A system and method for identifying 3D scanned core feature parameters

Abstract

The application relates to the technical field of engineering geology exploration, and discloses a 3D scanning core characteristic parameter identification system and method, which comprises a core storage module and a 3D scanning module; the method comprises the following steps: separating complete and fragmented cores by self-adaptive cleaning and separation of drilling core samples, forming a space-attribute binding initial data set; dynamically adjusting the height and levelness of a 3D laser scanner by an automatic numerical control platform, generating an optimal scanning path topology network covering all core boxes; the scanning path topology network feeds back point cloud coverage in real time, dynamically adjusts the rotation angle, and outputs a full-angle high-density core point cloud sequence; the full-angle high-density core point cloud sequence is fused by multi-scale features, and an automatic columnar chart, a quantitative parameter set and a joint parameter package are generated. Through automatic and high-precision scanning and data analysis, the application realizes digitalized collection of cores and accurate calculation of geological parameters, and improves the efficiency and quality of geological exploration.

Description

Technical Field

[0001] This invention relates to the field of engineering geological exploration technology, and in particular to a system and method for identifying characteristic parameters of 3D scanning rock cores. Background Technology

[0002] Core data collection from geological boreholes is of paramount importance in various fields, including geological exploration, engineering construction, and resource development. Core data obtained through borehole exploration helps understand the distribution and properties of underground soil and rock layers, predicting potential geological problems encountered during engineering construction. Furthermore, core data collection is a crucial means of directly obtaining information about the lithology of strata, which is essential for geological description and engineering design. Currently, engineering drilling data collection relies heavily on manual operation. Especially after obtaining the core, sample testing and logging are necessary. This includes image acquisition of the core, statistical analysis of data such as borehole depth, drilling depth, and core length, calculation of recovery rate and RQD, and assessment of joint development. This step requires significant time and effort from personnel and is susceptible to the subjective experience of data collectors, making human error prone to occur during collaborative data collection involving multiple people.

[0003] Prior art 1, Chinese Patent Application No. 202510419252.6, discloses a tunnel stratum identification method based on machine learning of tunnel boring machine (TBM) excavation parameters. This method includes: data acquisition, used to construct a project dataset, collecting core samples from boreholes within the project dataset, and identifying a learning sample set based on the core samples; data feature mining, used to obtain excavation parameters corresponding to various strata types from the learning sample set, and mining statistical feature sets responding to stratum changes based on these parameters, forming a first feature parameter matrix; data feature filtering, used to perform sensitivity analysis on the first feature parameter matrix, thereby filtering out a second feature parameter matrix; model construction, used to construct a stratum identification model based on the learning sample set and the second feature parameter matrix using the K-nearest neighbor algorithm; and stratum identification, used to identify the target stratum according to the stratum identification model. While this method improves the efficiency of stratum identification, the fixed angle cannot be adjusted, leading to feature omissions due to fixed-angle sampling.

[0004] Prior art two, Chinese patent application number: 202010878952.9, discloses a method, device, system, and storage medium for lithology identification based on vibration signals. The method includes the following steps: acquiring vibration signal samples of rock samples taken by the drill bit in the drilling area, and extracting signal feature parameters from the vibration signal samples; establishing a probability distribution relationship model between the rock lithology of the drilling area and the signal feature parameters of the drill bit's vibration signals based on the correspondence between the rock lithology of the rock samples and the signal feature parameters of the vibration signal samples; acquiring vibration signals generated by the drill bit breaking rock during the drilling process in the drilling area, and extracting signal feature parameters from the vibration signals. Although the lithology of the strata encountered by the drill bit in the drilling area can be inferred by using the probability distribution relationship model between the rock lithology of the drilling area and the signal feature parameters of the drill bit's vibration signals based on the signal feature parameters of the vibration signals, relying on indirect parameters such as vibration signals to estimate lithology makes the rock lithology acquisition process complex and cumbersome.

[0005] Prior art three, Chinese patent application number 201410037682.3, discloses a method for identifying lithology across an entire well section, comprising: extracting multiple core samples from multiple depth locations, determining the grain size data and lithology type of each core sample; determining the values ​​of various logging parameters at multiple depth locations, selecting N logging parameters that increase or decrease with increasing lithology sample grain size as feature parameters, where N≥2; determining the correspondence between each N-dimensional cell and the lithology type in an N-dimensional space with feature parameter coordinates based on the lithology type of the core sample points at multiple depth locations and the values ​​of the N feature parameters; and determining the representative lithology type at each depth location based on the values ​​of the feature parameters and their correspondences across the entire well section. Although this method integrates the advantages of traditional geological analysis and logging processing for identifying lithology types, and the established interactive multidimensional histogram lithology identification method is accurate and reliable, especially suitable when core data is limited, the large volume of discrete depth sampling data and the inability to perform batch processing lead to a decrease in the accuracy of lithology identification across the entire well section.

[0006] Current technologies 1, 2, and 3 suffer from time-consuming and labor-intensive processes during borehole data collection. Therefore, this invention provides a 3D scanning system and method for identifying core feature parameters, applicable to geological drilling in various fields such as geological exploration, engineering construction, and resource development. Summary of the Invention

[0007] The main objective of this invention is to provide a system and method for identifying characteristic parameters of 3D scanned rock cores, so as to solve the problems of time-consuming and labor-intensive drilling data collection in the prior art; through this invention, the time and labor costs of drilling data collection can be effectively reduced, and work efficiency and accuracy can be improved.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] A system for identifying characteristic parameters of 3D scanned rock cores, the system comprising:

[0010] The core collection module is used to automatically rotate the core at a fixed angle in a single rotation under the control of an automated CNC platform, enabling multi-angle scanning; it provides orderly core fixation, controllable rotation, and optimized scanning environment.

[0011] The 3D scanning module is used to scan and acquire high-precision point cloud data of the rock core surface; the data acquisition processor processes the point cloud to construct a complete three-dimensional model, automatically calculates key geological parameters, and assists in the identification of joints.

[0012] As a further improvement of the present invention, the core collection module includes: a core box, a fragmented core box, a core rotating wheel, a core plate, and a transmission structure;

[0013] The core box has a transmission structure installed on one side, which is connected to the automated CNC platform via a control line. The core rotating wheel is placed on the core box. The core rotating wheel is placed under each slot of the core box and is connected to the transmission structure via a pin structure, so that the core rotating wheel rotates in the same direction. The transmission structure is connected to the core rotating wheel via a pin structure, and drives the core rotating wheel to rotate by rotating.

[0014] As a further improvement of the present invention, one or more core boxes can be connected as needed to achieve overall scanning of the core.

[0015] As a further improvement of the present invention, the core plate includes an abrasion-resistant outer jacket, a cardboard plate, and a slot structure, which are placed on the core box; the abrasion-resistant outer jacket is located on the outside of the slot structure, and the cardboard plate is located in the middle of the abrasion-resistant outer jacket.

[0016] As a further improvement of the present invention, the 3D scanning module comprises: a 3D laser scanner, a scanning fixation structure, a data acquisition processor, and an automated numerical control platform.

[0017] The system comprises a 3D laser scanner for scanning the core model, controlled by an automated CNC platform, positioned directly above the core box; a scanning fixing structure for securing the 3D laser scanner, also controlled by the automated CNC platform; a data acquisition processor for processing the data obtained by the 3D laser scanner, which emits a laser beam and receives light signals reflected from the object's surface, calculating the propagation time and angle of the light to obtain the three-dimensional coordinate data of the object's surface; and an automated CNC platform for controlling the scanning fixing structure and the core rotating wheel, ensuring that the 3D laser scanner is positioned directly above the core each time it scans, by pre-setting the number and spacing of the core boxes and the slot spacing of the core boxes.

[0018] As a further improvement of the present invention, the four corner brackets are fixed to the ground and can extend and retract in height. A level is attached to the top of the four corner brackets. By adjusting the four corner brackets to align with the level, the 3D laser scanner on the top surface of the four corner brackets is kept horizontal. The part of the top surface of the four corner brackets that connects to the 3D laser scanner is connected to the pulley and then locked on the top four corner brackets. The horizontal position of the 3D laser scanner can be adjusted by adjusting the pulley.

[0019] To achieve the above objectives, the present invention also provides the following technical solution:

[0020] A method for identifying characteristic parameters of 3D scanned rock cores, applied to the aforementioned 3D scanned rock core characteristic parameter identification system, the method comprising:

[0021] The core samples from the borehole were separated into intact and fragmented cores through adaptive cleaning. The fragmented cores were stored in the fragmented core box, while the intact cores and the core tags were positioned together in the slots of the core box to form an initial dataset with spatial-attribute binding.

[0022] The core boxes are arranged, and based on point cloud registration, the automated CNC platform dynamically adjusts the height and level of the 3D laser scanner to generate the optimal scanning path topology network covering all core boxes.

[0023] The scanning path topology network triggers a reinforcement learning control loop, a 3D laser scanner collects point clouds, the core rotating wheel performs angle optimization rotation, the point cloud coverage is fed back in real time, the rotation angle is dynamically adjusted, and a full-angle high-density core point cloud sequence is output.

[0024] The full-angle high-density core point cloud sequence is processed through multi-scale feature fusion: point cloud filtering and registration are performed to remove noise and align multi-view data; a topologically complete model is generated through 3D reconstruction; and a convolutional neural network is used to identify joints and fractures, generating automatically drawn bar charts, quantized parameter sets, and joint parameter packages.

[0025] As a further improvement of the present invention, the process of generating automatically drawn bar charts, quantified parameter sets, and joint parameter packages includes the following steps:

[0026] A high-density core point cloud sequence from all angles was obtained. Using the core box slot matrix as the topological skeleton, the multi-angle point clouds were rigidly registered according to the core rotating wheel angle parameters. Spatial markers of the fragmented core box were then fused to generate a mask for the missing area.

[0027] Joint identification and parameter extraction, slicing along the groove axis to generate virtual core columns, detecting curvature discontinuities within the slice area, and adjusting detection sensitivity based on the lithological record of the core plate;

[0028] The column chart and parameter set are generated, the virtual core column length is accumulated and matched with the core card footage data, the joint attitude is converted into dip and dip angle parameter packages, and the core extraction length is calculated to be forced to be an integer multiple of the slot spacing.

[0029] The output includes a triple modeling result: a topologically complete model that preserves the entity of the card slot space topology, a quantified parameter set, and joint parameters including a dip distribution histogram and dip angle confidence intervals.

[0030] As a further improvement of the present invention, the process of preserving the entity of the card slot space topology in the topological complete model includes the following steps:

[0031] Using the core box slot matrix as the spatial grid reference, the point cloud surface is divided into equal-length segments according to the slot spacing, and the endpoints of each segment are forced to align with the center normal plane of the slot; the quantitative parameters for generating structural constraints are generated.

[0032] Joint parameters are encapsulated in an engineering manner; dip and tilt angle conversions are based on establishing an attitude reference system on the card slot method plane; confidence intervals are calculated based on the number of repeated observations of point cloud during eight-cycle scanning.

[0033] The output includes a triple modeling result: a topologically complete model that retains the entities of the slot number and the void marker of the fractured core box, and a quantified parameter set; and joint parameters that include a dip-angle joint distribution matrix.

[0034] As a further improvement of the present invention, the process of engineering encapsulating joint parameters includes the following steps:

[0035] The normal vector of the major axis of the core box slot is extracted as a horizontal reference datum, and the dip zero plane is established with the vertical bisector of the slot spacing; the angle between the joint surface and the normal vector of the major axis of the slot is defined as the dip angle, and the angle between the joint surface and the vertical bisector is directly used as the dip angle;

[0036] The number of times the same joint is independently detected in the eight-cycle point cloud is counted. When the number of detections is not less than the minimum number of observations required for the lithology to be recorded in the core, the confidence interval of the joint parameter is automatically expanded to the upper limit allowed by the project.

[0037] The row vector is the slot number index, the column vector is the joint attitude data cluster that has passed eight-period verification, and the output is the dip-pitch joint distribution matrix.

[0038] This invention achieves automated and high-precision data acquisition through the collaborative operation of a core collection module and a 3D scanning module. The system automatically controls the fixing and rotation of the core, enabling multi-angle scanning and ensuring automated and efficient data acquisition. Simultaneously, the precise positioning by the 3D laser scanner ensures high accuracy of the acquired core surface point cloud data. Multi-angle scanning: The core rotation wheel and transmission structure in the core collection module automatically rotate the core at specific angles, enabling multi-angle scanning, which helps obtain more comprehensive core feature information. Reduced scanning interference: The black / transparent material designed for the system reduces interference during the scanning process, improving scanning accuracy. Optimized scanning environment: The fragmented core box provides a solution for independently storing fragmented samples, avoiding potential interference from rotation and further optimizing the scanning environment. 3D modeling and parameter calculation: The 3D scanning module processes point cloud data through a data acquisition processor to construct a complete 3D model and automatically calculates key geological parameters such as borehole depth, recovery rate, and RQD. This significantly improves the efficiency and accuracy of geological parameter calculation and assists in joint identification. Geological parameter identification assistance: The system uses 3D modeling and analysis to assist in the identification of geological features such as joints, providing an important reference for geological research. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of the functional modules of the 3D scanning core feature parameter identification system of the present invention;

[0040] Figure 2 This is a schematic diagram of the structure of the 3D scanning core feature parameter identification system of the present invention;

[0041] Figure 3 This is a schematic diagram of the core box structure of the 3D scanning core feature parameter identification system of the present invention. Figure 1 ;

[0042] Figure 4 This is a schematic diagram of the core box structure of the 3D scanning core feature parameter identification system of the present invention. Figure 2 ;

[0043] Figure 5 This is a schematic diagram of the core box structure of the 3D scanning core feature parameter identification system of the present invention. Figure 3 ;

[0044] Figure 6 This is a schematic diagram of the fragmented core box structure of the 3D scanning core feature parameter identification system of the present invention. Figure 1 ;

[0045] Figure 7 This is a schematic diagram of the fragmented core box structure of the 3D scanning core feature parameter identification system of the present invention. Figure 2 ;

[0046] Figure 8 This is a schematic diagram of the fragmented core box structure of the 3D scanning core feature parameter identification system of the present invention. Figure 3 ;

[0047] Figure 9 This is a schematic diagram of the core plate structure of the 3D scanning core feature parameter identification system of the present invention. Figure 1 ;

[0048] Figure 10 This is a schematic diagram of the core plate structure of the 3D scanning core feature parameter identification system of the present invention. Figure 2 ;

[0049] Figure 11 This is a schematic diagram of the transmission structure of the 3D scanning core feature parameter identification system of the present invention. Figure 1 ;

[0050] Figure 12 This is a schematic diagram of the transmission structure of the 3D scanning core feature parameter identification system of the present invention. Figure 2 ;

[0051] Figure 13 This is a schematic flowchart illustrating the steps of an embodiment of the method for identifying characteristic parameters of 3D scanned rock cores according to the present invention.

[0052] Figure 14 This is a schematic diagram of the structure of an embodiment of the electronic device of the present invention;

[0053] Figure 15 This is a schematic diagram of the structure of one embodiment of the storage medium of the present invention. Detailed Implementation

[0054] 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 a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0055] The terms "first," "second," and "third" used in this invention are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified. All directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of this invention are only used to explain the relative positional relationships and movements between components in a specific orientation (as shown in the accompanying drawings). If the specific orientation changes, the directional indications also change accordingly. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0056] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0057] like Figure 1 As shown, this embodiment provides an example of a 3D scanning core feature parameter identification system. In this embodiment, the 3D scanning core feature parameter identification system specifically includes: a core collection module 1 and a 3D scanning module 2.

[0058] The core collection module 1 is used to fix the core and core tag 6 through the core box 3 slot, and integrates the core rotating wheel 5 and transmission structure 7. Under the control of the automated CNC platform 11, it automatically rotates the core at a fixed angle in a single rotation to achieve multi-angle scanning. Its black / transparent material design minimizes scanning interference, and the fragmented core box 4 independently stores broken samples to avoid the influence of rotation. It provides orderly core fixation, controllable rotation, and optimized scanning environment. The 3D scanning module 2 is used to automatically control the 3D laser scanner 8 to be precisely positioned directly above the core through the automated CNC platform 11, and scan to obtain high-precision point cloud data of the core surface. The data acquisition processor 10 processes the point cloud to build a complete three-dimensional model, automatically calculates key geological parameters such as borehole depth, recovery rate, and RQD, and assists in joint identification. It is responsible for automated and precise scanning and three-dimensional modeling analysis. The core collection module 1 and the 3D scanning module 2 work together to achieve fully automatic and high-precision digital acquisition of cores and calculation of geological parameters.

[0059] Preferably, this embodiment achieves automated and high-precision data acquisition through the collaborative work of the core collection module 1 and the 3D scanning module 2: the system can automatically control the fixing and rotation of the core, making multi-angle scanning possible and ensuring the automation and efficiency of data acquisition; simultaneously, with the precise positioning of the 3D laser scanner 8, the acquired core surface point cloud data has high precision. Multi-angle scanning: The core rotating wheel 5 and transmission structure 7 in the core collection module 1 can automatically rotate the core at specific angles to achieve multi-angle scanning, which helps to obtain more comprehensive core feature information. Reduced scanning interference: The black / transparent material designed in the system reduces interference during the scanning process and improves scanning accuracy. Optimized scanning environment: The fragmented core box 4 provides a solution for independently storing fragmented samples, avoiding the possible impact of rotation on scanning, and further optimizing the scanning environment. 3D Modeling and Parameter Calculation: The 3D scanning module 2 processes point cloud data through the data acquisition processor 10 to construct a complete 3D model and automatically calculates key geological parameters such as borehole depth, recovery rate, and RQD. This greatly improves the efficiency and accuracy of geological parameter calculation and assists in joint identification. Geological Parameter Assisted Identification: The system uses 3D modeling analysis to assist in the identification of geological features such as joints, providing important references for geological research.

[0060] In summary, this embodiment achieves digital core acquisition and accurate calculation of geological parameters through automated, high-precision scanning and data analysis, thereby improving the efficiency and quality of geological exploration.

[0061] Furthermore, such as Figures 2-11 As shown, the core collection module 1 specifically includes: core box 3, fragmented core box 4, core rotating wheel 5, core plate 6, and transmission structure 7.

[0062] The core box 3 has a transmission structure 7 installed on one side. The transmission structure 7 is connected to the automated CNC platform 11 via control lines. Core rotating wheels 5 are mounted on the core box 3 and are primarily controlled by the automated CNC platform 11. Core rotating wheels 5 are installed under each slot in the core box 3 and connected to the transmission structure 7 via pin structures 15. This ensures that the core rotating wheels 5 rotate in the same direction, causing the core to rotate synchronously, thus scanning the core from various angles and achieving overall core scanning. The transmission structure 7 is connected to the core rotating wheels 5 via pin structures 15, and its rotation drives the core rotating wheels 5 to rotate. One or more core boxes can be connected as needed to achieve overall core scanning.

[0063] The core plate 6 includes an abrasion-resistant outer cover 12, a cardboard plate 13, and a slot structure 14, and is placed on the core box 3. The abrasion-resistant outer cover 12 is located on the outside of the slot structure 14, and the cardboard plate 13 is located in the middle of the abrasion-resistant outer cover 12.

[0064] The core box 3 is a cuboid structure with length a+2t, width b+2t, height h, and side thickness t. It has an open top and is equipped with equally spaced slot structures 14 parallel to its long end. The cuboid structure b is divided by equal intervals d. The slot structures 14 are used to hold the cores. The core diameter is typically smaller than the intervals d in the slot structures 14, and the core box height h is typically larger than the intervals d in the slot structures 14. The core box 3 must be made of black light-absorbing material or transparent light-transmitting material to ensure that almost no point cloud data is generated during scanning, thus improving scanning quality. The fragmented core box 4 is a semi-cylindrical structure with a diameter slightly smaller than d and a length less than a. It is hung on the slot structures 14 of the core box 3 via hooks on the long axis of the semi-cylindrical structure. It is mainly used to hold loose or broken cores that cannot be pieced together completely, so that they do not need to rotate with the complete cores.

[0065] The core plate 6 is placed on the slot structure 14 as needed. Its structure engages with the slot structure 14 to ensure the core plate 6 remains in place, preventing loss and minimizing the risk of it being misplaced within the core during cataloging. Furthermore, its structure is typically thin, and since the height h of the core box 3 is usually greater than the gap d of the slot structure 14, the core plate 6 hardly obstructs the rotation of the core. The core plate 6 must include information such as sample number, borehole depth, depth, core number, core length, sampling rate, time, and recording personnel. The core plate 6 should be made of black light-absorbing material or transparent, light-transmitting, wear-resistant material to ensure minimal point cloud data generation during scanning, improving scanning quality and maximizing the plate's preservation time.

