CN113658245B - Method for acquiring rock mass structural plane occurrence information from drilling image - Google Patents
Method for acquiring rock mass structural plane occurrence information from drilling image Download PDFInfo
- Publication number
- CN113658245B CN113658245B CN202110749987.7A CN202110749987A CN113658245B CN 113658245 B CN113658245 B CN 113658245B CN 202110749987 A CN202110749987 A CN 202110749987A CN 113658245 B CN113658245 B CN 113658245B
- Authority
- CN
- China
- Prior art keywords
- image
- structural
- borehole
- training
- curve
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 239000011435 rock Substances 0.000 title claims abstract description 29
- 238000005553 drilling Methods 0.000 title claims abstract description 17
- 238000012549 training Methods 0.000 claims abstract description 37
- 238000013139 quantization Methods 0.000 claims abstract description 34
- 230000011218 segmentation Effects 0.000 claims abstract description 17
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000012360 testing method Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000004445 quantitative analysis Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 13
- 238000000605 extraction Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 238000013441 quality evaluation Methods 0.000 description 2
- 238000011002 quantification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Geometry (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses a method for acquiring rock mass structural plane occurrence information from a drilling image, which comprises the steps of taking ResNet as a pre-training model to extract image features of a training set, inputting the image features into a Unet network for training, and obtaining a Unet model after training; dividing the borehole photographic image by using Unet model to obtain the division result of the structural plane; processing the segmentation result of the structural surface through an image refinement algorithm to obtain a structural surface image skeleton, and processing the structural surface image skeleton in unit width to form a structural surface quantization curve; performing sine function fitting on the quantization curve of the structural surface to obtain a quantization curve sine function expression; and calculating structural plane attitude information according to the structural plane segmentation result and the quantized curve sine function expression, wherein the structural plane attitude information comprises structural plane inclination angle, inclination and thickness. And theoretical and method support is provided for quantitative analysis of the structural surface occurrence in the hydraulic engineering rock foundation.
Description
Technical Field
The invention belongs to the technical field of rock mass quality evaluation methods, and relates to a method for acquiring rock mass structural plane occurrence information from a drilling image.
Background
Hydraulic engineering is built in mountain gorge areas, and geological conditions are quite complex. Poor geological conditions can lead to instability or seepage damage of hydraulic engineering foundations, and influence stability of the dam body. Therefore, rock mass quality evaluation is necessary in hydraulic engineering construction. Drill hole core pictures with position information and drill hole photographing panoramic pictures are always reserved in geological survey engineering; on-site logging rock mass RQD values and pressurized water test data; and measuring and recording borehole acoustic data, seismic wave data and the like. And qualitatively analyzing the quality of the rock mass through the recorded and measured data of the survey site, and making corresponding treatment measures.
Only the rock mass quality can be comprehensively analyzed through the survey data, and the rock mass integrity can be comprehensively evaluated. By analyzing the borehole photography panoramic image, the position, depth, trend and other yield information of the fault, crack and other undesirable geological structures in the borehole can be obtained, different borehole yield information can be combined, the detailed information of the fault crack structural plane of the survey rock mass can be analyzed, and a geological engineer can realize the prevention and control of the undesirable geological problems of the hydraulic engineering according to the geological survey information.
With the progress of computing power of computers and the rapid development and application of deep learning, image analysis techniques have been widely used. To obtain accurate structural plane information from a borehole image, firstly, accurate identification and segmentation of structural planes in the borehole image are realized, and the conventional image processing method based on a threshold value or an edge is difficult to adapt to the processing of borehole photographic images under complex geology.
Disclosure of Invention
The invention aims to provide a method for acquiring the occurrence information of a rock mass structural plane from a drilling image, which solves the problem that the occurrence information of the rock mass structural plane cannot be automatically and accurately acquired in the prior art.
The technical scheme adopted by the invention is that the method for acquiring the occurrence information of the rock mass structural plane from the drilling image comprises the following steps:
step 1, extracting image features of a training set by taking ResNet as a pre-training model, inputting the image features into a Unet network for training to obtain a Unet model after training;
Step 2, segmenting a borehole photographic image by using a trained Unet model to obtain a segmentation result of a structural surface, processing the segmentation result of the structural surface by using a Zhang-Suen image refinement algorithm to obtain a structural surface image skeleton, and processing the structural surface image skeleton in unit width to form a structural surface quantization curve;
step 3, performing sine function fitting on the structure surface quantization curve to obtain a quantization curve sine function expression;
and 4, calculating structural plane attitude information according to the segmentation result of the structural plane obtained in the step 2 and the quantized curve sine function expression, wherein the structural plane attitude information comprises the structural plane inclination angle, the inclination and the thickness.
The invention is characterized in that:
The step 1 specifically comprises the following steps:
Step 1.1, acquiring a borehole photographic image with a structural surface, marking pixels of the structural surface part of the borehole photographic image, and dividing the marked borehole photographic image into a training set and a testing set;
And step 1.2, extracting image features of a training set by taking ResNet as a pre-training model, and inputting the image features into a Unet network for training to obtain a Unet model after training.
The step 1 may further comprise the step of,
And 1.3, segmenting each borehole photographic image in the test set by using the Unet model, further calculating an F1_score value and a IoU value, and judging the performance of the Unet model after training.
The formulas for the F1_score and IoU values are as follows:
In the above formula, precision is the Precision, recall is the Recall, TP represents the label as positive, and prediction is also positive; FP represents the label negative, predicted positive; TN represents the negative of the tag, predicted to be negative; FN represents a label positive, predicted negative;
In the above formula, DT represents a predicted result range, and GT represents a true result range of the object; and U represents the intersection and U represents the union.
The step 3 specifically comprises the following steps:
Step 3.1, assuming that the relative coordinates of each point on the structural plane quantization curve are expressed as (x i,yi), fitting the structural plane quantization curve by adopting a least square method, and adopting a sine function formula and fitting single-point errors as follows:
In the middle of Is the ordinate of the fitting point curve, A is the amplitude, ω is the angular velocity,/>B is the sine curve baseline position, delta is the fitting error;
in the above formula, T is a sine function period, and L 0 is the width of a borehole photographic image;
step 3.2, according to the deviation calculation formula in the formula (5), the whole deviation I is minimized, as shown in the formula (7), and A, B, C and D can be obtained, b:
In the step 4, the structural inclination angle alpha, the inclination beta and the thickness H are calculated as follows:
calculating dy/dx=0, and obtaining two extreme points of the sine function, which are respectively marked as (x min,ymin) and (x max,ymax);
the trend beta is determined by the lowest point azimuth of the fitting curve, and the calculation formula is as follows:
In the above formula, L 0 is the borehole photographic image width, x 0 is the initial pixel position of the borehole photographic image;
In the above formula, S is the area of the structural surface of the drilling hole, L is the length of the curve of the structural surface, and the calculation formula of L is as follows:
In the above formula, n is the number of coordinate points of the quantization curve of the structural plane, and x i、yi represents the relative abscissa and ordinate of the ith pixel in the quantization curve.
The beneficial effects of the invention are as follows:
According to the method for acquiring the occurrence information of the rock mass structural plane from the drilling image, the accurate identification of the drilling structural plane is realized by adopting a depth migration model algorithm, and the structural plane image segmentation result is evaluated by adopting a plurality of indexes, so that the influence of subjective factors is avoided, and the identification efficiency of the structural plane is improved; the drilling photographic image in the actual engineering is segmented by utilizing a Unet model, the structural face skeleton is extracted by a Zhang-Suen image refinement algorithm and subjected to unit width processing, a structural face quantization curve is obtained, the quantization curve is fitted by a least square method, automatic quantization analysis of the trend, the dip angle and the thickness of the structural face can be realized from a segmentation result and the fitted curve, the intelligent quantization process of the structural face information is realized, and the geological investigation efficiency is improved; and theoretical and method support is provided for quantitative analysis of the structural surface occurrence in the hydraulic engineering rock foundation.
Drawings
FIG. 1 is a flow chart of a method of acquiring rock mass structural plane attitude information from a borehole image in accordance with the present invention;
FIG. 2 is a schematic diagram of a model Unet in a method of obtaining formation face shape information of a rock mass from a borehole image according to the present invention;
FIG. 3 is a schematic diagram showing the variation of different evaluation indexes in Unet model training process in a method for acquiring the occurrence information of a rock mass structural plane from a borehole image according to the present invention;
FIG. 4 is a borehole photographic image to be subjected to face segmentation and quantification in a method of acquiring face occurrence information of a rock mass from a borehole image in accordance with the present invention;
FIG. 5 is a schematic diagram of training results of Unet models in a method of obtaining formation face shape information of a rock mass from a borehole image according to the present invention;
FIG. 6 is a schematic diagram of a Zhang-Suen refinement algorithm for extracting structural face skeleton in a method for acquiring structural face occurrence information of a rock mass from a borehole image according to the present invention;
FIG. 7 is a schematic representation of a partial borehole photographic image structural surface quantification curve and its fitted curve in a method of obtaining structural surface occurrence information of a rock mass from a borehole image in accordance with the present invention.
Detailed Description
The invention will be described in detail below with reference to the drawings and the detailed description.
A method for obtaining rock mass structural plane occurrence information from a borehole image, as shown in fig. 1, comprising the steps of:
step 1, extracting image features of a training set by taking ResNet as a pre-training model, inputting the image features into a Unet network for training to obtain a Unet model after training;
Step 1.1, acquiring a borehole photographic image with a structural surface, processing the length and width dimensions of the borehole photographic image to 160 multiplied by 1280, marking pixels of the structural surface part of the borehole photographic image as a segmentation prospect, and taking 80% of the marked borehole photographic image as a training set and 20% as a test set;
Step 1.2, extracting image features of the training set by taking ResNet as a pre-training model, inputting the image features into a Unet network for training to obtain a Unet model after training, as shown in figure 2,
Step 1.3, divide each drilling photographic image in the testing set by Unet models, and then calculate F1_score value and IoU value, judge Unet model performance after training, different evaluation index changes in Unet model training process are shown in figure 3, in this embodiment, it is set that Unet can be used for dividing drilling photographic image when F1_score value is above 95%, ioU value is above 75%;
The formulas for the F1_score and IoU values are as follows:
In the above formula, precision is the Precision, recall is the Recall, TP represents the label as positive, and prediction is also positive; FP represents the label negative, predicted positive; FN represents a label positive, predicted negative;
In the above formula, DT represents a predicted result range, and GT represents a true result range of the object; and U represents the intersection and U represents the union.
Step 2, segmenting a borehole photographic image in an actual engineering by utilizing a Unet model to obtain a segmentation result of a structural surface, processing the segmentation result of the structural surface by using a Zhang-Suen image refinement algorithm to obtain a structural surface image skeleton, and processing the structural surface image skeleton in unit width to form a structural surface quantization curve;
Step 2.1, selecting 10 drilling images in actual engineering as test cases, applying a trained Unet model to the test images as shown in fig. 4 to obtain a segmentation result, and respectively naming 17 structural faces in the 10 test images uniformly as shown in fig. 5, wherein the format is I i-j, I represents the sequence number of the image from left to right, and j represents the sequence number of the structural face from top to bottom in the image;
Step 2.2, obtaining a structural plane image skeleton from the test image segmentation result of fig. 5 through a Zhang-Suen image refinement algorithm, as shown in fig. 6;
And 2.3, firstly deriving skeleton coordinates, and then carrying out unit width processing on the skeleton coordinates to form a structural plane quantization curve, wherein the unit width processing method adopts a new ordinate value as a mean value for the condition that the same abscissa corresponds to a plurality of different ordinates.
Step 3, performing sine function fitting on the structure surface quantization curve to obtain a quantization curve sine function expression;
Step 3.1, assuming that the relative coordinates of each point on the structural plane quantization curve are expressed as (x i,yi), fitting the structural plane quantization curve by adopting a least square method, and adopting a sine function formula and fitting single-point errors as follows:
In the middle of For fitting point ordinate, A is amplitude, ω is angular velocity,/>For the initial phase, b is the sinusoidal baseline position, delta is the fitting error, and the undetermined parameters are A,/>And b;
in the above formula, T is a sine function period, and L 0 is the width of a borehole photographic image;
Step 4.2, according to the deviation calculation formula in the formula (5), the whole deviation I is minimized, as shown in the formula (7), and A, B, C and D can be obtained, b:
In this embodiment, function fitting is performed on the structural planes of I 4-1, I 9-1、I9-2 and I 9-3, as shown in formula (8), the quantization and fitting curves are shown in fig. 7, and as can be seen from fig. 7, the error of the fitting result is smaller;
and 4, calculating structural plane attitude information according to the segmentation result of the structural plane obtained in the step 2 and the quantized curve sine function expression, wherein the structural plane attitude information comprises the structural plane inclination angle, the inclination and the thickness.
In the step 4, the structural inclination angle alpha, the inclination beta and the thickness H are calculated as follows:
calculating dy/dx=0, and obtaining two extreme points of the sine function, which are respectively marked as (x min,ymin) and (x max,ymax);
the trend beta is determined by the lowest point azimuth of the fitting curve, and the calculation formula is as follows:
In the above formula, L 0 is the borehole photographic image width, x 0 is the initial pixel position of the borehole photographic image;
In the above formula, S is the area of the structural surface of the drilling hole, L is the length of the curve of the structural surface, and the calculation formula of L is as follows:
In the above formula, n is the number of coordinate points of the quantization curve of the structural plane, and x i、yi represents the relative abscissa and ordinate of the ith pixel in the quantization curve.
According to the method for acquiring the bearing surface attitude information of the rock mass structure from the drilling image, provided by the invention, the Unet model is utilized to segment the drilling photographic image in the actual engineering, the Zhang-Suen image refinement algorithm is utilized to extract the structure surface skeleton and process the unit width to obtain the structure surface quantization curve, the least square method is used for fitting the quantization curve, the automatic quantization analysis of the inclination, the inclination angle and the thickness of the structure surface can be realized from the segmentation result and the fitting curve, the surveying efficiency is improved, and the theory and method support is provided for the bearing surface attitude quantization analysis in the hydraulic engineering rock foundation.
Examples:
the method of the invention is adopted to extract the occurrence information of the structural surface of the borehole photographic image, and table 1 is the intelligent acquisition result of occurrence information of all structural surfaces in the test image, and the results are compared with the occurrence data of manual extraction, so that the effectiveness of the occurrence information extraction of the structural surface of the invention is proved.
Table 1 Manual extraction and automatic identification results comparison Table
Note that: relative error= | (automatic recognition-manual extraction) |/manual extraction)
Claims (2)
1. A method for obtaining rock mass structural plane occurrence information from a borehole image, comprising the steps of:
step1, extracting image features of a training set by taking ResNet as a pre-training model, and inputting the image features into a Unet network for training to obtain a Unet model after training;
step 2, segmenting a borehole photographic image by using a trained Unet model to obtain a segmentation result of a structural surface, processing the segmentation result of the structural surface by using a Zhang-Suen image refinement algorithm to obtain a structural surface image skeleton, and performing unit width processing on the structural surface image skeleton to form a structural surface quantization curve;
step 3, performing sine function fitting on the structure surface quantization curve to obtain a quantization curve sine function expression;
step 4, calculating structural plane shape information according to the segmentation result of the structural plane obtained in the step 2 and the quantized curve sine function expression, wherein the structural plane shape information comprises structural plane inclination angle, inclination and thickness;
The step 1 may further comprise the step of,
Dividing each borehole photographic image in the test set by using Unet model, further calculating F1_score value and IoU value, and judging the performance of the Unet model after training;
the calculation formulas of the F1_score value and IoU value are as follows:
In the above formula, precision is the Precision, recall is the Recall, TP represents the label as positive, and prediction is also positive; FP represents the label negative, predicted positive; FN represents a label positive, predicted negative;
in the above formula, DT represents a predicted result range, and GT represents a true result range of the object; u represents intersection, U represents union;
The step 3 specifically comprises the following steps:
Step 3.1, assuming that the relative coordinates of each point on the structural surface quantization curve are expressed as (x i,yi), fitting the structural surface quantization curve by using a least square method, and adopting a sine function formula and fitting single-point errors as follows:
in the method, in the process of the invention, For fitting point ordinate, A is amplitude, ω is angular velocity,/>B is the sine curve baseline position, delta is the fitting error;
in the above formula, T is a sine function period, and L 0 is the width of a borehole photographic image;
step 3.2, according to the deviation calculation formula in the formula (5), the whole deviation I is minimized, as shown in the formula (7), and A, B, C and D can be obtained, b:
In the step 4, the structural inclination angle alpha, the inclination beta and the thickness H are calculated as follows:
Calculating dy/dx=0, obtaining two zero points of the sine function, which are respectively marked as (x min,ymin) and (x max,ymax);
the trend beta is determined by the lowest point azimuth of the fitting curve, and the calculation formula is as follows:
In the above formula, L 0 is the borehole photographic image width, x 0 is the initial pixel position of the borehole photographic image;
In the above formula, S is the area of the structural surface of the drilling hole, L is the length of the curve of the structural surface, and the calculation formula of L is as follows:
In the above formula, n is the number of coordinate points of the quantization curve of the structural plane, and x i、yi represents the relative abscissa and ordinate of the ith pixel in the quantization curve.
2. A method for obtaining rock mass structural plane bearing information from a borehole image according to claim 1, wherein step 1 comprises the steps of:
step 1.1, acquiring a borehole photographic image with a structural surface, marking pixels of the structural surface part of the borehole photographic image, and dividing the marked borehole photographic image into a training set and a testing set;
and 1.2, extracting features of borehole photographic images in a training set by taking ResNet as a pre-training model, inputting the image features into a Unet network for training, and obtaining a Unet model after training.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110749987.7A CN113658245B (en) | 2021-07-01 | 2021-07-01 | Method for acquiring rock mass structural plane occurrence information from drilling image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110749987.7A CN113658245B (en) | 2021-07-01 | 2021-07-01 | Method for acquiring rock mass structural plane occurrence information from drilling image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113658245A CN113658245A (en) | 2021-11-16 |
CN113658245B true CN113658245B (en) | 2024-04-30 |
Family
ID=78489860
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110749987.7A Active CN113658245B (en) | 2021-07-01 | 2021-07-01 | Method for acquiring rock mass structural plane occurrence information from drilling image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113658245B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116258972B (en) * | 2023-05-16 | 2023-08-01 | 四川安信科创科技有限公司 | Deep learning-based rock high-steep slope structural surface extraction method |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104808248A (en) * | 2015-04-22 | 2015-07-29 | 长江大学 | Attitude of stratum detecting method based on electric imaging logging full hole image |
CN107169504A (en) * | 2017-03-30 | 2017-09-15 | 湖北工业大学 | A kind of hand-written character recognition method based on extension Non-linear Kernel residual error network |
GB201910300D0 (en) * | 2013-03-11 | 2019-09-04 | Reeves Wireline Tech Ltd | Method of and apparatuses for indentifying geological characteristics |
CN110360963A (en) * | 2019-07-22 | 2019-10-22 | 中国科学院武汉岩土力学研究所 | A kind of rock mass discontinuity three-dimensional roughness of borehole camera technology describes method |
CN112837269A (en) * | 2021-01-11 | 2021-05-25 | 西安理工大学 | Rock mass quality evaluation method based on deep learning model |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2511744B (en) * | 2013-03-11 | 2020-05-20 | Reeves Wireline Tech Ltd | Methods of and apparatuses for identifying geological characteristics in boreholes |
-
2021
- 2021-07-01 CN CN202110749987.7A patent/CN113658245B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB201910300D0 (en) * | 2013-03-11 | 2019-09-04 | Reeves Wireline Tech Ltd | Method of and apparatuses for indentifying geological characteristics |
CN104808248A (en) * | 2015-04-22 | 2015-07-29 | 长江大学 | Attitude of stratum detecting method based on electric imaging logging full hole image |
CN107169504A (en) * | 2017-03-30 | 2017-09-15 | 湖北工业大学 | A kind of hand-written character recognition method based on extension Non-linear Kernel residual error network |
CN110360963A (en) * | 2019-07-22 | 2019-10-22 | 中国科学院武汉岩土力学研究所 | A kind of rock mass discontinuity three-dimensional roughness of borehole camera technology describes method |
CN112837269A (en) * | 2021-01-11 | 2021-05-25 | 西安理工大学 | Rock mass quality evaluation method based on deep learning model |
Non-Patent Citations (2)
Title |
---|
Improving accuracy of automatic fracture detection in borehole images with deep learning and GPUs;Rommel Anatoli Quintanilla Cruz et al.;2017 30th SIBGRAPI Conference on Graphics, Patterns and Images;20171231;全文 * |
基于电成像测井图像识别求取地层产状的新方法;杨玉卿 等;石油物探;20170331;第56卷(第2期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN113658245A (en) | 2021-11-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107701160A (en) | Simulate hydraulically created fracture and the experimental provision and method of intrinsic fracture extension influence | |
WO2023185734A1 (en) | Mine digital model construction method based on three-dimensional point cloud database | |
CN104948176B (en) | A kind of method based on infiltration Magnification identification carbonate reservoir crack | |
CN110360963B (en) | Rock mass structural plane three-dimensional roughness description method based on drilling and shooting technology | |
CN114387328A (en) | RQD calculation method based on deep learning model and core image | |
CN107807387A (en) | Acquisition methods when seismic first break based on neutral net is walked | |
CN113658245B (en) | Method for acquiring rock mass structural plane occurrence information from drilling image | |
CN111189507B (en) | Karst water stratification discrimination and underground water level observation method | |
CN110596757B (en) | Method for correcting longitudinal wave and transverse wave velocities of shale formation | |
CN112241711A (en) | Intelligent method for identifying RQD from borehole core photo | |
CN110211142A (en) | Water level online test method outside a kind of hole of the deep basal pit based on video image processing | |
CN113515654B (en) | Pile foundation construction parameter visualization processing method and device based on building information modeling | |
CN109725354B (en) | Anisotropic speed modeling method and system | |
CN115577616A (en) | Carbonatite fracture-cave earthquake depicting method and device based on deep learning | |
CN115113281A (en) | Method, device, equipment and storage medium for determining marker well and marker layer thereof | |
CN108876011A (en) | A kind of coal seam macroscopic view crack advantage develops bearing prediction method | |
CN114200538B (en) | Ground stress direction determining method, device, storage medium and computer equipment | |
CN104462792B (en) | Log data lithological stratum numerical value reduction method | |
CN112282730A (en) | Real-time monitoring and evaluating method for deformation of underground casing induced by reservoir fracturing modification | |
CN113848209A (en) | Dam crack detection method based on unmanned aerial vehicle and laser ranging | |
CN113433123B (en) | Method for testing geological parameters in drill hole and intelligent device | |
CN111123373A (en) | Ground penetrating radar full waveform inversion method based on wave field expansion reconstruction | |
CN113376697B (en) | Static correction method and device based on three-dimensional near-surface model | |
CN117687096B (en) | Proxy model construction method for predicting small-scale fracture-cavity distribution | |
CN113482596B (en) | Real-time three-dimensional gamma imaging data processing method based on LWD while-drilling instrument |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |