Comprehensive quantitative discrimination method for alteration zoning of rock mass

Abstract

The invention relates to a comprehensive quantitative discrimination method for rock alteration zoning, and the method specifically comprises the following steps: obtaining a rock alteration zoning image in a region to be discriminated, converting a color image into a gray image, and building an original histogram according to the obtained image through the distribution condition of pixel values; carrying out smooth filtering on the histogram by adopting a mean filter; carrying out wavelet decomposition on the histogram by adopting secondary wavelet decomposition after smooth filtering; and obtaining an optimal threshold value according to a histogram of wavelet decomposition, and analyzing an extreme value of the histogram. The rock alteration zoning image segmentation method can greatly reduce the influence of light, noise and the like on the image segmentation precision in the aspect of rock alteration zoning image segmentation processing, thereby improving the calculation precision, has higher calculation precision and speed in the aspect of a crack width and inclination angle calculation method compared with a traditional algorithm, and is not only good in expression effect, but also high in efficiency. Moreover, the method is simple in operation in a computer, and is superior to a traditional algorithm in speed and segmentation effect.

Description

Technical Field

[0001] The present invention relates to the technical field of geological engineering, and particularly to a comprehensive quantitative discrimination method for rock mass alteration zoning. Background Art

[0002] Altered rocks are the products of hydrothermal metasomatism. The altered rocks in the fractured alteration zone are tectonites to varying degrees and are tectonic altered rocks formed by undergoing hydrothermal metasomatism to varying degrees. Since there are many types of tectonic altered rocks, according to their alteration types, protoliths and the degree of protolith fragmentation, and taking the main altered rocks as representatives, the altered rocks of various types of gold deposits are divided into zones.

[0003] Altered rocks are weakly expansive rocks with low water absorption. Their strength and deformation characteristics are jointly affected by the degree of alteration, weathering degree and rock properties, and are manifested as follows: with the increase of the degree of alteration, the compressive strength, shear strength and modulus value decrease, while the peak strain and Poisson's ratio increase; when the degree of alteration is the same, the strength value and modulus value of altered rocks are controlled by the weathering degree, and the higher the weathering degree, the lower the strength and modulus values; for altered rocks with the same degree of alteration and weathering, their strength and modulus values are affected by the properties of altered rocks.

[0004] For example, a comprehensive quantitative discrimination method for rock mass alteration zoning proposed in 201610691959.3 introduced in-situ testing technology into the discrimination of rock mass alteration zoning for the first time, solving the problems of single evaluation method, great influence of human factors and inability to quantitatively evaluate in the prior art. The testing means are advanced and the discrimination method is more novel. Compared with the traditional qualitative experience method, this method is more scientific and reasonable.

[0005] However, the prior art uses a combination of pattern recognition and acoustic wave tester for the comprehensive quantitative discrimination of rock mass alteration zoning, lacking a specific analysis of the characteristics of rock mass alteration zoning. And the geometric parameters, size and distribution law of rock mass alteration zoning directly affect the deformation, permeability and stability of rock mass, etc. Therefore, a comprehensive quantitative discrimination method for rock mass alteration zoning is proposed to solve the above problems. Summary of the Invention

[0006] Aiming at the deficiencies of the prior art, the present invention provides a comprehensive quantitative discrimination method for rock mass alteration zoning, which has the advantages of specifically analyzing the situation of rock mass alteration zoning and improving the accuracy of comprehensive quantitative discrimination, etc., and solves the problem that the prior art uses a combination of pattern recognition and acoustic wave tester for the comprehensive quantitative discrimination of rock mass alteration zoning, lacking a specific analysis of the characteristics of rock mass alteration zoning, and the geometric parameters, size and distribution law of rock mass alteration zoning directly affect the deformation, permeability and stability of rock mass, etc.

[0007] To achieve the above object, the present invention provides the following technical solution: a comprehensive quantitative discrimination method for rock mass alteration zoning, and the specific steps of the comprehensive quantitative discrimination method are as follows:

[0008] Step 1: Data acquisition

[0009] Obtain the rock mass alteration zoning image within the area to be discriminated, convert the color image into a grayscale image, and establish an original histogram based on the obtained image according to the pixel value distribution.

[0010] Step 2: Segmentation processing

[0011] 1) Use a mean filter to perform smoothing filtering on the histogram.

[0012] 2) After smoothing filtering, perform wavelet decomposition on the histogram using second-level wavelet decomposition.

