CN117132784A - Image recognition-based drilling rock core digital cataloging method - Google Patents
Image recognition-based drilling rock core digital cataloging method Download PDFInfo
- Publication number
- CN117132784A CN117132784A CN202311097258.3A CN202311097258A CN117132784A CN 117132784 A CN117132784 A CN 117132784A CN 202311097258 A CN202311097258 A CN 202311097258A CN 117132784 A CN117132784 A CN 117132784A
- Authority
- CN
- China
- Prior art keywords
- core
- image recognition
- information
- data
- rock
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 239000011435 rock Substances 0.000 title claims abstract description 63
- 238000005553 drilling Methods 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000012549 training Methods 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000010276 construction Methods 0.000 claims abstract description 5
- 238000013135 deep learning Methods 0.000 claims abstract description 4
- 238000005070 sampling Methods 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 6
- 230000001788 irregular Effects 0.000 claims description 6
- 239000002689 soil Substances 0.000 claims description 5
- 238000004140 cleaning Methods 0.000 claims description 4
- 239000000203 mixture Substances 0.000 claims description 4
- 230000009467 reduction Effects 0.000 claims description 4
- 239000000126 substance Substances 0.000 claims description 4
- 238000011161 development Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 3
- 238000003860 storage Methods 0.000 abstract description 5
- 238000011835 investigation Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 14
- 238000007726 management method Methods 0.000 description 7
- 238000005259 measurement Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 239000004575 stone Substances 0.000 description 5
- 239000010410 layer Substances 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000005520 cutting process Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000012530 fluid Substances 0.000 description 2
- 239000011521 glass Substances 0.000 description 2
- 239000011229 interlayer Substances 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000005406 washing Methods 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000004927 clay Substances 0.000 description 1
- 230000001010 compromised effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000945 filler Substances 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011707 mineral Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/191—Design or setup of recognition systems or techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
- G06V30/19147—Obtaining sets of training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention provides a method for digitally cataloging drill cores based on image recognition, relates to the technical field of geological investigation, and aims to solve the problem that other steps except RQD and related data are not integrated and unified in the prior art, and the method adopts the following technical scheme that: the method comprises the steps that various data information of a drilling rock core is recorded by adopting a mobile phone terminal on an exploration construction site; importing each item of data acquired by a mobile phone end to a computer end, performing preliminary processing on each item of imported data information, and generating a corresponding rock core relation chart by adopting a cloud image recognition algorithm; training a deep learning image recognition algorithm at a cloud; the electronic terminal equipment and the cloud database management system digitize and electronize the catalogued data, realize centralized storage, classification, retrieval and backup of the data, and improve the efficiency and reliability of information management.
Description
Technical Field
The invention relates to the technical field of geological investigation, in particular to a digital cataloging method of a drilling rock core based on image recognition.
Background
Borehole core cataloging is the process of systematically recording and describing cores obtained from a borehole. Borehole core catalogues generally include the following steps: 1. preparation: the tools and materials required for the catalog are determined, such as catalog tables, core boxes, measuring tools, scales, magnifying glasses, etc. 2. Core treatment: and numbering the extracted rock cores according to the drilling depth sequence, and performing preliminary treatment. This includes cleaning the earth, washing the core surface to remove drilling fluid, marking the starting and ending locations of each pass of the core, etc. 3. And (3) taking a rock core: a camera is used to take a record of each box of core and close up a particular part or feature. 4. And (3) catalog measurement: the length and diameter of the core are measured using a measuring tool, such as a scale or digital gauge. The measurement results for each core segment are recorded. 5. Observation and description: details of the core were observed using a magnifying glass and described. This includes rock color, structure, texture, mineral content, etc. At the same time, any specific structural, fracture, joint or core changes are recorded. 6. Sampling: and selecting a proper position for sampling according to the requirement and the sampling purpose, and recording the detailed information of sampling. 7. Filling in an inventory form: and filling the information of the measurement result, observation and description into the catalogue form. The table typically contains fields for borehole depth, return cut, diameter, rate of acquisition, RQD, view and description. 8. And (3) organizing the catalogued data: all of the catalog data is collated and aggregated and electronically for subsequent analysis and application.
