Method for constructing and visualizing dump space based on excavator positioning and visual analysis

Abstract

The application discloses a kind of based on excavator positioning and visual analysis's explosive heap space construction and visualization method, belong to intelligent mine and image processing field, comprising: the initial three-dimensional surface model of constructing explosive heap;In the process of excavating explosive heap material, the space coordinates of each bucket material and material image are synchronously acquired and are aligned with timestamp after forming data packet together;Based on the material image of each bucket, corresponding blasting effect score is generated by image segmentation and block degree analysis;The space coordinates of each bucket material is taken as its sampling center in explosive heap, and corresponding blasting effect score is taken as attribute and mapped to initial three-dimensional surface model, so that each vertex in mapped three-dimensional surface model is endowed with an attribute value;According to the attribute value of vertex, color mapping is carried out on the model according to the preset mapping rule, and a visual three-dimensional model reflecting the spatial distribution of blasting effect is obtained.The application can accurately and intuitively display the blasting effect of explosive heap, and provide direct basis for blasting parameter optimization.

Description

Technical Field

[0001] This invention belongs to the field of intelligent mining and image processing technology, specifically relating to a method for constructing and visualizing blasting space based on bucket positioning and visual analysis. Background Technology

[0002] Existing methods for evaluating the effects of blasting have two major limitations:

[0003] The global model lacks semantics: UAVs or lidar can quickly reconstruct the three-dimensional surface model of the blast pile, but the model is only a geometric shell and cannot reflect the fragmentation quality of different areas inside or on the surface.

[0004] Local evaluation lacks location: Although the size of materials in a single bucket can be analyzed by the excavator's camera, the original spatial location of the material in the blast pile is lacking, which makes it impossible to trace the evaluation results back to the blast pile itself.

[0005] Although some studies have attempted to combine mined data with 3D models, the key technical problem of "how to accurately back-project discrete, dynamic mined sampling points onto the surface of a static blast pile model and complete the spatial filling and smooth reconstruction of effect attributes" has not been solved. Therefore, there is an urgent need for a method that can connect the "sampling-localization-mapping-reconstruction" link. Summary of the Invention

[0006] To address the aforementioned problems in the existing technology, this invention provides a method for constructing and visualizing blast pile space based on bucket positioning and visual analysis. The technical problem to be solved by this invention is achieved through the following technical solution:

[0007] A method for constructing and visualizing the blast pile space based on bucket positioning and visual analysis includes:

[0008] Construct the initial three-dimensional surface model of the burst pile;

[0009] During the process of excavating explosively piled materials, the spatial coordinates of each bucket of material and the material image are acquired simultaneously and aligned with the timestamp to form a corresponding data packet.

[0010] Based on the material image of each bucket data packet, the blasting effect score of the material in that bucket is generated through image segmentation and block size analysis;

[0011] The spatial coordinates of each bucket of material are used as the sampling center of the material in the blast pile. The corresponding blasting effect score is used as an attribute and mapped to the initial three-dimensional surface model, so that each vertex in the obtained mapped three-dimensional surface model is assigned an attribute value.

[0012] Based on the attribute values ​​of the vertices, the mapped 3D surface model is color-mapped according to a preset mapping rule to obtain a visualized 3D model that reflects the spatial distribution of the blasting effect.

[0013] The beneficial effects of this invention are:

[0014] The blast pile space construction and visualization method based on bucket positioning and visual analysis provided in this invention first constructs an initial three-dimensional surface model of the blast pile. Then, during the excavator's excavation of the blast pile material, the bucket space coordinates and material image corresponding to each bucket of material are simultaneously acquired and aligned with a timestamp to form a corresponding data packet. Next, based on the material image of each bucket data packet, a blasting effect score for that bucket of material is generated through image segmentation and block size analysis. Then, the bucket space coordinates of each bucket of material are used as the sampling center of that bucket of material in the blast pile, and the corresponding blasting effect score is mapped as an attribute to the initial three-dimensional surface model, so that each vertex in the mapped three-dimensional surface model is assigned an attribute value. Finally, based on the attribute values ​​of the vertices, the mapped three-dimensional surface model is color-mapped according to a preset mapping rule to obtain a visualized three-dimensional model reflecting the spatial distribution of the blasting effect.

[0015] This method has the following beneficial effects:

[0016] 1. By using high-precision bucket positioning, a one-to-one spatial correspondence is achieved between the "excavation sample" and the "explosive pile model", enabling precise alignment between sampling and model;

[0017] 2. Visual analysis based on actual excavated rock pile samples is more accurate than pure geometric inference, achieving real material evaluation;

[0018] 3. Dynamically reconstruct the static model, transforming discrete, dynamic mining events into continuous, static burst attribute fields;

[0019] 4. Only the positioning and camera need to be installed on the existing excavator, without changing the work process, making the project highly feasible;

[0020] 5. It can pinpoint "which area did not blast well", providing direct basis for optimizing blasting parameters and enabling refined management. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating a method for constructing and visualizing blasting pile space based on bucket positioning and visual analysis, provided in an embodiment of the present invention.

[0022] Figure 2 This is a schematic diagram illustrating the principle and process of the method for constructing and visualizing blasting pile space based on bucket positioning and visual analysis provided in an embodiment of the present invention.

[0023] Figure 3 This is a schematic diagram of the system architecture of the method for constructing and visualizing blasting pile space based on bucket positioning and visual analysis provided in an embodiment of the present invention.

[0024] Figure 4 This is a rendering of a three-dimensional visualization model provided in an embodiment of the present invention from one angle.

[0025] Figure 5 This is an image showing another perspective of the visualized 3D model provided in an embodiment of the present invention;

[0026] Figure 6 This is another perspective view of the visualized 3D model provided in the embodiment of the present invention. Detailed Implementation

[0027] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0028] This invention provides a method for constructing and visualizing blast pile space based on bucket positioning and visual analysis, such as... Figure 1 and Figure 2 As shown, the method may include the following steps:

[0029] S1, Construct the initial three-dimensional surface model of the burst pile;

[0030] In this embodiment of the invention, the blast pile can be the blast pile after open-pit mine blasting.

[0031] In one optional implementation, S1 may include:

[0032] Before the blasting operation in the open-pit mine is completed and the loading operation has begun, an initial three-dimensional surface model of the blast pile is obtained by drone photogrammetry or lidar scanning.

[0033] Specifically, before the blasting operation in the open-pit mine is completed and before the loading operation begins, a drone equipped with a multi-view camera or a ground / vehicle-mounted LiDAR is used to scan the blast pile area; an initial three-dimensional surface model of the blast pile is generated through photogrammetry (SfM) or direct point cloud registration methods, represented as follows: ;in, This represents the set of vertices of the initial 3D surface model, with a total of vertices. , No. vertices ; This represents a set of triangular facets; and the coordinate system of the initial 3D surface model is unified to the mining area engineering coordinate system to ensure subsequent spatial alignment. The above processing procedure can be found in relevant technical explanations and will not be detailed here.

[0034] S2, during the process of the excavator digging the explosive material, the spatial coordinates of the bucket and the material image corresponding to each bucket of material are acquired simultaneously, and after being aligned with the timestamp, they are used to form the corresponding data packet;

[0035] When excavating explosive piles of materials, it is understandable that the excavator uses its bucket to excavate multiple buckets of material in bucket units.

[0036] In one optional implementation, S2 may include:

[0037] During the process of excavating explosively piled materials, the positioning system installed on the excavator is used to obtain the spatial coordinates of the bucket when excavating each bucket of material; and at the same time, the industrial camera installed on the bucket is used to collect the material image in the bucket after each excavation action is completed. The bucket spatial coordinates, material image and timestamp corresponding to the same bucket are aligned through the time synchronization mechanism to form the data packet corresponding to that bucket.

[0038] The positioning system installed on the excavator can be a high-precision positioning system, such as a GNSS (Global Navigation Satellite System) and IMU (Inertial Measurement Unit) combined navigation device, with positioning accuracy superior to [previous standard]. 5cm. An industrial camera mounted on the bucket automatically captures an image of the material inside the bucket after each digging action.

[0039] The bucket spatial coordinates are the three-dimensional spatial coordinates of the bucket tooth tip or the bucket center. The bucket's spatial coordinates are represented as follows: , They are the corresponding coordinate, coordinates and Coordinates, material image representation The timestamp is represented as The data packet is represented as .

[0040] S2 can be used to obtain multi-source synchronous data during the mining process. .

[0041] S3, based on the material image of each bucket data packet, generates a blasting effect score for the material in that bucket through image segmentation and block size analysis;

[0042] S3 evaluates the blasting effect of a single bucket of material based on visual perception. In one optional implementation, this step may include:

[0043] S31, using a pre-trained lightweight deep learning model, identify and segment each rock block region in the material image of the bucket, and obtain the rock block identification result;

[0044] The lightweight deep learning model can be any existing image recognition model, such as MobileNetV3 + Mask R-CNN (Convolutional Neural Network). This model is trained using a training dataset containing several labeled images of a batch of sample material. Each sample image is labeled with a rock region in the image and the rock identification result. The training process can be completed using existing deep learning model training methods, which will not be detailed here.

[0045] The identified rock block regions can be represented by the coordinates of the location box; the rock block identification results include the number of rock blocks detected in the material bucket and the pixel projection area of ​​each rock block.

[0046] S32, Based on the rock block identification results, calculate the equivalent particle size of the rock blocks in the bucket material, as well as the average equivalent particle size and large block ratio of the bucket material;

[0047] The formula for calculating the equivalent grain size of the rock block is as follows:

[0048] ;

[0049] in, Indicates the first The first batch of materials in the bucket The equivalent grain size of each rock block; Indicates the first The first batch of materials in the bucket The pixel projection area of ​​each rock block;

[0050] The formula for calculating the average equivalent particle size is as follows:

[0051] ;

[0052] in, Indicates the first The average equivalent particle size of the material in the bucket; The rock block identification results of the bucket material include the first Number of rock blocks detected in the bucket material , and the pixel projection area of ​​each rock block; This represents the summation function; the pixel projection area of ​​the rock block is calculated using existing methods. Indicates multiplication.

[0053] The formula for calculating the bulk ratio is:

[0054] ;

[0055] in, Indicates the first The proportion of large pieces of material in the bucket; It is a counting function; The preset threshold for large blocks can be, for example, 0.8m;

[0056] S33, calculate the blasting effect score of the material in the bucket based on the average equivalent particle size and the large particle ratio.

[0057] The formula for calculating the blasting effect score is as follows:

[0058] ;

[0059] in, Indicates multiplication; For the first Scoring of the blasting effect of the bucket material; For block-degree adaptation functions, they are known functions, such as the Sigmoid function; , The target block size is obtained by pre-setting; and These are the corresponding weighting coefficients, satisfying... ; This is a preset value, greater than 0, which can be set as needed; and It was eventually normalized to the interval [0, 100].

[0060] Understandably, through S31~S33, each bucket of material can obtain a blasting effect score. .

[0061] S4, take the bucket space coordinates of each bucket of material as the sampling center of the bucket of material in the blast pile, and map the corresponding blasting effect score as an attribute to the initial three-dimensional surface model, so that each vertex in the obtained mapped three-dimensional surface model is assigned an attribute value.

[0062] In one optional implementation, S4 may include:

[0063] S41, for each bucket of material, its bucket space coordinates are used as the sampling center of the bucket of material in the blast pile. In the initial three-dimensional surface model, all vertices in the sphere neighborhood centered on the bucket space coordinates of the bucket of material and with a radius of a preset radius are searched. The blasting effect score of the bucket of material is assigned to these vertices as the initial attribute value.

