CN113554071B - A method and system for identifying associated minerals in rock samples - Google Patents

Abstract

The invention discloses a method and a system for identifying associated minerals in a rock sample, which are implemented by collecting a cross-sectional image of a cross section of the rock sample; preprocessing, screening and segmenting the sectional images to obtain sub-mine area images, wherein the sub-mine area images form a first image set; classifying and identifying each sub-mine image in the first image set; calculating the number of associated mine images marked as associated mines in the first image set; when the number of the associated ore images exceeds a threshold value, the rock samples are judged to contain the associated ores, otherwise, the rock samples do not contain the associated ores, whether the rock samples contain the associated ores or not can be intelligently, quickly and accurately identified, so that the ores containing the associated ores can be quickly separated and screened, the mining value can be improved, and the method and the device are applied to the field of ore screening and classification.

Description

A method and system for identifying associated minerals in rock samples

technical field

The present disclosure belongs to the field of machine vision technology and ore identification, and in particular relates to a method and system for identifying associated minerals in rock samples.

Background technique

Associated minerals are found in a mineral deposit containing other minerals, and many ores contain associated minerals. Associated ore is in the same ore deposit (ore body) and does not have the value of independent mining, but can be exploited and utilized together with its associated main minerals. Associated minerals are relative to the main minerals. Because they have similar geochemical properties and common material sources, they are often associated in the same ore deposit (ore body).

Associated minerals generally do not exist independently in independent minerals, such as bauxite in porphyry copper ores; or mixed in minerals (main minerals) contained in main minerals, such as galena, tin in sphalerite, Indium, gallium, germanium, etc., but if the associated content is generally not too high, it is only mined and separated when its value is large, and the associated ore is difficult to mine. The reason is that in a large amount of ores, only a small part is attached. There are associated minerals, but most of the ores do not have associated minerals, so it is difficult to separate and screen out which ores have associated minerals from the existing ores, and it is difficult to quickly identify whether the ore has associated minerals with the current existing technical means. .

SUMMARY OF THE INVENTION

The purpose of the present invention is to provide a method and system for identifying associated minerals in rock samples, so as to solve one or more technical problems existing in the prior art, and at least provide a beneficial choice or create conditions.

In order to achieve the above object, according to an aspect of the present disclosure, a method for identifying associated minerals in a rock sample is provided, the method comprising the following steps:

S100, collecting cross-sectional images of rock sample cross-sections;

S200, performing preprocessing, screening and segmentation on the cross-sectional images to obtain images of each sub-mining area, and each sub-mining area image constitutes a first image set;

S300, classifying and identifying images of each sub-mining area in the first image set;

S400, calculating the number of associated ore images marked as associated ore in the first image set;

S500, when the number of associated ore images exceeds a threshold, it is determined that the rock sample contains associated ore, otherwise the rock sample does not contain associated ore.

Further, in S100, the method for collecting the cross-section image of the rock sample is: collecting the image of the cross-section of the rock sample by any one of a hyperspectral camera, a hyperspectral imager, a linear CCD industrial camera, and a near-infrared light image sensor. Obtain cross-sectional images.

Further, in S100, the rock samples include porphyry copper ore, wolframite, galena, sphalerite, iron ore, crystal ore, nickel ore, rare earth ore, tantalum niobium ore, zircon ore, phosphate ore , Uranium Ore, Thorium Ore.

Further, in S200, the cross-sectional images are preprocessed, screened and segmented to obtain images of each sub-mining area, and the method for each sub-mining area image to form the first image set is as follows: Gaussian filtering and graying of the cross-sectional image to obtain a grayscale image, The grayscale image is calculated by the watershed algorithm to obtain the boundary points, and the edge lines are obtained by connecting each boundary point.

Further, in S200, the cross-sectional images are preprocessed, screened and segmented to obtain images of each sub-mining area, and the method for each sub-mining area image to form the first image set is as follows: Gaussian filtering and graying of the cross-sectional image to obtain a grayscale image, In the grayscale image, edge lines are detected by the Sobel edge detection operator, and the closed image area formed by each edge line is used as a sub-mining area image, and each sub-mining area image constitutes a first image set.

Further, in S300, the method for classifying and identifying each sub-mining area image in the first image set is:

S301, let the first image set be G1={G1 _i }, let K be the number of sub-mining images in the first image set G1, set variables i, j, i∈[1,K], G1 _i is the first image The image of the ith sub-mining area in the set; let the value of i be 1;

S302, perform corner detection on G1 _i with a corner detection algorithm to obtain each corner point of G1 _i , and sort each corner point according to the distance from the corner point to the geometric center point of G1 _i from small to large to obtain an ordered set of corner points H1= {H _j }, H _j is the jth corner point in the corner point set H1 formed by the corner points of G1 _i ; let S be the number of corner points of G1 _i , let the value of j be 1, i∈[1, S]; Described corner detection algorithm is Harris corner detection algorithm or Shi-Tomasi corner detection algorithm;

S303, (due to the different mineral regions represented by the images of the sub-mining areas distributed on the ore, and the different densities between the mineral areas result in different stress, so the shape of the image of the sub-mining areas on the ore surface is fragmented and stretched, and the directly obtained The sub-mining area image is difficult to be accurately identified for the identification of associated minerals, and the misrecognition rate is very high. Therefore, if you want to accurately extract the sub-mining area image, the following processing should be performed on the sub-mining area image to highlight the characteristics of the associated ores and main minerals) ,

Connect H _j and H _j+1 to get line segment L1, connect H _j and H _j+2 to get line segment L2, connect H _j+1 and H _j+2 to get line segment L3, take H _j as vertex and L1, L2 as edges The included angle is ∠A, the included angle with H _j+1 as vertex and L1, L3 as sides is ∠B, the included angle with H _j+2 as vertex and L2, L3 as sides is ∠C; H The line segment of the vertical line of _j on the side L3 or the line segment of the line connecting H _j to the midpoint of L3 is C1;

S304, if any one of ∠A, ∠B, ∠C is an obtuse angle, (if sampling mineral identification is performed at this time, that is, the associated minerals and main minerals in the image shape of the sub-mining area are too narrow and entangled together, if the sampled sub-mining area is If the image is too small, it is easy to collect non-mineral areas, so the position of H _j needs to be corrected). If there is a projection point on the point H _j to the side L3, let the projection point from the point H _j to the side L3 be HP _j , if the point There is no projection point on the side L3 from H _j to the midpoint of L3 as HP _j , that is, from the point H _j to the intersection of the vertical line on the side L3 and the side L3 or the midpoint of L3 as HP _j ;

The connection point HP _j to the point H _j forms a straight line C1, and the direction from the point H _j to the point HP _j is the first direction, and the direction from the point HP _j to the point H _j is the second direction;

If ∠A is not an obtuse angle, move the position of H _j to the first direction along the straight line C1 by a distance ΔL to update the position coordinates of H _j , where ΔL=|Max(D2)-Min(D1)|, where , Max(D2) is the largest element in the calculation set D2, Min(D1) is the smallest element in the calculation set D1;

Sets D1 and D2 are respectively the maximum distance threshold set and the minimum distance threshold set between adjacent sampling points obtained according to steps S3041 to S3043;

S3041: Set the initial value of the variable k to 1, and set the empty sets D1 and D2;

S3042: Calculate the Euclidean distance d1 from the corner point H _k to the corner point H _k+1 in the ordered corner point set H1, the Euclidean distance d2 from the corner point H _k to the corner point H _k+2 , and the corner point H _k The Euclidean distance d3 from ₊₁ to the corner point H _k+2 ; H _k is the Kth corner point in H1; the maximum value of d1, d2 and d3 is added to the set D1, and the minimum value of d1, d2 and d3 is added to the set into set D2;

S3043: if k+2<S, increase the value of k by 1 and go to step S3042, otherwise output the obtained maximum distance threshold set D1 and minimum distance threshold set D2;

If ∠A is an obtuse angle, move the position of H _j along the straight line C1 to the second direction by a distance ΔL, thereby updating the position coordinates of H _j ;

S305, if ∠A, ∠B, and ∠C are all acute angles, (at this time, the sampling mineral identification is performed, and the sampling area is too small, which will confuse the sampling of the main mineral and the associated mineral and cause distortion, and the sampling area needs to be enlarged), let the point H _j The projected point of ₊₁ on the edge L2 is HP _j+1 , let the projected point of the point H _j+2 on the edge L1 be HP _j+2 , and connect the point HP _j+1 to the point H _j+1 to form a straight line C2, with The direction from point HP _j+1 to point H _j+1 is the third direction; connecting point HP _j+2 to point H _j+2 forms a straight line C3, and the direction from point HP _j+2 to point H _j+2 is the third direction. Four directions; move the position of H _j+1 along the straight line C2 to the third direction by a distance ΔL, thereby updating the position coordinates of H _j+1 ; move the position of H _j+2 along the straight line C3 to the fourth direction by a distance △L, thereby updating the position coordinates of H _j+2 ;

S306, the triangular area formed by the updated vertices H _j , H _j+1 and H _j+2 is recorded as the mining area to be identified;

S307, obtaining the spectral waveform data at the positions of the three vertexes H _j , H _j+1 and H _j+2 of the mining area to be identified, extracting the spectral waveform characteristics of the spectral waveform data at the three vertex positions respectively, according to the wavelength of the spectral waveform characteristic position Mineral type identification by spectral database matching;

S308, if the mineral type identified by the positions of any two of the three vertices H _j , H _j+1 and H _j+2 is the first mineral, mark the mining area to be identified as the first mineral partition; if the three vertices H The mineral type identified by the positions of any two vertices in _j , H _j+1 and H _j+2 is the second mineral, then mark the mining area to be identified as the second mineral partition; if j+2<S, increase the value of j 1 and go to step S303, otherwise go to step S309;

S309, if i<K, increase the value of i by 1 and go to step S302; otherwise, the classification and recognition of the images of each sub-mining area in the first image set ends.

Further, in S307, the method for acquiring the spectral waveform data of the mining area to be identified is as follows: acquiring the spectral waveform data through any one of a ground object spectrometer, a hyperspectral camera, a hyperspectral imager, and a near-infrared spectrometer.