[0066] Preferably, the core collection module 1 of this embodiment provides a highly efficient and accurate core scanning and storage system. The module's design integrates multiple functions, aiming to achieve automated processing, storage, and information recording of cores. First, the core box 3 is designed as a cuboid structure with an open top and internally spaced slots 14 for placing cores; this design facilitates both core storage and scanning. The core box 3 uses black light-absorbing material or transparent light-transmitting material, reducing point cloud data interference during scanning and improving scanning quality. Second, the core rotating wheel 5 allows the cores to rotate synchronously within the core box 3. Combined with the control of the automated CNC platform 11, it enables scanning of the cores from various angles. The rotation mechanism facilitates the comprehensive acquisition of three-dimensional information from the cores, providing more detailed data for subsequent geological analysis. Furthermore, the transmission structure 7 is connected to the core rotating wheel 5 via a pin structure 15, allowing connection of one or more core boxes as needed to achieve overall scanning of multiple cores, thus improving work efficiency. The design of Core Plate 6 takes into account abrasion resistance and interference with point cloud data during scanning. It uses black light-absorbing material or transparent light-transmitting abrasion-resistant material, and includes important information such as rock sample number and borehole depth for easy recording and retrieval of detailed core information. The design of Fragmented Core Box 4 addresses the storage problem of loose or broken cores that cannot be pieced together. Through a semi-cylindrical design and hook mechanism, these cores can be stored in an orderly manner, facilitating management and subsequent processing.

[0067] In summary, the design of the core collection module 1 in this embodiment makes the storage, scanning, and information recording of cores more efficient and accurate, improves the automation level of core processing, and provides strong technical support for geological exploration and research.

[0068] Furthermore, such as Figure 2 As shown, the 3D scanning module 1 specifically comprises: a 3D laser scanner 8, a scanning fixation structure 9, a data acquisition processor 10, and an automated numerical control platform 11.

[0069] The 3D laser scanner 8 is used to scan the core model and is controlled by the automated CNC platform 11. During scanning, the 3D laser scanner 8 must be positioned directly above the core box 3. The scanning fixing structure 9 is used to fix the 3D laser scanner 8 and is also controlled by the automated CNC platform 11. It is fixed to the ground by four corner brackets, which are telescopic in height. A level is attached to the top of each corner bracket. By adjusting the corner brackets to align with the level, the 3D laser scanner 8 on the top surface of the corner brackets is kept horizontal. The part of the top surface of the corner brackets that connects to the 3D laser scanner 8 is connected to pulleys and then locked onto the top four corner brackets, allowing adjustment of the horizontal position of the 3D laser scanner 8. The data acquisition processor 10 is used to process the data obtained by the 3D laser scanner 8. The 3D laser scanner 8 emits a laser beam and receives the light signal reflected from the object surface. By calculating the propagation time and angle of the light, it obtains the three-dimensional coordinate data of the object surface. The data usually exists in the form of a point cloud, containing the position, reflectivity, and texture information of each point. The acquired point cloud data needs to undergo a series of processing steps, including filtering, registration, and feature extraction. The data from repeated scans at different angles are merged into a complete 3D model. This 3D model accurately reflects the shape, size, and surface features of the object, and is used to statistically analyze data such as borehole depth, drilling depth, and core length. It also calculates data such as recovery rate and RQD, and assists in identifying joint conditions. The automated CNC platform 11 controls the scanning fixing structure 9 and the core rotating wheel 5. By presetting the number and spacing of the core boxes 3 and the slot spacing of the core boxes 3, the 3D laser scanner 8 is positioned directly above the core each time it scans the core, and the core rotating wheel 5 rotates normally each time, realizing an automatic cycle of scan-rotate-re-scan-re-rotate. With scan-rotate as one cycle, eight cycles are performed on the same core box 3. After that, the position of the 3D laser scanner 8 is automatically adjusted and the next core box 3 is scanned.

[0070] Preferably, this embodiment constitutes an automated 3D scanning system, which is significant in achieving efficient and accurate scanning and data acquisition of core models. Specifically, the automated CNC platform 11 controls the entire scanning process, including the fixed structure 9 and the core rotating wheel 5, reducing manual operation and improving scanning efficiency. Through preset parameters, the system can automatically complete the scanning-rotation-re-scanning cycle, achieving continuous operation. The three-dimensional coordinate data acquired by the 3D laser scanner 8, after being processed by the data acquisition processor 10, can accurately reflect the shape, size, and surface features of the core. The processing and feature extraction of point cloud data provide accurate basic data for subsequent analysis. Through multiple cycles of scanning, the system can acquire core data from different angles and finally merge them into a complete three-dimensional model, comprehensively reflecting the geometric and physical properties of the core. The data acquisition processor 10 not only processes the raw point cloud data but also includes filtering, registration, and feature extraction, providing support for subsequent statistical analysis. For example, it can statistically analyze data such as borehole depth, drilling depth, and core length, calculate indicators such as recovery rate and RQD, and assist in determining joint conditions. The design of the scanning fixing structure 9 allows it to be adjusted to a suitable position and kept horizontal by the four corner supports, adapting to core boxes 3 of different sizes and shapes, increasing the system's adaptability and flexibility.

[0071] In summary, this embodiment improves the efficiency and accuracy of core analysis through automated and precise scanning and data processing, providing strong technical support for geological exploration and research.

[0072] like Figure 13 As shown, this embodiment also provides an embodiment of a method for identifying 3D scanned core feature parameters. In this embodiment, the method for identifying 3D scanned core feature parameters is applied to the 3D scanned core feature parameter identification system as described in the above embodiment. The method for identifying 3D scanned core feature parameters includes the following steps:

[0073] S1: The borehole core samples are separated into intact and fragmented cores through adaptive cleaning. The fragmented cores are stored in the fragmented core box, and the intact cores and core tags are positioned together in the slot of the core box to form the initial dataset of spatial-attribute binding.

[0074] S2: Arrange core boxes, and based on point cloud registration, the automated CNC platform dynamically adjusts the height and level of the 3D laser scanner to generate the optimal scanning path topology network covering all core boxes;

[0075] S3: Scan the path topology network, trigger the reinforcement learning control loop, the 3D laser scanner collects point clouds, the core rotating wheel performs angle optimization rotation, provides real-time feedback on point cloud coverage, dynamically adjusts the rotation angle, and outputs a full-angle high-density core point cloud sequence.

[0076] S4: The full-angle high-density core point cloud sequence is fused with multi-scale features: point cloud filtering and registration are performed to remove noise and align multi-view data; a topologically complete model is generated through 3D reconstruction; convolutional neural network identifies joints and fissures, and generates automatically drawn bar charts, quantized parameter sets and joint parameter packages.