[0064] Specifically, taking the first Taking the bucket material as an example, its bucket spatial coordinates As the sampling center for the material in the explosive pile, in the initial three-dimensional surface model In the middle, the search is based on Centered on, with a radius equal to a preset radius value All vertices in the neighborhood of the ball Assign a score to the explosive effect of the material in the bucket at these vertices. As the initial value of the attribute; where You can set it as needed, for example, it can be 1m.

[0065] Understandably, through the processing of S41, the blasting effect score of each bucket of material is used as the initial value of the attribute and assigned to multiple vertices. However, since the sphere neighborhood may overlap, some vertices may be assigned the same value repeatedly. Of course, it is also possible that some vertices are not covered by any sphere neighborhood and therefore do not receive the attribute value.

[0066] S42, using a distance-weighted average method, determines a final attribute value for vertices with multiple initial attribute values;

[0067] In this embodiment of the invention, vertices with multiple initial attribute values ​​are used to determine a final attribute value using a distance-weighted average method. The calculation formula is as follows:

[0068] ;

[0069] in, For the first three-dimensional surface model One vertex; For the first The spatial coordinates of the bucket corresponding to the material in the bucket; To affect the vertex The set of all sample bucket indexes; For the first Scoring of the blasting effect of the bucket material; This is a preset constant, such as 0.01, used to avoid division by zero; Indicates calculation and The Euclidean distance between them.

[0070] Through the processing of S42, the vertex that is repeatedly assigned a value determines a final attribute value.

[0071] S43 uses interpolation to diffuse attributes for vertices that have not been assigned attribute values, so that each vertex in the mapped 3D surface model is assigned an attribute value.

[0072] The interpolation method used here can be radial basis function (RBF) interpolation or kriging interpolation. For details on the specific processing procedure, please refer to the relevant technical explanations. It will not be explained here.

[0073] It is understandable that by using S43 to perform attribute diffusion on vertices that have not yet received attribute assignments to assign attribute values, each vertex in the final mapped 3D surface model is assigned an attribute value, ensuring full coverage of vertex attributes in the 3D model.

[0074] S5. Based on the attribute values ​​of the vertices, color mapping is performed on the mapped 3D surface model according to the preset mapping rules to obtain a visualized 3D model that reflects the spatial distribution of the blasting effect.

[0075] The preset mapping rules are used to implement color encoding. In one optional implementation, the preset mapping rules include:

[0076] If the attribute value of a vertex in the mapped 3D surface model is greater than or equal to the first value, the corresponding coloring is green (RGB: 0, 176, 80), indicating good fragmentation;

[0077] If the attribute value of a vertex in the mapped 3D surface model is greater than or equal to the second value and less than the first value, the corresponding coloring is yellow (RGB: 255, 192, 0), indicating moderate fragmentation;

[0078] If the attribute value of a vertex in the mapped 3D surface model is less than the second value, the corresponding coloring is red (RGB: 255, 0, 0), indicating insufficient fragmentation or the presence of large pieces;

[0079] The second value is less than the first value. These two values ​​can be set as needed, for example, the first value is 85 and the second value is 60.

[0080] After color-coded mapping, the visualized 3D model can reflect the spatial distribution of blasting effects and carry the semantics of blasting effects. The spatial position of each vertex can be obtained, its attribute values ​​can be acquired to characterize the blasting effect score, and the blasting effect of the corresponding area can be understood through color.

[0081] In one optional implementation, after obtaining a visualized three-dimensional model reflecting the spatial distribution of the blasting effect, the method further includes:

[0082] View the spatial distribution of blasting effects in the visualized 3D model on the mine digital twin platform.

[0083] Specifically, the visualized 3D model can be loaded into the mine digital twin platform in the form of a 3D model file with color channels (PLY, OBJ or glTF format, etc.) for interactive display. Users can click on any location to query the corresponding blasting effect score and view the spatial distribution of the blasting effect.

