Automatic interpretation method and device for stratigraphic horizon of geological outcrop image

Abstract

The present application relates to a geological outcrop stratum interpretation technical field, and is a geological outcrop image stratum horizon automatic interpretation method and device, comprising: obtaining an original image of a selected area and performing pretreatment; combining complex shear wave transformation to perform segmentation and edge extraction on the pretreated original image, and extracting stratum horizon line segments in the original image; deleting block-shaped pixel points in the stratum horizon line segments, and using a minimum connected cost function of each stratum horizon line segment to perform clustering analysis, and combining a length threshold of each line segment to perform denoising processing, and obtaining final stratum horizon line segments. The present application embodiment solves the problems of large manual interpretation workload and low efficiency, realizes automatic interpretation of geological outcrop image stratum horizons, increases the efficiency and accuracy of image edge extraction through the use of complex shear wave transformation, and further improves the correctness of stratum horizon interpretation through the application of a double denoising mode.

Description

Technical Field

[0001] This invention relates to the field of geological outcrop stratigraphic interpretation technology, and to an automatic interpretation method and apparatus for the stratigraphic position of geological outcrop images. Background Technology

[0002] Surface geological outcrops are the most direct means of studying the stratigraphic attribution of basin-covered areas and analyzing basin structure and evolution. Through the interpretation of geological outcrops, remote sensing image identification, delineation, and research on the morphology, boundaries, occurrence, distribution patterns, assemblage relationships, and genetic connections of geological structures can be achieved. This will be beneficial in promoting oil and gas exploration in deep, complex structures and lithologies within the basin. Stratigraphic interpretation of geological outcrops can be applied in three aspects: 1. Stratigraphic research: Through the interpretation of geological outcrops, the boundaries of bedding planes, lithology, fossils, mineral composition, chemical composition, physical properties, and sedimentary discontinuities between strata can be accurately extracted and delineated. Furthermore, by tracing the origins of strata within the outcrop area... 1. Source-tracing and comparative studies effectively solve difficult problems such as stratigraphic attribution of high-yield oil layers, reservoir modeling of stratigraphic structure, and restoration of prototype basins; 2. Tectonic studies clarify the folds, faults, cleavages, and other planar and linear structures of rocks in key areas, providing support for in-depth understanding of the deep geological structure of basins, basin-mountain coupling, and the relationship between tectonic-sedimentary evolution and hydrocarbon generation and accumulation; 3. Geological capping: constructing high-precision surface geological models through three-dimensional surface modeling and geological interpretation, and calibrating surface geological outcrops based on surface geological structural characteristics, combined with lithological profiles that integrate field measured outcrop profiles and seismic profiles.

[0003] Currently, there is relatively more research on fracture extraction from geological outcrop images, but less research on the automatic extraction of stratigraphic horizons from geological outcrops. Most studies utilize 3D geological outcrop stratigraphic interpretation methods, and the challenges of 3D geological outcrop stratigraphic interpretation are as follows:

[0004] 1. Three-dimensional geological outcrop stratigraphic interpretation is a completely new technical field, and there is no mature experience to draw upon both domestically and internationally. Unlike traditional DOM / DEM interpretation which uses orthophoto projection planes, the three-dimensional geological outcrop model uses 3Dtiler data volumes, which have high data accuracy and can realistically reflect the elevation and concave surfaces of the geological outcrop, making the distribution of strata clearer. Geological interpretation needs to be carried out in a three-dimensional environment.

[0005] 2. Manual interpretation of 3D geological outcrops suffers from problems such as high workload, poor operability, insufficient accuracy, and incompatibility between interpreted data and 3D models. Manual interpretation requires manual identification of stratigraphic boundaries and manual drawing of stratigraphic boundary segments. Due to the high precision of 3D geological outcrop models, with data accuracy down to the centimeter level and a data area of ​​nearly 100 square kilometers, drawing these segments in 3D space exceeds the limits of manual work. Furthermore, according to Long et al.'s research on the variability of stratigraphic interpretation, interpreting a single image by geologists of varying skill levels revealed significant differences in the topological structure, orientation, and length distribution of stratigraphic horizons.

[0006] 3. The conventional Beamlet algorithm has insufficient support for automatic extraction of stratigraphic horizons. The Beamlet algorithm is a high-dimensional anisotropy analysis method. Currently, when this algorithm is used for stratigraphic horizon interpretation, it suffers from problems such as discontinuous gray values ​​in edge images, isolated feature lines that cannot well reflect direction information, poor noise resistance, large computational load, and increased difficulty in operator learning, which are insufficient to support large-scale engineering applications. Summary of the Invention

[0007] This invention provides an automatic interpretation method and apparatus for stratigraphic positions of geological outcrop images, which overcomes the shortcomings of the prior art and can effectively solve the problems of low efficiency and large workload in the existing manual interpretation methods of stratigraphic positions of geological outcrop images.

[0008] One of the technical solutions of this invention is achieved through the following measures: an automatic interpretation method for stratigraphic horizons of geological outcrop images, comprising:

[0009] Acquire the original image of the selected region and perform preprocessing;

[0010] The preprocessed original image is segmented and edge-extracted by combining complex shear wave transform, and stratigraphic horizon segments are extracted from the original image.

[0011] Blocky pixels in the stratigraphic horizon segments are deleted, and clustering analysis is performed using the minimum connectivity cost function of each stratigraphic horizon segment. Denoising is then performed by combining the length threshold of each segment to obtain the final stratigraphic horizon segments.

[0012] The following are further optimizations and / or improvements to the above-mentioned technical solution:

[0013] The preprocessed original image is segmented and edge-extracted to extract stratigraphic horizon lines from the original image, including:

[0014] Set the range of values ​​for the parameters in the complex shear wave transform, and use the complex shear wave transform to perform coarse segmentation on the original image;

[0015] Edge extraction algorithms are used to extract the stratigraphic horizon lines from the coarsely segmented original image.

[0016] The dilation operation in digital morphology is used to connect adjacent stratigraphic segments within the same stratigraphic level.

[0017] The above-mentioned removal of blocky pixels from stratigraphic horizon segments, clustering analysis using the minimum connectivity cost function of each stratigraphic horizon segment, and denoising processing combined with a length threshold for each segment yield the final stratigraphic horizon segments, including:

[0018] The minimum bounding matrix and the minimum connectivity cost function value of each stratigraphic horizon segment are obtained using the connected component analysis algorithm.

[0019] Obtain the corresponding aspect ratio and area based on the minimum bounding matrix of each stratigraphic segment, and delete pixels that meet the first denoising condition, where the aspect ratio and area are both greater than a set threshold.

[0020] The stratigraphic horizon segments are classified by cluster analysis based on the minimum connectivity cost function value. Stratigraphic horizon segments that meet the second denoising condition are found and deleted to obtain the final stratigraphic horizon segments. The second denoising condition includes the minimum connectivity cost function value being greater than the threshold obtained by clustering and the segment length being less than the set length threshold.

[0021] The above-mentioned process of acquiring the original image of the selected region and performing preprocessing includes:

[0022] Get the original image of the selected region;

[0023] Perform color space conversion on the original image;

[0024] The Sobel and Laplace operators are used to perform inverse distance-weighted sharpening on the image.

[0025] The second technical solution of the present invention is achieved through the following measures: an automatic interpretation device for stratigraphic positions of geological outcrop images, comprising:

[0026] The preprocessing unit acquires the original image of the selected region and performs preprocessing.

[0027] The line segment extraction unit combines complex shear wave transform to segment and extract edges from the preprocessed original image, extracting stratigraphic line segments from the original image.

[0028] The denoising analysis unit removes blocky pixels from the stratigraphic horizon segments, performs clustering analysis using the minimum connectivity cost function of each stratigraphic horizon segment, and combines the length threshold of each segment for denoising processing to obtain the final stratigraphic horizon segments.

[0029] The following are further optimizations and / or improvements to the above-mentioned technical solution:

[0030] The above-mentioned denoising analysis unit includes:

[0031] The feature extraction module uses a connected component analysis algorithm to obtain the minimum bounding matrix and the minimum connectivity cost function value of each stratigraphic horizon segment.

[0032] The first denoising module obtains the corresponding aspect ratio and area based on the minimum bounding matrix of each stratigraphic segment, and deletes pixels that meet the first denoising condition, wherein the first denoising condition includes that both the aspect ratio and area are greater than a set threshold.

[0033] The second denoising module classifies stratigraphic segments by cluster analysis based on the minimum connectivity cost function value, finds and deletes stratigraphic segments that meet the second denoising condition, and obtains the final stratigraphic segments. The second denoising condition includes the minimum connectivity cost function value being greater than the threshold obtained by clustering and the segment length being less than the set length threshold.

[0034] The above-mentioned line segment extraction unit includes:

[0035] The segmentation module sets the range of parameters in the complex shear wave transform and uses the complex shear wave transform to perform coarse segmentation on the original image.

[0036] The extraction module uses an edge extraction algorithm to extract the edges of the coarsely segmented original image, extracting stratigraphic line segments from the original image; and uses the dilation operation in digital morphology to connect adjacent stratigraphic line segments of the same layer.

[0037] The embodiments of the present invention solve the problems of large workload and low efficiency of manual interpretation, realize the automatic interpretation of stratigraphic positions in geological outcrop images, increase the efficiency and accuracy of image edge extraction by utilizing complex shear wave transform, and further improve the accuracy of stratigraphic position interpretation by applying dual denoising methods. Attached Figure Description

[0038] Appendix Figure 1 This is a flowchart of the interpretation method of the present invention.

[0039] Appendix Figure 2 This is a flowchart of the method for extracting stratigraphic horizon segments from the original image according to the present invention.

[0040] Appendix Figure 3This is a flowchart of the method for denoising to obtain the final stratigraphic horizon line segment in this invention.

[0041] Appendix Figure 4 This is the original experimental image of Example 3 in this invention.

[0042] Appendix Figure 5 This is a schematic diagram of the inverse distance weighted sharpening result in Example 3 of the present invention.

[0043] Appendix Figure 6 This is a schematic diagram of the coarse segmentation result in Example 3 of the present invention.

[0044] Appendix Figure 7 This is a schematic diagram of the endpoint expansion result in Embodiment 3 of the present invention.

[0045] Appendix Figure 8 This is a schematic diagram of the connected component analysis results in Example 3 of the present invention.

[0046] Appendix Figure 9 This is a schematic diagram of the final result of Embodiment 3 of the present invention.

[0047] Appendix Figure 10 This is a schematic diagram of the device structure of the present invention. Detailed Implementation

[0048] The present invention is not limited to the following embodiments, and specific implementation methods can be determined according to the technical solutions and actual conditions of the present invention.

[0049] The present invention will be further described below with reference to embodiments and accompanying drawings:

[0050] Example 1: As shown in the attached document Figure 1 As shown in the figure, this invention discloses an automatic interpretation method for stratigraphic horizons in geological outcrop images, including:

[0051] Step S101: Obtain the original image of the selected region and perform preprocessing;

[0052] In this embodiment, a selected area is defined, and tile images are obtained from the corresponding three-dimensional geological outcrop images as the original images for interpretation. The original images are then preprocessed. The main purpose of image preprocessing is to eliminate irrelevant information in the images, restore useful real information, enhance the detectability of relevant information, and simplify the data to the maximum extent, thereby improving the reliability of feature extraction, image segmentation, matching, and recognition.

[0053] Step S102: Combine complex shear wave transform to segment and extract edges from the preprocessed original image, and extract stratigraphic horizon segments from the original image.