[0077] Preferably, the adaptive cleaning process in this embodiment effectively separates intact and fragmented core samples, providing a high-quality initial dataset for subsequent precise scanning and analysis. Spatial-attribute binding technology ensures the accurate correlation between the spatial location and attribute information of the core samples. Point cloud registration and an automated CNC platform enable dynamic adjustment of the scanner's height and level, optimizing scanning efficiency and ensuring the generation of an optimal path topology network covering all core samples, laying the foundation for obtaining high-quality 3D scanning data. Reinforcement learning is used to control the loop, combining the point cloud data acquired by the 3D laser scanner with the optimized rotation of the core rotating wheel. The system rotates the core, providing real-time feedback on point cloud coverage and dynamically adjusting the rotation angle to ensure the generation of high-density core point cloud sequences from all angles, thus improving the integrity and accuracy of data acquisition. Multi-scale feature fusion is employed to filter and register the point cloud, effectively removing noise and aligning multi-view data, achieving high-precision 3D reconstruction of the core and providing a structured data foundation for subsequent feature parameter identification. Through deep learning technologies such as convolutional neural networks, joints / fractures in the core are identified, and histograms, quantified parameter sets, and joint parameter packages are automatically generated, enabling automatic identification and quantification of core feature parameters, significantly improving the automation level and efficiency of core feature analysis.

[0078] In summary, this embodiment integrates technologies such as adaptive cleaning, point cloud registration, reinforcement learning, multi-scale feature fusion, and deep learning to achieve full automation of the entire process from core sample preprocessing to 3D reconstruction and automatic identification of feature parameters, significantly improving the accuracy and efficiency of core feature parameter identification.

[0079] Furthermore, the process of separating intact and fragmented core samples from borehole samples in step S1 through adaptive cleaning specifically includes the following steps:

[0080] S11: Obtain the initial core flow containing surface overburden debris and potentially fractured sections. Using the slot spacing as the benchmark, construct a core axial continuity detection template, integrate the core plate preloaded hole depth data, and mark areas with suspected lithological abrupt changes.

[0081] S12: Dynamic fragmentation criterion generation. When the measured extension length in the suspected area is <0.8d, debris removal is activated. When the curvature gradient between adjacent slots exceeds the reconstruction threshold of subsequent scanning stages, a fragmentation marker is triggered.

[0082] S13: Self-organizing separation execution, complete core is forced to match slot space vector, fractured unit is oriented into the box and inherits the source slot topology coordinates.

[0083] Preferably, this embodiment achieves intelligent sorting of core samples; the axial continuity detection template established based on the slot spacing, through the fusion of preloaded borehole depth data, realizes spatial benchmark calibration of the core structural integrity, solving the misjudgment problem caused by the lack of a unified reference system in traditional manual sorting, and providing a standardized spatial coordinate system for subsequent processing; through the joint determination of the extension length threshold (0.8d) and the curvature gradient threshold, a hierarchical identification system of fragmentation features is established. The former targets macroscopic fragmentation features, while the latter captures microscopic structural distortions. The two work together to achieve accurate detection from millimeter-level debris to centimeter-level fragmented segments. The slot spatial vector matching of the intact core ensures the geometric integrity of the sample, while the fragmentation unit maintains the traceability of the original stratigraphic information through topological coordinate inheritance, achieving physical separation while completely preserving the spatial relationship data of the core.

[0084] In summary, this embodiment establishes a machine vision-based three-dimensional core structure analysis system. Through a technical chain of spatial benchmark calibration, dynamic threshold determination, and topology-preserving separation, it achieves automated identification and classification of fractured cores while ensuring the spatial integrity of geological information. The system output simultaneously meets the dual requirements of laboratory analysis for sample integrity and field geological surveys for spatial correlation.

[0085] Furthermore, the process of generating the optimal scanning path topology network covering all core boxes in step S2 specifically includes the following steps:

[0086] S21: Construct a point cloud registration benchmark for the structural constraints of the core box based on the initial dataset of spatial-attribute binding, use the slot matrix of the core box as a rigid frame, and exclude non-scanned areas based on the topological coordinates of the fragmented core box;

[0087] S22: Dynamic scanning parameter optimization, based on the slot spacing, the laser incident angle tolerance threshold is obtained, and the level reading of the fixed structure is scanned in real time to compensate for the tilt angle error.

[0088] S23: The CNC platform performs topology mapping, transforming the slot coordinate cluster into three-dimensional spatial path nodes, and integrates horizontal compensation parameters to generate a minimum energy scanning trajectory;

[0089] S24: Outputs the optimal scanning path topology network, with node density matching the borehole depth gradient recorded by the core card, and the trajectory envelope covering the rotational space domain of all valid slots.

[0090] Preferably, this embodiment achieves global optimization of the core box scanning path through multi-step collaboration. A rigid framework constructed based on the core box slot matrix, combined with the topological coordinates of the fragmented core box, establishes precise spatial constraints on the scanning area, eliminating invalid scanning areas and ensuring that the scanning path only covers valid slots, reducing redundant calculations. The laser incident angle tolerance threshold is dynamically adjusted based on the slot spacing and linked to the level reading to correct tilt errors in real time, ensuring the geometric consistency of the scanning data and avoiding measurement distortion caused by equipment vibration or assembly deviations. The slot coordinate cluster is transformed into three-dimensional path nodes, and after fusing horizontal compensation parameters, the optimal scanning trajectory is calculated, minimizing the robotic arm's motion energy consumption while satisfying coverage, thus improving scanning efficiency. The node density of the optimal scanning path is adaptively matched with the borehole depth gradient recorded by the core plate, ensuring that high-resolution scanning focuses on key areas, and the trajectory envelope covers the rotational space domain of all valid slots, guaranteeing no data omissions.

[0091] In summary, this embodiment constructs a scanning path planning system based on three-dimensional spatial constraints and dynamic parameter optimization. Through a technical chain of rigid frame registration, real-time error compensation, minimum energy trajectory generation, and gradient matching node density, it achieves efficient and high-precision scanning of core boxes. While ensuring data integrity, it optimizes the motion efficiency of the scanning equipment and is suitable for the automated detection needs of complex geological samples.

[0092] Furthermore, the process of outputting the optimal scan path topology network in step S24 specifically includes the following steps:

[0093] S241: Obtain the minimum energy scan trajectory, perform node densification driven by the borehole depth gradient, extract the borehole depth abrupt change points recorded by the core plate, and multiply the path node density in the corresponding section of the trajectory according to the borehole depth change rate.

[0094] S242: Slot rotation space envelope construction, with the center of the effective slot as the origin, generates a rotation envelope based on the rotation radius of the core rotating wheel, excluding the space domain occupied by the fragmented core box;

[0095] S243: Topology network synthesis, mapping encrypted nodes to the envelope body, and associating the vertical path of the fixed structure with the extremum constraint of the height of the scanning structure.

[0096] S244: Outputs the optimal scanning path topology network. The node distribution reflects the geological characteristics of the core sample, and the envelope boundary ensures collision-free scanning of the core rotating wheel at all angles.

