Vehicle intelligent unloading control method, system, device and medium

Abstract

The embodiment of the application provides a kind of vehicle intelligent unloading control method, system, equipment and medium, belong to industrial automation control technical field.The method comprises: according to the initial image information of the material carried by target vehicle is carried out image analysis and obtains data distribution information;According to the preset distribution information of current position information and data distribution information are compared to determine the first position state;When the first position state meets the preset state, the three-dimensional data of the material carried by target vehicle is collected to obtain three-dimensional point cloud data;According to three-dimensional point cloud data, the contour feature of the material carried by target vehicle is identified to obtain target contour information;According to target contour information and target transport type, determine the carrying volume information;According to the maximum carrying volume corresponding to current position information and carrying volume information, determine the second position state corresponding to target vehicle;According to the second position state, control target vehicle to unload material, obtain the target unloading result corresponding to target vehicle.

Description

Technical Field

[0001] This invention relates to the field of industrial automation control technology, and in particular to a vehicle intelligent unloading control method, system, device, and medium. Background Technology

[0002] In current material handling operations, the scheduling and management of unloading areas still heavily rely on human experience to determine material types and release decisions. This approach is not only inefficient and slow-responding, but also prone to misjudgments due to visual fatigue and subjective biases. In recent years, with the development of artificial intelligence, the introduction of image recognition technology for real-time analysis and classification of transported materials can significantly improve the intelligence level of the process. However, existing recognition models still face many challenges in practical implementation: complex ambient lighting, material surface reflections, vehicle obstructions, and the diversity of material shapes often lead to large fluctuations in recognition accuracy and insufficient robustness, resulting in frequent interruptions of automated processes and requiring manual intervention for verification. Ultimately, this prevents a fundamental improvement in the overall efficiency of unloading operations. Summary of the Invention

[0003] The main objective of this invention is to provide a vehicle intelligent unloading control method, system, device, and medium, aiming to solve the problem of low accuracy in the judgment process during automatic unloading of transported materials in the prior art, which leads to poor overall efficiency of unloading operations.

[0004] In a first aspect, embodiments of the present invention provide a vehicle intelligent unloading control method, comprising:

[0005] Initial image information is obtained by acquiring images of the materials being transported on the target vehicle, and image analysis is performed based on the initial image information to obtain data distribution information corresponding to the materials being transported on the target vehicle.

[0006] Collect the current location information corresponding to the target vehicle, and compare the preset distribution information corresponding to the current location information with the data distribution information to determine the first location state corresponding to the target vehicle;

[0007] When the first position state meets the preset state, the three-dimensional data of the transported materials of the target vehicle is collected to obtain the three-dimensional point cloud data corresponding to the target vehicle.

[0008] Based on the three-dimensional point cloud data, the contour features of the material carried by the target vehicle are identified to obtain the target contour information corresponding to the target vehicle.

[0009] The required carrying volume information for the target vehicle to unload is determined based on the target contour information and the target transport type.

[0010] Obtain the maximum load-bearing volume corresponding to the current location information, and determine the second location state of the target vehicle based on the maximum load-bearing volume and the load-bearing volume information;

[0011] Based on the second position status, the target vehicle is controlled to unload materials, and the target unloading result corresponding to the target vehicle is obtained.

[0012] In a second aspect, embodiments of the present invention provide a vehicle intelligent unloading control system, comprising:

[0013] The data analysis module is used to acquire images of the materials being transported by the target vehicle to obtain initial image information, and to perform image analysis based on the initial image information to obtain data distribution information corresponding to the materials being transported by the target vehicle.

[0014] The status analysis module is used to collect the current location information of the target vehicle and compare it with the preset distribution information and the data distribution information corresponding to the current location information to determine the first location status of the target vehicle.

[0015] The data acquisition module is used to acquire three-dimensional data of the transported materials of the target vehicle to obtain the three-dimensional point cloud data corresponding to the target vehicle when the first position state meets the preset state.

[0016] The contour recognition module is used to identify the contour features of the transported materials of the target vehicle based on the three-dimensional point cloud data, and obtain the target contour information corresponding to the target vehicle.

[0017] The volume determination module is used to determine the carrying volume information required by the target vehicle when unloading, based on the target contour information and the target transportation type.

[0018] The state determination module is used to obtain the maximum load-bearing volume corresponding to the current location information, and determine the second position state corresponding to the target vehicle based on the maximum load-bearing volume and the load-bearing volume information;

[0019] The material unloading module is used to control the target vehicle to unload materials according to the second position state, and to obtain the target unloading result corresponding to the target vehicle.

[0020] Thirdly, embodiments of the present invention also provide a terminal device, the terminal device including a processor, a memory, a computer program stored in the memory and executable by the processor, and a data bus for implementing communication between the processor and the memory, wherein when the computer program is executed by the processor, it implements the steps of any of the vehicle intelligent unloading control methods provided in this specification.

[0021] Fourthly, embodiments of the present invention also provide a storage medium for computer-readable storage, characterized in that the storage medium stores one or more programs, which can be executed by one or more processors to implement the steps of any of the vehicle intelligent unloading control methods provided in this specification.

[0022] This invention provides a vehicle intelligent unloading control method, system, device, and medium. The method includes: acquiring initial image information of the material being transported by a target vehicle; performing image analysis based on the initial image information to obtain data distribution information corresponding to the material being transported by the target vehicle; acquiring the current position information of the target vehicle; comparing the current position information with preset distribution information and data distribution information to determine a first position state of the target vehicle; when the first position state meets the preset state, acquiring three-dimensional data of the material being transported by the target vehicle to obtain three-dimensional point cloud data corresponding to the target vehicle; performing contour feature recognition on the material being transported by the target vehicle based on the three-dimensional point cloud data to obtain target contour information corresponding to the target vehicle; determining the carrying volume information required for unloading by the target vehicle based on the target contour information and the target transport type; obtaining the maximum carrying volume corresponding to the current position information; determining a second position state of the target vehicle based on the maximum carrying volume and carrying volume information; and controlling the target vehicle to unload the material based on the second position state to obtain the target unloading result corresponding to the target vehicle. This method first acquires and analyzes images of the transported materials in real time to obtain data distribution information. Then, it compares the current location information with a preset distribution to determine the first position state, effectively verifying whether the vehicle is in a compliant unloading area and avoiding misoperation. When conditions are met, it accurately calculates the material's carrying volume information through 3D point cloud data acquisition and contour feature recognition, and compares it with the maximum carrying volume at the current location to determine the second position state. This intelligently controls the unloading process, improving not only unloading efficiency and safety but also reducing human error through data-driven decision-making, achieving efficient and reliable material management. It also solves the problem of low accuracy in the judgment process during automatic unloading of transported materials in existing technologies, leading to poor overall unloading efficiency. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 A flowchart illustrating a vehicle intelligent unloading control method provided in an embodiment of the present invention;

[0025] Figure 2 A schematic diagram of the module structure of a vehicle intelligent unloading control system provided in an embodiment of the present invention;

[0026] Figure 3 This is a schematic block diagram of a terminal device provided in an embodiment of the present invention. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0029] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0030] This invention provides a vehicle intelligent unloading control method, system, device, and medium. The vehicle intelligent unloading control method can be applied to a terminal device, which can be an electronic device such as a tablet computer, laptop computer, desktop computer, personal digital assistant, or wearable device. The terminal device can be a server or a server cluster.

