Digital cataloging and duplicate checking method for exploration core based on panoramic image processing

Abstract

The present application relates to the technical field of geotechnical engineering, and in particular to a method for digital cataloging and duplicate checking of survey cores based on panoramic image processing. The present application first obtains a panoramic image of the drill core and preprocesses it, then segments the preprocessed panoramic image to obtain a series of sub-images, and generates corresponding image feature vectors and digital cataloging data based on these sub-images; then compares the extracted image feature vector with a preset image database, calculates the similarity, and sets a reasonable threshold according to different comparison scenarios to determine whether there is an abnormality; finally, based on the digital cataloging data, the similarity calculation results, and the abnormality determination results, a standardized cataloging report is automatically generated, which has a high degree of automation and can greatly improve work efficiency; reduce human errors; and introduce a comparison and duplicate checking link to avoid waste of resources; finally, a standardized report is generated with a high level of informatization.

Description

Technical Field

[0001] The present application relates to the technical field of geotechnical engineering, and in particular to a method for digital cataloging and duplicate checking of survey cores based on panoramic image processing. Background Art

[0002] At present, exploration and mining activities are of vital importance to the fields of energy and mineral resources. The core samples obtained from exploration drilling are an important basis for studying geological structures, ore body enrichment rules and mining conditions. Traditionally, visual interpretation and manual recording of drill cores is a time-consuming and laborious task with many shortcomings such as low efficiency and large human errors.

[0003] In order to improve the efficiency of exploration work, some automated core digital cataloging systems have emerged in the existing technology. They use camera equipment to take panoramic photos of the drill core to obtain high-resolution images, and then use computer vision and image processing technology to analyze the images, automatically identify the lithology, structure and other characteristics of the core, and generate structured digital cataloging data, which greatly improves work efficiency. In actual exploration projects, different exploration projects may explore the same geographical area or target ore body, or due to human error, repeated exploration work may be arranged. Due to the lack of effective information sharing and data comparison methods, it is easy to lead to repeated exploration of the explored area. However, the existing technology lacks the means to discover repeated exploration, resulting in repeated waste of resources. This situation needs to be further improved. Summary of the invention

[0004] In order to solve the problem that the existing core digital cataloging system lacks means to discover duplicate exploration, the present application provides a method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing, which adopts the following technical solutions:

[0005] In a first aspect, the present application provides a method for digitally cataloging and checking duplicates of exploration cores based on panoramic image processing, comprising the following steps:

[0006] Obtain panoramic images of drill cores and perform preprocessing;

[0007] Segmenting the preprocessed panoramic image to obtain a series of sub-images and generating image feature vectors and digital catalog data corresponding to the series of sub-images;

[0008] Comparing the image feature vector with a preset image database, calculating the similarity and determining whether there is an anomaly according to preset thresholds of different comparison scenarios, wherein the comparison scenarios include comparison between different boreholes in the same project and comparison between images in the current project and historical project databases;

[0009] A catalog report is generated based on the digital catalog data, similarity and anomaly determination results.

[0010] By adopting the above-mentioned technical scheme, the present application first obtains a panoramic image of the drill core and preprocesses it, then segments the preprocessed panoramic image to obtain a series of sub-images, and generates corresponding image feature vectors and digital cataloging data based on these sub-images; then compares the extracted image feature vector with a preset image database, calculates the similarity, and sets reasonable thresholds according to different comparison scenarios to determine whether there are any abnormalities; finally, based on the digital cataloging data, the similarity calculation results and the abnormality determination results, a standardized cataloging report is automatically generated, which has a high degree of automation and can greatly improve work efficiency; reduce human errors; and introduce a comparison and duplication checking link to avoid waste of resources; finally, a standardized report is generated with a high level of informatization.

[0011] Optionally, segmenting the preprocessed panoramic image to obtain a series of sub-images and generating image feature vectors and digital catalog data corresponding to the series of sub-images specifically includes the following steps:

[0012] Identify the depth scale in the panoramic image, extract the starting depth and ending depth of the core section corresponding to the panoramic image, and obtain depth information;

[0013] Use edge detection algorithms to identify the upper and lower boundaries of the core and determine the effective core area;

[0014] Based on the color, texture changes and visible geological interfaces of the core, the image segmentation algorithm is used to divide the core into different sections to obtain a series of sub-images;

[0015] Calculating the length of each sub-image in the series of sub-images, and judging whether it is a complete core segment according to a preset threshold, and determining the RQD value;

[0016] The color features, texture features and structural features of each sub-image are extracted, and the depth information and the RQD value are combined to generate a fixed-length image feature vector and structured digital catalog data.

[0017] By adopting the above technical solution, the present application first identifies the depth scale, effective core area and boundary of the core in the panoramic image based on image processing technology, so as to obtain the depth information of the core section and the rock quality index RQD value; then uses the image segmentation algorithm to divide the core into a series of sub-images according to the color, texture changes and visible geological interfaces of the core; then extracts the color, texture and structure characteristics of each sub-image respectively, and generates an image feature vector of fixed length in combination with the depth information and RQD value, and generates structured digital cataloging data at the same time, which can accurately obtain the spatial position and RQD rock quality information of the core, and the generated feature vector is convenient for subsequent comparison and analysis applications; the encoding process has a high degree of automation, which improves efficiency and consistency.

[0018] Optionally, the image feature vector is compared with a preset image database, similarity is calculated, and whether an abnormality exists is determined according to preset thresholds of different comparison scenarios, specifically including the following steps:

[0019] Extracting a reference image feature vector from a preset image database;

[0020] Calculating the distance between the image feature vector and each reference image feature vector as similarity;

[0021] Compare the calculated similarity with a preset threshold of the corresponding comparison scene to determine whether there is an anomaly;

[0022] When an exception exists, the exception is marked and the exception information is output.

[0023] By adopting the above technical solution, the present application compares the extracted core image feature vector with the reference feature vector in the preset image database, and calculates the distance between the two as a similarity evaluation index; then sets a reasonable similarity threshold according to different comparison scenarios, such as comparison between different boreholes in the same project, cross-project comparison, etc., and compares the calculated similarity with the threshold to determine whether there is an abnormality, thereby improving the accuracy of anomaly determination; once an anomaly is found, the anomaly can be marked and the anomaly information can be output to remind relevant personnel to pay attention to the risk of repeated exploration.

[0024] Optionally, generating a catalog report based on the digital catalog data, similarity and anomaly determination results specifically includes the following steps:

[0025] Associating the digital catalog data with the abnormality determination result to generate a catalog record;

[0026] Classifying and sorting the catalog records according to the similarity;

[0027] Generate a catalog report based on the sorted catalog records in a preset format.

[0028] By adopting the above technical solution, the present application first associates the digital cataloging data with the abnormality determination results to generate a complete cataloging record; then classifies and sorts the cataloging records according to the similarity between the cataloging records and the reference data; finally, according to the preset format requirements, the sorted cataloging records are used to generate a standardized cataloging report, which helps to quickly find and discover problems, and is convenient for manual review and analysis and decision-making, with better readability and practicality.

[0029] Optionally, the method further comprises the following steps:

[0030] Detecting whether there is a target sub-image with an incomplete image in the series of sub-images;

[0031] For a detected target sub-image with an incomplete image, an image interpolation algorithm is used to repair the target sub-image to generate a repaired sub-image;

[0032] The incomplete sub-image of the image is replaced by the restored sub-image, and the corresponding image feature vector and structured digital catalog data are regenerated according to the restored sub-image.

[0033] By adopting the above technical solution, after segmenting the panoramic image to obtain a sub-image sequence, the present application detects and identifies target sub-images with incomplete images; then, for these incomplete sub-images, image interpolation and other algorithms are used to repair them to generate complete repaired sub-images; finally, the repaired sub-images replace the original incomplete sub-images, and features are re-extracted and encoding data is generated based on the repaired images, which can effectively avoid the impact of missing information and ensure the integrity and accuracy of the encoding results.

[0034] Optionally, the target sub-image is repaired using an image interpolation algorithm, which specifically includes the following steps:

[0035] Searching the preset image database for a complete reference core image that is most similar to the target sub-image;

[0036] Extracting a corresponding complete sub-image region in the reference core image;

[0037] Performing image registration on the extracted complete sub-image area and the target sub-image;

[0038] Detect whether the incomplete area of ​​the image contains key features that divide the core sections;

[0039] If the key features are not included, the incomplete area of ​​the image is interpolated and repaired, otherwise the corresponding complete area is directly copied from the reference core image.

[0040] By adopting the above technical solution, the present application first searches for a complete reference core image that is most similar to the target incomplete sub-image from a preset image database; then extracts the corresponding complete sub-image area from the reference image, and accurately aligns it with the target sub-image; then detects whether the incomplete area contains the key features that divide the core sections, and if not, interpolates and repairs the area; if it contains the key features, directly copies the corresponding complete area from the reference image, which can retain the key geological features to the greatest extent and avoid feature loss or distortion; uses the most similar reference image for repair, and the repair result is more realistic and reliable, significantly improving the quality and reliability of exploration data.

[0041] Optionally, directly copying the corresponding complete area from the reference core image specifically includes the following steps:

[0042] Analyzing context information of the incomplete area of ​​the image to locate key feature areas for dividing core sections;

[0043] Searching the reference core image for a corresponding region that best matches the key feature region;

[0044] directly copying the matched corresponding area to the incomplete area of ​​the image as a repair result;

[0045] An interpolation algorithm is used to perform transition processing on the target sub-image so that the copy area is smoothly connected.

[0046] By adopting the above technical solution, the present application first analyzes the incomplete area of ​​the target sub-image and locates the area containing the key geological features; then searches the reference core image for the corresponding area that best matches the feature area; directly copies the matched area to the missing part of the target image as the result of the repair; finally, uses an interpolation algorithm to perform transition processing on the edge of the copied area so that it is smoothly connected to the surrounding image, which can maximize the retention of key geological features and facilitate subsequent segmentation and analysis.

[0047] In a second aspect, the present application provides a digital cataloging and duplication checking system for exploration cores based on panoramic image processing, comprising:

[0048] A panoramic image acquisition module is used to acquire a panoramic image of the drill core, perform image preprocessing and feature extraction based on the panoramic image, and obtain digital catalog data of the core;

[0049] A panoramic image segmentation module, used to segment the panoramic image based on a preset depth interval to obtain a series of sub-images and generate image feature vectors corresponding to the series of sub-images;

[0050] An image feature comparison module is used to compare the image feature vector with a preset image database, calculate the similarity and determine whether there is an abnormality according to preset thresholds of different comparison scenarios, wherein the comparison scenarios include comparison between different boreholes in the same project and comparison between images in the current project and historical project databases;

[0051] The catalog report generation module is used to generate a catalog report based on the digital catalog data, similarity and abnormality determination results.

[0052] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the above-mentioned method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing are implemented.

[0053] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned method for digital cataloging and duplicate checking of survey cores based on panoramic image processing.

[0054] In summary, the present application includes at least one of the following beneficial technical effects:

[0055] 1. The present application first obtains a panoramic image of the drill core and preprocesses it, then segments the preprocessed panoramic image to obtain a series of sub-images, and generates corresponding image feature vectors and digital cataloging data based on these sub-images; then compares the extracted image feature vector with a preset image database, calculates the similarity, and sets a reasonable threshold according to different comparison scenarios to determine whether there is an abnormality; finally, based on the digital cataloging data, the similarity calculation results and the abnormality determination results, a standardized cataloging report is automatically generated, which has a high degree of automation and can greatly improve work efficiency; reduce human errors; and introduce a comparison and duplication check link to avoid waste of resources; finally, a standardized report is generated with a high level of informatization;

[0056] 2. This application first identifies the depth scale, effective core area and core boundary in the panoramic image based on image processing technology, so as to obtain the depth information of the core section and the rock quality index RQD value; then uses the image segmentation algorithm to divide the core into a series of sub-images according to the color, texture changes and visible geological interfaces of the core; then extracts the color, texture and structure characteristics of each sub-image respectively, and generates a fixed-length image feature vector in combination with the depth information and RQD value, and generates structured digital cataloging data at the same time, which can accurately obtain the spatial position and RQD rock quality information of the core, and the generated feature vector is convenient for subsequent comparison and analysis applications; the encoding process has a high degree of automation, which improves efficiency and consistency;

[0057] 3. After segmenting the panoramic image to obtain a sub-image sequence, the present application detects and identifies target sub-images with incomplete images; then, for these incomplete sub-images, image interpolation and other algorithms are used to repair them to generate complete repaired sub-images; finally, the repaired sub-images replace the original incomplete sub-images, and features are re-extracted and encoding data is generated based on the repaired images, which can effectively avoid the impact of missing information and ensure the integrity and accuracy of the encoding results. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a flow chart of a method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing according to an embodiment of the present application;

[0059] Figure 2 It is a flowchart of step S120 in a method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing according to an embodiment of the present application;

[0060] Figure 3 It is a flowchart of step S130 in a method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing according to an embodiment of the present application;

[0061] Figure 4 It is a flowchart of step S140 in a method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing according to an embodiment of the present application;

[0062] Figure 5 It is a schematic diagram of a process of repairing an image in a method for digital cataloging and duplicate checking of survey cores based on panoramic image processing according to an embodiment of the present application;

[0063] Figure 6 It is a flowchart of step S220 in a method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing according to an embodiment of the present application;

[0064] Figure 7 It is a flowchart of step S225 in a method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing according to an embodiment of the present application;

[0065] Figure 8 This is a module schematic diagram of a digital cataloging and duplicate checking system for exploration cores based on panoramic image processing according to an embodiment of the present application;

[0066] Fig. 9 It is a diagram of the internal structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0067] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification and appended claims of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more listed items.

[0068] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0069] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0070] In the first aspect, the present application provides a method for digitally cataloging and checking duplicates of exploration cores based on panoramic image processing, referring to Figure 1 , including the following steps:

[0071] S110, obtaining a panoramic image of the drill core and performing preprocessing.

[0072] In this embodiment, a panoramic camera is used to capture the entire borehole core and stitch them together to generate a panoramic image, or a common camera is used to capture the core at multiple angles and stitch multiple local images together to generate a panoramic image.

[0073] Specifically, the preprocessing of the acquired panoramic image mainly includes: image denoising, contrast enhancement, deformation correction, etc., to improve the quality and effect of subsequent image processing.

[0074] S120 , segmenting the preprocessed panoramic image to obtain a series of sub-images and generate image feature vectors and digital catalog data corresponding to the series of sub-images.

[0075] In this embodiment, the panoramic image is segmented using a segmentation algorithm based on low-level visual features such as color and texture.

[0076] Specifically, after obtaining the sub-image sequence, the color histogram and texture energy features of each sub-image are extracted and spliced ​​to form an image feature vector of fixed length; at the same time, the RQD value is calculated according to the length of the sub-image, and combined with the depth information to generate structured digital catalog data.

[0077] S130, comparing the image feature vector with a preset image database, calculating the similarity and determining whether an abnormality exists according to preset thresholds of different comparison scenarios.

[0078] Among them, the comparison scenarios include the comparison between different boreholes in the same project and the comparison between the images in the current project and the historical project database.

[0079] In this embodiment, the calculation method of image similarity can be based on Euclidean distance, cosine similarity, etc.

[0080] Specifically, the distance between the current image feature vector and each reference vector in the database is calculated as the similarity, and then different similarity thresholds are set according to the two different scenarios of comparison within the same project or comparison across projects. If the similarity exceeds the threshold, it is determined that there is an anomaly. For example, the threshold can be set to 0.95 within the same project and 0.9 across projects, etc. This avoids the waste of resources caused by repeated exploration work.

[0081] S140: Generate a catalog report based on the digital catalog data, similarity and abnormality determination results.

[0082] Specifically, the catalog data is associated with the anomaly determination results to form a complete record, and then the records are sorted and classified according to the degree of similarity. Finally, the sorted records are written into a report file according to a preset template, and necessary explanatory information is added to generate an exploration catalog report.

[0083] In one embodiment, referring to Figure 2 In step S120, the pre-processed panoramic image is segmented to obtain a series of sub-images and generate image feature vectors and digital catalog data corresponding to the series of sub-images, which specifically includes the following steps:

[0084] S121, identifying the depth scale in the panoramic image, extracting the starting depth and ending depth of the core section corresponding to the panoramic image, and obtaining depth information.

[0085] In this embodiment, a method for identifying a depth scale is based on template matching.

[0086] Specifically, template images of various depth scales are collected in advance, and then sliding matching is performed in the panoramic image to find the best matching position, which is the scale position. Once the scale position is identified, the start and end depth values ​​of the corresponding core section are extracted according to the scale scale.

[0087] S122. Use edge detection algorithm to identify the upper and lower boundaries of the core and determine the effective core area.

[0088] In this embodiment, the image noise is first removed, the gradient amplitude and direction of each pixel are calculated, and then non-maximum suppression and double threshold detection are performed to obtain the edge, so as to detect the positions of the upper and lower boundaries of the core, and then determine the effective core area range.

[0089] S123. Based on the color, texture changes and visible geological interfaces of the core, the core is divided into different sections using an image segmentation algorithm to obtain a series of sub-images.

[0090] Specifically, first, some seed points are initialized according to the mutation of color and texture and the edge detection results of geological interface; then, starting from each seed point, a connected area is gradually grown with reference to the similarity of adjacent pixels in color and texture until it can no longer be expanded, and a segmentation mask is obtained; this process is repeated until all seed points are processed, and finally the union of all masks is the segmentation result.

[0091] S124, calculating the length of each sub-image in the series of sub-images, and judging whether it is a complete core segment according to a preset threshold, and determining the RQD value.

[0092] In this embodiment, the unit for calculating the length of the sub-image may be pixels, or may be an actual length unit such as meters or feet converted from a known scale.

[0093] Specifically, for pixel units, the number of pixel rows of the sub-image can be directly counted; for actual length units, the pixel value needs to be converted into the actual length value according to the scale. Once the actual length of the sub-image is obtained, it can be compared with the preset complete core segment length threshold of 10 meters. If it exceeds the threshold, it is judged as a complete segment, otherwise it is an incomplete segment. Finally, the length of the complete segment is divided by the total length of the drilled core to calculate the RQD value.

[0094] S125 , extracting the color features, texture features and structural features of each sub-image, combining the depth information and the RQD value, and generating a fixed-length image feature vector and structured digital catalog data.

[0095] In this embodiment, for each sub-image, the color features, texture features and structural features are calculated, and then the features are spliced ​​to form a feature vector of fixed length; at the same time, the depth value, length, RQD value, etc. corresponding to the sub-image are woven into structured digital catalog data.

[0096] In one embodiment, referring to Figure 3 In step S130, the image feature vector is compared with the preset image database, the similarity is calculated, and whether there is an abnormality is determined according to the preset thresholds of different comparison scenarios, which specifically includes the following steps:

[0097] S131. Extracting a reference image feature vector from a preset image database.

[0098] In this embodiment, the preset image database includes an image feature library based on drilling within a project and an overall feature library across projects.

[0099] Specifically, for the intra-project feature library, feature vectors of all borehole core images are extracted and indexed at the initial stage of the project; for the cross-project feature library, new feature samples are regularly mined from historical projects to update and expand the feature set in the library. When extracting reference feature vectors, the corresponding feature set is located according to the metadata of the current image (such as project ID, borehole number, etc.).

[0100] S132: Calculate the distance between the image feature vector and each reference image feature vector as the similarity.

[0101] In this embodiment, the similarity is calculated based on the Euclidean distance. Assuming that the image feature vector is x and the reference feature vector is y, the Euclidean distance between the image feature vector and each reference image feature vector is calculated. The Euclidean distance is used to evaluate the difference between the current image and each reference image in the database.

[0102] S133: Compare the calculated similarity with a preset threshold of the corresponding comparison scene to determine whether there is an abnormality.

[0103] In this embodiment, the comparison scenarios are divided into two situations: comparison within the same project and comparison across projects, and the abnormality determination threshold for each situation may be different.

[0104] Specifically, for comparison within the same project, since the sources are relatively consistent, the anomaly threshold is set to a higher value, such as 0.95. Only when the similarity between the current image and the reference image is higher than 0.95 is it considered an anomaly; for comparison across projects, since the sources are quite different, the anomaly threshold is lowered, such as 0.9. If the similarity is higher than this value, it is considered an anomaly. The settings of these thresholds need to be adjusted according to the actual data distribution.

[0105] S134. When an exception occurs, mark the exception and output the exception information.

[0106] In one embodiment, referring to Figure 4 In step S140, a catalog report is generated based on the digital catalog data, similarity and abnormality determination results, which specifically includes the following steps:

[0107] S141. Associate the digital catalog data with the abnormality determination result to generate a catalog record.

[0108] In this embodiment, the catalog record is a structured data format. By taking the previously generated digital catalog data as a basis, a new abnormality mark field is added to ensure the correct association between the abnormality determination result and the corresponding catalog data.

[0109] S142. Classify and sort the catalog records according to similarity.

[0110] Specifically, the records are divided into three lists according to the magnitude relationship between the similarity value and two preset thresholds. Within each list, the records are further sorted in descending order according to the similarity value.

[0111] S143: Generate a catalog report based on the sorted catalog records in a preset format.

[0112] After sorting, records with similar abnormal characteristics can be brought together. By placing highly similar catalog records in the front, manual inspection can focus on these records that are highly similar to known database samples to determine whether they are repeated explorations. For records with low similarity and placed at the back, they need to be reviewed in detail to determine whether they are new abnormal situations, making the manual review process more efficient and orderly.

[0113] In one embodiment, referring to Figure 5 , the method further comprises the following steps:

[0114] S210 , detecting whether there is an incomplete target sub-image in the series of sub-images.

[0115] In this embodiment, the incomplete image mainly refers to the missing core image information due to shooting angle, obstruction or probe movement.

[0116] Specifically, the presence of incompleteness is determined by detecting the gradient change in the edge area of ​​the sub-image. If the gradient value in the edge area is abnormally high and the direction changes suddenly, it is likely caused by image incompleteness.

[0117] S220 , for the detected target sub-image with incomplete image, use an image interpolation algorithm to repair the target sub-image to generate a repaired sub-image.

[0118] In this embodiment, the complete area is first subjected to wavelet transform to obtain low-frequency and high-frequency components, and then the high-frequency components of the missing area are estimated based on known edges and neighboring pixel values, and finally the repaired image is reconstructed.

[0119] S230: replacing the incomplete sub-image with the restored sub-image, and regenerating the corresponding image feature vector and structured digital catalog data according to the restored sub-image.

[0120] In this embodiment, for the restored image, the RGB histogram, LBP histogram and HOG feature vector are calculated in sequence and then spliced ​​into a new feature vector. Then, the image ID and other metadata fields corresponding to the restored image are updated to regenerate the digital catalog data.

[0121] In one embodiment, referring to Figure 6 In step S220, the target sub-image is repaired based on the image interpolation algorithm, which specifically includes the following steps:

[0122] S221. Searching for a complete reference core image that is most similar to the target sub-image in a preset image database.

[0123] Specifically, for the target sub-image, a set of images with the same metadata information such as borehole and drilling section ID is preliminarily screened out from the preset image database, and then the feature vector distance between each complete image and the target image is traversed and calculated in the set, and the one with the smallest distance is selected as the most similar reference image.

[0124] S222: Extract the corresponding complete sub-image area in the reference core image.

[0125] In this embodiment, the complete sub-image area is extracted in combination with the image coordinates of the target sub-image.

[0126] Specifically, the reference core image is first divided into sub-image blocks of equal length, and then the corresponding sub-image block index is located according to the start and end coordinates of the target sub-image to obtain the required complete sub-image area.

[0127] S223: Perform image registration between the extracted complete sub-image area and the target sub-image.

[0128] Specifically, feature points such as SIFT or ORB are detected on the complete sub-image area and the target sub-image respectively, and the complete sub-image area is remapped to the viewing angle coordinate system of the target sub-image.

[0129] S224, detecting whether the incomplete area of ​​the image contains key features for dividing core sections.

[0130] In this embodiment, the key features may be some obvious bedding planes, sudden changes in rock color or texture, etc.

[0131] Specifically, the present application pre-builds a semantic segmentation model, labels the pixel areas corresponding to the key features as specific categories, and then determines whether the incomplete area of ​​the target sub-image contains pixels of these categories. If so, it means that the area may be the interface of important geological bodies such as faults or bedding.

[0132] S225. If the key features are not included, the incomplete area of ​​the image is interpolated and repaired, otherwise the corresponding complete area is directly copied from the reference core image.

[0133] In one embodiment, referring to Figure 7 In step S225, directly copying the corresponding complete area from the reference core image specifically includes the following steps:

[0134] S2251. Analyze the context information of the incomplete area of ​​the image to locate the key feature areas for dividing the core sections.

[0135] Specifically, the context of the incomplete area is divided into multiple small blocks according to the image coordinates, and the color histogram and LBP texture histogram of each small block are extracted as features. Then, it is judged whether each small block belongs to the key feature area. Finally, adjacent small blocks of the same type are merged to locate the key area.

[0136] S2252. Search the reference core image for a corresponding area that best matches the key feature area.

[0137] Specifically, the color histogram, LBP texture histogram and HOG shape feature vector of the key feature area are spliced ​​into a comprehensive feature vector, and then a sliding window is performed in the reference image to traverse and calculate the feature vector distance between each window block area and the key feature area, and the one with the smallest distance is selected as the best matching area.

[0138] S2253. Copy the matched corresponding area directly to the incomplete area of ​​the image as the repair result.

[0139] Specifically, the matching area is first enlarged or reduced by interpolation to make it the same size as the incomplete area, and then copied according to the registration transformation to replace the original pixel value of the incomplete area.

[0140] S2254: Use an interpolation algorithm to perform transition processing on the target sub-image so that the copy area is smoothly connected.

[0141] Specifically, SIFT feature points are first extracted from the original complete area and the copied area respectively, and the position and range of the overlapping area on both sides are determined by point matching. Then, interpolation algorithms such as wavelet decomposition or Poisson fusion are used within the overlapping range to smoothly transition the images on both sides so that the colors and textures at the edges are naturally connected.

[0142] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0143] In the second aspect, the present application provides a digital cataloging and duplication checking system for exploration cores based on panoramic image processing. The digital cataloging and duplication checking system for exploration cores based on panoramic image processing of the present application is described below in combination with the above-mentioned digital cataloging and duplication checking method for exploration cores based on panoramic image processing.

[0144] Reference Figure 8 , a digital cataloging and duplicate checking system for exploration cores based on panoramic image processing, comprising:

[0145] A panoramic image acquisition module is used to acquire a panoramic image of the drill core, perform image preprocessing and feature extraction based on the panoramic image, and obtain digital catalog data of the core;

[0146] A panoramic image segmentation module, used to segment the panoramic image based on a preset depth interval, obtain a series of sub-images and generate image feature vectors corresponding to the series of sub-images;

[0147] An image feature comparison module is used to compare the image feature vector with a preset image database, calculate the similarity and determine whether there is an anomaly based on preset thresholds for different comparison scenarios, including comparisons between different boreholes in the same project and comparisons between images in the current project and historical project databases;

[0148] The catalog report generation module is used to generate a catalog report based on the digital catalog data, similarity and anomaly determination results.

[0149] In one embodiment, the present application provides an electronic device, which may be a server, and its internal structure diagram may be as follows: Fig. 9 As shown. The electronic device includes a processor, a memory and a network interface connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the electronic device is used to store data. The network interface of the electronic device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing is implemented.

[0150] Those skilled in the art will understand that Fig. 9 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the electronic device to which the scheme of the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0151] In one embodiment, an electronic device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0152] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the above-mentioned computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0153] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing, characterized in that: The steps include: Obtain panoramic images of drill cores and perform preprocessing; Segmenting the preprocessed panoramic image to obtain a series of sub-images and generating image feature vectors and digital catalog data corresponding to the series of sub-images; Comparing the image feature vector with a preset image database, calculating the similarity and determining whether there is an anomaly according to preset thresholds of different comparison scenarios, wherein the comparison scenarios include comparison between different boreholes in the same project and comparison between images in the current project and historical project databases; Generate a catalog report based on the digital catalog data, similarity and anomaly determination results; The pre-processed panoramic image is segmented to obtain a series of sub-images and generate image feature vectors and digital catalog data corresponding to the series of sub-images, which specifically includes the following steps: Identify the depth scale in the panoramic image, extract the starting depth and ending depth of the core section corresponding to the panoramic image, and obtain depth information; Use edge detection algorithms to identify the upper and lower boundaries of the core and determine the effective core area; Based on the color, texture changes and visible geological interfaces of the core, the image segmentation algorithm is used to divide the core into different sections to obtain a series of sub-images; Calculating the length of each sub-image in the series of sub-images, and judging whether it is a complete core segment according to a preset threshold, and determining the RQD value; The color features, texture features and structural features of each sub-image are extracted, and the depth information and the RQD value are combined to generate a fixed-length image feature vector and structured digital catalog data.

2. The method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing according to claim 1, characterized in that: The image feature vector is compared with a preset image database, similarity is calculated, and whether there is an abnormality is determined according to preset thresholds of different comparison scenarios, specifically including the following steps: Extracting a reference image feature vector from a preset image database; Calculating the distance between the image feature vector and each reference image feature vector as similarity; Compare the calculated similarity with a preset threshold of the corresponding comparison scene to determine whether there is an anomaly; When an exception exists, the exception is marked and the exception information is output.

3. The method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing according to claim 1, characterized in that: Generating a catalog report based on the digital catalog data, similarity and anomaly determination results specifically includes the following steps: Associating the digital catalog data with the abnormality determination result to generate a catalog record; Classifying and sorting the catalog records according to the similarity; Generate a catalog report based on the sorted catalog records in a preset format.

4. The method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing according to claim 1, characterized in that: The method further comprises the steps of: Detecting whether there is a target sub-image with an incomplete image in the series of sub-images; For a detected target sub-image with an incomplete image, an image interpolation algorithm is used to repair the target sub-image to generate a repaired sub-image; The incomplete sub-image of the image is replaced by the restored sub-image, and the corresponding image feature vector and structured digital catalog data are regenerated according to the restored sub-image.

5. The method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing according to claim 4 is characterized in that: The target sub-image is repaired using an image interpolation algorithm, which specifically includes the following steps: Searching the preset image database for a complete reference core image that is most similar to the target sub-image; Extracting a corresponding complete sub-image region in the reference core image; Performing image registration on the extracted complete sub-image area and the target sub-image; Detect whether the incomplete area of ​​the image contains key features that divide the core sections; If the key features are not included, the incomplete area of ​​the image is interpolated and repaired, otherwise the corresponding complete area is directly copied from the reference core image.

6. The method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing according to claim 5, characterized in that: Directly copying the corresponding complete area from the reference core image specifically includes the following steps: Analyzing context information of the incomplete area of ​​the image to locate key feature areas for dividing core sections; Searching the reference core image for a corresponding region that best matches the key feature region; directly copying the matched corresponding area to the incomplete area of ​​the image as a repair result; An interpolation algorithm is used to perform transition processing on the target sub-image so that the copy area is smoothly connected.

7. A digital cataloging and duplicate checking system for exploration cores based on panoramic image processing, characterized in that: include: A panoramic image acquisition module is used to acquire a panoramic image of the drill core, perform image preprocessing and feature extraction based on the panoramic image, and obtain digital catalog data of the core; A panoramic image segmentation module, used to segment the panoramic image based on a preset depth interval to obtain a series of sub-images and generate image feature vectors corresponding to the series of sub-images; An image feature comparison module is used to compare the image feature vector with a preset image database, calculate the similarity and determine whether there is an abnormality according to preset thresholds of different comparison scenarios, wherein the comparison scenarios include comparison between different boreholes in the same project and comparison between images in the current project and historical project databases; A catalog report generation module, used to generate a catalog report based on the digital catalog data, similarity and abnormality determination results; The panoramic image acquisition module specifically performs the following steps: Acquire a panoramic image of the drill core, identify a depth scale in the panoramic image, extract a starting depth and an ending depth of a core section corresponding to the panoramic image, and obtain depth information; Use edge detection algorithms to identify the upper and lower boundaries of the core and determine the effective core area; Based on the color, texture changes and visible geological interfaces of the core, the image segmentation algorithm is used to divide the core into different sections to obtain a series of sub-images; Calculating the length of each sub-image in the series of sub-images, and judging whether it is a complete core segment according to a preset threshold, and determining the RQD value; The color features, texture features and structural features of each sub-image are extracted, and the depth information and the RQD value are combined to generate a fixed-length image feature vector and structured digital catalog data.

8. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing described in any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for digital cataloging and duplicate checking of exploration cores based on panoramic image processing described in any one of claims 1 to 6 are implemented.

Images