[0097] Preferably, this embodiment achieves refined construction of the optimal scanning path topology network through multi-step collaboration. By extracting abrupt changes in borehole depth recorded on the core plate, the node density is multiplied according to the borehole depth change rate in the corresponding trajectory segment, achieving automatic matching between scanning resolution and the degree of geological feature variation, ensuring higher-precision data acquisition of key geological interfaces. A rotating envelope constructed with the effective slot center as the origin, combined with the fragmented core box spatial domain exclusion mechanism, establishes a complete collision-free scanning spatial model; this ensures the motion freedom of the scanning equipment while avoiding spatial interference with the sample container. The mapping process of encrypted nodes to the envelope, while simultaneously associating the extreme values ​​of the extension and contraction heights of the scanning fixed structure, realizes the feasibility verification of path nodes in three-dimensional space, ensuring that the scanning trajectory simultaneously meets the horizontal and vertical motion constraints. The final output topology network node distribution directly reflects the geological feature changes of the core plate, and the envelope boundary completely covers the full-angle scanning requirements of the core rotating wheel, achieving a unity of the geological significance and mechanical feasibility of the scanning path.

[0098] In summary, this embodiment establishes a scanning path optimization system based on a dual-drive approach of geological features and mechanical constraints. Through a technical chain encompassing borehole depth gradient node encryption, collision-free spatial modeling, 3D path constraint integration, and geological feature mapping, it achieves high-resolution data acquisition of key geological areas while ensuring scanning safety. It is particularly suitable for the automated inspection of core samples with complex geological features.

[0099] Furthermore, the process of outputting the full-angle high-density core point cloud sequence in step S3 specifically includes the following steps:

[0100] S31: Using the network nodes of the optimal scanning path topology network as the positioning reference for the 3D laser scanner, the initial rotation angle base is set according to the radius of the core rotating wheel; the current slot point cloud void rate is obtained in real time after a single scan.

[0101] When the void ratio is greater than the allowable threshold for lithology recorded in the core, the rotation angle of the next cycle is dynamically adjusted, with the angle base multiplied by the void ratio compensation coefficient.

[0102] S32: Each core box performs eight scan-rotation cycles, and terminates early when the cumulative rotation angle reaches the core circumference coverage requirement;

[0103] S33: The void ratio after single-box point cloud superposition and fusion is not greater than the error tolerance of subsequent 3D reconstruction. The spatial coordinates of the sequence strictly match the slot topology network, and output a full-angle high-density core point cloud sequence.

[0104] Preferably, this embodiment achieves efficient acquisition and quality control of high-density core point cloud sequences from all angles through multi-step collaboration. Using the nodes of the optimal scanning path topology network as a benchmark, the rotation angle base is dynamically adjusted based on the real-time acquired slot point cloud void ratio, ensuring automatic improvement in scanning density in lithologically complex areas. Adaptive optimization of scanning resolution is achieved through a void ratio compensation coefficient, avoiding data loss caused by fixed-angle scanning. In eight scan-rotation cycles, if the cumulative rotation angle reaches the core circumference coverage requirement, the scan is terminated early, balancing scanning efficiency and data integrity, avoiding redundant scanning, and ensuring full-angle coverage of key areas. After single-box point cloud overlay and fusion, the void ratio does not exceed the 3D reconstruction error tolerance, and the sequence coordinates strictly match the slot topology network, ensuring the geometric accuracy and topological consistency of the point cloud data, providing high-fidelity input for subsequent 3D reconstruction.

[0105] In summary, this embodiment constructs a dynamic scanning system based on real-time feedback. Through a technical chain of path benchmark positioning, void ratio-driven angle correction, loop termination condition judgment, and point cloud quality verification, it achieves efficient acquisition and quality control of core point clouds. While ensuring full-angle coverage, it optimizes scanning efficiency and ensures that the point cloud data meets the accuracy requirements of 3D reconstruction.

[0106] Furthermore, the process of generating automatically drawn bar charts, quantization parameter sets, and joint parameter packages in step S4 specifically includes the following steps:

[0107] S41: Obtain a full-angle high-density core point cloud sequence, use the core box slot matrix as the topological skeleton, rigidly register the multi-angle point cloud according to the core rotating wheel angle parameters, and fuse the fragmented core box spatial markers to generate a missing area mask;

[0108] S42: Joint identification and parameter extraction, generating virtual core columns by slicing along the slot axis, detecting curvature discontinuities within the slice area, and adjusting the detection sensitivity based on the lithological record of the core plate;

[0109] S43: Generate column charts and parameter sets, accumulate virtual core column lengths to match core plate footage data, convert joint orientation into dip and dip angle parameter packages, and force core extraction length calculation to align with integer multiples of slot spacing;

[0110] S44: Outputs triple modeling results, a topologically complete model that preserves the entity of the card slot space topology, a quantified parameter set, and joint parameters including a dip distribution histogram and dip angle confidence intervals.

[0111] Preferably, this embodiment achieves three-dimensional digital reconstruction of rock cores and automated extraction of joint parameters through multimodal data fusion and structured processing; based on the topological skeleton constraint of the core box slot matrix, the spatial positioning problem of fragmented rock cores is solved by rigid registration of point clouds from multiple angles, and the generation of missing area masks ensures the topological integrity of the model; the virtual core column generation technology transforms discrete point clouds into structured axial slices, establishing a standardized data carrier for subsequent parameter extraction; the curvature discontinuity detection algorithm runs within the slice domain, and the sensitivity parameters modulated by the lithological record realize lithology-related joint identification, while the axial slicing strategy preserves the joints. The spatial correspondence between the foliation plane and the original core ensures the geometric accuracy of the attitude conversion; the length accumulation algorithm and the forced matching of the footage data eliminate the scale error between the scanned data and the field records; the joint parameter package realizes the mathematical standardization description of the attitude of the structural plane through dual-parameter (dip / dip angle) characterization; the length integer multiple alignment mechanism ensures the metrological compatibility between the output data and the physical slots; the topological solid model maintains the original slot spatial relationship to meet the geometric accuracy requirements of geological interpretation; the quantitative parameter set provides a machine-readable structured data interface; and the dip histogram and dip angle confidence interval constitute a dual verification mechanism for joint statistics.

[0112] In summary, this embodiment achieves a closed-loop conversion from unstructured point clouds to standardized geological parameters through a three-layer verification mechanism of geometric constraints, parameter transformation, and data alignment, thus resolving the key technical contradiction between topology preservation and parameter extraction during the digitization of fractured rock cores.

[0113] Furthermore, the process of preserving the entity of the card slot space topology in the topologically complete model in step S44 specifically includes the following steps:

[0114] S441: Using the core box slot matrix as the spatial grid reference, the point cloud surface is divided into equal-length segments according to the slot spacing, and the endpoints of each segment are forced to align with the center normal plane of the slot; the quantitative parameters for generating structural constraints are generated.

