Method for evaluating and partitioning rock mass quality of reservoir drawdown zone based on machine vision

Abstract

This invention discloses a machine vision-based method for evaluating and zoning the quality of rock masses in the drawdown zone of a reservoir area. It relates to the fields of rock engineering and machine vision intelligent recognition technology, and includes the following steps: collecting rock mass image information covering the study area's drawdown zone, and dividing the rock mass image information according to a unified spatial scale to form multiple rock mass evaluation regions within the study area, each corresponding to a specific spatial location, and establishing a correspondence between the rock mass evaluation regions and their spatial locations. This invention achieves automatic evaluation and scoring of rock mass quality by spatially dividing the rock mass images of the drawdown zone and identifying fracture structure parameters. Simultaneously, it organizes the scoring results based on the spatial location of the rock mass evaluation regions to form a rock mass quality zoning representation, allowing for a direct presentation of rock mass quality differences within the study area. This improves the efficiency of rock mass surveys in the drawdown zone of the reservoir area and reduces labor costs.

Description

Technical Field

[0001] This invention relates to the fields of rock mass engineering and machine vision intelligent recognition technology, specifically to a machine vision-based method for evaluating and zoning the quality of rock mass in the drawdown zone of a reservoir. Background Technology

[0002] Machine vision-based rock mass quality assessment and zoning in the drawdown zone of a reservoir refers to the use of image acquisition equipment to capture large-scale images of the rock mass surface in the drawdown zone. Machine vision and image processing technologies are then used to automatically identify and extract fissures from the rock mass images. Furthermore, parameters reflecting the structural characteristics of the rock mass, such as fissure length, fissure width, spacing within fissure groups, and apparent degree of fragmentation, are calculated. Based on this, mechanical indicators such as the uniaxial compressive strength of the rock are incorporated to adaptively optimize the traditional rock mass quality assessment system. This allows the fissure structure information acquired from the images to directly participate in the rock mass quality assessment calculation, resulting in a rock mass quality score. The rock mass in the study area is then spatially divided and visualized according to the score level. This automates the entire process from rock mass image acquisition, fissure identification, structural parameter extraction to rock mass quality assessment and regional zoning. It is used to quickly identify weak areas and quality differences in the rock mass within the drawdown zone of the reservoir, providing data for reservoir bank stability analysis and engineering risk identification.

[0003] The existing technology has the following shortcomings: In the process of evaluating and zoning rock mass quality in the drawdown zone of reservoir areas based on machine vision, some practical problems still exist: On the one hand, although existing machine vision technology can automatically identify rock mass fractures and extract geometric parameters, most methods only stay at the level of fracture segmentation and parameter statistics, lacking a technical path to further transform fracture image information into rock mass quality evaluation indicators, making it difficult for fracture identification results to directly serve engineering quality judgment; on the other hand, key parameters in the traditional rock mass quality evaluation system, such as RQD, joint spacing, and structural surface characteristics, usually rely on borehole cores or field surveys, lacking a direct correspondence with image data, making it difficult to directly apply the fracture parameters extracted by machine vision to rock mass quality grading and spatial zoning analysis, thus limiting the development of automated rock mass quality evaluation and large-scale rapid zoning technology, and making it difficult for existing systems to meet the engineering application needs of rock mass environments such as the drawdown zone of reservoir areas, which are wide-ranging, rapidly changing, and have significant fracture control.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a machine vision-based method for evaluating and zoning the rock mass quality of the drawdown zone in a reservoir area, in order to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a machine vision-based method for evaluating and zoning the rock mass quality of the drawdown zone in a reservoir area, comprising the following steps: Collect rock mass image information covering the drawdown zone of the study area, and divide the rock mass image information according to a unified spatial scale. Multiple rock mass evaluation areas with one-to-one spatial location are formed within the study area, and the correspondence between the rock mass evaluation areas and spatial location is established. The corresponding rock mass image information is read around the rock mass evaluation area. The distribution of fractures is identified pixel by pixel within the rock mass evaluation area. Fracture trajectory information is extracted from the rock mass evaluation area. Fracture morphology is formed by sorting out the fracture trajectory information. Based on the fracture trajectory information, the fracture morphology inside the rock mass evaluation area is continuously read. The spatial position relationship is read point by point along the fracture trajectory. The fracture extension distance is calculated based on the distance between adjacent positions. The opening distance between the two sides of the fracture is also counted to obtain the fracture length information and fracture width information corresponding to the rock mass evaluation area, forming a set of fracture geometric information. The spatial distribution relationship of fractures is read from the set of fracture geometric information, the fracture extension direction is sorted to form fracture grouping results, the spatial distance between adjacent fractures is measured according to the fracture grouping results, and the fracture spacing ratio is calculated according to the fracture spacing to obtain the corresponding apparent degree of fracture information of the rock mass evaluation area. By combining the information on fracture length, fracture width, fracture spacing ratio, and apparent degree of fragmentation of the rock mass evaluation area, a comprehensive score is given to the rock mass evaluation area. Based on the score results, spatial correspondence is organized within the study area to form the rock mass quality zoning results.

[0007] Preferably, the process of collecting image information of the drawdown zone rock mass covering the study area and forming the rock mass evaluation area is as follows: Continuous images of the rock mass surface in the drawdown zone are acquired along the spatial distribution direction of the study area to form a set of rock mass image information covering the study area. The rock mass image information is then organized according to the acquisition order to establish a correspondence between the rock mass image information and the spatial location within the study area. The rock mass image information is organized into a unified spatial scale based on the rock mass image information set, so that the spatial dimensions in the rock mass image information are expressed according to a unified scale, and the rock mass image information in the study area is expressed in the same spatial scale. Based on a unified spatial scale, rock mass image information is regularly divided to form multiple regularly arranged rock mass evaluation areas within the study area, and each rock mass evaluation area contains rock mass image information corresponding to its spatial location. The spatial locations of the rock mass evaluation areas within the study area are organized around the rock mass evaluation areas to establish a one-to-one correspondence between the rock mass evaluation areas and their spatial locations within the study area, forming a continuously arranged set of rock mass evaluation areas within the study area.

[0008] Preferably, the rock mass evaluation areas are spatially arranged according to the acquisition order of rock mass image information within the study area, and spatial correlation is carried out around the rock mass image information contained within the rock mass evaluation areas, so that the rock mass evaluation areas correspond to the spatial locations within the study area, and a continuous distribution structure of rock mass evaluation areas is formed within the study area.

[0009] Preferably, the steps for reading the corresponding rock mass image information and forming the fracture morphology around the rock mass evaluation area are as follows: The image information of the rock mass within the rock mass evaluation area is read around the rock mass evaluation area. The image content within the rock mass evaluation area is expanded pixel by pixel to form a set of pixel distribution within the rock mass evaluation area, and then a continuous pixel sequence is formed according to the spatial arrangement order. The pixel positions are read pixel by pixel around the pixel distribution set within the rock mass evaluation area. The fracture distribution is identified at the pixel positions to form a fracture pixel set. The fracture pixel set is then organized according to the spatial adjacency relationship to form a continuous fracture region. The fracture pixel set is read point by point according to the spatial arrangement order, the fracture pixel positions are connected and organized to form a fracture trajectory path, and a fracture trajectory information set is formed within the rock mass evaluation area. The information set of fracture trajectories is arranged and organized according to their spatial distribution, and the extension direction of the fracture trajectories is continuously organized to form the expression result of fracture morphology within the rock mass evaluation area.

[0010] Preferably, the steps for obtaining fracture length and fracture width information based on fracture trajectory information and forming a set of fracture geometric information are as follows: The fracture trajectory information within the rock mass evaluation area is continuously read, and the trajectory points in the fracture trajectory are arranged sequentially according to the spatial extension direction of the fracture trajectory within the rock mass evaluation area. The spatial position corresponding to the trajectory point is read to form a continuous trajectory point set. The spatial relationship between adjacent trajectory points is read according to the order of the trajectory points, the spatial distance between adjacent trajectory points is recorded segment by segment, and the spatial distance is organized according to the trajectory extension order to form the crack length information. The positions of the crack boundaries on both sides of the trajectory points are read around the set of trajectory points. The spatial positional relationship between the trajectory points and the two sides of the crack boundaries is recorded. The spatial distance between the trajectory points and the two sides of the crack boundaries is arranged according to the order of the trajectory points to form crack width information. The fracture length and width information are combined and organized according to the correspondence of fracture trajectory to form a set of fracture geometric information within the rock mass evaluation area.

[0011] Preferably, the spatial distance between adjacent trajectory points is continuously recorded according to the spatial arrangement order of the trajectory points around the trajectory point set, and the crack length information is formed according to the trajectory extension order. At the same time, the positions of the crack boundaries on both sides of the trajectory points are read around the trajectory point set, and the spatial distance between the trajectory points and the two sides of the crack boundaries is sorted to form crack width information. The crack length information and crack width information are combined and sorted according to the crack trajectory correspondence to form the crack geometric information expression result.

[0012] Preferably, the steps for obtaining apparent fragmentation information based on the set of fracture geometry information are as follows: Based on the geometric information set of fractures within the rock mass evaluation area, the spatial distribution of each fracture within the rock mass evaluation area is read, the spatial location of the fracture initiation, extension and termination positions is organized, and a fracture spatial distribution sequence is formed according to the spatial arrangement order. The fracture extension paths are read around the spatial distribution sequence of fractures, the fracture extension directions are continuously sorted, and the fractures are grouped according to the extension direction to form multiple fracture grouping results. Based on the fracture grouping results, the fractures within the same fracture group are spatially arranged and read, and the spatial positional relationship between adjacent fractures is measured and recorded to form a fracture spacing sequence; The fracture spacing sequence was statistically analyzed, and the spatial distribution ratio of fracture spacing within the rock mass evaluation area was statistically expressed to obtain information on the apparent degree of fragmentation in the rock mass evaluation area.

[0013] Preferably, the steps for forming rock mass quality zoning results around the rock mass evaluation area are as follows: Information on fracture length, fracture width, fracture spacing ratio, and apparent fragmentation degree is collected around the rock mass evaluation area. The fracture length, fracture width, fracture spacing ratio, and apparent fragmentation degree information are then organized according to the spatial location of the fractures within the rock mass evaluation area to form a fracture structure information set. Based on the fracture structure information set, the fracture length information, fracture width information, fracture spacing ratio and apparent degree of fragmentation information are comprehensively read, and the fracture structure state inside the rock mass evaluation area is comprehensively scored and sorted to form the corresponding rock mass quality score result of the rock mass evaluation area. The rock mass quality score results are read and organized according to the spatial arrangement of the rock mass evaluation area within the study area. The rock mass quality score results are correlated with the spatial location of the rock mass evaluation area within the study area to form a spatial distribution sequence of rock mass quality scores. The rock mass evaluation area is divided and organized according to the spatial distribution sequence of rock mass quality scores. Rock mass evaluation areas with scores within the same score range are spatially arranged to form the rock mass quality zoning results corresponding to the study area.

[0014] The technical effects and advantages provided by the present invention in the above technical solution are as follows: This invention performs a unified spatial scale division on rock mass images of the drawdown zone, and conducts pixel-by-pixel identification of fractures, extraction of fracture trajectories, and acquisition of fracture geometric information within the rock mass evaluation area. This enables fracture length, fracture width, fracture spacing ratio, and apparent fragmentation information to be fully expressed within the rock mass evaluation area and further transformed into rock mass quality scoring results. This achieves an automated processing procedure that directly generates rock mass quality evaluation results from rock mass image information, allowing rock mass fracture image information to directly serve the rock mass quality evaluation process, reducing the degree of manual intervention, and improving the automation level and processing efficiency of the rock mass quality evaluation process.