[0084] The blast pile space construction and visualization method based on bucket positioning and visual analysis provided in this invention first constructs an initial three-dimensional surface model of the blast pile. Then, during the excavator's excavation of the blast pile material, the bucket space coordinates and material image corresponding to each bucket of material are simultaneously acquired and aligned with a timestamp to form a corresponding data packet. Next, based on the material image of each bucket data packet, a blasting effect score for that bucket of material is generated through image segmentation and block size analysis. Then, the bucket space coordinates of each bucket of material are used as the sampling center of that bucket of material in the blast pile, and the corresponding blasting effect score is mapped as an attribute to the initial three-dimensional surface model, so that each vertex in the mapped three-dimensional surface model is assigned an attribute value. Finally, based on the attribute values ​​of the vertices, the mapped three-dimensional surface model is color-mapped according to a preset mapping rule to obtain a visualized three-dimensional model reflecting the spatial distribution of the blasting effect.

[0085] This method has the following beneficial effects:

[0086] 1. By using high-precision bucket positioning, a one-to-one spatial correspondence is achieved between the "excavation sample" and the "explosive pile model", enabling precise alignment between sampling and model;

[0087] 2. Visual analysis based on actual excavated rock pile samples is more accurate than pure geometric inference, achieving real material evaluation;

[0088] 3. Dynamically reconstruct the static model, transforming discrete, dynamic mining events into continuous, static burst attribute fields;

[0089] 4. Only the positioning and camera need to be installed on the existing excavator, without changing the work process, making the project highly feasible;

[0090] 5. It can pinpoint "which area did not blast well", providing direct basis for optimizing blasting parameters and enabling refined management.

[0091] To facilitate understanding of the implementation process of the method of this invention, a specific example is given below. Please refer to the example provided. Figures 1-3 understand.

[0092] After a blasting operation was completed and before the loading operation began, an aerial survey of the blast pile area was conducted using a multi-rotor drone (such as a DJIM 300 RTK). The drone flew at an altitude of 50 meters, with the image overlap set to 80% in the forward direction and 70% in the side direction, acquiring a total of 120 high-resolution RGB (Red, Green, Blue) images. On a modeling server, photogrammetry software (such as Pix4Dmapper) was used to process the images, generating an initial 3D surface model of the blast pile with a spatial resolution of 2 cm. And unified to the WGS-84 / UTM engineering coordinate system of the mining area.

[0093] Subsequently, a high-precision positioning system was installed on the PC800 hydraulic excavator, including: 1) a dual-frequency GNSS receiver (such as NovAtel PwrPak7) that supports RTK differential positioning; 2) an inertial measurement unit (IMU) (such as SBG Ellipse-N) to compensate for attitude changes caused by the movement of the excavator's boom.

[0094] The positioning calculation module of the high-precision positioning system outputs the three-dimensional coordinates of the bucket tooth tip in real time. The update frequency is 10 Hz, and the static positioning accuracy is better than ±3 cm.

[0095] Meanwhile, an industrial camera (such as a Basler ace acA2440-35uc, 5 megapixels, global shutter) is installed above the bucket, and image acquisition is triggered by a digging motion sensor (installed in the hydraulic circuit) to ensure that material images are automatically captured after each digging operation. All data is synchronized in time via an onboard industrial control computer (Intel i7 processor, 16 GB RAM), forming synchronization data packets. .

[0096] A lightweight rock segmentation model (based on the MobileNetV3 backbone and Mask R-CNN, with a model size of less than 15 MB) is deployed in the edge computing module to segment each frame of material images. Perform real-time processing. Set the target block size. =0.5 m, large block threshold = 0.8 m, calculate the average equivalent particle size of the material in the bucket. With large block ratio And substitute it into the formula for calculating the explosion effect score:

[0097] ;

[0098] The calculation results are linearly normalized to the interval [0, 100] and used as the blasting effect score for the material in the bucket.