[0115] Among them, the core extraction length calculation is: the number of effective segments in the slot × the spacing d; the recovery rate calculation is: extraction length ÷ core plate footage data × 100%; the RQD value calculation is: the sum of complete segments with a cumulative length ≥ 10cm ÷ extraction length.

[0116] S442: Joint parameters are engineered and encapsulated. The dip and tilt angle conversion is based on the slot method plane to establish an attitude reference system. The confidence interval is calculated based on the number of repeated observations of the point cloud in an eight-cycle scan.

[0117] S443: Outputs triple modeling results. The topologically complete model retains the entities of the slot number and the void marker of the fractured core box, and quantifies the parameter set; the joint parameters include the dip-angle joint distribution matrix.

[0118] Preferably, this embodiment achieves the engineering application of the core topology model through spatial constraints and parameter standardization encapsulation; using the slot matrix as a rigid grid reference, the endpoints of point cloud segments are forced to align with the center normal plane of the slot, ensuring spatial consistency between the digital model and the physical core box; the empty space markers of the fractured core box are inherited to the topological entity model, maintaining the geometric representation of the original sampling missing state; the sampling length calculation eliminates the cumulative error between the scan data and the physical slot by using the integer multiple relationship between the slot spacing d and the number of effective segments; the sampling rate and RQD value are calculated based on the same spatial reference, making the three types of parameters (sampling length / sampling rate / RQD) have self-consistent measurement logic; the dip / pitch angle conversion is bound to the slot method plane reference system, converting the geological attitude into an engineering coordinate system that can be identified for construction; the repeated observation data of eight-cycle scanning supports the calculation of confidence intervals, improving the statistical significance of the attitude parameters; the topological entity model retains the slot numbering system, realizing reverse tracing between the digital model and the field records; the dip-pitch angle joint distribution matrix replaces the traditional histogram, revealing the spatial coupling relationship of structural parameters.

[0119] In summary, this embodiment establishes a deterministic mapping relationship from point cloud data to workable geological parameters through three layers of processing: spatial grid constraints, parameter calculation normalization, and engineering coordinate system transformation. This solves the problem of unifying geometric realism and engineering applicability in the digital model of fractured rock cores.

[0120] Furthermore, the process of engineering and encapsulating the joint parameters in step S442 specifically includes the following steps:

[0121] S4421: Extract the normal vector of the major axis of the core box slot as the horizontal reference benchmark, and establish the dip zero plane with the vertical bisector of the slot spacing; the angle between the joint surface and the normal vector of the major axis of the slot is defined as the dip angle, and the angle between the joint surface and the vertical bisector is directly used as the dip angle;

[0122] S4422: Count the number of times the same joint is independently detected in the eight-period point cloud. When the number of detections is greater than or equal to the minimum number of observations required for the lithology record of the core, automatically expand the confidence interval of the joint parameter to the upper limit allowed by the project.

[0123] S4423: The row vector is the slot number index, the column vector is the joint attitude data cluster that has passed eight periods of verification, and the output is the joint distribution matrix of dip and dip angle.

[0124] Preferably, the engineering encapsulation technology of joint parameters in this embodiment, when combined, forms a standardized joint parameter generation system based on spatial reference constraints and statistical verification. Using the major axis normal vector of the slot as the horizontal reference, and combining it with the vertical bisector to define the dip zero-plane, the measurement reference of joint attitude parameters (dip angle, dip angle) is strictly aligned with the engineering coordinate system, eliminating the conversion error between traditional geological attitude and construction coordinate systems. The reliability of joint parameters is verified through eight-cycle point cloud independent detection times. When the observation frequency reaches the minimum sample size required by the lithological record, the confidence interval is automatically widened to the upper limit of the engineering tolerance, balancing data accuracy and engineering applicability. A matrix-style joint attitude database is constructed using the slot number as the row vector index and the eight-cycle verification data cluster as the column vector, supporting subsequent spatial distribution analysis of parameters (such as the generation of a joint dip-dip angle distribution matrix).

[0125] In summary, this embodiment integrates three technologies—benchmark unification, statistical verification standardization, and data storage structuring—to achieve efficient conversion from raw point cloud joint detection to attitude parameters that can be directly used in engineering design.

[0126] Furthermore, the process of establishing the tilt zero-position plane using the perpendicular bisector of the slot spacing in step S4421 specifically includes the following steps:

[0127] S44211: Obtain the physical characteristics of the core box slot matrix, determine the confidence domain of the normal vector based on the machining tolerance of the long side parallelism of the slot, and compensate for the assembly tilt by integrating the historical readings of the scanning fixed structure level.

[0128] S44212: Geometric generation of the perpendicular bisector plane. Using the line connecting the center points of adjacent slots as the baseline, a perpendicular plane to the minor axis of the slot is drawn through the midpoint of the baseline. The intersection of the perpendicular plane and the bottom surface of the slot is defined as the zero-angle baseline.

[0129] S44213: Dual reference solidification output, horizontal reference major axis normal vector compensation value, tilt zero plane, plane where the tilt baseline is located.

[0130] Preferably, this embodiment, by combining the technical features of establishing the tilt zero-angle plane, forms a high-precision, interference-resistant spatial reference definition system. Its technical effects are as follows: Based on the machining tolerance of the long side parallelism of the card slot, the reliability domain of the normal vector is limited; historical data from the level instrument is integrated to compensate for assembly errors, ensuring the engineering applicability of the horizontal reference datum (major axis normal vector); a minor axis perpendicular plane is constructed by connecting the midpoints of adjacent card slot centers, physically solidifying the intersection line as a zero-tilt baseline, giving the tilt zero-angle plane a traceable geometric generation logic; the horizontal reference datum (major axis normal vector) and the tilt zero-angle plane (zero-tilt baseline plane) form an orthogonal spatial coordinate system, which is dynamically correlated through compensation values ​​to jointly resist local disturbances in the scanned data.

[0131] In summary, this embodiment achieves a balance between the physical interpretability and engineering robustness of the tilt measurement datum through three technical means: machining tolerance constraints, geometric structural determinism, and dynamic coupling of dual datums.

[0132] like Figure 14 As shown, this embodiment provides an embodiment of an electronic device 4, which includes a processor 161 and a memory 162 coupled to the processor 161.

[0133] The memory 162 stores program instructions for implementing the method for identifying 3D scan core feature parameters in any of the above embodiments.

[0134] The processor 161 is used to execute program instructions stored in the memory 162 to identify characteristic parameters of the 3D scan core.

[0135] The processor 161 can also be referred to as a CPU (Central Processing Unit). The processor 161 may be an integrated circuit chip with signal processing capabilities. The processor 161 can also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. A general-purpose processor can be a microprocessor or any conventional processor.

[0136] Furthermore, Figure 15 This is a schematic diagram of the structure of a storage medium according to an embodiment of this application. The storage medium 17 of this embodiment stores program instructions 171 capable of implementing all the methods described above. These program instructions 171 can be stored in the storage medium in the form of a software product, including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks, or terminal devices such as computers, servers, mobile phones, and tablets.