[0013] 3) Obtain the optimal threshold according to the wavelet-decomposed histogram, analyze the histogram extrema, establish an extremum impulse response function, and perform peak detection.

[0014] 4) Analyze the response characteristics to determine the fracture area. Use the depth-first search algorithm to extract each connected domain of the segmented binary image, and then calculate the area of each connected region. When the area of the connected region is less than the threshold, define this region as background noise and clear it.

[0015] Step 3: Feature extraction of rock mass alteration zoning

[0016] 1) Discretely collect the rock mass alteration zoning image, count the total number of pixels in the area to obtain the total pixel area, multiply the total pixel area by the scale factor, and convert it to the area of the rock mass alteration zoning in the region.

[0017] 2) Use the Zhang-Suen thinning algorithm to extract the skeleton of the rock mass alteration zoning map, and then use the skeleton pruning algorithm to delete the noise from the skeleton to obtain the rock mass alteration zoning skeleton map.

[0018] 3) Use the outer boundary tracking algorithm to obtain the rock mass alteration zoning contour map.

[0019] 4) Calculate the width of each point on the skeleton according to the intersection points of the normal line segments and the edges through the rock mass alteration zoning skeleton map in 2). Calculate the width of each point on the rock mass alteration zoning skeleton map to obtain the average value of the fracture widths, and then the average width of the entire rock mass alteration zoning can be obtained.

[0020] 5) Gradually approximate the length of the rock mass alteration zoning by means of differential accumulation, and multiply the obtained length data by the scale factor to obtain the actual length of the rock mass alteration zoning.

[0021] 6) Solve for the tangent direction vectors of each point on the skeleton of the rock mass alteration zoning skeleton diagram in 2), then calculate the average tangent direction vector of the entire skeleton, and perform an inverse trigonometric function transformation on the obtained average tangent direction vector to obtain the dip angle of the rock mass alteration zoning.

[0022] Preferably, in the first step, a rock mass alteration zoning image within the area to be discriminated is obtained through satellite images, and the image is in the RGB565 format.

[0023] Preferably, after performing a smoothing filtering process on the histogram in the second step, peak detection is carried out to obtain the optimal threshold, and the optimal threshold obtained after the second-level wavelet decomposition is the minimum value before the second maximum peak.

[0024] Preferably, when performing peak detection by establishing an extreme value impulse response function in the second step, when the extreme value impulse response function drops to 0 for the second time, the previous minimum value is the threshold sought, and this minimum value is restored to the original histogram through the inverse second-level wavelet transform as the optimal image segmentation threshold.

[0025] Preferably, the background noise removal method is specifically as follows:

[0026] 1) Find the unselected seed points;

[0027] 2) Use the depth-first search algorithm to obtain the connected domain area S;

[0028] 3) If S > the threshold, there is a rock mass alteration zoning in this area; if S < the threshold, this area is noise, and delete this area;

[0029] 4) Determine whether there are unselected seed points on the histogram. If there are, loop back to step 1); if not, end.

[0030] Preferably, in the third step, the pixel area is regarded as a differential approximation of the actual rock mass alteration zoning area, and the pixel area calculation formula is:

[0031]

[0032] A = S × k b

[0033] where ds = k b × dx × dy, dx × dy is the area occupied by a unit pixel, k b is the scale factor, S is the total number of pixels in the area, and A is the area of the area sought.

[0034] Preferably, the steps of the Zhang-Suen thinning algorithm are as follows:

[0035] Step 1. Loop to delete all pixel points that meet the following conditions:

[0036] ①2 ≤ N(p0) ≤ 6

[0037] ②p1·p3·p5 = 0

[0038] ③p3·p5·p7 = 0

[0039] Wherein, N represents the number of pixels with a value of 1 in the domain, P represents the respective orientations of the current pixel point, and p1 - 7 represent different orientations.

[0040] Preferably, the process of the skeleton pruning algorithm is as follows:

[0041] 1) Extract the endpoints of the skeleton of the rock mass alteration zoning.

[0042] 2) Conduct an inward search on the skeleton.

[0043] 3) Determine whether a crossover point is reached. If it is reached, delete the crossover point. If the crossover point is not reached, determine whether the search length is greater than the threshold.

[0044] 4) If it is greater than the threshold, retain the branch. If the search length is less than the threshold, loop back to step 1).

[0045] Preferably, the outer boundary tracking algorithm is specifically as follows:

[0046] 1) Find the starting point of the contour. The starting search direction starts from 0 and searches clockwise until the next boundary point is found, and update the current search direction to the direction of the boundary point found.