[0099] bucket space coordinates Considered as the sampling center of the material in the blast pile, in the initial three-dimensional surface model Search in China Center and radius of the sphere All vertices within a sphere with a radius of 1.0 m are identified, and their attributes are assigned using a distance-weighted average method. For regions not covered by any sampling points, radial basis function (RBF) interpolation is used to fill in the full model attributes.

[0100] Finally, based on vertex attribute values Perform color mapping on the model:

[0101] like ≥ 85, colored green (RGB: 0, 176, 80), indicating good breakage;

[0102] If 60 ≤ <85, colored yellow (RGB: 255, 192, 0), indicating moderate breakage;

[0103] like <60, colored red (RGB: 255, 0, 0), indicates insufficient fragmentation or the presence of large chunks.

[0104] Output a 3D model in PLY format with color channels (a visualized 3D model). Load the model using the Unity engine or other 3D engines that support PLY format onto the mine's digital twin platform (such as a visualization system customized based on Unity3D) for rendering and display. Through interactive viewing, technicians discovered a large red area in the northwest corner of the blast pile. On-site verification confirmed this to be a large concentrated area. Based on this, the hole spacing, row spacing, and explosive consumption for the next round of blasting were adjusted to achieve closed-loop optimization of blasting parameters.

[0105] This invention first acquires an initial 3D surface model of the blast pile after blasting and before loading. Then, a high-precision positioning system and industrial camera are integrated and installed on the excavator to simultaneously acquire the bucket spatial coordinates and material images of each bucket of material during excavation. Based on image analysis, a blasting effect score for each bucket of material is generated. This score is then mapped to the corresponding area of ​​the initial 3D surface model according to the bucket spatial coordinates, and interpolation is used to achieve full model attribute coverage, resulting in a mapped 3D surface model. Finally, the model is color-coded according to attribute levels to generate a visual 3D model that intuitively reflects the spatial distribution of blasting quality, which is interactively displayed on a mine digital twin platform. This invention combines high-precision bucket positioning, visual analysis of excavated materials, and spatial registration methods with the 3D surface model of the blast pile to accurately map local blasting effect evaluation results to the global blast pile model, enabling refined, location-based, and visual representation of blasting quality. It achieves a precise transformation of blasting effect from "local evaluation" to "global visibility" without changing existing operational processes, providing a high-precision spatial decision-making basis for open-pit mine blasting optimization.

[0106] It should be noted that in the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0107] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. In addition, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.

[0108] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A method for constructing and visualizing blast pile space based on bucket positioning and visual analysis, characterized in that, include: Construct the initial three-dimensional surface model of the burst pile; During the process of excavating explosively piled materials, the spatial coordinates of each bucket of material and the material image are acquired simultaneously and aligned with the timestamp to form a corresponding data packet. Using a pre-trained lightweight deep learning model, the rock block regions in the material image of the bucket are identified and segmented to obtain the rock block identification results; Based on the rock block identification results, the equivalent particle size of the rock blocks in the bucket material is calculated, as well as the average equivalent particle size and large block ratio of the bucket material. The blasting effect score of the material in the bucket is calculated based on the average equivalent particle size and the proportion of large pieces. The spatial coordinates of each bucket of material are used as the sampling center of the material in the blast pile, and the corresponding blasting effect score is used as an attribute to be mapped to the initial three-dimensional surface model, so that each vertex in the obtained mapped three-dimensional surface model is assigned an attribute value. Based on the attribute values ​​of the vertices, the mapped 3D surface model is color-mapped according to a preset mapping rule to obtain a visualized 3D model that reflects the spatial distribution of the blasting effect. The formula for calculating the equivalent grain size of the rock block is as follows: ； in, Indicates the first The first batch of materials in the bucket The equivalent grain size of each rock block; Indicates the first The first batch of materials in the bucket The pixel projection area of ​​each rock block; The formula for calculating the average equivalent particle size is as follows: ； in, Indicates the first The average equivalent particle size of the material in the bucket; The rock block identification results of the bucket material include the first Number of rock blocks detected in the bucket material , and the pixel projection area of ​​each rock block; This represents the summation function; Indicates multiplication; The formula for calculating the bulk ratio is: ； in, Indicates the first The proportion of large pieces of material in the bucket; It is a counting function; The preset large block threshold; The formula for calculating the blasting effect score is as follows: ； in, For the first Scoring of the blasting effect of the bucket material; For block-level adaptation functions; , Target block size; and These are the corresponding weighting coefficients, satisfying... , It is a preset value; and It was eventually normalized to the interval [0, 100].