Further, in S307, the method for identifying the mineral type through spectral database matching according to the wavelength of the characteristic position of the spectral waveform is:

Extract the spectral waveform features of the spectral waveform data at the positions of the three vertices, and the spectral waveform features include the first-order differential, second-order differential, wave peak, and wave trough of the spectral waveform;

Wavelength matching is performed between each spectral waveform feature and the spectral waveform feature of each mineral in the spectral database, and matching minerals that are consistent with the spectral waveform features of the spectral waveform data of the three vertices are obtained.

Further, in S307, the spectral database specifically includes the USGS spectral database, the ASD atomic spectral database, the JPL standard spectral database, the ASTER spectral database, the HIPAS spectral database, and the JHU spectral database, as well as the literature: Zhang Yingtong, Xiao Qing, Wen Jianguang , et al. Construction progress and application status of ground object spectral database [J]. Journal of Remote Sensing, 2017, 21(001): 12-26, any kind of spectral database involved in this document.

Further, in S400, the method for calculating the number of associated ore images marked as associated ore in the first image set is: respectively calculating the number Ta1 marked as the first mineral partition in the first image set and the number Ta1 marked as the first image set in the first image set. The number of second mineral partitions Ta2, when Ta1>Ta2, the first mineral is the main mineral, and the second mineral is the associated mineral; when Ta1≤Ta2, the first mineral is the associated mineral, and the second mineral is the main mineral Minerals; calculate the first mineral partition or the second mineral partition that is an associated ore in the first image set as the number of associated ore images.

Further, in S308, the main minerals include any one of copper ore, wolframite, galena, sphalerite, iron ore, crystal ore, and nickel ore; associated minerals include rare earth ore, tantalum niobium ore, zirconium ore Any one of British Ore, Phosphate Ore, Uranium Ore, and Thorium Ore.

Further, in S500, the value range of the threshold is set to [0.2, 0.5] times the number of images of main minerals in the first image set or the threshold is set to [5, 20].

The present invention also provides a system for identifying associated minerals in a rock sample, the system comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the The computer program described runs in units of the following systems:

an image acquisition unit, used for acquiring cross-sectional images of rock sample cross-sections;

The image segmentation unit is used for preprocessing, screening and segmenting the cross-sectional image to obtain images of each sub-mining area, and each sub-mining area image constitutes a first image set;

A classification and identification unit, used for classifying and identifying images of each sub-mining area in the first image set;

An associated ore identification unit, used to calculate the number of associated ore images marked as associated ore in the first image set;

Associated ore judgment unit, used for judging that the rock sample contains associated ore when the number of associated ore images exceeds the threshold, otherwise the rock sample does not contain associated ore.

The beneficial effects of the present disclosure are as follows: the present invention provides a method and system for identifying associated minerals in rock samples, which can intelligently, quickly and accurately identify whether each rock sample contains associated minerals, so as to quickly separate and screen out the associated minerals containing associated minerals. , can improve the mining value.

Description of drawings

The above-mentioned and other features of the present disclosure will become more apparent from the detailed description of the embodiments shown in conjunction with the accompanying drawings, in which the same reference numerals refer to the same or similar elements of the present disclosure. The drawings are only some embodiments of the present disclosure. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without creative efforts. In the drawings:

Figure 1 shows a flowchart of a method for identifying associated minerals in a rock sample;

Figure 2 shows the structure diagram of a system for identifying associated minerals in a rock sample.

Detailed ways

The concept, specific structure and technical effects of the present disclosure will be clearly and completely described below with reference to the embodiments and accompanying drawings, so as to fully understand the purpose, solutions and effects of the present disclosure. It should be noted that the embodiments in the present application and the features of the embodiments may be combined with each other in the case of no conflict.

Figure 1 is a flow chart of a method for identifying associated minerals in a rock sample. The following describes a method for identifying associated minerals in a rock sample according to an embodiment of the present invention with reference to Figure 1. The method includes the following steps:

S100, collecting cross-sectional images of rock sample cross-sections;

S200, performing preprocessing, screening and segmentation on the cross-sectional images to obtain images of each sub-mining area, and each sub-mining area image constitutes a first image set;

S300, classifying and identifying images of each sub-mining area in the first image set;

S400, calculating the number of associated ore images marked as associated ore in the first image set;

S500, when the number of associated ore images exceeds a threshold, it is determined that the rock sample contains associated ore, otherwise the rock sample does not contain associated ore.

Further, in S100, the method for collecting the cross-section image of the rock sample is: collecting the image of the cross-section of the rock sample by any one of a hyperspectral camera, a hyperspectral imager, a linear CCD industrial camera, and a near-infrared light image sensor. Obtain cross-sectional images.

Further, in S100, the rock samples include porphyry copper ore, wolframite, galena, sphalerite, iron ore, crystal ore, nickel ore, rare earth ore, tantalum niobium ore, zircon ore, phosphate ore , Uranium Ore, Thorium Ore.

Further, in S200, the cross-sectional images are preprocessed, screened and segmented to obtain images of each sub-mining area, and the method for each sub-mining area image to form the first image set is as follows: Gaussian filtering and graying of the cross-sectional image to obtain a grayscale image, The grayscale image is calculated by the watershed algorithm to obtain the boundary points, and the edge lines are obtained by connecting each boundary point.

