Method for quickly obtaining core recovery and RQD value

Abstract

The present application belongs to geological exploration technical field, disclose a kind of method for quickly obtaining core recovery and RQD value, to solve the problem of inaccuracy of core recovery and RDQ value caused by the large error of artificial measurement of core.The present application includes: (1) the core box that stores core is photographed, to obtain the picture with the outline of core box and the outline of several core outer surfaces: (2) the outline of core outer surface is preprocessed to obtain preprocessed core outer surface outline; (3) the dark area between adjacent preprocessed core outer surface outlines is deleted; (4) the length data of core box is obtained to calculate the length value of each preprocessed core outer surface outline; (5) the length value of the obtained core outer surface outline is used to calculate the core recovery and RQD value.The present application can obtain core recovery and RQD value by photographing, with the characteristics of fast speed and more accurate calculation of core recovery and RQD value.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of geological exploration, and particularly relates to a method for quickly obtaining core recovery rate and RQD value. BACKGROUND

[0002] In the geological exploration work, a large amount of manpower, material resources, financial resources and time are consumed for the purpose of core drilling, and the ultimate goal is to obtain cores in quality and quantity as the first physical data for geological research and evaluation of the exploration area. Therefore, the core recovery rate is an important indicator for measuring the quality of drilling.

[0003] The core recovery rate refers to the ratio of the total length of the complete rock and the broken rock taken out by drilling to the footage of a round trip (the depth drilled in a round trip, that is, the drilling depth in a cycle of tripping in and out), which is an index reflecting the drilling quality and is related to the degree of rock fragmentation, drilling technology and quality control. The core recovery rate in a round trip is referred to as the round trip recovery rate, and the core recovery rate in a stratum is referred to as the stratified core recovery rate.

[0004] The rock RQD is an index of rock quality, which is obtained by drilling into the rock with a diamond drill bit with a diameter of 75 mm and a double-layer core barrel, continuously taking cores, and calculating the ratio of the sum of the lengths of the core segments with a length greater than or equal to 10 cm in the cores taken in a round trip to the footage of the round trip.

[0005] Therefore, as important quality indicators in the exploration field, the core recovery rate and RQD are generally stored in core boxes during the preservation process. The core boxes are generally made of wood, aluminum alloy and plastic.

[0006] At present, the core recovery rate and RQD are calculated by manually measuring the cores and then calculating the core recovery rate and RQD value. However, manual measurement has a large error in measuring the cores, resulting in a large error between the calculated core recovery rate and RQD value and the true value. SUMMARY

[0007] The present application provides a method for quickly obtaining the core recovery rate and RQD value to solve the problem of inaccurate core recovery rate and RQD value caused by manual measurement of the cores, which can improve the speed of obtaining the core recovery rate and RQD value and the accuracy of the core recovery rate and RQD value.

[0008] To solve the technical problem, the technical scheme adopted by the present application is as follows:

[0009] A method for quickly obtaining the core recovery rate and RQD value, characterized in that it comprises:

[0010] (1) Taking a photo of the core box in which the core is stored, thereby obtaining a picture with the outline of the core box and the outlines of several core outer surfaces:

[0011] (2) Preprocessing the core outer surface outline to obtain a preprocessed core outer surface outline;

[0012] (3) Deleting the dark area between adjacent preprocessed core outer surface outlines;

[0013] (4) Obtaining the length data of the core box, comparing each preprocessed core outer surface outline with the outline of the core box to obtain the contrast ratio of each preprocessed core outer surface outline with the outline of the core box, and calculating the length value of each preprocessed core outer surface outline according to the obtained length data of the core box and the contrast ratio;

[0014] (5) Superimposing the length values of each preprocessed core outer surface outline to obtain the total length of the core, and obtaining the core recovery rate by comparing the total length of the core with the footage of the round; selecting the length value greater than or equal to 10 cm from the length values of each preprocessed core outer surface outline and superimposing them to obtain the RQD value.

[0015] In some embodiments, the core outer surface outline includes the outline of the complete rock and the outline of the broken rock.

[0016] In some embodiments, the preprocessing of the core outer surface outline to obtain a preprocessed core outer surface outline includes:

[0017] de-noising the core outer surface outline.