[0015] Based on the rock mass quality scoring results corresponding to each rock mass evaluation area, this invention unifies and organizes the spatial location relationships of the rock mass evaluation areas within the study area, enabling the rock mass quality scoring results to form a continuous spatial distribution expression within the study area. Furthermore, it forms rock mass quality zoning results, thereby achieving an intuitive expression of the spatial distribution of rock mass quality within the study area. This allows different rock mass quality areas to form a clear spatial correspondence within the study area, facilitating engineers to quickly identify areas with differences in rock mass quality and potentially weak areas, improving the efficiency of rock mass investigation in the reservoir drawdown zone and reducing the cost of manual investigation. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a flowchart of the method for evaluating and zoning the rock mass quality of the drawdown zone in the reservoir area based on machine vision, as described in this invention. Detailed Implementation

[0018] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0019] This invention provides, for example Figure 1 The machine vision-based method for evaluating and zoning the rock mass quality of the reservoir drawdown zone, as shown, includes the following steps: Collect rock mass image information covering the drawdown zone of the study area, and divide the rock mass image information according to a unified spatial scale. Multiple rock mass evaluation areas with one-to-one spatial location are formed within the study area, and the correspondence between the rock mass evaluation areas and spatial location is established. By uniformly acquiring and spatially organizing rock mass image information within the study area, multiple rock mass evaluation regions can be formed according to a unified spatial scale, and a correspondence between rock mass evaluation regions and spatial locations can be established within the study area. The specific implementation steps are as follows: A continuous image acquisition process was conducted on the surface of the rock mass in the drawdown zone within the study area, enabling a complete image record of all rock mass surface information within the study area. During the process, the rock mass surface was photographed segment by segment along the spatial distribution direction of the study area, forming a set of rock mass image information covering the entire study area through continuous image acquisition. In the image acquisition process, the rock mass surface was photographed region by region, ensuring that each acquired rock mass image corresponds to a specific spatial location within the study area, while maintaining continuous spatial coverage between adjacent images, thus creating a continuous spatial distribution of rock mass surface images within the study area. Subsequently, all acquired rock mass image information was organized according to the acquisition sequence, establishing an initial correspondence between each rock mass image and a specific spatial location within the study area. Through this process, a set of rock mass image information containing multiple rock mass surface images was formed within the study area, providing a complete data foundation for subsequent unified spatial scaling processing.

[0020] After creating the rock mass image information set, the rock mass image information is standardized in terms of spatial scale, ensuring that rock mass image information at different locations is represented within the same scale system. In the implementation process, firstly, the spatial dimensions in the rock mass image information are standardized, ensuring that the spatial dimensions of different rock mass image information are represented according to a unified scale, thus guaranteeing consistency in spatial dimension representation across different rock mass image information. Subsequently, the rock mass image information is comprehensively organized according to the unified spatial scale, enabling rock mass image information at different locations within the study area to be represented under the same spatial scale conditions. Through this standardized spatial scale organization process, all rock mass image information within the study area is placed under unified spatial representation conditions, providing a unified reference basis for the subsequent rock mass evaluation area delineation process and ensuring consistency in spatial dimensions among various rock mass image information.

[0021] Under a unified spatial scale, rock mass image information within the study area is regularly divided to form multiple rock mass evaluation regions. In the implementation process, a regular division sequence is first established within the study area, allowing the rock mass image information to be spatially divided according to a unified division order. Subsequently, the rock mass image information is continuously divided along one direction within the study area, forming multiple consecutively arranged image regions in that direction. Then, continuous division continues along another direction within the study area, resulting in multiple regularly arranged rock mass evaluation regions. Each rock mass evaluation region contains rock mass image information corresponding to its spatial location, while maintaining a regular arrangement between the evaluation regions. Through this division process, the rock mass image information within the study area is transformed from a continuous image state into a set of multiple regularly arranged rock mass evaluation regions. Each evaluation region can independently reflect the surface image information of the rock mass at its corresponding spatial location, providing an independent evaluation unit for subsequent fracture identification processing.

[0022] After establishing multiple rock mass evaluation zones, the spatial locations of each evaluation zone within the study area are mapped to establish a stable correspondence. In the process, the spatial locations of each evaluation zone are first recorded according to their order within the study area, ensuring that each evaluation zone corresponds to a specific spatial location within the study area. Subsequently, the rock mass image information contained within each evaluation zone is correlated with its corresponding spatial location, enabling the evaluation zone to accurately represent the surface condition of the rock mass at that location. Finally, all rock mass evaluation zones within the study area are organized holistically according to their spatial order, maintaining a continuous spatial relationship between them and establishing stable spatial adjacency between adjacent evaluation zones. Through the above-mentioned sorting process, multiple rock mass evaluation areas corresponding one-to-one with spatial locations are formed within the study area. At the same time, the correspondence between the rock mass evaluation areas and spatial locations is established, thereby providing a unified reference basis for subsequent crack identification processing, crack geometric information acquisition, and rock mass quality evaluation processes around the rock mass evaluation areas, and ensuring that the rock mass quality evaluation results can be expressed in an orderly manner according to the spatial location within the study area.

[0023] The corresponding rock mass image information is read around the rock mass evaluation area. The distribution of fractures is identified pixel by pixel within the rock mass evaluation area. Fracture trajectory information is extracted from the rock mass evaluation area. Fracture morphology is formed by sorting out the fracture trajectory information. By performing pixel-by-pixel analysis of the rock mass image information within the evaluation area, the distribution of fractures within the evaluation area can be continuously identified, and stable fracture trajectory information can be formed. Based on this, the fracture trajectory information is spatially organized to form a complete fracture morphology representation, thus providing a unified fracture representation basis for subsequent fracture geometric information acquisition. The specific implementation steps are as follows: Around the established spatial correspondence of rock mass evaluation areas, the image information of the rock mass contained within each evaluation area is read region by region, enabling the image content within each evaluation area to form a continuous image representation. During implementation, the rock mass image information within each evaluation area is completely read according to their arrangement within the study area, maintaining the original spatial distribution of the image content. Subsequently, the image content within each evaluation area is unfolded point by point according to its pixel spatial location, making each pixel an independent image representation unit and establishing a correspondence between each pixel and its corresponding spatial location within the evaluation area. After pixel-by-pixel unfolding, all pixel locations within the evaluation area are sequentially arranged according to their spatial arrangement, forming a continuous pixel sequence within the evaluation area. Through pixel-by-pixel unfolding and spatial ordering, the image information within the evaluation area is transformed from a holistic image representation to a pixel-level spatial representation, thus forming a pixel distribution set within the evaluation area and providing a complete image information foundation for subsequent fracture distribution identification.

[0024] After establishing the pixel distribution set within the rock mass evaluation area, fracture distribution is identified at the pixel locations within the evaluation area, allowing each pixel belonging to a fracture region to be individually identified. During implementation, pixel positions are read one by one around the pixel sequence formed within the rock mass evaluation area. Combining the pixel positions with their representational characteristics in the rock mass image, fracture distribution identification is performed, recording pixels belonging to fracture regions and forming fracture pixel sets. As the pixel-by-pixel identification process continues, multiple fracture pixel locations within the rock mass evaluation area gradually form a continuous distribution. Subsequently, the identified fracture pixel locations are organized according to spatial adjacency, forming continuous fracture regions from adjacent fracture pixel locations while maintaining independent spatial distribution relationships between different fracture regions. Through this continuous organization of fracture pixel locations, multiple fracture pixel sets are formed within the rock mass evaluation area, each corresponding to a fracture distribution region within the rock mass evaluation area. This allows for a clear spatial representation of the fracture distribution within the rock mass evaluation area at the pixel level.

[0025] After the fracture pixel set is formed, it is continuously read within the rock mass evaluation area, gradually forming fracture trajectory information. During implementation, each fracture pixel set is read point-by-point according to its spatial arrangement, connecting the pixel positions according to spatial adjacency, thus gradually forming a continuous path representation from the originally discrete fracture pixel positions. As the point-by-point connection process continues, the fracture pixel set gradually forms a complete fracture trajectory path, resulting in multiple fracture trajectory information sets within the rock mass evaluation area. Subsequently, each fracture trajectory path is organized according to its spatial extension order, forming a continuous trajectory sequence of pixel positions along the extension direction while maintaining a stable spatial arrangement. Through point-by-point reading and trajectory organization of the fracture pixel sets, all fracture distributions within the rock mass evaluation area can be transformed into multiple fracture trajectory information sets, each corresponding to a fracture extension path within the rock mass evaluation area, thus forming a trajectory representation structure of the fracture distribution within the rock mass evaluation area.

[0026] After forming the fracture trajectory information set, all fracture trajectory information within the rock mass evaluation area is comprehensively organized to further refine the fracture morphology. In the process, the fracture trajectory information is first arranged sequentially according to its spatial distribution within the rock mass evaluation area, ensuring that spatially adjacent fracture trajectories remain adjacent. Then, the extension direction of each fracture trajectory is continuously organized, creating a continuous extension within the rock mass evaluation area and maintaining a stable spatial distribution. After completing the direction organization, all fracture trajectories within the rock mass evaluation area are organized as a whole according to their spatial arrangement, forming a complete fracture distribution structure and thus representing the fracture morphology within the rock mass evaluation area. Through this process, the fracture distribution within the rock mass evaluation area is gradually transformed from the original image state to a continuous fracture trajectory representation state, ensuring that each fracture trajectory accurately reflects the spatial distribution of fractures within the rock mass evaluation area. This provides a unified foundation for subsequent fracture length and width acquisition processes based on the fracture trajectory information.

[0027] Based on the fracture trajectory information, the fracture morphology inside the rock mass evaluation area is continuously read. The spatial position relationship is read point by point along the fracture trajectory. The fracture extension distance is calculated based on the distance between adjacent positions. The opening distance between the two sides of the fracture is also counted to obtain the fracture length information and fracture width information corresponding to the rock mass evaluation area, forming a set of fracture geometric information. By segmentally analyzing the fracture trajectory information already formed within the rock mass evaluation area, the fracture trajectory can be transformed into fracture length and width information that describe the spatial extension and opening characteristics of the fractures. This forms a complete set of fracture geometric information within the rock mass evaluation area, thus providing a unified parameter basis for subsequent rock mass structure evaluation. The specific implementation steps are as follows: Based on the fracture trajectory information already formed within the rock mass evaluation area, the fracture trajectories are continuously read, forming a complete trajectory reading sequence according to their spatial extension order. During implementation, each fracture trajectory within the rock mass evaluation area is read one by one, and all trajectory points within the fracture trajectory are arranged sequentially according to the spatial extension direction within the rock mass evaluation area, ensuring that each trajectory point forms a continuous trajectory sequence according to spatial arrangement. Subsequently, the spatial location of each trajectory point in the fracture trajectory sequence is read, ensuring that each trajectory point corresponds to a specific spatial location within the rock mass evaluation area, and establishing continuous spatial adjacency relationships between trajectory points. Through continuous reading, all trajectory points in the fracture trajectory sequence form a continuous trajectory point set according to spatial order, maintaining a stable spatial arrangement relationship between the trajectory points. After completing the trajectory point reading, the fracture trajectory sequence is organized according to the trajectory extension order, forming a complete fracture morphology expression structure within the rock mass evaluation area, thus providing a continuous trajectory basis for the subsequent acquisition of fracture extension distance.

[0028] After completing the point-by-point reading of the fracture trajectory, the spatial distances between adjacent trajectory points in the fracture trajectory sequence are read, enabling a continuous spatial representation of the fracture extension distance within the rock mass evaluation area. During implementation, the spatial relationships between trajectory points are read point-by-point according to their arrangement in the fracture trajectory sequence, establishing a clear spatial correspondence between adjacent trajectory points. Subsequently, the spatial distance between every two adjacent trajectory points is recorded segment by segment, ensuring that each segment of the fracture trajectory sequence can be expressed through the spatial distance between trajectory points. As the segment-by-segment reading of the fracture trajectory sequence continues, the spatial distances between all adjacent trajectory points in the trajectory sequence gradually form a continuous distance sequence. Then, the spatial distances of each segment in the fracture trajectory sequence are organized according to the trajectory extension order, forming a complete extension distance expression structure. Through the above segment-by-segment reading and organization process, each fracture trajectory within the rock mass evaluation area can form a continuous fracture extension distance expression result, thereby obtaining the fracture length information corresponding to each fracture within the rock mass evaluation area.

[0029] After obtaining the fracture length information, the positions of the fracture boundaries on both sides are read around the trajectory points in the fracture trajectory sequence, so that the fracture opening distance can form a continuous spatial expression within the rock mass evaluation area. During implementation, each trajectory point in the fracture trajectory sequence is used as the center position, and the fracture boundary positions are read in both directions on either side of the trajectory point, establishing a clear spatial relationship between the trajectory point and the fracture boundaries. Subsequently, the spatial positions from the trajectory point to the fracture boundaries are continuously recorded, so that each trajectory point forms a corresponding set of fracture boundary positions. As the trajectory point reading process continues, a continuous record of fracture boundary positions gradually forms within the fracture trajectory sequence. Then, the spatial distances between each trajectory point and the corresponding fracture boundaries are organized according to the trajectory point arrangement order, so that the fracture trajectory sequence forms a continuous opening distance expression sequence. Through the above reading and organization process, each fracture trajectory within the rock mass evaluation area can form a corresponding fracture opening distance expression result, thereby obtaining the fracture width information corresponding to the fractures within the rock mass evaluation area.

[0030] After obtaining the fracture length and width information, all fracture geometric information within the rock mass evaluation area is uniformly organized to form a complete set of fracture geometric information within the evaluation area. In the implementation process, firstly, the fracture length information corresponding to all fracture trajectories within the rock mass evaluation area is arranged according to the spatial position of the fracture trajectories, so that each fracture trajectory generates a corresponding fracture length expression result. Then, the fracture width information corresponding to each trajectory point is organized according to the arrangement order of the trajectory points, forming a continuous fracture width expression sequence within the fracture trajectory. Finally, the fracture length and width information are combined according to the correspondence of the fracture trajectories, so that each fracture trajectory generates a fracture geometric expression result containing both fracture length and width information. As the geometric information corresponding to all fracture trajectories within the rock mass evaluation area is organized, a unified set of fracture geometric information is gradually formed within the rock mass evaluation area. This allows each fracture within the rock mass evaluation area to be expressed through fracture length and fracture width information, thus providing a complete foundation of fracture geometric information for subsequent fracture spatial distribution analysis and rock mass structure evaluation.

[0031] The spatial distribution relationship of fractures is read from the set of fracture geometric information, the fracture extension direction is sorted to form fracture grouping results, the spatial distance between adjacent fractures is measured according to the fracture grouping results, and the fracture spacing ratio is calculated according to the fracture spacing to obtain the corresponding apparent degree of fracture information of the rock mass evaluation area. By analyzing the spatial relationships of the geometric information set of existing fractures within the rock mass evaluation area, the spatial distribution characteristics of the fractures can be systematically expressed. Based on this, fracture grouping results and fracture spacing ratios are generated, thereby obtaining information on the apparent degree of fragmentation within the rock mass evaluation area. This provides a structural integrity parameter basis for the rock mass quality evaluation process. The specific implementation steps are as follows: Based on the existing set of fracture geometric information within the rock mass evaluation area, the spatial distribution of each fracture within the evaluation area is continuously read, enabling a clear spatial representation of their positional relationships. During implementation, each fracture within the evaluation area is read individually according to the fracture length and width information recorded in the fracture geometric information set. The starting, extension, and ending positions of each fracture are then spatially organized, forming a complete spatial distribution path for each fracture within the evaluation area. Subsequently, the spatial positions of all fractures within the evaluation area are uniformly arranged, forming a spatial distribution sequence. This spatial ordering clearly expresses the relative positional relationships between different fractures. As the spatial organization process continues, a complete fracture spatial distribution structure gradually forms within the evaluation area, allowing the spatial distribution relationship of each fracture to be expressed through positional arrangement. This provides a spatial relationship basis for subsequent fracture extension direction organization.

[0032] After organizing the spatial distribution of fractures, the fracture extension directions in the fracture geometry information set are continuously read and organized to group fractures with similar extension directions. In the process, firstly, the extension path of each fracture within the rock mass evaluation area is read and organized sequentially according to its extension direction, ensuring each fracture has a corresponding extension direction. Then, all fractures within the rock mass evaluation area are arranged sequentially according to their extension directions, creating adjacency relationships between fractures with similar extension directions. Next, the fractures are grouped according to their extension direction arrangement, ensuring that fractures with consistent extension directions form the same group, and fractures with different extension directions form different group sets. Through this extension direction organization process, multiple fracture groupings are gradually formed within the rock mass evaluation area, each containing multiple fractures with consistent extension directions, thus enabling a structured representation of the fracture distribution within the rock mass evaluation area according to their extension directions.