[0047] 2) Take the newly found point as a new starting point, set the starting search direction to the direction obtained by rotating the current search direction counterclockwise by 90°, start searching clockwise from the starting search direction until the next boundary point is found, and update the current search direction to the direction of the boundary point found.

[0048] 3) Repeat step 2) until the next boundary point coincides with the starting point to obtain the contour map of the rock mass alteration zoning.

[0049] Compared with the prior art, the technical solution of this application has the following beneficial effects:

[0050] 1. The present invention can greatly reduce the influence of light, noise, etc. on the image segmentation accuracy in the image segmentation processing of the rock mass alteration zoning, thereby improving the calculation accuracy. The calculation method of the fracture width and dip angle also has higher calculation accuracy and speed compared with the traditional algorithm.

[0051] 2. Based on the histogram wavelet analysis, the present invention establishes an automatic threshold segmentation algorithm. This algorithm not only has a good performance effect, but also is simple to operate in the computer, and it is superior to the traditional algorithm in terms of speed and segmentation effect. Detailed implementation manners

[0052] Next, in combination with the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0053] Embodiment 1

[0054] A comprehensive quantitative discrimination method for rock mass alteration zoning in this embodiment is as follows specifically:

[0055] Step 1: Data acquisition

[0056] Obtain the rock mass alteration zoning image in the area to be discriminated, convert the color image into a grayscale image, and establish an original histogram according to the obtained image through the pixel value distribution.

[0057] Step 2: Segmentation processing

[0058] 1) Use a mean filter to perform smoothing filtering on the histogram;

[0059] 2) After smoothing filtering, perform wavelet decomposition on the histogram using two-level wavelet decomposition;

[0060] 3) Obtain the optimal threshold according to the wavelet-decomposed histogram, analyze the histogram extreme values, establish an extreme value impulse response function, and perform peak detection;

[0061] 4) Analyze the response characteristics to determine the fissure area. Use the depth-first search algorithm to extract each connected domain of the segmented binary image, and then calculate the area of each connected region. When the area of the connected region is less than the threshold, define this region as background noise and clear it;

[0062] Step 3: Rock mass alteration zoning feature extraction

[0063] 1) Discretely collect the rock mass alteration zoning image, count the total number of pixels in the area to obtain the total pixel area, multiply the total pixel area by the scale factor, and convert it to the area of the rock mass alteration zoning in the area class;

[0064] 2) Use the Zhang-Suen thinning algorithm to extract the skeleton of the rock mass alteration zoning map, and then use the skeleton pruning algorithm to delete the noise of the skeleton to obtain the rock mass alteration zoning skeleton map;

[0065] 3) Use the outer boundary tracking algorithm to obtain the rock mass alteration zoning contour map;

[0066] 4) Calculate the width of each point on the skeleton by calculating the intersection points of the normal line segments and the edge according to the rock mass alteration zoning skeleton diagram in 2). Calculate the width of each point on the rock mass alteration zoning skeleton diagram, and obtain the average value of the fracture widths. Then, the average width of the entire rock mass alteration zoning can be obtained.

[0067] 5) Gradually approximate the length of the rock mass alteration zoning by using the differential accumulation method, and multiply the obtained length data by the proportionality coefficient to obtain the actual length of the rock mass alteration zoning.

[0068] 6) Solve the tangent direction vector of each point on the skeleton of the rock mass alteration zoning skeleton diagram in 2), and then calculate the average tangent direction vector of the entire skeleton. Perform inverse trigonometric function transformation on the obtained average tangent direction vector to obtain the dip angle of the rock mass alteration zoning.

[0069] Example Two

[0070] In this example, in step one, the rock mass alteration zoning image within the area to be discriminated is obtained through satellite images, and the image is in the RGB565 format. In step two, after performing smoothing filtering on the histogram, peak detection is carried out to obtain the optimal threshold. After two-level wavelet decomposition, the optimal threshold is taken as the minimum value before the second maximum peak. When performing peak detection by establishing an extreme value impulse response function in step two, when the extreme value impulse response function drops to 0 for the second time, the previous minimum value is the threshold sought. This minimum value is restored to the original histogram through inverse two-level wavelet transformation as the optimal image segmentation threshold.

[0071] In this example, the filtering formula of the mean filter is specifically as follows:

[0072]

[0073] where m is the filtering kernel radius Hist old is the original histogram, and Hist new is the filtered histogram.