[0031] The following detailed description of some embodiments of the present invention is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0032] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a vehicle intelligent unloading control method provided in an embodiment of the present invention.

[0033] like Figure 1 As shown, the intelligent unloading control method for vehicles includes steps S101 to S107.

[0034] Step S101: Acquire initial image information by image acquisition of the materials carried by the target vehicle, and perform image analysis based on the initial image information to obtain data distribution information corresponding to the materials carried by the target vehicle.

[0035] For example, the target vehicle is a vehicle carrying materials that need to be unloaded. The target vehicle is parked in the unloading area. An image acquisition device is installed above the unloading area. The image acquisition device is used to acquire images of the materials carried on the target vehicle to obtain initial image information.

[0036] For example, edge analysis algorithms are used to perform edge recognition on the initial image information to obtain the corresponding edge image. Then, target segmentation is performed based on the edge image to obtain the target object in the initial image information. The size or volume information of the target object is calculated based on the edge image. Then, data distribution analysis is performed based on the size or volume information of the target object to obtain the data distribution information corresponding to the material being transported on the target vehicle.

[0037] In some embodiments, the step of obtaining data distribution information corresponding to the materials transported on the target vehicle by performing image analysis based on the initial image information includes: performing bilateral filtering on the initial image information to obtain filtered image information corresponding to the initial image information; performing edge detection on the filtered image information to obtain edge image information corresponding to the filtered image information; obtaining edge information corresponding to each sub-edge in the edge image information, and obtaining adjacent information corresponding to the sub-edge from the edge image information; and performing data analysis based on the edge information and the adjacent information to obtain the data distribution information corresponding to the materials transported on the target vehicle.

[0038] For example, the parameters selected for the bilateral filter include the spatial domain standard deviation and the value domain standard deviation, and then the weighted average of the pixels in the neighborhood of each pixel in the initial image information is calculated. The weights are jointly determined by the spatial Gaussian function and the value domain Gaussian function: the spatial weights are based on the distance between pixels, and the value domain weights are based on the difference in pixel values, thereby outputting the filtered image information after bilateral filtering based on the weighted average.

[0039] For example, edge detection algorithms (such as Canny edge detection) are used to process the filtered image information, calculate the gradient magnitude and direction of each pixel in the filtered image information, and then refine the edges by retaining only the pixels with the largest local gradient magnitude. Thus, double thresholds are used to connect the edge pixels to generate continuous edges and obtain edge image information.

[0040] For example, edge image information is used to form continuous sub-edges by connecting adjacent edge pixels, which are then converted into edge chains or contour representations to obtain edge information corresponding to each sub-edge. The edge information includes the coordinates of the start point, end point, and midpoint of the sub-edge, as well as the geometric features of the sub-edge such as length (number of pixels), direction (average angle), curvature (degree of bending) or shape descriptor, and statistical features such as average gradient magnitude or intensity.

[0041] For example, for each sub-edge, the distance between sub-edges is calculated using Euclidean distance (e.g.), or a neighborhood search is used based on bounding boxes or spatial indexes to find other spatially close sub-edges in the edge image information. This yields the neighbor information corresponding to each sub-edge, which includes the neighboring edges adjacent to the sub-edge and the edge information corresponding to those neighboring edges.

[0042] For example, information distribution analysis is performed based on the edge information corresponding to adjacent edges and the edge information corresponding to sub-edges. For example, by statistically analyzing the average size, volume estimation, or edge shape distribution of sub-edges using edge information, data distribution information corresponding to the materials transported on the target vehicle can be obtained.

[0043] Specifically, by revealing the spatial relationships and layout patterns between materials, the data distribution information corresponding to the materials transported on the target vehicle is obtained. This overall process can efficiently and automatically quantify the size and spatial arrangement of materials, providing reliable data support for subsequent material type identification, thereby improving the accuracy and efficiency of image analysis.

[0044] In some embodiments, the step of performing edge detection processing on the filtered image information to obtain edge image information corresponding to the filtered image information includes: determining a multi-directional convolution template, and performing convolution processing on the filtered image information according to the multi-directional convolution template to obtain multiple target convolution images; fusing the multiple target convolution images to determine the gradient angle corresponding to each pixel in the filtered image information; obtaining the neighboring points corresponding to the pixel, and constructing a first target line according to the neighboring points; constructing a second target line according to the gradient angle and the pixel, and calculating the intersection point of the first target line and the second target line to obtain target intersection point information; from The relevant points associated with the target intersection information are obtained from the adjacent points, and the target pixel value corresponding to the target intersection information is determined according to the pixel information corresponding to the relevant points; the filtered image information is interpolated according to the target pixel value and the target intersection information to obtain an interpolated image; the interpolated image is thresholded to obtain a first segmented image, and the filtered image information is edge segmented to obtain a second segmented image; the segmentation difference map between the first segmented image and the second segmented image is calculated, and the second segmented image is supplemented with data according to the segmentation difference map to obtain the edge image information corresponding to the filtered image information.

[0045] For example, a set of convolution kernels with four directions such as 0°, 45°, 90° and 135° is designed, and then this set of multi-directional convolution templates is used to perform convolution operations on the input filtered image information. Each convolution will produce a target convolution image, where the brightness value of each pixel represents the edge intensity of the original image at that pixel point along the corresponding template direction.

[0046] For example, for each pixel in the fine-grained system of the filtered image, its response value in all target convolutional images is compared, and then the direction of the maximum gradient intensity of the pixel is determined. This direction is the gradient angle corresponding to the pixel.

[0047] For example, for the currently processed pixel, the pixels in its surrounding 3x3 or 5x5 neighborhood are obtained, and then the pixels in the horizontal or vertical direction are connected to form multiple horizontal or vertical straight lines, thereby obtaining multiple first target straight lines. Furthermore, the pixels in the surrounding 3x3 or 5x5 neighborhood of the pixel are taken as the neighboring points of that pixel.

[0048] For example, a pixel is identified as a necessary point for the second straight line, and its calculated gradient angle is used as the direction to construct the second straight line, called the second target straight line. This straight line represents the normal direction of the edge. Subsequently, the intersection points of multiple first and second target straight lines are calculated. This target intersection point information is the coordinate of the line intersection point. The target intersection point information is located at the sub-pixel level, which can more accurately locate the actual edge position than the original pixel grid.

[0049] For example, on the first target line, the pixel closest to the target intersection point information, i.e., the correlation point, is found. Based on the pixel values ​​of these correlation points, a more precise target pixel value at the target intersection point is calculated. This process is repeated for each pixel in the filtered image information, calculating a sub-pixel intersection point and its target pixel value for each point. All calculated target pixel values ​​are then integrated to form a higher-resolution interpolated image. This image contains sub-pixel detail.

[0050] For example, thresholding is performed on the interpolated image to generate a binary image, resulting in a first segmented image. This first segmented image may contain finer edges. Simultaneously, a conventional edge segmentation method, such as the Canny algorithm, is applied directly to the original filtered image information to generate another binary image, resulting in a second segmented image.

[0051] For example, a high threshold and a low threshold are used. Pixels with an intensity higher than the high threshold are identified as strong edges. Pixels with an intensity between the low and high thresholds are marked as weak edges. Only when a weak edge pixel is connected to a strong edge pixel is it accepted as a true edge. The result of processing the image using only the low threshold contains very rich edge information, including many weak edges and noise, but the edges are coarse and discontinuous.

[0052] For example, a segmentation difference map is calculated by comparing the first segmented image and the second segmented image. This segmentation difference map clearly indicates edge regions present in the first segmented image but missing in the second segmented image. This segmentation difference map serves as a candidate material library, containing all pixels that were discarded by the second segmented image due to insufficient strength or disconnection from the main edge, but which may still belong to real edges. Subsequent patching uses materials drawn from this library.

[0053] For example, clockwise tracing is performed along the contour of the second segmented image. When an edge line reaches its end and no next connected edge pixel is found, this terminating pixel is identified as a breakpoint. This breakpoint is the starting position for edge repair. The breakpoint is then used to find edge points in the 5×5 neighborhood of the segmentation difference map. If an edge point exists, it is repaired onto the second segmented image to obtain the edge image information corresponding to the filtered image information.

[0054] Specifically, by connecting broken edges through intelligent, direction-based local search, while effectively avoiding the introduction of irrelevant noise, the continuity and integrity of the edges can be significantly improved, thereby recovering the true weak edges and ultimately obtaining a higher quality edge image that is more conducive to subsequent analysis.

[0055] Step S102: Collect the current location information corresponding to the target vehicle, and compare the preset distribution information corresponding to the current location information with the data distribution information to determine the first location state corresponding to the target vehicle.

[0056] For example, the current location information of the target vehicle is determined by the image acquisition device triggered by the target vehicle, that is, the current location information is the unloading area currently corresponding to the target vehicle.

[0057] For example, the preset distribution information corresponding to the stored materials under the current location information is obtained from the database. For example, if the stored material is sand, the preset distribution information is the data distribution information corresponding to sand; if the stored material is gravel, the preset distribution information is the data distribution information corresponding to gravel.

[0058] For example, the similarity between the preset distribution information and the data distribution information is calculated using the KL distribution. If the similarity is greater than the preset value, the first position state corresponding to the target vehicle is determined to be unloadable; otherwise, the first position state corresponding to the target vehicle is determined to be unloadable.

[0059] In some implementations, determining the first location state of the target vehicle by comparing the preset distribution information corresponding to the current location information with the data distribution information includes: performing type identification on the transported materials of the target vehicle based on the initial image information to obtain type distribution information; obtaining a preset type corresponding to the current location information, and obtaining first probability information corresponding to the preset type from the type distribution information; calculating distribution similarity based on the preset distribution information and the data distribution information to obtain second probability information; fusing the first probability information and the second probability information to determine the target type of the target vehicle, and determining the first location state of the target vehicle based on the target type.

[0060] For example, a pre-trained image recognition model, such as a deep learning classification network, is used to perform type analysis on the initial image information to obtain type distribution information. The type distribution information indicates the probability that the material in the image belongs to each predefined type, such as sand, gravel, container, stone, etc.

[0061] For example, based on the current location information, the expected material type at that location, i.e., the preset type, is queried from a preset rule base. For instance, the preset type might be sand or stone. From the type distribution information, the probability value corresponding to the preset type is extracted as the first probability information. This probability reflects the degree of consistency between the image recognition result and prior location knowledge.

[0062] For example, the preset distribution information and the data distribution information have the same data type. The preset distribution type is the data information stored after the material type is known under the current location information. Then, the Jensen-Shannon divergence is used to combine the preset distribution information and the data distribution information to calculate the similarity and obtain the second probability information. This probability represents the overall matching degree between the actual observed distribution and the expected distribution.

[0063] For example, a weighted average and Bayesian inference are used to fuse the first probability information and the second probability information to determine the target probability information when the material loaded by the target vehicle is of a preset type. When the target probability information is greater than or equal to the preset probability, the target type of the target vehicle is determined to be the preset type; when the target probability information is less than the preset probability, the target type of the target vehicle is determined to be not the preset type.

[0064] For example, when the target type is a preset type, it is determined that it is consistent with the preset type of the current location information, and the first location status of the target vehicle is marked as normal; otherwise, it is marked as abnormal.

[0065] Specifically, this step determines the status by judging whether the materials loaded on the target vehicle are consistent with known preset types. By deeply integrating real-time perception data with domain prior knowledge, it achieves a leap from passive identification to active verification. This provides a higher level of reliability and accuracy for status determination in complex real-world scenarios.

[0066] Step S103: When the first position state meets the preset state, three-dimensional data acquisition is performed on the transported materials of the target vehicle to obtain the three-dimensional point cloud data corresponding to the target vehicle.

[0067] For example, if the preset state is normal, then when the first position state is normal, the lidar corresponding to the current position information is used to collect three-dimensional data of the material being transported by the target vehicle, thereby obtaining the three-dimensional point cloud data corresponding to the target vehicle.