The prior art currently has the following problems: 1. at present, borehole cataloging mainly depends on manual operation and manual paper recording. Firstly, the related recorded information needs to be tidied and archived in the later period, is inconvenient to retrieve, is easy to cause information management confusion, and has the risk of data loss. Secondly, paper recordings are susceptible to human factors, including human error, ambiguous handwriting and readability problems, leading to compromised data accuracy and reliability. Finally, the transfer and sharing of paper records requires physical media, resulting in difficult collaboration, communication delays, and low work efficiency. 2. The existing core cataloging method requires a large amount of manual operation and mechanical repetition, so that the on-site measurement workload is huge, and the efficiency is low; secondly, the mode of using a steel tape or a tape to carry out the measurement one by one results in lower measurement accuracy, and is easily influenced by the accuracy of the measuring tool and human subjective factors. The method can not provide data accurately reflecting the rock quality, and can not meet the requirement of accurately evaluating the engineering rock quality. In the prior art CN113806681A, CN112241711A, CN114387328A, the RQD is automatically processed in the process of the cataloging by means of image recognition and the like, but the systematic engineering that the borehole cataloging is a comprehensive process of a plurality of flow steps is ignored, and other steps except the RQD and related data are not integrated and unified.
The prior art CN202210682472.4 discloses a method, a system and a device for automatically cataloging a drilling core RQD, which relate to the method, the system and the device for RQD, wherein the above patent relies on single RQD image recognition and data processing, and the whole process of digital cataloging cannot be realized through multi-platform collaboration; the above patent only counts cores with core lengths of 10cm (including 10 cm), and has certain disadvantages; in addition, the measurement basis and calculation method of the data such as RQD in image recognition in the prior art do not strictly follow the methods recommended by the International Society of Rock Mechanics (ISRM) and the laboratory and field test standardization committee, which results in overestimation of the quality of rock mass.
Disclosure of Invention
In view of the problems in the prior art, the invention discloses a digital cataloging method of a drilling rock core based on image recognition, which comprises each link of the cataloging of the rock core, comprehensively utilizes the mobile terminal, the desktop terminal and the cloud technology, and realizes comprehensive cooperation and efficient completion of the cataloging of the drilling rock core through multi-platform cooperation. Firstly, by utilizing the convenience and mobility of a mobile phone end, a geological engineer can acquire core images through mobile equipment to finish a part of cataloging task; secondly, completing more complex task processing and data preprocessing by means of a larger screen space provided by a computer end; and finally, completing data analysis processing of all secondary core relations through an image recognition model trained by a cloud and stronger computing power, and simultaneously being capable of being used as a center for data storage and sharing to ensure the safety and accessibility of data.
The technical scheme adopted is that the method comprises the following steps:
step 1, adopting a mobile phone terminal to record various data information of a drilling rock core on an exploration construction site;
step 2, importing each item of data acquired by a mobile phone end to a computer end, performing preliminary processing on each item of imported data information, and generating a corresponding rock core relation chart by adopting a cloud image recognition algorithm;
step 3, training an image recognition algorithm of deep learning at a cloud; according to different shape outlines and combination relations thereof, core relation data of each round are calculated according to corresponding formulas, and core image recognition results and the core relation data are output to a computer end, so that information management and storage optimization are improved: the electronic terminal equipment and the cloud database management system digitize and electronize the catalogued data, realize centralized storage, classification, retrieval and backup of the data, and improve the efficiency and reliability of information management.
As a preferred technical solution of the present invention, the step 1 further includes the following steps:
step 1.1, cleaning the surface of a rock core by using water;
step 1.2, inserting a rock core plate with a marking function in the depth of each termination hole;
step 1.3, recording project information of the rock core;
step 1.4, inputting drilling information of the core;
step 1.5, photographing the rock core, and performing preliminary treatment on the photographing;
step 1.6, describing the rock core in sections according to the composition of substances and the weathering degree;
step 1.7, recording crack development conditions of each section of the core;
step 1.8, selecting representative rock and soil for sampling;
step 1.9, exporting the recorded information to a computer end in the form of compressed packets.
As a preferred technical solution of the present invention, the step 2 further includes the following steps:
step 2.1, performing secondary treatment on the core photo;
step 2.2, adding a title to the core photo after secondary treatment, and generating a title bar at the top;
step 2.3, importing the secondary information of the drill hole;
step 2.4, calling a cloud training recognition algorithm to perform core recognition processing on the current drilling;
step 2.5, performing manual auxiliary error correction on the information of the identification errors;
and 2.6, generating corresponding core relation chart data, and storing and backing up the data.
As a preferred technical solution of the present invention, the step 3 further includes the following steps:
step 3.1, carrying out recognition training on the rock core plate at the cloud end to ensure that the rock core plate can be accurately recognized, and recognizing character information on the rock core plate through OCR characters to ensure that the information is correct;
step 3.2, carrying out recognition training on the core box at the cloud end, ensuring that the core box and the core can be accurately segmented and recognized, and establishing a plane coordinate network according to the length and width dimensions of the core box, so as to lay a foundation for core length recognition; and 3.3, carrying out recognition training on cores with different forms of contours at the cloud, accurately dividing the contours recognized by the algorithm, and calculating the sampling rate, the acquisition rate and the RQD of each time by adopting different formulas. And 3.4, outputting the core identification result and the calculated core relation table of each round to a computer end, carrying out contour color marking according to the category of each core, and marking information in the core contour.
As a preferable technical scheme of the invention, the step 3.3 comprises a quasi-rectangular contour, a quasi-triangular contour, a split contour and an irregular contour,
the mathematical calculation expression of the rectangular-like outline is as follows:
rate of collection=ln/Lm
Yield = Ln/Lm
The mathematical computational expression of the triangle-like profile is:
rate of collection= (ln1+ln2)/Lm
Yield = (ln1+ln2)/Lm
The mathematical calculation expression of the split profile is as follows:
rate of collection= (ln1+ln2)/(2×lm)
Yield = (ln1+ln2)/(2×lm)
The mathematical computational expression of the irregular profile is:
rate of collection= (Ln a)/Lm
Yield = 0
RQD＝0
m represents a round number; lm represents the mth drilling length; ln, ln 1 、Ln 2 Is a single sheetThe length of each core; a represents a reduction coefficient.
The invention has the beneficial effects that: the invention improves information management and storage optimization: the electronic terminal equipment and the cloud database management system digitize and electronize the catalogued data, realize centralized storage, classification, retrieval and backup of the data, and improve the efficiency and reliability of information management.
Furthermore, the invention improves the data quality and accuracy: the digital recording eliminates the handwriting and readability problems, reduces the occurrence of human errors, and ensures the accuracy and reliability of data.
Furthermore, the invention facilitates data sharing and collaboration: the digital records can realize instant data sharing and cooperation through the network and the cloud platform, so that real-time communication, cooperation and decision among team members are promoted, and the working efficiency and team cooperation capability are improved.
Furthermore, the invention can reduce the cost and enhance the efficiency: the problem of on-site artificial machinery measure the core length is solved, through core image recognition technology, the core length is automatically recognized and processed, the automatic calculation of the core relation related data is realized, the working efficiency is improved, and the labor cost is reduced.
Furthermore, the invention can realize the fine recognition calculation: through core contour recognition, various types of cores are subdivided, different formulas are adopted to calculate various data of the cores according to different conditions, accuracy of the data is improved, and accurate basis is provided for rock integrity evaluation, dam foundation rock engineering geology classification, tunnel surrounding rock classification and side slope rock structure classification.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. Like elements or portions are generally identified by like reference numerals throughout the several figures. In the drawings, elements or portions thereof are not necessarily drawn to scale.
FIG. 1 is a schematic diagram of a mobile phone end borehole catalog of the present invention;
FIG. 2 is a schematic diagram of a mobile phone end borehole catalog of the present invention;
FIG. 3 is a schematic diagram III of each module of the cell phone end borehole catalog of the present invention;
FIG. 4 is a diagram of a mobile phone end borehole catalog of the present invention;
FIG. 5 is a schematic diagram of each module of the mobile phone end borehole catalog of the present invention;
FIG. 6 is a diagram of a mobile phone end borehole catalog of the present invention;
FIG. 7 is a diagram of a computer-side borehole logging interface and recognition results according to the present invention;
FIG. 8 is a diagram of a computer-side borehole logging interface and recognition results according to the present invention;
FIG. 9 is a first diagram of the identification effect of the core plate of the present invention;
FIG. 10 is a second view of the core plate and its identification effect according to the present invention;
FIG. 11 is a graph I of a core contour image recognition situation according to the present invention;
FIG. 12 is a second illustration of the core profile image recognition of the present invention;
FIG. 13 is a third view of the core profile image recognition of the present invention;
FIG. 14 is a fourth illustration of the core profile image recognition scenario of the present invention;
FIG. 15 is a diagram of a core contour image recognition situation of the present invention;
FIG. 16 is a diagram of a core profile image recognition scenario six according to the present invention;
FIG. 17 is a graph seven of the core contour image recognition case of the present invention;
FIG. 18 is a graph eight of the core contour image recognition case of the present invention;
FIG. 19 is a diagram of a core contour image recognition situation according to the present invention;
FIG. 20 is a schematic diagram of a core profile image recognition scenario according to the present invention;
FIG. 21 is a diagram of a core profile image recognition situation eleventh in the present invention;
FIG. 22 is a graph twelve for identifying the contour image of a core according to the present invention;
FIG. 23 is a thirteenth view of a core profile image recognition scenario in accordance with the present invention;
FIG. 24 is a flow chart of the present invention.
Detailed Description
Example 1
As shown in fig. 1 to 24, the invention discloses a digital catalogue method of a drilling rock core based on image recognition, which adopts the following technical scheme that:
step 1, a mobile phone terminal is adopted to record drilling rock cores on an exploration construction site, wherein the logging comprises the steps of inputting information such as engineering projects, exploration drilling and the like, photographing the rock cores, describing cracks, sampling the information and the like, and the operation interfaces of all modules and the information related to drilling and recording are specifically shown in figures 1 to 6. And after the on-site cataloging work is completed, each item of data collected by the mobile phone end is exported to the computer end.
And 2, importing collected related data at a computer end, firstly, rotating and correcting part of the drilling photographs, ensuring that each box of core photographs is a overlook vertical shot image, editing information such as project names, investigation stages, drilling names, hole depth intervals, core box numbers and the like into picture titles, and generating pictures with titles. And secondly, invoking an image recognition algorithm of the cloud, recognizing the outline and the length of the rock core, and finally generating a corresponding rock core relation chart. The corresponding operation interface and recognition result of the computer end are shown in fig. 7 and 8.
And 3, training a deep learning image recognition algorithm at the cloud, wherein the training algorithm comprises a core plate, a core box, a core contour and the like. According to core graphic data uploaded by a computer end, the identified core cards are divided into different rounds, a plane coordinate network is established according to length, width and other data information of a core box to reversely calculate the length and the round footage information of a single core, the core outline identified by the image is subdivided and identified by different colors, the rock and the covering layer are divided according to the composition of substances, and the rock outline can be further subdivided into quasi-rectangular, quasi-parallelogram, quasi-trapezoid, quasi-triangle and irregular broken stone. And calculating core relation data of each round according to different shape outlines and combination relations thereof and corresponding formulas, and finally outputting core image recognition results and the core relation data to a computer end.
Step 1 further comprises the steps of:
step 1.1, cleaning the rock core: and the impurities such as soil, drilling fluid and the like on the surface of the rock core are removed by water washing, so that the clean, attractive and convenient photographing of the rock core is ensured.
Step 1.2, inserting a rock core plate: and a special core plate is inserted into the depth of each time of termination hole, and the secondary information and the hole depth information are arranged on the core plate, so that the core plate is convenient to identify and read, and the color of the core plate is distinguished from that of a common core plate, so that the core plate is convenient and accurate to identify.
Step 1.3, item information input: creating a project at a mobile phone end, inputting the project name, the industries and the investigation stage, and importing an EXCEL table corresponding to the stratum lithology.
Step 1.4, drilling information input: newly creating a drilling hole in the belonging project, and inputting information such as a drilling hole number, a hole inclination, kong Xiang, a hole depth, a drilling unit, a catalogue, a site checking person and the like.
Step 1.5, taking a picture of the rock core: all the core photos are shot from the position right above each core box in sequence, and the core photos are subjected to preliminary cutting and correction.
Step 1.6, core description: the method is characterized in that the recorded drilling holes are described in a segmented mode according to the composition of substances and the weathering degree, information such as the depth of the bottom of each segment, stratum codes, lithology names, hardness degree, structure and construction is recorded, special bad geological phenomena such as faults, karst, weak interlayers, interlayer shear bands and the like can be recorded in a focused and detailed mode, and finally the coring state is summarized.