2. The method according to claim 1, characterized in that, The initial three-dimensional surface model for constructing the burst pile includes: Before the blasting operation in the open-pit mine is completed and the loading operation has begun, an initial three-dimensional surface model of the blast pile is obtained by drone photogrammetry or lidar scanning.

3. The method according to claim 1, characterized in that, During the process of excavating explosively piled materials by an excavator, the spatial coordinates of each bucket of material and the material image are acquired simultaneously, and after being aligned with the timestamp, they are combined to form a corresponding data packet, including: During the process of excavating explosively piled materials, the positioning system installed on the excavator is used to obtain the spatial coordinates of the bucket when excavating each bucket of material; and at the same time, the industrial camera installed on the bucket is used to collect the material image in the bucket after each excavation action is completed. The bucket spatial coordinates, material image and timestamp corresponding to the same bucket are aligned through the time synchronization mechanism to form the data packet corresponding to that bucket. Wherein, the bucket spatial coordinates are the three-dimensional spatial coordinates of the bucket tooth tip or the bucket center, the first... The bucket's spatial coordinates are represented as follows: , They are the corresponding coordinate, coordinates and Coordinates, material image representation The timestamp is represented as The data packet is represented as .

4. The method according to claim 1, characterized in that, The method involves using the spatial coordinates of each bucket of material as the sampling center of that bucket in the blast pile, and mapping the corresponding blasting effect score as an attribute to the initial three-dimensional surface model. This results in each vertex in the mapped three-dimensional surface model being assigned an attribute value, including: For each bucket of material, its bucket spatial coordinates are used as the sampling center of the bucket of material in the blast pile. In the initial three-dimensional surface model, all vertices in the sphere neighborhood centered on the bucket spatial coordinates of the bucket of material and with a radius of a preset radius are searched. The blasting effect score of the bucket of material is assigned to these vertices as the initial attribute value. For vertices with multiple initial attribute values, a final attribute value is determined by using a distance-weighted average method. For vertices that have not been assigned attribute values, attribute diffusion is performed using interpolation, so that each vertex in the mapped 3D surface model is assigned an attribute value.

5. The method according to claim 4, characterized in that, The method of determining a final attribute value for vertices with multiple initial attribute values ​​using a distance-weighted average is as follows: ； in, For the first three-dimensional surface model One vertex; For the first The spatial coordinates of the bucket corresponding to the material in the bucket; To affect the vertex All sample bucket index sets; For the first Scoring of the blasting effect of the bucket material; This is a preset constant; Indicates calculation and The Euclidean distance between them.

6. The method according to claim 1, characterized in that, The preset mapping rules include: If the attribute value of a vertex in the mapped 3D surface model is greater than or equal to the first value, the corresponding coloring is green, indicating good breakage; If the attribute value of a vertex in the mapped 3D surface model is greater than or equal to the second value and less than the first value, the corresponding coloring is yellow, indicating moderate fragmentation; If the attribute value of a vertex in the mapped 3D surface model is less than the second value, it will be colored red, indicating insufficient fragmentation or the presence of large blocks. The second value is less than the first value.

7. The method according to claim 6, characterized in that, The first value is 85, and the second value is 60.

8. The method according to claim 1, characterized in that, After obtaining a visualized 3D model reflecting the spatial distribution of the blasting effect, the method further includes: View the spatial distribution of blasting effects in the visualized 3D model on the mine digital twin platform.

Images