[0068] For example, when the first position status is abnormal, the loading material type with the highest probability is obtained from the type distribution information, and this highest probability is determined as the third probability information. Then, historical data distribution information corresponding to the loading material type is obtained from the database, and the similarity between the historical data distribution information and the data distribution information corresponding to the target vehicle is calculated and determined as the fourth probability information. The third and fourth probability information are then fused to obtain the final probability information corresponding to the loading material type. When the final probability information is greater than or equal to a preset probability, the material unloading position corresponding to the loading material type is obtained, and a control command to depart for the material unloading position is sent to the target vehicle so that the target vehicle unloads the material at the material unloading position. When the final probability information is less than the preset probability, the actual material type corresponding to the target vehicle is determined manually, thereby obtaining the material unloading position corresponding to the actual material type. Then, a control command to depart for the material unloading position is sent to the target vehicle so that the target vehicle unloads the material at the material unloading position.

[0069] Step S104: Based on the three-dimensional point cloud data, perform contour feature recognition on the transported materials of the target vehicle to obtain the target contour information corresponding to the target vehicle.

[0070] For example, statistical filtering is applied to the 3D point cloud data to identify and remove discrete noise points whose average distance to their neighbors exceeds a reasonable threshold, thus laying the foundation for subsequent accurate segmentation and contour extraction. Subsequently, based on spatial relative position, geometric structural features, or clustering algorithms, the preprocessed 3D point cloud data is divided into several independent point clusters. During this process, based on prior knowledge that the vehicle body typically has a regular geometric shape and fixed position, point cloud clusters belonging to the vehicle body are identified and removed. Finally, the remaining point cloud clusters within the vehicle compartment are determined to be the point cloud of the transported materials.

[0071] For example, the 3D convex hull of the point cloud containing the transported material is calculated. The 3D convex hull is the smallest convex polyhedron containing all points in the point cloud, which constructs the most compact convex boundary enclosing the material. The closed triangular mesh surface defined by this 3D convex hull is the desired target contour information.

[0072] In some embodiments, the step of identifying the contour features of the transported materials of the target vehicle based on the three-dimensional point cloud data to obtain the target contour information corresponding to the target vehicle includes: performing curvature feature analysis on the three-dimensional point cloud data to obtain curvature distribution information corresponding to the three-dimensional point cloud data; performing contour filtering based on the curvature distribution information and the three-dimensional point cloud data to obtain the initial contour information corresponding to the target vehicle; and performing contour correction on the initial contour information based on surface fitting on the three-dimensional point cloud data to obtain the target contour information corresponding to the target vehicle.

[0073] For example, for each point in the 3D point cloud data, its K nearest neighbors are selected, and then within this neighborhood, the normal vector and the rate of change of the surface at that point are estimated through surface fitting. After calculating for all points, the curvature value corresponding to each point is obtained, and feature analysis is performed based on all curvature values ​​to obtain the curvature distribution information corresponding to the 3D point cloud data.

[0074] For example, curvature distribution information is used as a guide to initially extract a set of feature points representing the basic contour of the target vehicle from 3D point cloud data. For instance, a suitable curvature threshold is set. This threshold is used to distinguish between feature regions and non-feature regions that may belong to the contour. The 3D point cloud data is then traversed to filter out all points with curvature values ​​higher than this threshold. These high-curvature points constitute the approximate boundaries, edges, and other contour feature points of the object. This set of filtered contour feature points is then used as the initial contour information.

[0075] For example, guided by initial contour information, surface fitting is performed under the constraints of 3D point cloud data. This process essentially approximates the real object surface with a smooth, continuous mathematical surface. For instance, in sparse or missing areas of the point cloud, the complete contour shape is inferred from the fitted surface model, ensuring the contour lines strictly conform to the fitted smooth geometric surface, thus achieving sub-pixel accuracy. Then, from the corrected smooth surface model, clear, continuous, and accurate contour lines are re-extracted by calculating surface intersections, thereby obtaining the target contour information. The target contour information is represented as a series of continuous 3D spatial curves or polylines, providing a reliable geometric basis for subsequent applications such as volume calculation and deformation analysis.

[0076] For example, target contour information is used to characterize the contour information of the target vehicle's onboard object above the truck bed.

[0077] In some embodiments, the step of performing surface fitting to correct the initial contour information based on the three-dimensional point cloud data to obtain the target contour information corresponding to the target vehicle includes: obtaining relevant point cloud data involved in the initial contour information from the three-dimensional point cloud data; performing vector calculation on each initial sub-point cloud data in the relevant point cloud data to obtain target direction information corresponding to the initial sub-point cloud data; obtaining initial neighboring point cloud information corresponding to the initial sub-point cloud data from the three-dimensional point cloud data based on the target direction information; performing morphological feature analysis and filtering on the initial sub-point cloud data based on the initial neighboring point cloud information to obtain target sub-point cloud data; obtaining target neighboring point cloud information corresponding to the target sub-point cloud data, and performing surface fitting based on the target neighboring point cloud information to correct the initial contour information to obtain the target contour information corresponding to the target vehicle.

[0078] For example, relevant point cloud data associated with the initial contour information is extracted from the 3D point cloud data; then, for each independent initial sub-point cloud, the target direction information representing the local geometric extension direction is determined by vector calculation using the relevant point cloud data.

[0079] For example, based on the target direction information, the initial neighboring point cloud information that is spatially adjacent and geometrically continuous with the initial sub-point cloud is intelligently retrieved and obtained from the three-dimensional point cloud data; then, by performing morphological feature analysis on the initial neighboring point clouds, such as evaluating point cloud density, curvature consistency or plane fitting degree, a set of points with significant features and excellent quality is selected from the initial sub-point cloud to constitute the target sub-point cloud data.

[0080] For example, the target neighboring point cloud information corresponding to the target sub-point cloud data is obtained from the 3D point cloud data, and high-order surface fitting is performed based on this. This fitting process can globally optimize and smooth the local geometry, thereby achieving accurate correction of the initial contour information, and finally outputting high-precision target contour information that can accurately describe the loading material of the target vehicle.

[0081] Step S105: Determine the carrying volume information required for the target vehicle to unload cargo based on the target contour information and the target transportation type.

[0082] For example, by performing three-dimensional geometric analysis on the target contour information, the protrusion height and indentation depth of the material surface relative to the standard truck bed reference surface are identified, accurately quantifying the actual shape changes of the loaded material. Subsequently, the standard capacity information of the truck bed for this vehicle model, including key parameters such as geometric dimensions and rated volume, is retrieved from a pre-set database based on the vehicle's unique identifier.

[0083] For example, the acquired capacity information and contour change information are spatially mapped and correlated. Simultaneously, the target transport type, such as gravel or sand, is determined based on the first position status, and a type-adaptive volume calculation model is established. For protruding portions, the incremental volume is calculated using solid geometric integration, while for recessed portions, volume is deducted. Furthermore, the bulk density and porosity are compensated and corrected using a material type coefficient, ultimately synthesizing and calculating high-precision load-bearing volume information.

[0084] Step S106: Obtain the maximum carrying volume corresponding to the current location information, and determine the second location state corresponding to the target vehicle based on the maximum carrying volume and the carrying volume information.

[0085] For example, the maximum carrying capacity of the unloading area under the current location information can be obtained from the database. When the maximum carrying capacity is greater than or equal to the carrying capacity information, the second location status of the target vehicle under the current location information is determined to be normal; when the maximum carrying capacity is less than or equal to the carrying capacity information, the second location status of the target vehicle under the current location information is determined to be abnormal.

[0086] Step S107: Control the target vehicle to unload materials according to the second position state, and obtain the target unloading result corresponding to the target vehicle.

[0087] For example, when the second position status is normal, the target vehicle is controlled to unload materials, and the target unloading result corresponding to the target vehicle is obtained as unloaded; when the second position status is abnormal, the relevant unloading area corresponding to the target transportation type is obtained from the database, and the unloading capacity of the relevant unloading area is ensured to be greater than the carrying capacity information, thereby obtaining the final unloading area, and the target vehicle is controlled to reach the final unloading area to unload materials, and the target unloading result corresponding to the target vehicle is obtained as unloaded.

[0088] In some embodiments, after obtaining the target unloading result, the method further includes: when the target unloading result is unloaded, obtaining the latest location information corresponding to the target vehicle; determining the distance information between the target vehicle and the target unloading area corresponding to the current location information based on the latest location information and the current location information; when the distance information meets a preset condition, controlling the barrier bar of the target unloading area to be lowered so that subsequent vehicles can perform accidental unloading operations.

[0089] For example, when it is confirmed that the target vehicle has been unloaded, the latest location information of the target vehicle is obtained through a positioning system such as GPS. Then, the latest location information is compared with the current location information (i.e., the coordinates of the target unloading area) to calculate the distance between the two. The key here is to determine whether the target vehicle has safely left the unloading operation area.

[0090] For example, distance information is matched against preset conditions. These preset conditions are typically defined as the distance between the vehicle and the unloading area being greater than a preset safety threshold. This threshold ensures that the vehicle has sufficient space to leave without obstructing subsequent vehicles. Therefore, when the distance information meets the preset conditions, meaning the vehicle has safely left, a lowering command is automatically sent to the barrier control system of the target unloading area. After the barrier is lowered, the lock on the unloading position is released, allowing subsequent queuing vehicles to enter the position for unloading operations, thus ensuring the continuity and safety of the unloading process.

[0091] In addition, it should be noted that when the first position state or the second position state is abnormal, the current position information will be updated according to the changes in the unloading area. The current position information is used to characterize the position information corresponding to the final unloading of the target vehicle.

[0092] Please see Figure 2 , Figure 2 This application provides an intelligent vehicle unloading control system 200, which includes a data analysis module 201, a state analysis module 202, a data acquisition module 203, a contour recognition module 204, a volume determination module 205, a state determination module 206, and a material unloading module 207. The data analysis module 201 is used to acquire images of the material being transported by the target vehicle to obtain initial image information, and to perform image analysis based on the initial image information to obtain data distribution information corresponding to the material being transported on the target vehicle. The state analysis module 202 is used to acquire the current position information corresponding to the target vehicle, and to determine a first position state corresponding to the target vehicle by comparing the preset distribution information corresponding to the current position information with the data distribution information. The data acquisition module 203 is used to acquire three-dimensional data of the material being transported by the target vehicle to obtain three-dimensional point cloud data corresponding to the target vehicle when the first position state meets the preset state. The contour recognition module... Block 204 is used to identify the contour features of the transported material of the target vehicle based on the three-dimensional point cloud data, and obtain the target contour information corresponding to the target vehicle; Volume determination module 205 is used to determine the carrying volume information required for the target vehicle to unload based on the target contour information and the target transport type; State determination module 206 is used to obtain the maximum carrying volume corresponding to the current position information, and determine the second position state corresponding to the target vehicle based on the maximum carrying volume and the carrying volume information; Material unloading module 207 is used to control the target vehicle to unload materials based on the second position state, and obtain the target unloading result corresponding to the target vehicle.

[0093] In some implementations, during the process of obtaining data distribution information corresponding to the materials transported on the target vehicle by performing image analysis based on the initial image information, the data analysis module 201 performs the following:

[0094] The initial image information is subjected to bilateral filtering to obtain the filtered image information corresponding to the initial image information;

[0095] Edge detection processing is performed on the filtered image information to obtain the edge image information corresponding to the filtered image information;

[0096] Obtain edge information corresponding to each sub-edge in the edge image information, and obtain adjacent information corresponding to the sub-edge from the edge image information;

[0097] Data distribution information corresponding to the materials transported on the target vehicle is obtained by performing data analysis based on the edge information and the adjacent information.

[0098] In some implementations, during the process of performing edge detection processing on the filtered image information to obtain the edge image information corresponding to the filtered image information, the data analysis module 201 performs the following:

[0099] A multi-directional convolution template is determined, and the filtered image information is convolved according to the multi-directional convolution template to obtain multiple target convolution images;

[0100] By fusing multiple target convolutional images, the gradient angle corresponding to each pixel in the filtered image information is determined;

[0101] Obtain the neighboring points corresponding to the pixel, and construct a first target straight line based on the neighboring points;

[0102] A second target line is constructed based on the gradient angle and the pixel, and the target intersection point information is obtained by calculating the intersection point of the first target line and the second target line.

[0103] Obtain relevant points associated with the target intersection information from the adjacent points, and determine the target pixel value corresponding to the target intersection information based on the pixel information corresponding to the relevant points;

[0104] The filtered image information is interpolated based on the target pixel value and the target intersection information to obtain an interpolated image;

[0105] The interpolated image is subjected to threshold segmentation to obtain a first segmented image, and the filtered image information is subjected to edge segmentation to obtain a second segmented image;

[0106] Calculate the segmentation difference map between the first segmented image and the second segmented image, and supplement the second segmented image with data based on the segmentation difference map to obtain the edge image information corresponding to the filtered image information.

[0107] In some implementations, during the process of determining the first location state of the target vehicle by comparing the preset distribution information corresponding to the current location information with the data distribution information, the state analysis module 202 performs the following:

[0108] Based on the initial image information, the type of materials carried by the target vehicle are identified to obtain type distribution information;

[0109] Obtain the preset type corresponding to the current location information, and obtain the first probability information corresponding to the preset type from the type distribution information;

[0110] The second probability information is obtained by calculating the distribution similarity based on the preset distribution information and the data distribution information;

[0111] The target type corresponding to the target vehicle is determined by fusing the first probability information and the second probability information, and the first position state corresponding to the target vehicle is determined based on the target type.

[0112] In some implementations, during the process of recognizing the contour features of the transported materials of the target vehicle based on the three-dimensional point cloud data to obtain the target contour information corresponding to the target vehicle, the contour recognition module 204 performs the following:

[0113] Curvature distribution information corresponding to the three-dimensional point cloud data is obtained by performing curvature feature analysis on the three-dimensional point cloud data.

[0114] Based on the curvature distribution information and the three-dimensional point cloud data, the initial contour information corresponding to the target vehicle is obtained by contour filtering.

[0115] Based on the three-dimensional point cloud data, surface fitting is performed to correct the initial contour information, thereby obtaining the target contour information corresponding to the target vehicle.

[0116] In some embodiments, during the process of performing contour correction on the initial contour information based on the three-dimensional point cloud data to obtain the target contour information corresponding to the target vehicle, the contour recognition module 204 performs the following:

[0117] Obtain the relevant point cloud data involved in the initial contour information from the three-dimensional point cloud data;

[0118] Vector calculation is performed on each initial sub-point cloud data in the relevant point cloud data to obtain the target direction information corresponding to the initial sub-point cloud data;

[0119] Based on the target direction information, the initial adjacent point cloud information corresponding to the initial sub-point cloud data is obtained from the three-dimensional point cloud data;

[0120] Based on the initial adjacent point cloud information, the initial sub-point cloud data is analyzed and filtered for morphological features to obtain the target sub-point cloud data;

[0121] Obtain the target neighboring point cloud information corresponding to the target sub-point cloud data, and perform surface fitting based on the target neighboring point cloud information to perform contour correction on the initial contour information, thereby obtaining the target contour information corresponding to the target vehicle.

[0122] In some implementations, after obtaining the target unloading result, the vehicle intelligent unloading control system 200 further performs the following:

[0123] If the target unloading result is "unloaded", then the latest location information corresponding to the target vehicle is obtained;

[0124] Based on the latest location information and the current location information, determine the distance information between the target vehicle and the target unloading area corresponding to the current location information;

[0125] When the distance information meets the preset conditions, the barrier bar of the target unloading area is lowered to prevent subsequent vehicles from unloading.

[0126] In some implementations, the vehicle intelligent unloading control system 200 can be applied to terminal equipment.

[0127] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the vehicle intelligent unloading control system 200 described above can be referred to the corresponding process in the aforementioned vehicle intelligent unloading control method embodiment, and will not be repeated here.

[0128] Please see Figure 3 , Figure 3 This is a schematic block diagram of the structure of a terminal device provided in an embodiment of the present invention.

[0129] like Figure 3 As shown, the terminal device 300 includes a processor 301 and a memory 302, which are connected via a bus 303, such as an I2C (Inter-integrated Circuit) bus.

[0130] Specifically, processor 301 provides computing and control capabilities to support the operation of the entire terminal device. Processor 301 can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0131] Specifically, the memory 302 can be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a portable hard drive, etc.

[0132] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the embodiments of the present invention, and does not constitute a limitation on the terminal device to which the embodiments of the present invention are applied. A specific server may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0133] The processor is used to run a computer program stored in a memory, and when executing the computer program, implements any of the vehicle intelligent unloading control methods provided in the embodiments of the present invention.

[0134] In one embodiment, the processor is configured to run a computer program stored in memory, and when executing the computer program, perform the following steps:

[0135] Initial image information is obtained by acquiring images of the materials being transported on the target vehicle, and image analysis is performed based on the initial image information to obtain data distribution information corresponding to the materials being transported on the target vehicle.

[0136] Collect the current location information corresponding to the target vehicle, and compare the preset distribution information corresponding to the current location information with the data distribution information to determine the first location state corresponding to the target vehicle;

[0137] When the first position state meets the preset state, the three-dimensional data of the transported materials of the target vehicle is collected to obtain the three-dimensional point cloud data corresponding to the target vehicle.

[0138] Based on the three-dimensional point cloud data, the contour features of the material carried by the target vehicle are identified to obtain the target contour information corresponding to the target vehicle.

[0139] The required carrying volume information for the target vehicle to unload is determined based on the target contour information and the target transport type.

[0140] Obtain the maximum load-bearing volume corresponding to the current location information, and determine the second location state of the target vehicle based on the maximum load-bearing volume and the load-bearing volume information;

[0141] Based on the second position status, the target vehicle is controlled to unload materials, and the target unloading result corresponding to the target vehicle is obtained.

[0142] In some embodiments, during the process of obtaining data distribution information corresponding to the materials transported on the target vehicle by performing image analysis based on the initial image information, the processor 301 executes:

[0143] The initial image information is subjected to bilateral filtering to obtain the filtered image information corresponding to the initial image information;

[0144] Edge detection processing is performed on the filtered image information to obtain the edge image information corresponding to the filtered image information;

[0145] Obtain edge information corresponding to each sub-edge in the edge image information, and obtain adjacent information corresponding to the sub-edge from the edge image information;

[0146] Data distribution information corresponding to the materials transported on the target vehicle is obtained by performing data analysis based on the edge information and the adjacent information.

[0147] In some embodiments, during the process of performing edge detection processing on the filtered image information to obtain the edge image information corresponding to the filtered image information, the processor 301 executes:

[0148] A multi-directional convolution template is determined, and the filtered image information is convolved according to the multi-directional convolution template to obtain multiple target convolution images;

[0149] By fusing multiple target convolutional images, the gradient angle corresponding to each pixel in the filtered image information is determined;

[0150] Obtain the neighboring points corresponding to the pixel, and construct a first target straight line based on the neighboring points;

[0151] A second target line is constructed based on the gradient angle and the pixel, and the target intersection point information is obtained by calculating the intersection point of the first target line and the second target line.