[0018] In some embodiments, the de-noising of the core outer surface outline includes: extracting features of the core outer surface outline using an encoding network to obtain core outer surface outline feature information, wherein the core outer surface outline feature information includes noise information; inputting the core outer surface outline feature information into a decoding network for de-noising to obtain a preprocessed core outer surface outline after de-noising.

[0019] In some embodiments, the encoding network and the decoding network are convolutional neural networks, and the convolutional neural network includes a plurality of convolutional layers, each convolutional layer being configured to perform convolution operations for compression and feature reconstruction of the core outer surface outline; the decoding network includes a plurality of convolutional layers with the same number of layers as the encoding network, each convolutional layer being configured to perform convolution operations for restoration and feature reconstruction of the input core outer surface outline feature information to obtain a de-noised preprocessed core outer surface outline.

[0020] In some embodiments, the deleting of the dark area between adjacent preprocessed core outer surface outlines includes:

[0021] (1) Obtain color information of the core box profile, the color information including red component value, green component value and blue component value;

[0022] (2) Obtain color information of each pretreated core outer surface profile, the color information including red component value, green component value and blue component value;

[0023] (3) Compare the red component value, green component value and blue component value of each pretreated core outer surface profile with the red component value, green component value and blue component value of the core box profile respectively, and obtain the comparison value of the color of each pretreated core outer surface profile and the color of the core box profile respectively;

[0024] (4) Obtain color information of the color in the photographed picture except the color of the core box profile and the color of the core outer surface profile, and compare the color information with the red component value, green component value and blue component value of the core box profile;

[0025] (5) Delete color information of the color except the comparison value of the color of each pretreated core outer surface profile and the color of the core box profile, so as to delete dark color area of adjacent pretreated core outer surface.

[0026] In some embodiments, when the core outer surface profile includes the outer surface profile of broken rock, first, path recognition is performed on each outer surface profile of broken rock to obtain path information, and then the path information of each outer surface profile of broken rock is matched, and each broken rock with the same path information is spliced to form the outer surface profile of complete rock.

[0027] In some embodiments, when the core box storing the core is photographed, the core should be rotated to take pictures, so as to obtain pictures of four surfaces of the core in the core box; when the path information of each outer surface profile of broken rock is matched, the path information of the outer surface profile of one broken rock is matched with the path of the outer surface profile of the four surfaces of the remaining broken rocks.

[0028] In some embodiments, the matching of the path information of the outer surface profile of one broken rock with the path of the outer surface profile of the four surfaces of the remaining broken rocks further includes matching the broken rocks with the same color information.

[0029] Compared with the prior art, the present application has the following beneficial effects:

[0030] The method for quickly obtaining core recovery and RQD value of the application takes photos of four surfaces of the core in the core box, then obtains the outer surface contour of the complete rock, the outer surface contour of the broken rock and the outer surface contour of the core box, then obtains the color information of the outer surface contour of the complete rock, the color information of the outer surface contour of the broken rock and the color information of the outer surface contour of the core box, then performs path recognition on the outer surface contour of the broken rock and performs matching to splice into the outer surface contour information of the complete rock, then performs denoising pretreatment on the obtained outer surface contour of the complete rock, then compares the color information of the pretreated core outer surface contour with the color information of the outer surface contour of the core box to obtain the comparison value of each pretreated core outer surface contour, then deletes the dark color area of the adjacent pretreated core outer surface; the length value of each pretreated core outer surface contour is calculated according to the length data of the core box; the length values of each pretreated core outer surface contour are superimposed to obtain the total length of the core, and the ratio of the total length of the core to the footage per round is obtained to obtain the core recovery; the length value greater than or equal to 10cm is selected from the length value of each pretreated core outer surface contour, superimposed and compared with the obtained footage per round to obtain the RQD value. Compared with the prior art, the core recovery and RQD value can be obtained by taking photos, and the core recovery and RQD value calculated by the method have the characteristics of high speed and high accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1 The flow chart of an embodiment of the application. DETAILED DESCRIPTION

[0032] The application will be further described below in conjunction with the embodiments. The described embodiments are only a part of the embodiments of the application, and are not all the embodiments. Based on the embodiments in the application, other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0033] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application; the terms "first", "second", "third" are only for description purposes, and cannot be understood as indicating or implying relative importance; in addition, unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connecting" should be understood in a broad sense, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected; it can be directly connected, or indirectly connected through an intermediate medium. For those skilled in the art, the specific meaning of the above terms in the present application can be understood in conjunction with the specific circumstances.

[0034] The method for quickly obtaining core recovery rate and RQD value of the present application in combination with the drawings comprises:

[0035] (1) Taking a picture of the core box in which the cores are stored, thereby obtaining a picture with the outline of the core box and the outlines of the outer surfaces of the cores: during the photographing process, the cores should be placed vertically, that is, the center lines of the complete cores and the broken cores are on the same straight line, in order to improve the accuracy of calculation and avoid placing the broken cores horizontally.

[0036] In order to obtain more information, the four directions of the cores should be photographed, that is, after taking a picture, each core (complete core and broken core) is rotated 90° and photographed again, and the same is repeated until the circumferential surface of the core is completely photographed.

[0037] (2) Preprocessing the outer surface contour of the core to obtain a pretreated core outer surface contour, the purpose of preprocessing is to denoise the outer surface contour of the core, thereby improving the accuracy of subsequent processing;

[0038] (3) Deleting the dark area between adjacent pretreated core outer surface contours, since the length of the core box is generally fixed, and a core box contains multiple rows of cores, each row of cores is not placed at zero distance, therefore, the gap between adjacent cores needs to be removed, thereby ensuring the accuracy of subsequent calculation of core recovery rate and RQD value;

[0039] (4) Obtain the length data of the core box, compare each pretreated core outer surface profile with the core box profile to obtain a comparison ratio of each pretreated core outer surface profile to the core box profile, and calculate the length value of each pretreated core outer surface profile according to the obtained length data of the core box and the comparison ratio; wherein the length of the core box is generally fixed, even if the length of the core box is not fixed, the length data of the core box can be obtained very conveniently through measurement.

[0040] (5) Superimpose the length values of each pretreated core outer surface profile to obtain the total length of the core, and obtain the core recovery rate by the ratio of the total length of the core to the footage per round.

[0041] In some embodiments, the core outer surface profile includes an outer surface profile of intact rock and an outer surface profile of broken rock.

[0042] In some embodiments, the pretreatment of the core outer surface profile to obtain a pretreated core outer surface profile includes:

[0043] de-noising the core outer surface profile.

[0044] In some embodiments, the de-noising of the core outer surface profile includes: extracting features of the core outer surface profile using an encoding network to obtain core outer surface profile feature information, wherein the core outer surface profile feature information includes noise information; inputting the core outer surface profile feature information into a decoding network for de-noising to obtain a de-noised pretreated core outer surface profile.

[0045] In some embodiments, the encoding network and the decoding network are convolutional neural networks, and the convolutional neural network includes a plurality of convolutional layers, each convolutional layer being configured to perform convolution operations for compression and feature reconstruction of the core outer surface profile; the decoding network includes a plurality of convolutional layers with the same number of layers as the encoding network, each convolutional layer being configured to perform convolution operations for restoration and feature reconstruction of the input core outer surface profile feature information to obtain a de-noised pretreated core outer surface profile. In a specific implementation process, the convolutional neural network is a deep convolutional neural network (CNN), which is a machine learning model under deep supervised learning, has strong adaptability, is good at mining local features of data, extracting global training features and classification, and its weight sharing structure network makes it more similar to a biological neural network, and has achieved good results in various fields of pattern recognition.

[0046] The encoding network and the decoding network can include a plurality of convolutional layers (for example, 5 convolutional layers), each of which includes a convolutional block for compressing a size and a convolutional block for reorganizing features. It is known and understood by those skilled in the art that the use of the encoding network and the decoding network for denoising the picture, and thus will not be described here.

[0047] In some embodiments, the deleting of the dark color region between the adjacent pre-processed core outer surface profiles comprises:

[0048] (1) obtaining color information of the core tank profile, the color information comprising red component value, green component value and blue component value;

[0049] (2) obtaining color information of each pre-processed core outer surface profile, the color information comprising red component value, green component value and blue component value;

[0050] (3) comparing the red component value, the green component value and the blue component value of each pre-processed core outer surface profile with the red component value, the green component value and the blue component value of the core tank profile, respectively, to obtain the comparison value of the color of each pre-processed core outer surface profile with the color of the core tank profile;

[0051] (4) obtaining color information of the color in the photographed picture except the color of the core tank profile and the color of the core outer surface profile, and comparing the color information with the red component value, the green component value and the blue component value of the core tank profile;

[0052] (5) deleting the color information of the color of each pre-processed core outer surface profile except the comparison value of the color of each pre-processed core outer surface profile with the color of the core tank profile, so as to delete the dark color region between the adjacent pre-processed core outer surface profiles.

[0053] In the present application, the red component value, the green component value and the blue component value of the core tank profile and the core outer surface profile can be used to determine the core of different colors and identify the core and the core tank.

[0054] In some embodiments, when the core outer surface profile includes the outer surface profile of the broken rock, the path information of each broken rock outer surface profile is obtained by performing path recognition on each broken rock outer surface profile, and the broken rocks with the same path information are spliced to form the outer surface profile of the complete rock.

[0055] In some embodiments, when taking a picture of the core box in which the core is stored, the core should be rotated to take a picture, so as to obtain the pictures of the four surfaces of the core in the core box; when matching the path information of the outer surface profile of each broken rock, the path information of the outer surface profile of a broken rock is matched with the paths of the outer surface profiles of the four surfaces of the remaining broken rocks.

[0056] In some embodiments, the matching the path information of the outer surface profile of a broken rock with the paths of the outer surface profiles of the four surfaces of the remaining broken rocks further comprises: matching the broken rocks with the same color information.

[0057] When the path information of the outer surface profile of a broken rock is matched with the paths of the outer surface profiles of the four surfaces of the remaining broken rocks, the outer surface profiles of the remaining broken rocks are rotated to find the remaining broken rocks with the same path as the broken rock, and in the case of the same color, the broken rocks with the same path can indicate that the two broken rocks should belong to one whole, because of the structure of the rock itself or the breaking during the taking and transporting of the core. The present application matches the broken rocks by the above color information matching, path identification and matching to form a complete rock.

[0058] In summary, when containing broken cores, the method for quickly obtaining the core recovery rate and RQD value of the present application comprises:

[0059] (1) Taking a picture of the core box in which the core is stored, so as to obtain a picture with the core box profile and the outer surface profiles of a plurality of cores: during the photographing process, the outer surface profiles of the four surfaces of the core should be obtained;

[0060] (2) When the outer surface profile of the core contains the outer surface profile of a broken rock, the broken rock is spliced to form the outer surface profile of a complete rock;

[0061] (3) Preprocessing the outer surface profile of the core to obtain a pretreated outer surface profile of the core;

[0062] (4) Deleting the dark area between adjacent pretreated outer surface profiles of the core;

[0063] (5) Obtaining the length data of the core box, comparing each pretreated outer surface profile of the core with the core box profile to obtain the contrast ratio of each pretreated outer surface profile of the core with the core box profile, and calculating the length value of each pretreated outer surface profile of the core according to the obtained length data of the core box and the contrast ratio;

[0064] (6) superimpose the length values of the outer surface profiles of the pretreated cores to obtain a total length of the cores, and the ratio of the total length of the cores to the footage of the turns obtains a core recovery rate; select the length values greater than or equal to 10 cm from the length values of the outer surface profiles of the pretreated cores, superimpose the selected length values, and compare the superimposed length values with the footage of the turns to obtain RQD values.

[0065] The method for quickly obtaining a core recovery rate and RQD values of the present application comprises the following steps: taking pictures of four surfaces of the cores in the core barrel, then obtaining an outer surface profile of complete rocks, an outer surface profile of broken rocks, and an outer surface profile of the core barrel; then obtaining color information of the outer surface profile of the complete rocks, color information of the outer surface profile of the broken rocks, and color information of the outer surface profile of the core barrel; then performing path recognition on the outer surface profile of the broken rocks and performing matching to splice the outer surface profile information of the complete rocks; then performing denoising pretreatment on the obtained outer surface profile of the complete rocks; then comparing the color information of the pretreated core outer surface profile with the color information of the outer surface profile of the core barrel to obtain comparison values of the pretreated core outer surface profiles; then deleting adjacent dark areas of the pretreated core outer surfaces; calculating length values of the pretreated core outer surface profiles according to the length data of the core barrel; superimposing the length values of the pretreated core outer surface profiles to obtain a total length of the cores, and the ratio of the total length of the cores to the footage of the turns obtains a core recovery rate; selecting length values greater than or equal to 10 cm from the length values of the pretreated core outer surface profiles, superimposing the selected length values, and comparing the superimposed length values with the footage of the turns to obtain RQD values. Compared with the prior art of manual measurement and calculation, the present application can obtain core recovery rates and RQD values through photographing, and has the characteristics of fast speed and more accurate calculation of core recovery rates and RQD values.

Claims

1. A method for quickly obtaining core recovery and RQD values, characterized in that, The method comprises the following steps: (1) taking a photo of the core box containing cores, thereby obtaining a picture with the outline of the core box and the outlines of the surfaces of the cores: when taking a photo of the core box containing cores, the cores should be rotated to take photos, thereby obtaining photos of the four surfaces of the cores in the core box; the surface outline of the cores comprises the surface outline of the complete rock and the surface outline of the broken rock; the path information of the surface outline of each broken rock is obtained by path recognition, and then the path information of the surface outline of each broken rock is matched, and each broken rock with the same path information is spliced to form the surface outline of the complete rock; when matching the path information of the surface outline of each broken rock, the path information of the surface outline of one broken rock is matched with the surface outlines of the four surfaces of the remaining broken rocks; (2) pre-treating the surface outline of the cores to obtain a pre-treated surface outline of the cores; (3) deleting the dark area between adjacent pre-treated surface outlines of the cores, specifically comprising the following steps: (3.1) obtaining the color information of the outline of the core box, wherein the color information comprises red component value, green component value and blue component value; (3.2) obtaining the color information of each pre-treated surface outline of the cores, wherein the color information comprises red component value, green component value and blue component value; (3.3) comparing the red component value, green component value and blue component value of each pre-treated surface outline of the cores with the red component value, green component value and blue component value of the outline of the core box, respectively, to obtain the comparison value of the color of each pre-treated surface outline of the cores with the color of the outline of the core box; (3.4) obtaining the color information in the taken picture except the color information of the outline of the core box and the surface outline of the cores, and comparing the color information with the red component value, green component value and blue component value of the outline of the core box; (3.5) deleting the color information except the comparison value of the color of each pre-treated surface outline of the cores with the color of the outline of the core box, thereby deleting the dark area between adjacent pre-treated surface outlines of the cores; (4) obtaining the length data of the core box, comparing each pre-treated surface outline of the cores with the outline of the core box to obtain the comparison value of each pre-treated surface outline of the cores with the outline of the core box, and calculating the length value of each pre-treated surface outline of the cores according to the obtained length data of the core box and the comparison value; (5) superimposing the length values of each pre-treated surface outline of the cores to obtain the total length of the cores, and obtaining the core recovery rate by comparing the total length of the cores with the footage of the round; selecting the length value greater than or equal to 10 cm from the length values of each pre-treated surface outline of the cores, superimposing the length values, and comparing the superimposed length values with the footage of the round to obtain the RQD value.

2. The method of quickly obtaining core recovery and RQD values of claim 1, wherein, The pre-treating the surface outline of the cores comprises denoising the surface outline of the cores.

3. The method of quickly obtaining core recovery and RQD values of claim 2, wherein, The denoising processing of the core outer surface profile includes: using an encoding network to extract features of the core outer surface profile to obtain core outer surface profile feature information, wherein the core outer surface profile feature information includes noise information; and inputting the core outer surface profile feature information into a decoding network for denoising processing to obtain a denoised preprocessed core outer surface profile.

4. The method of quickly obtaining core recovery and RQD values of claim 3, wherein, The encoding network and the decoding network are convolutional neural networks, and the convolutional neural networks include multiple convolutional layers, each of which is used for convolution operation of compression and feature reconstruction of the core outer surface profile; the decoding network includes multiple convolutional layers with the same number of layers as the encoding network, each of which is used for convolution operation of restoration and feature reconstruction of the input core outer surface profile feature information to obtain the denoised preprocessed core outer surface profile.

5. The method of quickly obtaining core recovery and RQD values of claim 1, wherein, The matching of the path information of the outer surface profile of the broken rock with the paths of the outer surface profiles of the four sides of the remaining broken rocks further includes: matching the broken rocks with the same color information.

Images