[0074] Example Three

[0075] In this example, the background noise removal method is specifically as follows:

[0076] 1) Find the unselected seed points;

[0077] 2) Use the depth-first search algorithm to obtain the connected domain area S;

[0078] 3) If S > the threshold, then there is a rock mass alteration zoning in this area. If S < the threshold, then this area is noise, and delete this area;

[0079] 4) Determine whether there are unselected seed points on the histogram. If there are, loop back to step 1); if not, end.

[0080] Example 4

[0081] In this example, the pixel area is regarded as a differential approximation of the actual rock mass alteration zoning area. The pixel area calculation formula is:

[0082]

[0083] A = S × k b

[0084] Where ds = k b × dx × dy, dx × dy is the area occupied by a unit pixel, and k b is the scale factor, S is the total number of pixels in the region, and A is the area of the region to be found.

[0085] In this example, the steps of the Zhang-Suen thinning algorithm are as follows:

[0086] Step 1: Loop to delete all pixel points that meet the following conditions:

[0087] ① 2 ≤ N(p0) ≤ 6

[0088] ② p1·p3·p5 = 0

[0089] ③ p3·p5·p7 = 0

[0090] Where N represents the number of pixels with a value of 1 in the neighborhood, P represents the various orientations of the current pixel point, and p1-7 represent different orientations.

[0091] In this example, the flow of the skeleton pruning algorithm is as follows:

[0092] 1) Extract the endpoints of the skeleton of the rock mass alteration zoning;

[0093] 2) Search inward along the skeleton;

[0094] 3) Determine whether a crossing point is reached. If it is reached, delete the crossing point; if not, determine whether the search length is greater than the threshold;

[0095] 4) If it is greater than the threshold, retain the branch; if the search length is less than the threshold, loop back to step 1).

[0096] In this example, the outer boundary tracking algorithm is as follows:

[0097] 1) Find the starting point of the contour. The starting search direction starts from 0 and searches clockwise until the next boundary point is found, and update the current search direction to the direction of the boundary point found;

[0098] 2) Take the newly found point as a new starting point, set the starting search direction to the direction after rotating the current search direction counterclockwise by 90°, start searching clockwise from the starting search direction until the next boundary point is found, and update the current search direction to the direction of the boundary point found.

[0099] 3) Repeat step 2) until the next boundary point coincides with the starting point to obtain the contour map of the rock mass alteration zoning.

[0100] By adopting the above technical solutions, the present invention can greatly reduce the influence of light, noise, etc. on the image segmentation accuracy in the rock mass alteration zoning image segmentation process, thereby improving the calculation accuracy. In the calculation method of the fracture width and dip angle, it also has higher calculation accuracy and speed compared with the traditional algorithm. Based on the histogram wavelet analysis, an automatic threshold segmentation algorithm is established. This algorithm not only has good performance, but also is simple to operate in the computer. It is superior to the traditional algorithm in terms of speed and segmentation effect, and solves the problem that the existing technology uses the combination of pattern recognition and acoustic wave tester to conduct comprehensive quantitative discrimination of the rock mass alteration zoning, lacking specific analysis of the characteristics of the rock mass alteration zoning, while the geometric parameters, size and distribution law of the rock mass alteration zoning directly affect the deformation, permeability and stability of the rock mass.

[0101] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0102] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principle and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A comprehensive quantitative identification method for rock mass alteration zoning, characterized in that: The comprehensive quantitative discrimination method is as follows: Step 1: Data Collection Obtain the rock mass alteration zoning image in the area to be identified, convert the color image into a grayscale image, and establish an original histogram based on the pixel value distribution of the acquired image; Step 2: Segmentation 1) Use a mean filter to smooth the histogram; 2) After smoothing filtering, the histogram is decomposed using secondary wavelet decomposition; 3) Obtain the optimal threshold value based on the histogram of wavelet decomposition, analyze the extreme value of the histogram, and establish the extreme value impulse response function to perform peak detection; 4) Analyze the response characteristics, determine the crack area, use the depth-first search algorithm to extract the connected domains of the segmented binary graph, and then calculate the area of ​​each connected area. When the area of ​​the connected area is less than the threshold, the area is defined as background noise and cleared; Step 3: Extraction of rock alteration zoning characteristics 1) Discretize and collect the rock mass alteration zoning images, count the total number of pixels in the area, obtain the total pixel area, multiply the total pixel area by the scale factor, and convert the area of ​​the regional rock mass alteration zoning; 2) The Zhang-Suen thinning algorithm is used to extract the skeleton of the rock mass alteration zoning map, and then the skeleton pruning algorithm is used to remove the noise of the skeleton to obtain the rock mass alteration zoning skeleton map; 3) Use the outer boundary tracking algorithm to obtain the contour map of rock mass alteration zones; 4) According to the rock mass alteration zone skeleton diagram in 2), the width of each point on the skeleton is calculated by the intersection point of the normal line segment and the edge. The width of each point on the rock mass alteration zone skeleton diagram is calculated, and the average width of the crack is calculated to obtain the average width of the rock mass alteration zone as a whole; 5) Use the differential accumulation method to gradually approximate the length of the rock mass alteration zone, and multiply the obtained length data by the proportional coefficient to obtain the actual length of the rock mass alteration zone; 6) For the rock mass alteration zoning skeleton diagram in 2), solve the tangent direction vector of each point on the skeleton, and then calculate the average tangent direction vector of the entire skeleton. Perform inverse trigonometric function transformation on the solved average tangent direction vector to obtain the inclination angle of the rock mass alteration zoning.

2. A comprehensive quantitative identification method for rock mass alteration zones according to claim 1, characterized in that: In the step 1, a rock alteration zoning image in the area to be identified is obtained through satellite images, and the image is in RGB565 format.

3. A comprehensive quantitative identification method for rock mass alteration zones according to claim 1, characterized in that: In the step 2, the histogram is smoothed and filtered before peak detection is performed to obtain an optimal threshold value. The optimal threshold value obtained after the secondary wavelet decomposition is the minimum value before the second maximum peak value.

4. A comprehensive quantitative identification method for rock mass alteration zones according to claim 1, characterized in that: In the step 2, when the extreme impulse response function is established to perform peak detection, when the extreme impulse response function drops to 0 for the second time, its previous minimum value is the threshold value to be found. The minimum value is restored to the original histogram through the secondary wavelet inverse transform as the optimal image segmentation threshold.

5. A comprehensive quantitative identification method for rock mass alteration zones according to claim 1, characterized in that: The background noise removal method is specifically as follows: 1) Find unselected seed points; 2) Use the depth-first search algorithm to find the area S of the connected domain; 3) If S>threshold, then the area has rock alteration zoning; if S<threshold, then the area has no noise and is deleted; 4) Determine whether there are unselected seed points on the histogram. If yes, loop to step 1), otherwise end.

6. A comprehensive quantitative identification method for rock mass alteration zones according to claim 1, characterized in that: In step 3, the pixel area is regarded as a differential approximation of the actual rock mass alteration zone area, and the pixel area calculation formula is: A=S×k b Where ds = k b ×dx×dy, dx×dy is the area occupied by a unit pixel, k b is the scale factor, S is the total number of pixels in the region, and A is the area of ​​the required region.

7. A comprehensive quantitative identification method for rock mass alteration zones according to claim 1, characterized in that: The Zhang-Suen thinning algorithm steps are as follows: Step 1: Loop and delete all pixels that meet the following conditions: ①2≤N(p0)≤6 ②p1·p3·p5=0 ③p3·p5·p7=0 Where N represents the number of pixels with a value of 1 in the area, P represents the positions of the current pixel, and p1-7 represents different positions.

8. A comprehensive quantitative identification method for rock mass alteration zones according to claim 1, characterized in that: The skeleton pruning algorithm process is as follows: 1) Extract the endpoints of the rock mass alteration zoning skeleton; 2) Search the skeleton inwards; 3) Determine whether the intersection is reached, if so, delete the intersection; if not, determine whether the search length is greater than a threshold; 4) If the search length is greater than the threshold, the side branch is retained; if the search length is less than the threshold, step 1) is repeated.

9. A comprehensive quantitative identification method for rock mass alteration zones according to claim 1, characterized in that: The outer boundary tracking algorithm is as follows: 1) Find the starting point of the contour, start the search direction from 0 and search clockwise until the next boundary point is found, and update the current search direction to the direction of the boundary point; 2) The new point found is used as a new starting point, and the starting search direction is set to the direction of the current search direction rotated 90° counterclockwise, and the search starts from the starting search direction clockwise until the next boundary point is found, and the current search direction is updated to the direction of the search to the boundary point; 3) Repeat step 2) until the next boundary point coincides with the starting point to obtain a contour map of rock alteration zones.

Images