Further, in S200, the cross-sectional images are preprocessed, screened and segmented to obtain images of each sub-mining area, and the method for each sub-mining area image to form the first image set is as follows: Gaussian filtering and graying of the cross-sectional image to obtain a grayscale image, In the grayscale image, edge lines are detected by the Sobel edge detection operator, and the closed image area formed by each edge line is used as a sub-mining area image, and each sub-mining area image constitutes a first image set.

Further, in S300, the method for classifying and identifying each sub-mining area image in the first image set is:

S301, let the first image set be G1={G1 _i }, let K be the number of sub-mining images in the first image set G1, set variables i, j, i∈[1,K], G1 _i is the first image The image of the ith sub-mining area in the set; let the value of i be 1;

S302, perform corner detection on G1 _i with a corner detection algorithm to obtain each corner point of G1 _i , and sort each corner point according to the distance from the corner point to the geometric center point of G1 _i from small to large to obtain an ordered set of corner points H1= {H _j }, H _j is the jth corner point in the corner point set H1 formed by the corner points of G1 _i ; let S be the number of corner points of G1 _i , let the value of j be 1, i∈[1, S]; Described corner detection algorithm is Harris corner detection algorithm or Shi-Tomasi corner detection algorithm;

S303, (due to the different mineral regions represented by the images of the sub-mining areas distributed on the ore, and the different densities between the mineral areas result in different stress, so the shape of the image of the sub-mining areas on the ore surface is fragmented and stretched, and the directly obtained The sub-mining area image is difficult to be accurately identified for the identification of associated minerals, and the misrecognition rate is very high. Therefore, if you want to accurately extract the sub-mining area image, the following processing should be performed on the sub-mining area image to highlight the characteristics of the associated ores and main minerals) ,

Connect H _j and H _j+1 to get line segment L1, connect H _j and H _j+2 to get line segment L2, connect H _j+1 and H _j+2 to get line segment L3, take H _j as vertex and L1, L2 as edges The included angle is ∠A, the included angle with H _j+1 as vertex and L1, L3 as sides is ∠B, the included angle with H _j+2 as vertex and L2, L3 as sides is ∠C; H The line segment of the vertical line of _j on the side L3 or the line segment of the line connecting H _j to the midpoint of L3 is C1;

S304, if any one of ∠A, ∠B, ∠C is an obtuse angle, (if sampling mineral identification is performed at this time, that is, the associated minerals and main minerals in the image shape of the sub-mining area are too narrow and entangled together, if the sampled sub-mining area is If the image is too small, it is easy to collect non-mineral areas, so the position of H _j needs to be corrected). If there is a projection point on the point H _j to the side L3, let the projection point from the point H _j to the side L3 be HP _j , if the point There is no projection point on the side L3 from H _j to the midpoint of L3 as HP _j , that is, from the point H _j to the intersection of the vertical line on the side L3 and the side L3 or the midpoint of L3 as HP _j ;

The connection point HP _j to the point H _j forms a straight line C1, and the direction from the point H _j to the point HP _j is the first direction, and the direction from the point HP _j to the point H _j is the second direction;

If ∠A is not an obtuse angle, move the position of H _j along the straight line C1 to the first direction by a distance ΔL, so as to update the position coordinates of H _j , where ΔL=|Max(D2)-Min(D1)|, Wherein, Max(D2) is the largest element in the calculation set D2, and Min(D1) is the smallest element in the calculation set D1;

Sets D1 and D2 are respectively the maximum distance threshold set and the minimum distance threshold set between adjacent sampling points obtained according to steps S3041 to S3043;

S3041: Set the initial value of the variable k to 1, and set the empty sets D1 and D2;

S3042: Calculate the Euclidean distance d1 from the corner point H _k to the corner point H _k+1 in the ordered corner point set H1, the Euclidean distance d2 from the corner point H _k to the corner point H _k+2 , and the corner point H _k The Euclidean distance d3 from ₊₁ to the corner point H _k+2 ; H _k is the Kth corner point in H1; the maximum value of d1, d2 and d3 is added to the set D1, and the minimum value of d1, d2 and d3 is added to the set into set D2;

S3043: if k+2<S, increase the value of k by 1 and go to step S3042, otherwise output the obtained maximum distance threshold set D1 and minimum distance threshold set D2;

If ∠A is an obtuse angle, move the position of H _j along the straight line C1 to the second direction by a distance ΔL, thereby updating the position coordinates of H _j ;

S305, if ∠A, ∠B, and ∠C are all acute angles, (at this time, the sampling mineral identification is performed, and the sampling area is too small, which will confuse the sampling of the main mineral and the associated mineral and cause distortion, and the sampling area needs to be enlarged), let the point H _j The projected point of ₊₁ on the edge L2 is HP _j+1 , let the projected point of the point H _j+2 on the edge L1 be HP _j+2 , and connect the point HP _j+1 to the point H _j+1 to form a straight line C2, with The direction from point HP _j+1 to point H _j+1 is the third direction; connecting point HP _j+2 to point H _j+2 forms a straight line C3, and the direction from point HP _j+2 to point H _j+2 is the third direction. Four directions; move the position of H _j+1 along the straight line C2 to the third direction by a distance ΔL, thereby updating the position coordinates of H _j+1 ; move the position of H _j+2 along the straight line C3 to the fourth direction by a distance △L, thereby updating the position coordinates of H _j+2 ;