Step 1.7, crack description: the crack development condition of each section is recorded, including the number of cracks, the developed hole depth, the crack inclination angle, the straight degree, the roughness degree, the width and the filler characteristics, and the crack density of the section is automatically calculated.
Step 1.8, sampling and recording: and selecting a representative rock-soil body for sampling, recording the serial number, the type and the field name of the sampling, and taking corresponding pictures.
Step 1.9, data export: and exporting all the recorded information to a computer end in the form of a compressed packet for corresponding operation.
The step 2 further comprises the following steps:
step 2.1, correcting the picture: and importing the computer end catalogue data, carrying out secondary correction on the drill hole core photo, clicking a correction picture, sequentially selecting four corner points of the picture (the selected point is red), and automatically cutting to form a new picture after the four points are selected.
Step 2.2, adding a picture title: clicking a function button 'custom picture title' at the upper right corner, selecting text information to be displayed in a title bar in a popup window to obtain corresponding title content, and automatically generating the title bar at the top of a drilled photo.
Step 2.3, importing secondary information: clicking the 'lead-in round' button, selecting the round information file of the current drilling hole and leading in.
Step 2.4, image recognition: after clicking the 'identify' button, the cloud training identification algorithm can be invoked to identify the core of the current drilling hole. The intelligent recognition progress bar can change along with the change of the recognition progress in the recognition process, and when the intelligent recognition progress bar is 100%, the recognition process is finished.
Step 2.5, manual auxiliary error correction: after the identification is finished, the identification error caused by the reason of the definition of the photo can be corrected by manual assistance, so that the information is ensured to be accurate.
Step 2.6, data storage and backup: and generating a corresponding core relation chart, and storing all electronic data (drilling core pictures with titles, core descriptions, crack descriptions, sampling information and the like) of the drilling holes in a cloud project folder, so that later-stage consulting and calling are facilitated.
The step 3 further comprises the following steps:
and 3.1, carrying out recognition training on a special rock core plate at a cloud end, ensuring that the rock core plate can be accurately recognized at any position and in any placement mode in a rock core box, unifying the standard colors, determining the drilling position of each pass, determining the hole depth and the footage of each pass through the rock core plate character information recognized by OCR characters or the rock core pass table imported from the outside, and corresponding to the rock core plate one by one. The core plate and the identification effect thereof are shown in figures 9 and 10.
Step 3.2, carrying out recognition training on the core box at the cloud, wherein the core box comprises hollow core boxes made of various color materials and different layers and core boxes filled with various color pattern cores, and ensuring that the core boxes and the cores can be accurately segmented and recognized. And then a plane coordinate network is established according to the length and width dimensions of the core box (generally, the lower left corner of the core box is taken as an origin), so as to lay a foundation for core length identification.
And 3.3, carrying out recognition training on cores with different shapes of contours at the cloud, and accurately dividing the contours recognized by the algorithm into quasi-rectangles, quasi-parallelograms, quasi-trapezoids, quasi-triangles and irregular broken stones and covering layers. According to the International Society of Rock Mechanics (ISRM) and the methods recommended by the laboratory and field test standardization committee, the length of a single core should be measured along the centerline and statistics should be taken into account for RQDs for cores that can be spliced into columns and are greater than 10cm under certain conditions, rather than for columnar cores that are greater than 10cm at each footage in a national general sense. The length of the contour center line of each rock core is accurately read in the plane coordinate network to serve as the length of the rock core, and the sampling rate, the obtaining rate and the RQD of each time are calculated by adopting different formulas according to various actual conditions.
Case 1: for the situation that a similar rectangle, a similar trapezoid, a similar parallelogram and the like have a partially complete column shape and the situation that the three parts are mutually spliced left and right, see fig. 11 to 16, the sampling rate, the obtaining rate and the RQD are all independently calculated according to the central line length of a single rock core, and the mathematical calculation formula can be uniformly expressed as follows:
rate of collection=ln/Lm
Yield = Ln/Lm
Case 2: when the triangle-like shape, the trapezoid-like shape, the parallelogram-like shape and the like are spliced to form a complete or more complete column shape, see fig. 17 and fig. 18, the two cores are taken as a whole, and the collection rate, the acquisition rate and the RQD are considered by the accumulated value of the lengths of the center lines of the two cores. The mathematical calculation formula can be expressed as:
rate of collection= (ln1+ln2)/Lm
Yield = (ln1+ln2)/Lm
Case 3: when the core is split into upper and lower parts by the near-vertical slit, as shown in fig. 19, the two semi-cylindrical cores should be taken as a whole, and the average value of the lengths of the center lines of the two semi-cylindrical cores is used for considering the sampling rate, the obtaining rate and the RQD.