[0054] In this embodiment, the complex shear wave transform is used to coarsely segment the preprocessed original image into low-frequency and high-frequency components. The low-frequency components primarily represent the image's contour information, while the high-frequency components mainly represent its edges and texture information. Because the shear wave transform is multi-directional, it can obtain high-frequency and low-frequency information from noisy images in multiple directions, effectively capturing image details and making subsequent edge extraction more accurate. Compared to directly using edge detection algorithms in the spatial domain, the method of extracting candidate edges from the filtered image using gradients is more accurate. Furthermore, compared to the operators in spatial domain edge detection algorithms, the complex shear wave transform algorithm has lower computational cost and better noise resistance.

[0055] Step S103: Delete the blocky pixels in the stratigraphic horizon line segment, perform cluster analysis using the minimum connectivity cost function of each stratigraphic horizon line segment, and perform denoising processing in combination with the length threshold of each line segment to obtain the final stratigraphic horizon line segment.

[0056] In this embodiment, two different methods are used for secondary denoising processing, which further improves the accuracy of stratigraphic interpretation.

[0057] This invention discloses an automatic interpretation method for stratigraphic positions in geological outcrop images, which solves the problems of large workload and low efficiency of manual interpretation. It realizes automatic interpretation of stratigraphic positions in geological outcrop images. By utilizing complex shear wave transform, the efficiency and accuracy of image edge extraction are increased. At the same time, the application of dual denoising method of cluster analysis further improves the accuracy of stratigraphic position interpretation.

[0058] Example 2: As shown in the attached document Figure 2 , 3 As shown in the figure, this invention discloses an automatic interpretation method for stratigraphic horizons in geological outcrop images, including:

[0059] Step S201: Obtain the original image of the selected region and perform preprocessing;

[0060] Specifically, it includes:

[0061] (1) Obtain the original image of the selected region;

[0062] (2) Perform color space conversion on the original image; the color space conversion here can use the HSV color space to convert the RGB color space to the HSV color space;

[0063] (3) The image is sharpened by inverse distance weighting using the Sobel and Laplace operators. Here, image sharpening means increasing the difference between pixels in the neighborhood to make the abrupt changes in the image more obvious, thereby strengthening the edges and contours of the image. In this embodiment, the Sobel and Laplace operators can be used. The Laplace operator is to define a discrete formula of second-order differential, then construct a filter template based on the formula, and then convolve the template with the original image to achieve sharpening. The Sobel operator is the most commonly used edge detection operator. It is a discrete difference operator that calculates the approximate value of the gray level of the image brightness function.

[0064] Step S202: Combine complex shear wave transform to segment and extract edges from the preprocessed original image, and extract stratigraphic horizon segments from the original image;

[0065] As attached Figure 2 As shown, it specifically includes:

[0066] Step S2021: Set the range of values ​​for the parameters in the complex shear wave transform, and use the complex shear wave transform to perform coarse segmentation on the original image. The range of values ​​for the parameters in the complex shear wave transform is set according to the actual situation. By setting the range of values ​​for the parameters in the complex shear wave transform, a complex domain shear transform system is constructed. All combinations of the shear transform system are iterated to complete the coarse segmentation of the original image.

[0067] The parameters and standard values ​​of the above complex shear wave transform are shown in the table below:

[0068]

[0069] Step S2022: Use an edge extraction algorithm to extract the edges of the coarsely segmented original image and extract the stratigraphic horizon line segments in the original image;

[0070] Step S2023: Use the dilation operation in digital morphology to connect adjacent stratigraphic segments of the same stratum.

[0071] Due to noise, some stratigraphic layers may be fragmented during extraction, resulting in a single stratigraphic layer becoming several segments. In such cases, pixel connection operations are needed to reduce the number of segments within the same stratigraphic layer. Therefore, this embodiment utilizes dilation operations from digital morphology for image thinning. Specifically, dilation connects adjacent stratigraphic segments within the same layer. To improve connection efficiency and avoid disrupting the stratigraphic shape, dilation is only applied to the endpoints of the segments. The connectivity information of each pixel within a segment is used to determine if it is an endpoint; if so, dilation is performed. After all extracted stratigraphic layer pixels have been determined, a thinning algorithm is used to restore the linear shape.

[0072] Step S202: Delete the blocky pixels in the stratigraphic horizon line segment, perform cluster analysis using the minimum connectivity cost function of each stratigraphic horizon line segment, and perform denoising processing in combination with the length threshold of each line segment to obtain the final stratigraphic horizon line segment.

[0073] As attached Figure 3 As shown, it specifically includes:

[0074] Step S2031: Use the connected component analysis algorithm to obtain the minimum outer matrix of each stratigraphic horizon segment and the minimum connectivity cost function value of each stratigraphic horizon segment;

[0075] The connected component analysis algorithm in this image processing operation can use the Two-Pass scanning method or the Seed-Filling method to perform connected component analysis on the extracted pixels, calculate the outer matrix of the connected pixels, and calculate the minimum connected cost function value based on the image features.

[0076] Step S2032: Obtain the corresponding aspect ratio and area based on the minimum bounding matrix of each stratigraphic segment, and delete pixels that meet the first denoising condition, wherein the first denoising condition includes that both the aspect ratio and area are greater than a set threshold.

[0077] Since the pixel points that appear as blocks are not part of the geological strata, a first denoising condition is set in this embodiment. Pixel points that meet the threshold are pixel points that appear as blocks. Pixel points that meet the first denoising condition are then deleted, thus completing the first denoising.

[0078] The formula for calculating the aspect ratio of the minimum bounding rectangle is shown below:

[0079] T = D / W

[0080] Where D and W are the length and width of the smallest rectangle in the connected region, respectively;

[0081] The above connectivity cost function is shown below.

[0082] cost=|c1-c2|+0.25(|c1-start|+|c1-end|+|c2-start|+|c2-end|)

[0083] Where c1 represents the current pixel; c2 represents the neighboring pixels; start represents the starting pixel; and end represents the ending pixel.

[0084] Step S2033: Combine the minimum connectivity cost function value with cluster analysis to classify the stratigraphic horizon segments, find and delete stratigraphic horizon segments that meet the second denoising condition, and obtain the final stratigraphic horizon segments. The second denoising condition includes the minimum connectivity cost function value being greater than the threshold obtained by clustering and the segment length being less than the set length threshold.

[0085] Since all line segments are divided into two categories, this embodiment uses binary clustering to classify them. Here, the representative inter-class maximum variance method is used for feature clustering, taking the cost value of each stratigraphic segment as its feature value. Because the more similar the colors, the smaller the cost function value; and the larger the cost function value, the more likely it is a noise segment, and short features have no geological significance, this embodiment calculates the cost value between each pixel in each stratigraphic segment. The median, mean, and variance represent the cost value of the entire stratigraphic segment, respectively. Binary clustering is performed once using the median, variance, and mean. Based on the threshold obtained from the binary clustering, segments greater than the threshold are grouped into one category. Then, the length of each segment determines whether it is a noise segment; that is, a stratigraphic segment is considered a noise segment only when its cost value is greater than the clustering threshold and its length is less than a given length threshold, and thus the noise segment is deleted.

[0086] Example 3: Obtaining as shown in the attached document Figure 4 The original image to be interpreted shown is used to interpret the attached image according to the above embodiments. Figure 4 The interpretation includes:

[0087] 1. Preprocess the original image, including the image after inverse distance weighted sharpening, as shown in the attached image. Figure 5 As shown;

[0088] 2. The original image is coarsely segmented using complex shear wave transform, and the image after edge extraction is obtained by using an edge extraction algorithm is shown in the attached figure. Figure 6 As shown;

[0089] 3. After processing by connecting adjacent stratigraphic segments within the same stratigraphic level using the dilation operation in digital morphology, the image is shown in the attached figure. Figure 7 As shown;

[0090] 4. Using connected component analysis algorithms, blocky pixels in the stratigraphic line segments were removed. Cluster analysis was performed using the minimum connected component cost function of each linear feature, and the image after denoising was combined with the length threshold of each line segment is shown in the attached image. Figure 8 As shown;

[0091] 5. The final stratigraphic horizon line is shown in the attached figure. Figure 9 As shown, where I cpt =81.8%, I crt=85.2%, the red line represents manually extracted stratigraphic positions, and the yellow line represents automatically extracted stratigraphic positions.

[0092] Example 4: As shown in the appendix Figure 10 As shown in the figure, an embodiment of the present invention discloses an automatic interpretation device for stratigraphic horizons of geological outcrop images, comprising:

[0093] The preprocessing unit acquires the original image of the selected region and performs preprocessing.

[0094] The line segment extraction unit combines complex shear wave transform to segment and extract edges from the preprocessed original image, extracting stratigraphic line segments from the original image.

[0095] The denoising analysis unit removes blocky pixels from the stratigraphic horizon segments, performs clustering analysis using the minimum connectivity cost function of each stratigraphic horizon segment, and combines the length threshold of each segment for denoising processing to obtain the final stratigraphic horizon segments.

[0096] The noise analysis unit includes:

[0097] The feature extraction module uses a connected component analysis algorithm to obtain the minimum bounding matrix and the minimum connectivity cost function value of each stratigraphic horizon segment.

[0098] The first denoising module obtains the corresponding aspect ratio and area based on the minimum bounding matrix of each stratigraphic segment, and deletes pixels that meet the first denoising condition, wherein the first denoising condition includes that both the aspect ratio and area are greater than a set threshold.

[0099] The second denoising module classifies stratigraphic segments by cluster analysis based on the minimum connectivity cost function value, finds and deletes stratigraphic segments that meet the second denoising condition, and obtains the final stratigraphic segments. The second denoising condition includes the minimum connectivity cost function value being greater than the threshold obtained by clustering and the segment length being less than the set length threshold.

[0100] The line segment extraction unit includes:

[0101] The segmentation module sets the range of parameters in the complex shear wave transform and uses the complex shear wave transform to perform coarse segmentation on the original image.

[0102] The extraction module uses an edge extraction algorithm to extract the edges of the coarsely segmented original image, extracting stratigraphic line segments from the original image; and uses the dilation operation in digital morphology to connect adjacent stratigraphic line segments of the same layer.

[0103] Example 5: This embodiment of the invention discloses a storage medium storing a computer program that can be read by a computer. The computer program is configured to execute an automatic interpretation method of stratigraphic positions of geological outcrop images during runtime.

[0104] The aforementioned storage media may include, but are not limited to, USB flash drives, read-only memory, portable hard drives, magnetic disks, optical disks, and other media capable of storing computer programs.

[0105] Example 6: This embodiment of the invention discloses an electronic device, including a processor and a memory. The memory stores a computer program, which is loaded and executed by the processor to realize an automatic interpretation method of stratigraphic horizons of geological outcrop images.

[0106] The processor described above can be a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an ASIC, an FPGA, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. It can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc. The memory can include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, portable hard drives, magnetic disks, or optical disks.

[0107] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of this application can be implemented in various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0108] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0109] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0110] The above technical features constitute the preferred embodiment of the present invention, which has strong adaptability and optimal implementation effect. Unnecessary technical features can be added or removed according to actual needs to meet the requirements of different situations.

Claims

1. A method for automatically interpreting stratigraphic horizons in geological outcrop images, characterized in that, include: Acquire the original image of the selected region and perform preprocessing; The preprocessed original image is segmented and edge extracted by combining complex shear wave transform, and stratigraphic horizon segments are extracted from the original image. Blocky pixels in the stratigraphic horizon segments are removed, and clustering analysis is performed using the minimum connectivity cost function for each stratigraphic horizon segment. Denoising is then performed using a length threshold for each segment to obtain the final stratigraphic horizon segments, including: The minimum bounding rectangle of each stratigraphic horizon segment and the minimum connectivity cost function value of each stratigraphic horizon segment are obtained using the connected component analysis algorithm. Obtain the corresponding aspect ratio and area based on the minimum bounding rectangle of each stratigraphic segment, and delete pixels that meet the first denoising condition, where the aspect ratio and area are both greater than a set threshold. The stratigraphic horizon segments are classified by cluster analysis based on the minimum connectivity cost function value. Stratigraphic horizon segments that meet the second denoising condition are found and deleted to obtain the final stratigraphic horizon segments. The second denoising condition includes the minimum connectivity cost function value being greater than the threshold obtained by clustering and the segment length being less than the set length threshold.

2. The automatic interpretation method for stratigraphic horizons of geological outcrop images according to claim 1, characterized in that, The preprocessed original image is segmented and edge-extracted to extract stratigraphic horizon lines from the original image, including: Set the range of values ​​for the parameters in the complex shear wave transform, and use the complex shear wave transform to perform coarse segmentation on the original image; Edge extraction algorithms are used to extract the stratigraphic horizon lines from the coarsely segmented original image. The dilation operation in digital morphology is used to connect adjacent stratigraphic segments within the same stratigraphic level.

3. The method for automatic interpretation of stratigraphic horizons in geological outcrop images according to claim 1 or 2, characterized in that, The process of acquiring the original image of the selected region and performing preprocessing includes: Get the original image of the selected region; Perform color space conversion on the original image; The Sobel and Laplace operators are used to perform inverse distance-weighted sharpening on the image.

4. An automatic interpretation device for the stratigraphic horizon of a geological outcrop image using the method described in any one of claims 1 to 3, characterized in that, include: The preprocessing unit acquires the original image of the selected region and performs preprocessing. The line segment extraction unit combines complex shear wave transform to segment and extract edges from the preprocessed original image, extracting stratigraphic line segments from the original image. The denoising analysis unit removes blocky pixels from the stratigraphic horizon segments, performs clustering analysis using the minimum connectivity cost function for each stratigraphic horizon segment, and combines this with a length threshold for each segment to perform denoising processing, resulting in the final stratigraphic horizon segments, including: The feature extraction module uses a connected component analysis algorithm to obtain the minimum bounding rectangle of each stratigraphic horizon segment and the minimum connectivity cost function value of each stratigraphic horizon segment; The first denoising module obtains the corresponding aspect ratio and area based on the minimum bounding rectangle of each stratigraphic segment, and deletes pixels that meet the first denoising condition, wherein the first denoising condition includes that both the aspect ratio and area are greater than a set threshold. The second denoising module classifies stratigraphic segments by cluster analysis based on the minimum connectivity cost function value, finds and deletes stratigraphic segments that meet the second denoising condition, and obtains the final stratigraphic segments. The second denoising condition includes the minimum connectivity cost function value being greater than the threshold obtained by clustering and the segment length being less than the set length threshold.

5. The automatic interpretation method for stratigraphic horizons of geological outcrop images according to claim 4, characterized in that, The line segment extraction unit includes: The segmentation module sets the range of parameters in the complex shear wave transform and uses the complex shear wave transform to perform coarse segmentation on the original image. The extraction module uses an edge extraction algorithm to extract the edges of the coarsely segmented original image, extracting stratigraphic line segments from the original image; and uses the dilation operation in digital morphology to connect adjacent stratigraphic line segments of the same layer.

6. A storage medium, characterized in that, The storage medium stores a computer program that can be read by a computer, and the computer program is configured to execute the automatic interpretation method of the stratigraphic position of the geological outcrop image as described in any one of claims 1 to 3 when it is run.

7. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program, which is loaded and executed by the processor to implement the automatic interpretation method of stratigraphic horizons of geological outcrop images as described in any one of claims 1 to 3.

Images