S306, the triangular area formed by the updated vertices H _j , H _j+1 and H _j+2 is recorded as the mining area to be identified;

S307, obtaining the spectral waveform data at the positions of the three vertexes H _j , H _j+1 and H _j+2 of the mining area to be identified, extracting the spectral waveform characteristics of the spectral waveform data at the three vertex positions respectively, according to the wavelength of the spectral waveform characteristic position Mineral type identification by spectral database matching;

S308, if the mineral type identified by the positions of any two of the three vertices H _j , H _j+1 and H _j+2 is the first mineral, mark the mining area to be identified as the first mineral partition; if the three vertices H The mineral type identified by the positions of any two vertices in _j , H _j+1 and H _j+2 is the second mineral, then mark the mining area to be identified as the second mineral partition; if j+2<S, increase the value of j 1 and go to step S303, otherwise go to step S309;

S309, if i<K, increase the value of i by 1 and go to step S302; otherwise, the classification and recognition of the images of each sub-mining area in the first image set ends.

Further, in S307, the method for acquiring the spectral waveform data of the mining area to be identified is as follows: acquiring the spectral waveform data through any one of a ground object spectrometer, a hyperspectral camera, a hyperspectral imager, and a near-infrared spectrometer.

Further, in S307, the method for identifying the mineral type through spectral database matching according to the wavelength of the characteristic position of the spectral waveform is:

Extract the spectral waveform features of the spectral waveform data at the positions of the three vertices, and the spectral waveform features include the first-order differential, second-order differential, wave peak, and wave trough of the spectral waveform;

Wavelength matching is performed between each spectral waveform feature and the spectral waveform feature of each mineral in the spectral database, and matching minerals that are consistent with the spectral waveform features of the spectral waveform data of the three vertices are obtained.

Further, in S307, the spectral database specifically includes the USGS spectral database, the ASD atomic spectral database, the JPL standard spectral database, the ASTER spectral database, the HIPAS spectral database, and the JHU spectral database, as well as the literature: Zhang Yingtong, Xiao Qing, Wen Jianguang , et al. Construction progress and application status of ground object spectral database [J]. Journal of Remote Sensing, 2017, 21(001): 12-26, any kind of spectral database involved in this document.

Further, in S400, the method for calculating the number of associated ore images marked as associated ore in the first image set is: respectively calculating the number Ta1 marked as the first mineral partition in the first image set and the number Ta1 marked as the first image set in the first image set. The number of second mineral divisions Ta2, when Ta1>Ta2, the first mineral is the main mineral, and the second mineral is the associated mineral; when Ta1≤Ta2, the first mineral is the associated mineral, and the second mineral is the main mineral Minerals; calculate the first mineral subregion or the second mineral subregion that is an associated mineral in the first image set as the number of associated mineral images.

Further, in S308, the main minerals include any one of copper ore, wolframite, galena, sphalerite, iron ore, crystal ore, and nickel ore; associated minerals include rare earth ore, tantalum niobium ore, zirconium ore Any one of British Ore, Phosphate Ore, Uranium Ore, and Thorium Ore.

Further, in S500, the value range of the threshold is set to [0.2, 0.5] times the number of images of main minerals in the first image set or the threshold is set to [5, 20].

An embodiment of the present disclosure provides a system for identifying associated minerals in a rock sample. FIG. 2 is a structural diagram of a system for identifying associated minerals in a rock sample of the present disclosure. In this embodiment, a system for identifying associated minerals in a rock sample is shown. The system includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, when the processor executes the computer program, an embodiment of the above-mentioned system for identifying associated minerals in a rock sample is implemented steps in .

The system includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the computer program and running in elements of the following system:

an image acquisition unit, used for acquiring cross-sectional images of rock sample cross-sections;

The image segmentation unit is used for preprocessing, screening and segmenting the cross-sectional image to obtain images of each sub-mining area, and each sub-mining area image constitutes a first image set;

A classification and identification unit, used for classifying and identifying images of each sub-mining area in the first image set;

An associated ore identification unit, used to calculate the number of associated ore images marked as associated ore in the first image set;

Associated ore judgment unit, used for judging that the rock sample contains associated ore when the number of associated ore images exceeds the threshold, otherwise the rock sample does not contain associated ore.

The system for identifying associated minerals in a rock sample can be run in computing devices such as a desktop computer, a notebook, a palmtop computer and a cloud server. In the system for identifying associated minerals in a rock sample, the operable system may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the example is only an example of an identification system for associated minerals in a rock sample, and does not constitute a limitation on an identification system for associated minerals in a rock sample, which may include more or less proportions of Components, or a combination of certain components, or different components, for example, the identification system for associated minerals in a rock sample may also include input and output devices, network access devices, buses, and the like.

The processor may be a central processing unit (Central Processing Unit, CPU), or other general-purpose processors, digital signal processors (Digital Signal Processors, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field Programmable Gate Array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The processor is the control center of the operating system of the associated mineral identification system in the one rock sample, using various interfaces and circuits Linking the associated minerals identification system throughout a rock sample operates the various parts of the system.