[0137] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between apparatuses or units through some interfaces, and may be electrical, mechanical, or other forms.

[0138] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units. The above are merely embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

[0139] The specific embodiments of the invention have been described in detail above, but these are merely examples, and the invention is not limited to the specific embodiments described above. For those skilled in the art, any equivalent modifications or substitutions to the invention are also within the scope of this invention. Therefore, all equivalent transformations, modifications, and improvements made without departing from the spirit and principles of this invention should be included within the scope of this invention.

Claims

1. A system for identifying characteristic parameters of 3D scanned rock cores, characterized in that, The identification system includes: The core collection module is used to automatically rotate the core at a fixed angle in a single rotation under the control of an automated CNC platform, enabling multi-angle scanning; it provides orderly core fixation, controllable rotation, and optimized scanning environment. The 3D scanning module is used to scan and acquire high-precision point cloud data of the rock core surface; the data acquisition processor processes the point cloud to construct a complete three-dimensional model, automatically calculates key geological parameters, and assists in the identification of joints; The core collection module includes: a core box, a fragmented core container, a core rotating wheel, a core label, and a transmission structure; The core box has a transmission structure installed on one side, which is connected to the automated CNC platform via a control line. The core rotating wheel is mounted on the core box. The core rotating wheel is installed under each slot of the core box and is connected to the transmission structure via a pin structure, so that the core rotating wheel rotates in the same direction. The transmission structure is connected to the core rotating wheel via a pin structure and drives the core rotating wheel to rotate by rotating. Connect one or more core boxes as needed to achieve overall core scanning; The core plate consists of an abrasion-resistant outer cover, a cardboard plate, and a slot structure, and is placed on the core box. The abrasion-resistant outer cover is located on the outside of the slot structure, and the cardboard plate is placed in the middle of the abrasion-resistant outer cover. The 3D scanning module consists of a 3D laser scanner, a scanning fixture, a data acquisition processor, and an automated numerical control platform. The system comprises a 3D laser scanner for scanning the core model, controlled by an automated CNC platform, positioned directly above the core box; a scanning fixing structure for securing the 3D laser scanner, also controlled by the automated CNC platform; a data acquisition processor for processing the data obtained by the 3D laser scanner, which emits a laser beam and receives light signals reflected from the object's surface, calculating the propagation time and angle information of the light to obtain the three-dimensional coordinate data of the object's surface; and an automated CNC platform for controlling the scanning fixing structure and the core rotating wheel, ensuring that the 3D laser scanner is positioned directly above the core each time it scans, by pre-setting the number and spacing of the core boxes and the slot spacing of the core boxes.

2. The 3D scanning core feature parameter identification system according to claim 1, characterized in that, The scanner is fixed to the ground by four corner brackets, which are telescopic in height. A level is attached to the top of each corner bracket. By adjusting the corner brackets to align with the level, the 3D laser scanner on the top of the corner brackets is kept horizontal. The part of the top of the corner brackets that connects to the 3D laser scanner is connected to pulleys and then locked onto the top corner brackets. The horizontal position of the 3D laser scanner can be adjusted by these pulleys.

3. A method for identifying characteristic parameters of 3D scanned rock cores, applied to the 3D scanned rock core characteristic parameter identification system as described in any one of claims 1 to 2, characterized in that, The method for identifying the characteristic parameters of the 3D scanned rock core includes: The core samples from the borehole were separated into intact and fragmented cores through adaptive cleaning. The fragmented cores were stored in the fragmented core box, while the intact cores and the core tags were positioned together in the slots of the core box to form an initial dataset with spatial-attribute binding. The core boxes are arranged, and based on point cloud registration, the automated CNC platform dynamically adjusts the height and level of the 3D laser scanner to generate the optimal scanning path topology network covering all core boxes. The scanning path topology network triggers a reinforcement learning control loop, a 3D laser scanner collects point clouds, the core rotating wheel performs angle optimization rotation, the point cloud coverage is fed back in real time, the rotation angle is dynamically adjusted, and a full-angle high-density core point cloud sequence is output. The full-angle high-density core point cloud sequence is processed through multi-scale feature fusion: point cloud filtering and registration are performed to remove noise and align multi-view data; a topologically complete model is generated through 3D reconstruction; and a convolutional neural network is used to identify joints and fractures, generating automatically drawn bar charts, quantized parameter sets, and joint parameter packages.

4. The method for identifying characteristic parameters of 3D scanned rock cores according to claim 3, characterized in that, The process of generating automatically drawn bar charts, quantization parameter sets, and joint parameter packages includes the following steps: A high-density core point cloud sequence from all angles was obtained. Using the core box slot matrix as the topological skeleton, the multi-angle point clouds were rigidly registered according to the core rotating wheel angle parameters. Spatial markers of the fragmented core box were then fused to generate a mask for the missing area. Joint identification and parameter extraction, slicing along the groove axis to generate virtual core columns, detecting curvature discontinuities within the slice area, and adjusting detection sensitivity based on the lithological record of the core plate; The column chart and parameter set are generated, the virtual core column length is accumulated and matched with the core card footage data, the joint attitude is converted into dip and dip angle parameter packages, and the core extraction length is calculated to be forced to be an integer multiple of the slot spacing. The output includes a triple modeling result: a topologically complete model that preserves the entity of the card slot space topology, a quantified parameter set, and joint parameters including a dip distribution histogram and dip angle confidence intervals.

5. The method for identifying characteristic parameters of 3D scanned rock cores according to claim 4, characterized in that, The process of preserving the entity of the card slot space topology in the topological complete model includes the following steps: Using the core box slot matrix as the spatial grid reference, the point cloud surface is divided into equal-length segments according to the slot spacing, and the endpoints of each segment are forced to align with the center normal plane of the slot; the quantitative parameters for generating structural constraints are generated. Joint parameters are encapsulated in an engineering manner; dip and tilt angle conversions are based on establishing an attitude reference system on the card slot method plane; confidence intervals are calculated based on the number of repeated observations of point cloud during eight-cycle scanning. The output includes a triple modeling result: a topologically complete model that retains the entities of the slot number and the void marker of the fractured core box, and a quantified parameter set; and joint parameters that include a dip-angle joint distribution matrix.

6. The method for identifying characteristic parameters of 3D scanned rock cores according to claim 5, characterized in that, The process of engineering encapsulating joint parameters includes the following steps: The normal vector of the major axis of the core box slot is extracted as a horizontal reference datum, and the dip zero plane is established with the vertical bisector of the slot spacing; the angle between the joint surface and the normal vector of the major axis of the slot is defined as the dip angle, and the angle between the joint surface and the vertical bisector is directly used as the dip angle; The number of times the same joint is independently detected in the eight-cycle point cloud is counted. When the number of detections is not less than the minimum number of observations required for the lithology to be recorded in the core, the confidence interval of the joint parameter is automatically expanded to the upper limit allowed by the project. The row vector is the slot number index, the column vector is the joint attitude data cluster that has passed eight-period verification, and the output is the dip-pitch joint distribution matrix.

Images