[0152] Obtain relevant points associated with the target intersection information from the adjacent points, and determine the target pixel value corresponding to the target intersection information based on the pixel information corresponding to the relevant points;

[0153] The filtered image information is interpolated based on the target pixel value and the target intersection information to obtain an interpolated image;

[0154] The interpolated image is subjected to threshold segmentation to obtain a first segmented image, and the filtered image information is subjected to edge segmentation to obtain a second segmented image;

[0155] Calculate the segmentation difference map between the first segmented image and the second segmented image, and supplement the second segmented image with data based on the segmentation difference map to obtain the edge image information corresponding to the filtered image information.

[0156] In some embodiments, during the process of determining the first location state corresponding to the target vehicle by comparing the preset distribution information corresponding to the current location information with the data distribution information, the processor 301 executes:

[0157] Based on the initial image information, the type of materials carried by the target vehicle are identified to obtain type distribution information;

[0158] Obtain the preset type corresponding to the current location information, and obtain the first probability information corresponding to the preset type from the type distribution information;

[0159] The second probability information is obtained by calculating the distribution similarity based on the preset distribution information and the data distribution information;

[0160] The target type corresponding to the target vehicle is determined by fusing the first probability information and the second probability information, and the first position state corresponding to the target vehicle is determined based on the target type.

[0161] In some embodiments, during the process of performing contour feature recognition on the transported materials of the target vehicle based on the three-dimensional point cloud data to obtain the target contour information corresponding to the target vehicle, the processor 301 executes:

[0162] Curvature distribution information corresponding to the three-dimensional point cloud data is obtained by performing curvature feature analysis on the three-dimensional point cloud data.

[0163] Based on the curvature distribution information and the three-dimensional point cloud data, the initial contour information corresponding to the target vehicle is obtained by contour filtering.

[0164] Based on the three-dimensional point cloud data, surface fitting is performed to correct the initial contour information, thereby obtaining the target contour information corresponding to the target vehicle.

[0165] In some embodiments, during the process of performing contour correction on the initial contour information based on the three-dimensional point cloud data to obtain the target contour information corresponding to the target vehicle, the processor 301 executes:

[0166] Obtain the relevant point cloud data involved in the initial contour information from the three-dimensional point cloud data;

[0167] Vector calculation is performed on each initial sub-point cloud data in the relevant point cloud data to obtain the target direction information corresponding to the initial sub-point cloud data;

[0168] Based on the target direction information, the initial adjacent point cloud information corresponding to the initial sub-point cloud data is obtained from the three-dimensional point cloud data;

[0169] Based on the initial adjacent point cloud information, the initial sub-point cloud data is analyzed and filtered for morphological features to obtain the target sub-point cloud data;

[0170] Obtain the target neighboring point cloud information corresponding to the target sub-point cloud data, and perform surface fitting based on the target neighboring point cloud information to perform contour correction on the initial contour information, thereby obtaining the target contour information corresponding to the target vehicle.

[0171] In some implementations, after obtaining the target unloading result, the processor 301 further performs the following:

[0172] If the target unloading result is "unloaded", then the latest location information corresponding to the target vehicle is obtained;

[0173] Based on the latest location information and the current location information, determine the distance information between the target vehicle and the target unloading area corresponding to the current location information;

[0174] When the distance information meets the preset conditions, the barrier bar of the target unloading area is lowered to prevent subsequent vehicles from unloading.

[0175] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the terminal device described above can be referred to the corresponding process in the aforementioned vehicle intelligent unloading control method embodiment, and will not be repeated here.

[0176] This invention also provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs that can be executed by one or more processors to implement the steps of any of the vehicle intelligent unloading control methods provided in the specification of this invention.

[0177] The storage medium can be an internal storage unit of the terminal device described in the foregoing embodiments, such as the hard drive or memory of the terminal device. Alternatively, the storage medium can be an external storage device of the terminal device, such as a plug-in hard drive, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the terminal device.

[0178] Those skilled in the art will understand that all or some of the steps, systems, or apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware embodiments, the division between functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0179] It should be understood that the term "and / or" as used in this specification and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations. It should be noted that, herein, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0180] The sequence numbers of the above embodiments of the present invention are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. The above descriptions are only specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A vehicle intelligent unloading control method, characterized in that, The method includes: Initial image information is obtained by acquiring images of the materials being transported on the target vehicle, and image analysis is performed based on the initial image information to obtain data distribution information corresponding to the materials being transported on the target vehicle; the initial image information corresponds to the target object, and the data distribution information includes the size information, volume information, type distribution information of the target object, as well as the spatial relationship and layout pattern between the transported materials; The current location information of the target vehicle is collected, and the first location state of the target vehicle is determined by comparing the preset distribution information and the data distribution information corresponding to the current location information. The preset distribution information is the data distribution information corresponding to the stored materials under the current location information. The first location state is determined by the following method: the similarity between the preset distribution information and the data distribution information is calculated using the KL distribution. If the similarity is greater than the preset value, the first location state of the target vehicle is determined to be an unloadable state; otherwise, the first location state of the target vehicle is determined to be an unloadable state. When the first position state meets the preset state, the three-dimensional data of the transported materials of the target vehicle is collected to obtain the three-dimensional point cloud data corresponding to the target vehicle. Based on the three-dimensional point cloud data, the contour features of the material carried by the target vehicle are identified to obtain the target contour information corresponding to the target vehicle. The required carrying volume information for the target vehicle to unload is determined based on the target contour information and the target transport type. The maximum load-bearing volume corresponding to the current location information is obtained, and the second location state of the target vehicle is determined based on the maximum load-bearing volume and the load-bearing volume information; wherein, when the maximum load-bearing volume is greater than or equal to the load-bearing volume information, the second location state of the target vehicle corresponding to the current location information is determined to be normal; when the maximum load-bearing volume is less than the load-bearing volume information, the second location state of the target vehicle corresponding to the current location information is determined to be abnormal. Based on the second position status, the target vehicle is controlled to unload materials, and the target unloading result corresponding to the target vehicle is obtained.

2. The method according to claim 1, characterized in that, The step of obtaining data distribution information corresponding to the materials transported on the target vehicle by performing image analysis based on the initial image information includes: The initial image information is subjected to bilateral filtering to obtain the filtered image information corresponding to the initial image information; Edge detection processing is performed on the filtered image information to obtain the edge image information corresponding to the filtered image information; Obtain edge information corresponding to each sub-edge in the edge image information, and obtain adjacent information corresponding to the sub-edge from the edge image information; Data distribution information corresponding to the materials transported on the target vehicle is obtained by performing data analysis based on the edge information and the adjacent information.

3. The method according to claim 2, characterized in that, The step of performing edge detection processing on the filtered image information to obtain the edge image information corresponding to the filtered image information includes: A multi-directional convolution template is determined, and the filtered image information is convolved according to the multi-directional convolution template to obtain multiple target convolution images; By fusing multiple target convolutional images, the gradient angle corresponding to each pixel in the filtered image information is determined; Obtain the neighboring points corresponding to the pixel, and construct a first target straight line based on the neighboring points; A second target line is constructed based on the gradient angle and the pixel, and the target intersection point information is obtained by calculating the intersection point of the first target line and the second target line. Obtain relevant points associated with the target intersection information from the adjacent points, and determine the target pixel value corresponding to the target intersection information based on the pixel information corresponding to the relevant points; The filtered image information is interpolated based on the target pixel value and the target intersection information to obtain an interpolated image; The interpolated image is subjected to threshold segmentation to obtain a first segmented image, and the filtered image information is subjected to edge segmentation to obtain a second segmented image; Calculate the segmentation difference map between the first segmented image and the second segmented image, and supplement the second segmented image with data based on the segmentation difference map to obtain the edge image information corresponding to the filtered image information.

4. The method according to claim 2, characterized in that, The step of determining the first location state of the target vehicle by comparing the preset distribution information corresponding to the current location information with the data distribution information includes: Based on the initial image information, the type of materials carried by the target vehicle are identified to obtain type distribution information; Obtain the preset type corresponding to the current location information, and obtain the first probability information corresponding to the preset type from the type distribution information; The second probability information is obtained by calculating the distribution similarity based on the preset distribution information and the data distribution information; The target type corresponding to the target vehicle is determined by fusing the first probability information and the second probability information, and the first position state corresponding to the target vehicle is determined based on the target type.

5. The method according to claim 1, characterized in that, The step of performing contour feature recognition on the transported materials of the target vehicle based on the three-dimensional point cloud data to obtain the target contour information corresponding to the target vehicle includes: Curvature distribution information corresponding to the three-dimensional point cloud data is obtained by performing curvature feature analysis on the three-dimensional point cloud data. Based on the curvature distribution information and the three-dimensional point cloud data, the initial contour information corresponding to the target vehicle is obtained by contour filtering. Based on the three-dimensional point cloud data, surface fitting is performed to correct the initial contour information, thereby obtaining the target contour information corresponding to the target vehicle.

6. The method according to claim 5, characterized in that, The step of performing surface fitting based on the three-dimensional point cloud data to correct the initial contour information and obtain the target contour information corresponding to the target vehicle includes: Obtain the relevant point cloud data involved in the initial contour information from the three-dimensional point cloud data; Vector calculation is performed on each initial sub-point cloud data in the relevant point cloud data to obtain the target direction information corresponding to the initial sub-point cloud data; Based on the target direction information, the initial adjacent point cloud information corresponding to the initial sub-point cloud data is obtained from the three-dimensional point cloud data; Based on the initial adjacent point cloud information, the initial sub-point cloud data is analyzed and filtered for morphological features to obtain the target sub-point cloud data; Obtain the target neighboring point cloud information corresponding to the target sub-point cloud data, and perform surface fitting based on the target neighboring point cloud information to perform contour correction on the initial contour information, thereby obtaining the target contour information corresponding to the target vehicle.

7. The method according to any one of claims 1-6, characterized in that, After obtaining the target uninstallation result, the method further includes: If the target unloading result is "unloaded", then the latest location information corresponding to the target vehicle is obtained; Based on the latest location information and the current location information, determine the distance information between the target vehicle and the target unloading area corresponding to the current location information; When the distance information meets the preset conditions, the barrier bar of the target unloading area is lowered to allow subsequent vehicles to unload.

8. A vehicle intelligent unloading control system, characterized in that, include: The data analysis module is used to acquire images of the materials being transported by the target vehicle to obtain initial image information, and to perform image analysis based on the initial image information to obtain data distribution information corresponding to the materials being transported by the target vehicle. The initial image information corresponds to a target object, and the data distribution information includes the target object's size information, volume information, type distribution information, as well as the spatial relationship and layout pattern between the transported materials; The status analysis module is used to collect the current location information of the target vehicle and compare it with the preset distribution information and the data distribution information corresponding to the current location information to determine the first location status of the target vehicle. The preset distribution information is the data distribution information corresponding to the stored materials under the current location information; the first location state is determined by the following method: the similarity between the preset distribution information and the data distribution information is calculated using the KL distribution, and when the similarity is greater than the preset value, the first location state corresponding to the target vehicle is determined to be an unloadable state; Otherwise, the first position status corresponding to the target vehicle will be determined as an unloadable state; The data acquisition module is used to acquire three-dimensional data of the transported materials of the target vehicle to obtain the three-dimensional point cloud data corresponding to the target vehicle when the first position state meets the preset state. The contour recognition module is used to identify the contour features of the transported materials of the target vehicle based on the three-dimensional point cloud data, and obtain the target contour information corresponding to the target vehicle. The volume determination module is used to determine the carrying volume information required by the target vehicle when unloading, based on the target contour information and the target transportation type. The status determination module is used to obtain the maximum load-bearing volume corresponding to the current location information, and determine the second location status of the target vehicle based on the maximum load-bearing volume and the load-bearing volume information; wherein, when the maximum load-bearing volume is greater than or equal to the load-bearing volume information, the second location status of the target vehicle corresponding to the current location information is determined to be normal; when the maximum load-bearing volume is less than the load-bearing volume information, the second location status of the target vehicle corresponding to the current location information is determined to be abnormal. The material unloading module is used to control the target vehicle to unload materials according to the second position state, and to obtain the target unloading result corresponding to the target vehicle.

9. A terminal device, characterized in that, The terminal device includes a processor and a memory; The memory is used to store computer programs; The processor is used to execute the computer program and, in executing the computer program, implement the intelligent vehicle unloading control method as described in any one of claims 1 to 7.

10. A computer storage medium for computer storage, characterized in that, The computer storage medium stores one or more programs, which can be executed by one or more processors to implement the steps of the intelligent vehicle unloading control method according to any one of claims 1 to 7.

Images