos have interference between them and their corresponding interference regions, and the cargo interference pressure threshold is generated based on the simulation contact area and pressure distribution data. According to the simulation contact area and pressure distribution data, the pressure of the contact area is threshold-mapped. A certain pressure threshold is set to determine which regions have pressure values exceeding the threshold, and it is considered that the contact in these regions will lead to interference. Through this process, the cargo interference pressure threshold is generated for subsequent interference conflict detection. This mapping process determines a reasonable pressure threshold by statistical analysis of the pressure data, and detects conflict points based on the cargo interference pressure threshold and the simulation cargo interference data. The detection method checks whether the contact pressure value exceeds the set threshold at each point in the simulation contact area. If the threshold is exceeded, it is considered that interference has occurred at that location, and the interference point and its interference intensity are recorded to generate interference conflict data. This data will be used in subsequent optimization of the loading scheme to ensure that interference problems between cargos can be discovered and adjusted in a timely manner.

[0151] Preferably, step S53 comprises the following steps:

[0152] According to the pressure distribution data, the pressure concentration area of the simulated contact area is located to obtain a key contact point;

[0153] The key contact point is subjected to force direction identification to generate pressure transmission direction data; the key contact point is subjected to pressure value detection to obtain contact point pressure data;

[0154] Based on the pressure transmission direction data and the contact point pressure data, the key contact point is subjected to cargo contact simulation to generate simulated cargo contact data;

[0155] The simulated cargo contact data is subjected to local deformation analysis to obtain a local deformation data set; the local deformation data set is subjected to deformation propagation range definition to generate a deformation propagation range;

[0156] According to the local deformation data set and the deformation propagation range, relative displacement analysis is performed to generate a cargo displacement amount;

[0157] Based on the cargo displacement amount, cargo interference mapping is performed to obtain simulated cargo interference data.

[0158] In this embodiment, based on the pressure distribution data within the simulated contact area, the areas with higher pressure are located, and the areas with concentrated pressure in the contact surface are determined through pressure data mapping. These areas are considered as key contact points. The positioning process adopts a grid division method to divide the contact area into several grid units, calculates the pressure value of each unit and compares it with the global pressure data to find the areas with higher pressure and exceeding the preset threshold, and then determines the key contact points. Through the analysis of the pressure distribution map, the position coordinates of these key contact points and their corresponding pressure values are obtained. By analyzing the pressure distribution data of the key contact points, the direction of pressure transmission is identified, and the propagation path of the force is determined. The direction of the force is calculated according to the pressure size and direction of its adjacent points, and the transmission direction of the force is determined by vector analysis method to establish a data model of the force transmission direction. Through the force propagation path, the pressure transmission direction data is generated to represent the stress direction transmitted on the contact surface. The pressure values of each key contact point are detected in detail to obtain the actual stress condition of the point during the contact process. Through further refined grid analysis, combined with the finite element analysis (FEA) model, the pressure values of the key contact points are measured, and these pressure data are recorded. The pressure values of each contact point are compared with other areas within the simulated contact area to confirm the pressure change trend with the adjacent areas, generate the contact point pressure data, and simulate the cargo contact based on the pressure transmission direction data and the contact point pressure data to generate the simulated cargo contact data. According to the pressure transmission direction data and the contact point pressure data generated as described above, the cargo contact simulation is performed. This simulation models the pressure change of the contact points, uses contact mechanics theory (such as Hertz contact theory) for numerical simulation, and simulates the deformation of the cargo under stress on the contact surface. During the simulation, the pressure transmission direction and the pressure data are combined to calculate the deformation of the cargo surface during the contact process, and the relevant data of the simulated contact are recorded to finally obtain the simulated cargo contact data. Local deformation analysis is performed on the simulated cargo contact data to obtain a local deformation data set. According to the simulated cargo contact data, the finite element analysis method (FEA) is used to analyze the local deformation of the contact area to detect the deformation degree and distribution of the contact area. By calculating the displacement of the contact points, the local deformation of the cargo during contact is analyzed, and a local deformation data set is obtained. This data set contains the displacement and deformation of each contact point, and combined with the pressure data, the geometric changes of the deformation area are calculated. The local deformation data set is further processed to analyze the propagation range of the deformation on the cargo surface. By gradually propagating the displacement of each contact point, the influence range of the deformation on the entire cargo model is determined, and then the propagation range of the deformation is defined. Using the step-by-step recursive method, the influence range of each deformation point is combined with the surrounding contact points to finally generate complete deformation propagation range data.The relative displacement analysis is performed by combining the local deformation dataset and the deformation propagation range. The analysis determines the overall displacement of the cargo under pressure by calculating the displacement change of each contact point relative to other points. Using the relative displacement formula, combined with the simulated contact data, the displacement of the cargo is obtained, and the cargo displacement data is generated, and then the interference situation that may occur to the cargo is analyzed. According to the cargo displacement generated as described above, cargo interference mapping is performed. By comparing with the initial loading model, the interference region occurring between the cargos is identified. The mapping process determines the location and size of the interference region by mapping the displacement, and the simulated cargo interference is accurately calculated using a 3D modeling tool (such as MeshLab), to obtain the simulated cargo interference data.

[0159] Preferably, step S6 comprises the following steps:

[0160] Step S61 : locate the conflict point position according to the interference conflict data, and generate the conflict point position coordinates; perform point cluster concentration calculation on the conflict point position coordinates, and generate the conflict point position concentration degree;

[0161] Step S62: locate the conflict focus based on the conflict point position concentration degree, and obtain the conflict focus region;

[0162] Step S63: perform region mapping on the container skeleton model according to the conflict focus region, and generate the container model conflict region;

[0163] Step S64: perform container restriction analysis on the container model conflict region based on the interference conflict data, and obtain the container restriction factor;

[0164] Step S65: perform variant optimization on the container skeleton model based on the container restriction factor, and generate the adjusted container model.

[0165] In this embodiment, based on the aforementioned simulation cargo interference data, the positioning of the interference conflict points is carried out, the contact points or interference points between the cargos and between the cargos and the container skeleton are identified, the entire container loading area is divided into a plurality of grid units through a three-dimensional gridding method, and the geometric overlap calculation is adopted to determine each conflict point in the interference area. The coordinates of the conflict points are obtained by solving the spatial relationship between the cargo model and the container skeleton, and the specific coordinates of each conflict point are determined through comparison with the grid model and collision detection algorithm (such as AABB collision detection). Through the analysis of the spatial distribution of the conflict points, the point group concentration calculation is carried out. Firstly, the clustering analysis of the conflict points is carried out, the K-means clustering algorithm is adopted to group the conflict points according to the spatial distribution, the concentration of each group of conflict points is calculated, and the degree of concentration reflects the degree of interference between the cargos. For each cluster group, the density of the point group, the geometric center of the point group, the radius of the point group, etc. are calculated to determine the concentration area of each conflict point, and the corresponding conflict point concentration degree data is generated. According to the previously generated conflict point concentration degree data, the clustering result is used to locate the area with high concentration degree, the center position of these conflict point groups is identified using the centroid positioning algorithm, and the conflict focus area is determined. The size of the conflict focus area is directly related to the conflict point concentration degree. The higher the concentration degree is, the larger the conflict focus area is. The determination process of the conflict focus area is realized through the expansion calculation of the cluster center and its surrounding points, which specifically includes defining the local optimal area of each cluster to obtain the boundary of the conflict focus, and according to the data of the conflict focus area, the mapping algorithm is used to compare the corresponding area of the conflict focus with the spatial position of the container skeleton model to determine the area in which interference will occur in the container. The three-dimensional space mapping technology (such as space transformation matrix) is used to map the conflict focus area to the coordinate system of the container skeleton model to generate the conflict area of the container model. The conflict area is delimited by comparing the focus area with the geometric shape of the container skeleton, and then the interference area of the container skeleton is obtained. According to the conflict area of the container model and the interference conflict data, the restriction analysis of the container is carried out. Firstly, the bearing capacity, size, shape and other restriction conditions of the container are evaluated, the spatial analysis tool such as Convex Hull algorithm is used to analyze the relative position relationship between the conflict area and the container model, and the physical restriction factors that need to be followed by the container in handling the conflict are determined.These limiting factors include available space range, maximum load of goods, geometric structure and rigidity constraints of the container, etc. Through these data, the limiting factors of the container are generated, and the container model is optimized by using the limiting factors. First, the shape, size and carrying capacity of the container skeleton model are compared with the limiting factors to judge the optimal adaptability of the container in the current form. The container skeleton is fine-tuned through a variant optimization algorithm (such as a shape optimization based on a genetic algorithm), and the size, structure or form of the container skeleton model is adjusted to maximize the adaptation to the goods loading demand and eliminate interference. In the optimization process, the geometric shape of the container is modified according to the limiting factors to generate a new adjusted container model. Finally, the verification algorithm checks whether the new model meets all the loading and carrying requirements.

[0166] Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting, the scope of the application being defined by the appended claims and not by the above description, therefore all variations falling within the meaning and scope of the equivalent elements of the application file are intended to be included in the present application.

[0167] The above description is merely one specific implementation of the application, which enables those skilled in the art to understand or implement the application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the application. Therefore, the application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for constructing a container model based on three-dimensional software, characterized by, The method comprises the following steps: Step S1: performing three-dimensional laser scanning on the container to obtain container point cloud data; performing feature segmentation on the container point cloud data to generate a container feature set; Step S2: performing feature point spatial positioning on the container feature set to obtain a vertex coordinate set; performing geometric constraint modeling based on the vertex coordinate set to generate a container skeleton model; Step S3: obtaining a target cargo image; performing basic contour extraction on the target cargo image to generate a cargo basic contour image; performing three-dimensional reconstruction on the cargo basic contour image to obtain a reconstructed cargo model; Step S4: performing spatial planning processing on the container skeleton model based on the reconstructed cargo model to obtain an initial cargo planning model; performing barycenter distribution calculation on the initial cargo planning model to generate cargo balance parameters; Step S5: performing interference simulation on the initial cargo planning model based on the cargo balance parameters to obtain simulated cargo interference data; performing conflict detection on the simulated cargo interference data to generate interference conflict data; wherein, step S5 comprises the following steps: Step S51: performing adjacent cargo distance calculation on the initial cargo planning model to obtain adjacent cargo distance parameters; performing translation contact simulation according to the adjacent cargo distance parameters to generate a simulated contact area; Step S52: performing contact pressure distribution analysis based on the cargo balance parameters to generate pressure distribution data; Step S53: performing cargo interference simulation on the simulated contact area according to the pressure distribution data to obtain simulated cargo interference data; Step S54: performing pressure threshold mapping based on the simulated contact area and the pressure distribution data to generate cargo interference pressure thresholds; Step S55: performing conflict point detection on the simulated cargo interference data according to the cargo interference pressure thresholds to generate interference conflict data; Step S6: performing conflict focal point positioning on the interference conflict data to obtain a conflict focal point region; performing variant design on the container skeleton model according to the conflict focal point region to generate an adjusted container model.

2. The three-dimensional software-based container model building method of claim 1, wherein, Step S1 comprises the following steps: Step S11: performing multi-angle scanning on the container to obtain original container point cloud data; performing noise filtering on the original container point cloud data to generate filtered point cloud data; Step S12: performing coordinate conversion processing on the filtered point cloud data to generate container point cloud data; Step S13: performing geometric feature point identification on the container point cloud data to obtain a container feature point set; performing topological relationship analysis on the container feature point set to obtain a feature point topological relationship; Step S14: connecting the container feature point set based on the feature point topological relationship to obtain a feature point connection graph; performing geometric feature analysis on the feature point connection graph to generate a container feature set.

3. The three-dimensional software-based container model building method of claim 1, wherein, Step S2 comprises the following steps: Step S21: performing spatial coordinate mapping on the container feature set to obtain a three-dimensional coordinate point set; performing vertex clustering analysis on the three-dimensional coordinate point set to generate a candidate vertex set; Step S22: performing accurate positioning calculation on the candidate vertex set according to the three-dimensional coordinate point set to obtain a vertex coordinate set; Step S23: Geometric relationship extension is performed on the vertex coordinate set to generate surface domain relationship data; geometric constraint analysis is performed on the surface domain relationship data to generate geometric constraint data; Step S24: A framework structure is established based on the geometric constraint data to obtain an initial skeleton model; the initial skeleton model is filled with node connection based on the surface domain relationship data to generate a container skeleton model.

4. The three-dimensional software-based container model building method of claim 3, wherein, Step S23 includes the following steps: Adjacent relationship analysis is performed on the vertex coordinate point set to obtain a point pair relationship set; edge line connection processing is performed on the point pair relationship set to generate edge line data; Parallel degree detection is performed on the edge line data to obtain parallel edge data; perpendicularity detection is performed on the parallel edge data to generate an orthogonal edge data set; Surface element relationship construction is performed based on the orthogonal edge data set to generate surface domain relationship data; Dimension parameter extraction is performed on the surface domain relationship data to obtain dimension constraint data; Angle relationship analysis is performed on the dimension constraint data to generate angle constraint data; symmetry calculation is performed on the angle constraint data to obtain symmetry constraint data; Constraint integration is performed on the dimension constraint data, the angle constraint data and the symmetry constraint data to generate geometric constraint data.

5. The three-dimensional software-based container model building method of claim 1, wherein, Step S3 includes the following steps: Step S31: Multi-angle shooting is performed on the target cargo to obtain a target cargo image; cargo edge detection is performed on the target cargo image to obtain an edge data set; Step S32: Contour line segmentation processing is performed on the edge data set to obtain a line segment data set; basic contour cutting is performed on the target cargo image based on the line segment data set to generate a cargo basic contour image; Step S33: Stereoscopic matching is performed on the cargo basic contour image to obtain a cargo disparity map set; depth calculation is performed on the cargo disparity map set to generate a depth data set; Step S34: Point cloud conversion is performed on the depth data set to generate a space point set; surface fitting is performed on the space point set to generate a complete surface set; Step S35: Three-dimensional model reconstruction is performed on the complete surface set to obtain a reconstructed cargo model.

6. The three-dimensional software-based container model building method of claim 1, wherein, Step S4 includes the following steps: Step S41: Voxelization processing is performed on the container skeleton model to obtain a space grid set; available space labeling is performed on the space grid set to generate loading partition data; Step S42: Dimension calculation is performed on the reconstructed cargo model to generate cargo parameters; loading sequence sorting is performed based on the cargo parameters to generate a cargo loading sequence; Step S43: Space matching planning is performed on the cargo loading sequence based on the loading partition data to generate an initial cargo planning model; Step S44: Centroid positioning is performed on the initial cargo planning model to obtain cargo centroid data; moment calculation is performed on the cargo centroid data to generate a moment data set; Step S45: Stability analysis is performed on the moment data set to generate cargo balance parameters.

7. The three-dimensional software-based container model building method of claim 6, wherein, Step S43 includes the following steps: Available volume calculation is performed on the loading partition data to generate partition available volume data; boundary restriction mapping is performed based on the partition available volume data to generate partition boundary restriction data; Connectivity analysis is performed on the loading partition data to obtain partition connectivity data; partition bearing capacity evaluation is performed on the loading partition data based on the partition connectivity data to obtain partition bearing capacity; Adaptation capability judgment is performed based on the partition bearing capacity and partition boundary limit data, and a regional adaptation capability value is generated. Bearing demand calculation is performed on the cargo loading sequence, and cargo bearing demand data is obtained. According to the regional adaptation capability value and the cargo bearing demand data, spatial matching planning is performed, and an initial cargo planning model is generated.

8. The three-dimensional software-based container model building method of claim 1, wherein, Step S53 includes the following steps: According to the pressure distribution data, the pressure concentration area of the simulated contact area is located, and the key contact point is obtained. The force direction of the key contact point is identified, and pressure transmission direction data is generated. The key contact point is detected for pressure value, and contact point pressure data is obtained. Based on the pressure transmission direction data and the contact point pressure data, the key contact point is simulated for cargo contact, and simulated cargo contact data is generated. Local deformation analysis is performed on the simulated cargo contact data, and a local deformation data set is obtained. The deformation propagation range is defined based on the local deformation data set, and a deformation propagation range is generated.

9. The three-dimensional software-based container model building method of claim 1, wherein, According to the local deformation data set and the deformation propagation range, relative displacement analysis is performed, and cargo displacement is generated. Based on the cargo displacement, cargo interference mapping is performed to obtain simulated cargo interference data. Step S6 includes the following steps: Step S61: According to the interference conflict data, the conflict point position of the initial cargo planning model is located, and the conflict point coordinate is generated. The point group concentration degree of the conflict point coordinate is calculated, and the conflict point concentration degree is generated. Step S62: Based on the conflict point concentration degree, the conflict focus is located, and the conflict focus area is obtained. Step S63: According to the conflict focus area, the region mapping of the container skeleton model is performed, and the container model conflict area is generated. Step S64: Based on the interference conflict data, the container restriction analysis of the container model conflict area is performed, and the container restriction factor is obtained. Step S65: Based on the container restriction factor, the variant optimization of the container skeleton model is performed, and the adjusted container model is generated.

Images