[0033] After obtaining the fracture grouping results, the spatial distance between adjacent fractures is measured around each fracture group, enabling a continuous spatial representation of fracture spacing within the rock mass evaluation area. In the process, each fracture group is first read line by line according to the fracture grouping results, and then the fractures are sequentially arranged according to their spatial arrangement within the rock mass evaluation area, ensuring a continuous spatial arrangement of fractures within the same group. Subsequently, within the same fracture group, the spatial relationship between adjacent fractures is read, allowing the spatial distance between adjacent fractures to be recorded segment by segment. As the process of reading adjacent fractures within a fracture group continues, the spatial distances between all adjacent fractures in each group gradually form a continuous fracture spacing sequence. Then, the fracture spacing sequence corresponding to each fracture group is arranged according to its spatial arrangement, enabling a continuous spatial distribution of fracture spacing within the rock mass evaluation area. Through the above measurement and arrangement process, each group of fractures within the rock mass evaluation area can generate a complete fracture spacing representation, thus providing a spatial distance data foundation for subsequent fracture spacing ratio statistics.

[0034] After obtaining the fracture spacing sequence corresponding to each fracture group, the fracture spacing within the rock mass evaluation area is statistically organized to achieve a unified expression of the fracture spacing ratio and further obtain information on the apparent fragmentation degree of the rock mass evaluation area. In the implementation process, firstly, the fracture spacing sequences of all fracture groups within the rock mass evaluation area are uniformly organized to form a complete statistical sequence of fracture spacing information corresponding to each fracture group. Then, based on the distribution of fracture spacing within the rock mass evaluation area, the fracture spacing sequence is statistically analyzed to achieve a continuous expression of the spatial distribution ratio of fracture spacing within the rock mass evaluation area. Next, the statistically obtained fracture spacing ratios are organized according to the spatial distribution relationship of fractures within the rock mass evaluation area to form a unified proportional expression structure within the rock mass evaluation area. With the completion of the fracture spacing ratio organization process, the density of the spatial distribution of fractures within the rock mass evaluation area can be expressed through the fracture spacing ratio, thereby obtaining information on the apparent fragmentation degree of the rock mass evaluation area and providing a basis for structural integrity evaluation in the subsequent comprehensive rock mass quality scoring process.

[0035] Combining the information on fracture length, fracture width, fracture spacing ratio, and apparent degree of fragmentation of the rock mass evaluation area, a comprehensive score is given to the rock mass evaluation area. Based on the score results, spatial correspondence is organized within the study area to form the rock mass quality zoning results. By uniformly organizing the information on fracture length, fracture width, fracture spacing ratio, and apparent degree of fragmentation within the rock mass evaluation area, the fracture structure characteristics within the evaluation area can be transformed into rock mass quality evaluation results. Furthermore, a continuous rock mass quality zoning representation is formed within the study area based on the spatial location relationships of the evaluation area. The specific implementation steps are as follows: The process involves systematically reading and spatially organizing the fracture length, width, spacing ratio, and apparent fracturing degree information already obtained within the rock mass evaluation area. This ensures that fracture structure information from different sources can form a complete information representation within the rock mass evaluation area. During implementation, all fracture length information within the rock mass evaluation area is read line by line, arranging the fracture length information of each fracture according to its spatial location within the rock mass evaluation area to form a continuous fracture length sequence. Subsequently, the fracture width information recorded within the rock mass evaluation area is read point by point, organizing the fracture width information corresponding to each fracture at different trajectory positions according to the extension order of the fracture trajectory to form a continuous fracture width representation sequence. Next, the fracture spacing ratio statistically obtained within the rock mass evaluation area is uniformly read, ensuring that the fracture spacing ratio reflects the spatial density of fractures within the rock mass evaluation area. Finally, the apparent fracturing degree information already obtained within the rock mass evaluation area is correlated with the fracture spacing ratio, establishing a correspondence between the apparent fracturing degree information and the fracture spacing ratio within the rock mass evaluation area. Through the above-mentioned step-by-step sorting process, the information on fracture length, fracture width, fracture spacing ratio, and apparent degree of fragmentation is integrated into a unified set of fracture structure information within the rock mass evaluation area, thereby enabling the fracture development characteristics within the rock mass evaluation area to be expressed in a multi-dimensional information form.

[0036] After completing the collection and organization of fracture structure information, the fracture structure information within the rock mass evaluation area is comprehensively read to generate a corresponding rock mass quality score. In the process, firstly, the fracture length sequence within the rock mass evaluation area is read, creating a continuous spatial representation of fracture extension within the evaluation area. Subsequently, the fracture width sequence is read point-by-point, creating a spatial representation of fracture opening degree within the evaluation area. Then, the fracture spacing ratio information and apparent fragmentation degree information are jointly read to form a complete representation of the spatial distribution of fractures within the evaluation area. As the process of reading fracture length, width, spacing ratio, and apparent fragmentation degree information continues, a comprehensive representation of the fracture development degree within the rock mass evaluation area is gradually formed. Subsequently, based on the fracture extension state reflected by fracture length information, the fracture opening state reflected by fracture width information, the fracture spatial distribution state reflected by fracture spacing ratio, and the rock mass structure integrity reflected by apparent fragmentation information, the fracture structure state within the rock mass evaluation area is comprehensively scored and organized. This allows the degree of fracture development within the rock mass evaluation area to be converted into a rock mass quality score result, and ensures that each rock mass evaluation area has a unique rock mass quality score expression.

[0037] After obtaining the rock mass quality scores for each rock mass evaluation area, the scores for all rock mass evaluation areas within the study area are spatially organized to form a continuous spatial representation of the rock mass quality scores according to their spatial location within the study area. In the process, firstly, the scores for each rock mass evaluation area are read region by region according to their arrangement within the study area, ensuring a correspondence between the score for each rock mass evaluation area and its spatial location. Then, the scores for all rock mass evaluation areas within the study area are uniformly organized according to their spatial arrangement, forming a continuous sequence of rock mass quality scores within the study area. Finally, the scores for each rock mass evaluation area are holistically correlated with its spatial location within the study area, ensuring that the score for each rock mass evaluation area accurately reflects the rock mass quality state at its corresponding spatial location. Through this process, the rock mass quality scores for all rock mass evaluation areas within the study area form a complete spatial representation structure, and the scores exhibit a continuous spatial distribution within the study area.

[0038] After spatial organization of the rock mass quality scoring, the scoring results corresponding to all rock mass evaluation areas within the study area are regionalized to form rock mass quality zoning results within the study area. In the implementation process, firstly, the continuously arranged rock mass quality scoring results within the study area are read sequentially, and the rock mass evaluation areas are classified and organized according to the spatial distribution of the scoring results within the study area, forming continuous distribution areas for rock mass evaluation areas with scores within the same scoring interval. Subsequently, rock mass evaluation areas within the same scoring interval are merged according to spatial adjacency, forming multiple rock mass quality regions within the study area. Then, each rock mass quality region is arranged as a whole according to its spatial location within the study area, forming a continuous spatial representation for each rock mass quality region within the study area. Through the above organization process, a complete rock mass quality zoning result is gradually formed within the study area, establishing a clear spatial distribution relationship between different rock mass quality regions within the study area. This completes the process of comprehensive scoring and rock mass quality zoning of rock mass evaluation areas, providing a spatial distribution basis for identifying rock mass quality differences and analyzing rock mass stability within the study area.

[0039] This invention performs a unified spatial scale division on rock mass images of the drawdown zone, and conducts pixel-by-pixel identification of fractures, extraction of fracture trajectories, and acquisition of fracture geometric information within the rock mass evaluation area. This enables fracture length, fracture width, fracture spacing ratio, and apparent fragmentation information to be fully expressed within the rock mass evaluation area and further transformed into rock mass quality scoring results. This achieves an automated processing procedure that directly generates rock mass quality evaluation results from rock mass image information, allowing rock mass fracture image information to directly serve the rock mass quality evaluation process, reducing the degree of manual intervention, and improving the automation level and processing efficiency of the rock mass quality evaluation process.

[0040] Based on the rock mass quality scoring results corresponding to each rock mass evaluation area, this invention unifies and organizes the spatial location relationships of the rock mass evaluation areas within the study area, enabling the rock mass quality scoring results to form a continuous spatial distribution expression within the study area. Furthermore, it forms rock mass quality zoning results, thereby achieving an intuitive expression of the spatial distribution of rock mass quality within the study area. This allows different rock mass quality areas to form a clear spatial correspondence within the study area, facilitating engineers to quickly identify areas with differences in rock mass quality and potentially weak areas, improving the efficiency of rock mass investigation in the reservoir drawdown zone and reducing the cost of manual investigation.

[0041] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A machine vision-based method for evaluating and zoning the rock mass quality of the drawdown zone in a reservoir area, characterized in that, Includes the following steps: Collect rock mass image information covering the drawdown zone of the study area, and divide the rock mass image information according to a unified spatial scale. Multiple rock mass evaluation areas with one-to-one spatial location are formed within the study area, and the correspondence between the rock mass evaluation areas and spatial location is established. The corresponding rock mass image information is read around the rock mass evaluation area. The distribution of fractures is identified pixel by pixel within the rock mass evaluation area. Fracture trajectory information is extracted from the rock mass evaluation area. Fracture morphology is formed by sorting out the fracture trajectory information. Based on the fracture trajectory information, the fracture morphology inside the rock mass evaluation area is continuously read. The spatial position relationship is read point by point along the fracture trajectory. The fracture extension distance is calculated based on the distance between adjacent positions. The opening distance between the two sides of the fracture is also counted to obtain the fracture length information and fracture width information corresponding to the rock mass evaluation area, forming a set of fracture geometric information. The spatial distribution relationship of fractures is read from the set of fracture geometric information, the fracture extension direction is sorted to form fracture grouping results, the spatial distance between adjacent fractures is measured according to the fracture grouping results, and the fracture spacing ratio is calculated according to the fracture spacing to obtain the corresponding apparent degree of fracture information of the rock mass evaluation area. By combining the information on fracture length, fracture width, fracture spacing ratio, and apparent degree of fragmentation of the rock mass evaluation area, a comprehensive score is given to the rock mass evaluation area. Based on the score results, spatial correspondence is organized within the study area to form the rock mass quality zoning results.

2. The method for evaluating and zoning the rock mass quality of the reservoir drawdown zone based on machine vision according to claim 1, characterized in that, The process of collecting image information of the drawdown zone rock mass covering the study area and forming the rock mass evaluation area is as follows: Continuous images of the rock mass surface in the drawdown zone are acquired along the spatial distribution direction of the study area to form a set of rock mass image information covering the study area. The rock mass image information is then organized according to the acquisition order to establish a correspondence between the rock mass image information and the spatial location within the study area. The rock mass image information is organized into a unified spatial scale based on the rock mass image information set, so that the spatial dimensions in the rock mass image information are expressed according to a unified scale, and the rock mass image information in the study area is expressed in the same spatial scale. Based on a unified spatial scale, rock mass image information is regularly divided to form multiple regularly arranged rock mass evaluation areas within the study area, and each rock mass evaluation area contains rock mass image information corresponding to its spatial location. The spatial locations of the rock mass evaluation areas within the study area are organized around the rock mass evaluation areas to establish a one-to-one correspondence between the rock mass evaluation areas and their spatial locations within the study area, forming a continuously arranged set of rock mass evaluation areas within the study area.

3. The machine vision-based quality evaluation and zoning method of the rock mass in the drawdown zone of the stockyard according to claim 2, characterized in that, Based on the acquisition order of rock mass image information within the study area, the rock mass evaluation areas are spatially arranged, and spatial correlation is carried out around the rock mass image information contained within the rock mass evaluation areas, so that the rock mass evaluation areas correspond to the spatial locations within the study area, and a continuous distribution structure of rock mass evaluation areas is formed within the study area.

4. The machine vision-based quality evaluation and zoning method of the rock mass in the drawdown zone of the stockyard according to claim 2, characterized in that, The steps for reading corresponding rock mass image information and forming fracture morphology around the rock mass evaluation area are as follows: The image information of the rock mass within the rock mass evaluation area is read around the rock mass evaluation area. The image content within the rock mass evaluation area is expanded pixel by pixel to form a set of pixel distribution within the rock mass evaluation area, and then a continuous pixel sequence is formed according to the spatial arrangement order. The pixel positions are read pixel by pixel around the pixel distribution set within the rock mass evaluation area. The fracture distribution is identified at the pixel positions to form a fracture pixel set. The fracture pixel set is then organized according to the spatial adjacency relationship to form a continuous fracture region. The fracture pixel set is read point by point according to the spatial arrangement order, the fracture pixel positions are connected and organized to form a fracture trajectory path, and a fracture trajectory information set is formed within the rock mass evaluation area. The information set of fracture trajectories is arranged and organized according to their spatial distribution, and the extension direction of the fracture trajectories is continuously organized to form the expression result of fracture morphology within the rock mass evaluation area.

5. The machine vision-based quality evaluation and zoning method of the rock mass in the drawdown zone of the stockyard according to claim 4, characterized in that, The steps for obtaining fracture length and width information based on fracture trajectory information and forming a set of fracture geometric information are as follows: The fracture trajectory information within the rock mass evaluation area is continuously read, and the trajectory points in the fracture trajectory are arranged sequentially according to the spatial extension direction of the fracture trajectory within the rock mass evaluation area. The spatial position corresponding to the trajectory point is read to form a continuous trajectory point set. The spatial relationship between adjacent trajectory points is read according to the order of the trajectory points, the spatial distance between adjacent trajectory points is recorded segment by segment, and the spatial distance is organized according to the trajectory extension order to form the crack length information. The positions of the crack boundaries on both sides of the trajectory points are read around the set of trajectory points. The spatial positional relationship between the trajectory points and the two sides of the crack boundaries is recorded. The spatial distance between the trajectory points and the two sides of the crack boundaries is arranged according to the order of the trajectory points to form crack width information. The fracture length and width information are combined and organized according to the correspondence of fracture trajectory to form a set of fracture geometric information within the rock mass evaluation area.

6. The machine vision-based quality evaluation and zoning method of the rock mass in the drawdown zone of the stockyard according to claim 5, characterized in that, The spatial distance between adjacent trajectory points is continuously recorded according to the spatial arrangement order of the trajectory points around the trajectory point set, and the crack length information is formed according to the trajectory extension order. At the same time, the positions of the crack boundaries on both sides of the trajectory points are read around the trajectory point set, and the spatial distance between the trajectory points and the two sides of the crack boundaries is sorted to form crack width information. The crack length information and crack width information are combined and sorted according to the crack trajectory correspondence to form the crack geometric information expression result.

7. The machine vision-based quality evaluation and zoning method of the rock mass in the drawdown zone of the stockyard according to claim 5, characterized in that, The steps to obtain apparent fracture degree information based on the set of fracture geometry information are as follows: Based on the geometric information set of fractures within the rock mass evaluation area, the spatial distribution of each fracture within the rock mass evaluation area is read, the spatial location of the fracture initiation, extension and termination positions is organized, and a fracture spatial distribution sequence is formed according to the spatial arrangement order. The fracture extension paths are read around the spatial distribution sequence of fractures, the fracture extension directions are continuously sorted, and the fractures are grouped according to the extension direction to form multiple fracture grouping results. Based on the fracture grouping results, the fractures within the same fracture group are spatially arranged and read, and the spatial positional relationship between adjacent fractures is measured and recorded to form a fracture spacing sequence; The fracture spacing sequence was statistically analyzed, and the spatial distribution ratio of fracture spacing within the rock mass evaluation area was statistically expressed to obtain information on the apparent degree of fragmentation in the rock mass evaluation area.

8. The machine vision-based method for evaluating and zoning the rock mass quality of the reservoir drawdown zone according to claim 7, characterized in that, The steps for generating rock mass quality zoning results around the rock mass evaluation area are as follows: Information on fracture length, fracture width, fracture spacing ratio, and apparent fragmentation degree is collected around the rock mass evaluation area. The fracture length, fracture width, fracture spacing ratio, and apparent fragmentation degree information are then organized according to the spatial location of the fractures within the rock mass evaluation area to form a fracture structure information set. Based on the fracture structure information set, the fracture length information, fracture width information, fracture spacing ratio and apparent degree of fragmentation information are comprehensively read, and the fracture structure state inside the rock mass evaluation area is comprehensively scored and sorted to form the corresponding rock mass quality score result of the rock mass evaluation area. The rock mass quality score results are read and organized according to the spatial arrangement of the rock mass evaluation area within the study area. The rock mass quality score results are correlated with the spatial location of the rock mass evaluation area within the study area to form a spatial distribution sequence of rock mass quality scores. The rock mass evaluation area is divided and organized according to the spatial distribution sequence of rock mass quality scores. Rock mass evaluation areas with scores within the same score range are spatially arranged to form the rock mass quality zoning results corresponding to the study area.

Images