The memory can be used to store the computer program and/or module, and the processor implements the one by running or executing the computer program and/or module stored in the memory and calling the data stored in the memory. Various functions of associated mineral identification systems in rock samples. The memory may mainly include a stored program area and a stored data area, wherein the stored program area may store an operating system, an application program required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the storage data area may store Data (such as audio data, phonebook, etc.) created according to the usage of the mobile phone, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, internal memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device.

Although the present disclosure has been described in considerable detail and with particular reference to a few of the described embodiments, it is not intended to be limited to any of these details or embodiments or to any particular embodiment so as to effectively encompass the intended scope of the present disclosure. Furthermore, the foregoing description of the present disclosure in terms of embodiments foreseen by the inventors is intended to provide a useful description, and those insubstantial modifications of the present disclosure that are not presently foreseen may nevertheless represent equivalent modifications of the present disclosure.

Claims (9)

1. A method for identifying associated minerals in a rock sample, wherein the method comprises the following steps: S100, collecting cross-sectional images of rock sample cross-sections; S200, performing preprocessing, screening and segmentation on the cross-sectional images to obtain images of each sub-mining area, and each sub-mining area image constitutes a first image set; S300, classifying and identifying images of each sub-mining area in the first image set; S400, calculating the number of associated ore images marked as associated ore in the first image set; S500, when the number of associated ore images exceeds a threshold, it is judged that the rock sample contains associated ore, otherwise the rock sample does not contain associated ore; Among them, the method for classifying and identifying each sub-mining area image in the first image set is: S301, let the first image set be G1={G1 _i }, let K be the number of sub-mining images in the first image set G1, set variables i, j, i∈[1,K], G1 _i is the first image The image of the ith sub-mining area in the set; let the value of i be 1; S302, perform corner detection on G1 _i with a corner detection algorithm to obtain each corner of G1 _i , and sort each corner according to the distance from the corner to the geometric center point of G1 _i from small to large to obtain an ordered set of corners H1= {H _j }, H _j is the jth corner point in the corner point set H1 formed by the corner points of G1 _i ; let S be the number of corner points of G1 _i , let the value of j be 1, i∈[1, S]; Described corner detection algorithm is Harris corner detection algorithm or Shi-Tomasi corner detection algorithm; S303, connect H _j and H _j+1 to obtain line segment L1, connect H _j and H _j+2 to obtain line segment L2, connect H _j+1 and H _j+2 to obtain line segment L3, take H _j as a vertex and take L1, L2 The angle between the sides is ∠A, the angle with H _j+1 as the vertex and the sides L1 and L3 is ∠B, the angle with H _j+2 as the vertex and the sides L2 and L3 is ∠C ; The line segment of the vertical line of H _j on the side L3 or the line segment of the line connecting H _j to the midpoint of L3 is C1; S304, if any one of ∠A, ∠B, and ∠C is an obtuse angle, if there is a projection point on the point H _j to the side L3, let the projection point from the point H _j to the side L3 be HP _j , if the point H _j to the side L3 If there is no projection point on side L3, take the midpoint of L3 as HP _j , that is, take the point H _j to the intersection of the vertical line on side L3 and side L3 or take the midpoint of L3 as HP _j ; The connection point HP _j to the point H _j forms a straight line C1, and the direction from the point H _j to the point HP _j is the first direction, and the direction from the point HP _j to the point H _j is the second direction; If ∠A is not an obtuse angle, move the position of H _j to the first direction along the straight line C1 by a distance ΔL, so as to update the position coordinates of H _j , where ΔL=|Max(D2)-Min(D1)|, Wherein, Max(D2) is the largest element in the calculation set D2, and Min(D1) is the smallest element in the calculation set D1; Sets D1 and D2 are respectively the maximum distance threshold set and the minimum distance threshold set obtained according to steps S3041 to S3043; S3041: Set the initial value of the variable k to 1, and set the empty sets D1 and D2; S3042: Calculate the Euclidean distance d1 from the corner point H _k to the corner point H _k+1 in the ordered corner point set H1, the Euclidean distance d2 from the corner point H _k to the corner point H _k+2 , and the corner point H _k The Euclidean distance d3 from ₊₁ to the corner point H _k+2 ; H _k is the Kth corner point in H1; the maximum value of d1, d2 and d3 is added to the set D1, and the minimum value of d1, d2 and d3 is added to the set into set D2; S3043: if k+2<S, increase the value of k by 1 and go to step S3042, otherwise output the obtained maximum distance threshold set D1 and minimum distance threshold set D2; If ∠A is an obtuse angle, move the position of H _j along the straight line C1 to the second direction by a distance ΔL, thereby updating the position coordinates of H _j ; S305, if ∠A, ∠B, and ∠C are all acute angles, let the projection point of the point H _j+1 on the side L2 be HP _j+1 , and let the projection point of the point H _j+2 on the side L1 be HP _{j+ 2.} Connect the point HP _j+1 to the point H _j+1 to form a straight line C2, and take the direction from the point HP _j+1 to the point H _j+1 as the third direction; connect the point HP _j+2 to the point H _j+2 to form a straight line C2 Straight line C3, taking the direction from point HP _j+2 to point H _j+2 as the fourth direction; move the position of H _j+1 along the straight line C2 to the third direction by a distance ΔL, thereby updating the position of H _j+1 Coordinates; move the position of H _j+2 along the straight line C3 to the fourth direction by a distance ΔL, thereby updating the position coordinates of H _j+2 ; S306, the triangular area formed by the updated vertices H _j , H _j+1 and H _j+2 is recorded as the mining area to be identified; S307, obtaining the spectral waveform data at the positions of the three vertexes H _j , H _j+1 and H _j+2 of the mining area to be identified, extracting the spectral waveform characteristics of the spectral waveform data at the three vertex positions respectively, according to the wavelength of the spectral waveform characteristic position Mineral type identification by spectral database matching; S308, if the mineral type identified by the positions of any two of the three vertices H _j , H _j+1 and H _j+2 is the first mineral, mark the mining area to be identified as the first mineral partition; if the three vertices H The mineral type identified by the positions of any two vertices in _j , H _j+1 and H _j+2 is the second mineral, then mark the mining area to be identified as the second mineral partition; if j+2<S, increase the value of j 1 and go to step S303, otherwise go to step S309; S309, if i<K, increase the value of i by 1 and go to step S302; otherwise, the classification and recognition of the images of each sub-mining area in the first image set ends. 2 . The method for identifying associated minerals in a rock sample according to claim 1 , wherein, in S100 , the method for collecting cross-sectional images of the rock sample is: using a hyperspectral camera, a hyperspectral imager, a linear CCD Any one of an industrial camera and a near-infrared light image sensor is used to acquire a cross-section image of the rock sample to obtain a cross-section image. 3. The method for identifying associated minerals in a rock sample according to claim 1, wherein in S100, the rock sample comprises porphyry copper ore, wolframite, galena, sphalerite, iron ore, Crystal ore, nickel ore, rare earth ore, tantalum niobium ore, zircon ore, phosphate ore, uranium ore, thorium ore. 4 . The method for identifying associated minerals in a rock sample according to claim 1 , wherein in S200 , the cross-sectional images are preprocessed, screened, and segmented to obtain images of sub-mining areas, and the images of each sub-mining area constitute a first image set. 5 . The method is as follows: Gaussian filtering and graying of the cross-sectional image to obtain a grayscale image, the grayscale image is calculated by the watershed algorithm to obtain the boundary points, and each boundary point is connected to obtain an edge line, and the catchment area formed by each edge line is used as the boundary point. Sub-mining area images, each sub-mining area image constitutes a first image set. 5 . The method for identifying associated minerals in a rock sample according to claim 1 , wherein in S200 , the cross-sectional images are preprocessed, screened, and segmented to obtain images of each sub-mining area, and each sub-mining area image constitutes a first image set. 6 . The method is as follows: Gaussian filtering and graying of the cross-sectional image to obtain a grayscale image, the grayscale image is detected by the Sobel edge detection operator to obtain edge lines, and the closed image area formed by each edge line is used as the sub-mining area image, A first set of images is formed by images of the respective sub-mining areas. 6 . The method for identifying associated minerals in a rock sample according to claim 1 , wherein, in S307 , the method for obtaining spectral waveform data of the mining area to be identified is: through a ground object spectrometer, a hyperspectral camera, a hyperspectral The spectral waveform data can be acquired by any device among the imager and the near-infrared spectrometer. 7. The method for identifying associated minerals in a rock sample according to claim 1, wherein, in S307, the method for identifying mineral types by spectral database matching according to the wavelength of the spectral waveform characteristic position is: Extract the spectral waveform features of the spectral waveform data at the positions of the three vertices, and the spectral waveform features include the first-order differential, second-order differential, wave peak, and wave trough of the spectral waveform; Wavelength matching is performed between each spectral waveform feature and the spectral waveform feature of each mineral in the spectral database, and matching minerals that are consistent with the spectral waveform features of the spectral waveform data of the three vertices are obtained. 8 . The method for identifying associated minerals in a rock sample according to claim 1 , wherein, in S400 , the method for calculating the number of associated mineral images marked as associated minerals in the first image set is: respectively calculating the first The number Ta1 marked as the first mineral partition in the image set and the number Ta2 marked as the second mineral partition in the first image set, when Ta1>Ta2, the first mineral is the main mineral, and the second mineral is the associated mineral; When Ta1≤Ta2, the first mineral is the associated mineral, and the second mineral is the main mineral; the first mineral partition or the second mineral partition that is the associated mineral in the first image set is calculated as the number of associated mineral images. 9. A system for identifying associated minerals in a rock sample, wherein the system comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing The computer program runs in units of the following systems: an image acquisition unit, used for acquiring cross-sectional images of rock sample cross-sections; The image segmentation unit is used for preprocessing, screening and segmenting the cross-sectional images to obtain images of each sub-mining area, and each sub-mining area image constitutes a first image set; A classification and identification unit, used for classifying and identifying images of each sub-mining area in the first image set; An associated ore identification unit, used to calculate the number of associated ore images marked as associated ore in the first image set; Associated ore judgment unit, used to judge that the rock sample contains associated ore when the number of associated ore images exceeds the threshold, otherwise the rock sample does not contain associated ore; Among them, the method for classifying and identifying each sub-mining area image in the first image set is: S301, let the first image set be G1={G1 _i }, let K be the number of sub-mining images in the first image set G1, set variables i, j, i∈[1,K], G1 _i is the first image The image of the ith sub-mining area in the set; let the value of i be 1; S302, perform corner detection on G1 _i with a corner detection algorithm to obtain each corner of G1 _i , and sort each corner according to the distance from the corner to the geometric center point of G1 _i from small to large to obtain an ordered set of corners H1= {H _j }, H _j is the jth corner point in the corner point set H1 formed by the corner points of G1 _i ; let S be the number of corner points of G1 _i , let the value of j be 1, i∈[1, S]; Described corner detection algorithm is Harris corner detection algorithm or Shi-Tomasi corner detection algorithm; S303, connect H _j and H _j+1 to obtain line segment L1, connect H _j and H _j+2 to obtain line segment L2, connect H _j+1 and H _j+2 to obtain line segment L3, take H _j as a vertex and take L1, L2 The angle between the sides is ∠A, the angle with H _j+1 as the vertex and the sides L1 and L3 is ∠B, the angle with H _j+2 as the vertex and the sides L2 and L3 is ∠C ; The line segment of the vertical line of H _j on the side L3 or the line segment of the line connecting H _j to the midpoint of L3 is C1; S304, if any one of ∠A, ∠B, and ∠C is an obtuse angle, if there is a projection point on the point H _j to the side L3, let the projection point from the point H _j to the side L3 be HP _j , if the point H _j to the side L3 If there is no projection point on side L3, take the midpoint of L3 as HP _j , that is, take the point H _j to the intersection of the vertical line on side L3 and side L3 or take the midpoint of L3 as HP _j ; The connection point HP _j to the point H _j forms a straight line C1, and the direction from the point H _j to the point HP _j is the first direction, and the direction from the point HP _j to the point H _j is the second direction; If ∠A is not an obtuse angle, move the position of H _j to the first direction along the straight line C1 by a distance ΔL, so as to update the position coordinates of H _j , where ΔL=|Max(D2)-Min(D1)|, Wherein, Max(D2) is the largest element in the calculation set D2, and Min(D1) is the smallest element in the calculation set D1; Sets D1 and D2 are respectively the maximum distance threshold set and the minimum distance threshold set obtained according to steps S3041 to S3043; S3041: Set the initial value of the variable k to 1, and set the empty sets D1 and D2; S3042: Calculate the Euclidean distance d1 from the corner point H _k to the corner point H _k+1 in the ordered corner point set H1, the Euclidean distance d2 from the corner point H _k to the corner point H _k+2 , and the corner point H _k The Euclidean distance d3 from ₊₁ to the corner point H _k+2 ; H _k is the Kth corner point in H1; the maximum value of d1, d2 and d3 is added to the set D1, and the minimum value of d1, d2 and d3 is added to the set into set D2; S3043: if k+2<S, increase the value of k by 1 and go to step S3042, otherwise output the obtained maximum distance threshold set D1 and minimum distance threshold set D2; If ∠A is an obtuse angle, move the position of H _j along the straight line C1 to the second direction by a distance ΔL, thereby updating the position coordinates of H _j ; S305, if ∠A, ∠B, and ∠C are all acute angles, let the projection point of the point H _j+1 on the side L2 be HP _j+1 , and let the projection point of the point H _j+2 on the side L1 be HP _{j+ 2.} Connect the point HP _j+1 to the point H _j+1 to form a straight line C2, and take the direction from the point HP _j+1 to the point H _j+1 as the third direction; connect the point HP _j+2 to the point H _j+2 to form a straight line C2 Straight line C3, taking the direction from point HP _j+2 to point H _j+2 as the fourth direction; move the position of H _j+1 along the straight line C2 to the third direction by a distance ΔL, thereby updating the position of H _j+1 Coordinates; move the position of H _j+2 along the straight line C3 to the fourth direction by a distance ΔL, thereby updating the position coordinates of H _j+2 ; S306, the triangular area formed by the updated vertices H _j , H _j+1 and H _j+2 is recorded as the mining area to be identified; S307, obtaining the spectral waveform data at the positions of the three vertexes H _j , H _j+1 and H _j+2 of the mining area to be identified, extracting the spectral waveform characteristics of the spectral waveform data at the three vertex positions respectively, according to the wavelength of the spectral waveform characteristic position Mineral type identification by spectral database matching; S308, if the mineral type identified by the positions of any two of the three vertices H _j , H _j+1 and H _j+2 is the first mineral, mark the mining area to be identified as the first mineral partition; if the three vertices H The mineral type identified by the positions of any two vertices in _j , H _j+1 and H _j+2 is the second mineral, then mark the mining area to be identified as the second mineral partition; if j+2<S, increase the value of j 1 and go to step S303, otherwise go to step S309; S309, if i<K, increase the value of i by 1 and go to step S302; otherwise, the classification and recognition of the images of each sub-mining area in the first image set ends.

Images