Rate of collection= (ln1+ln2)/(2×lm)
Yield = (ln1+ln2)/(2×lm)
Case 4: the broken stone which cannot be spliced into columnar irregular shapes and the covering layers with relatively mixed material components (such as powdery clay, sand, broken stone soil, broken stone, gravel pebbles and the like) can be simplified into rectangular-like shapes and the length reduction is considered because only the sampling rate is counted. See fig. 20 to 23.
Rate of collection= (Ln a)/Lm
Yield = 0
RQD＝0
Wherein m represents a round number; lm represents the mth drilling length; ln, ln 1 、Ln 2 Is a single core length; a represents a reduction coefficient. The above 4 cases basically cover all cases in the actual exploration production process, all the above sampling rate, obtaining rate and RQD are only independent data of a single block (or two blocks are spliced), and the sampling rate, obtaining rate and RQD of a single round are calculated by only accumulating all the cases (independent sampling rate, independent obtaining rate and independent RQD) of the round.
And 3.4, outputting the core identification result and the calculated core relation table of each round to a computer end, carrying out contour color marking according to the category of each core, and marking information such as round, block number, length and the like in the core contour.
In the technical scheme, a plurality of tools for realizing the image recognition algorithm are available, such as CNN, openCV, tensorFlow, pyTorch.
Although the specific embodiments of the present invention have been described in detail, the present invention is not limited to the above embodiments, and various changes and modifications without inventive labor may be made within the scope of the present invention without departing from the spirit of the present invention, which is within the scope of the present invention.
Claims (5)
1. The digital catalogue method of the drill core based on the image recognition is characterized by comprising the following steps:
step 1, adopting a mobile phone terminal to record various data information of a drilling rock core on an exploration construction site;
step 2, importing each item of data acquired by a mobile phone end to a computer end, performing preliminary processing on each item of imported data information, and generating a corresponding rock core relation chart by adopting a cloud image recognition algorithm;
step 3, training an image recognition algorithm of deep learning at a cloud; according to different shape outlines and combination relations thereof, core relation data of each round are calculated according to corresponding formulas, and core image recognition results and the core relation data are output to a computer end.
2. The method for digitally cataloging drill cores based on image recognition according to claim 1, wherein the step 1 further comprises the steps of:
step 1.1, cleaning the surface of a rock core by using water;
step 1.2, inserting a rock core plate with a marking function in the depth of each termination hole;
step 1.3, recording project information of the rock core;
step 1.4, inputting drilling information of the core;
step 1.5, photographing the rock core, and performing preliminary treatment on the photographing;
step 1.6, describing the rock core in sections according to the composition of substances and the weathering degree;
step 1.7, recording crack development conditions of each section of the core;
step 1.8, selecting representative rock and soil for sampling;
step 1.9, exporting the recorded information to a computer end in the form of compressed packets.
3. The method for digitally cataloging drill cores based on image recognition according to claim 1, wherein the step 2 further comprises the steps of:
step 2.1, performing secondary treatment on the core photo;
step 2.2, adding a title to the core photo after secondary treatment, and generating a title bar at the top;
step 2.3, importing the secondary information of the drill hole;
step 2.4, calling a cloud training recognition algorithm to perform core recognition processing on the current drilling;
step 2.5, performing manual auxiliary error correction on the information of the identification errors;
and 2.6, generating corresponding core relation chart data, and storing and backing up the data.
4. The method for digitally cataloging drill cores based on image recognition according to claim 1, wherein the step 3 further comprises the steps of:
step 3.1, carrying out recognition training on the rock core plate at the cloud end to ensure that the rock core plate can be accurately recognized, and recognizing character information on the rock core plate through OCR characters to ensure that the information is correct;
step 3.2, carrying out recognition training on the core box at the cloud end, ensuring that the core box and the core can be accurately segmented and recognized, and establishing a plane coordinate network according to the length and width dimensions of the core box, so as to lay a foundation for core length recognition;
and 3.3, carrying out recognition training on cores with different forms of contours at the cloud, accurately dividing the contours recognized by the algorithm, and calculating the sampling rate, the acquisition rate and the RQD of each time by adopting different formulas.
And 3.4, outputting the core identification result and the calculated core relation table of each round to a computer end, carrying out contour color marking according to the category of each core, and marking information in the core contour.
5. The image recognition-based drilling core digital cataloguing method is characterized by comprising the following steps of: the step 3.3 includes a rectangular-like contour, a triangular-like contour, a split-shaped contour and an irregular-shaped contour, and the mathematical calculation expression of the rectangular-like contour is as follows:
rate of collection=ln/Lm
Yield = Ln/Lm
The mathematical computational expression of the triangle-like profile is:
rate of collection= (ln1+ln2)/Lm
Yield = (ln1+ln2)/Lm
The mathematical calculation expression of the split profile is as follows:
rate of collection= (ln1+ln2)/(2×lm)
Yield = (ln1+ln2)/(2×lm)
The mathematical computational expression of the irregular profile is:
rate of collection= (Ln a)/Lm
Yield = 0
RQD＝0
m represents a round number; lm meterShowing the mth drilling length; ln, ln 1 、Ln 2 Is a single core length; a represents a reduction coefficient.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311097258.3A CN117132784A (en) | 2023-08-29 | 2023-08-29 | Image recognition-based drilling rock core digital cataloging method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311097258.3A CN117132784A (en) | 2023-08-29 | 2023-08-29 | Image recognition-based drilling rock core digital cataloging method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117132784A true CN117132784A (en) | 2023-11-28 |
Family
ID=88857903
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311097258.3A Pending CN117132784A (en) | 2023-08-29 | 2023-08-29 | Image recognition-based drilling rock core digital cataloging method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117132784A (en) |
-
2023
- 2023-08-29 CN CN202311097258.3A patent/CN117132784A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111858980B (en) | Method for identifying and recording rock core in rock core disc | |
| CN103412967B (en) | The application process of field geology information acquisition system, acquisition method and acquisition system | |
| CN112241711B (en) | Intelligent method for identifying RQD from drilling core photo | |
| CN103955558A (en) | Method for collecting and processing engineering investigation data of different industries | |
| CN115131486B (en) | Engineering exploration data acquisition system and method | |
| CN114662915A (en) | Method for rapidly acquiring rock core sampling rate and RQD value | |
| CN109270567A (en) | A method of exploratory trench gamma-radiation dose rate isopleth map being made in uranium exploration | |
| CN114387328A (en) | RQD calculation method based on deep learning model and core image | |
| CN112381920A (en) | Three-dimensional geological logging method and system | |
| CN105069162A (en) | Informatized exploratory adit catalog data acquisition system and method | |
| CN111784803A (en) | Automatic acquisition system and method for drill core correlation data | |
| CN109973033A (en) | Wire line coring geological drilling rock core arranges and edits and records method | |
| CN115035297A (en) | Automatic recording method, system, device and medium for drilling core RQD | |
| CN113190910B (en) | Slice analysis system and method based on BIM three-dimensional geological tunnel model | |
| CN111192359B (en) | Method for extracting borehole lithology data by using geological section borehole space track morphology | |
| CN117132784A (en) | Image recognition-based drilling rock core digital cataloging method | |
| CN105577827A (en) | Highway engineering geological drilling information acquisition and real-time transmission apparatus based on Internet | |
| CN117056284A (en) | Management system for drilling achievements of water transport engineering | |
| CN113592823B (en) | Rock stratum porosity distribution calculation method | |
| CN113626546B (en) | Digital storage and application method for railway bridge geological graphics | |
| CN106777155B (en) | Geological logging method based on electronic centimeter paper | |
| CN106504319A (en) | Reservoir Three Dimensional Contrast map generalization method and device between well | |
| CN111651549A (en) | Similarity retrieval method based on geological drilling data | |
| CN110647554B (en) | Structural surface spacing calculation method based on pole isopycnic graph containing space position information | |
| CN205356411U (en) | Highway engineering geological drilling information acquisition and real time transport device based on internet |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |