CN104849276A - Granite three-dimensional microstructure re-building method based on pixel statistics - Google Patents
Granite three-dimensional microstructure re-building method based on pixel statistics Download PDFInfo
- Publication number
- CN104849276A CN104849276A CN201510258725.5A CN201510258725A CN104849276A CN 104849276 A CN104849276 A CN 104849276A CN 201510258725 A CN201510258725 A CN 201510258725A CN 104849276 A CN104849276 A CN 104849276A
- Authority
- CN
- China
- Prior art keywords
- mineral
- growth
- kinds
- cell element
- principle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Analysing Materials By The Use Of Radiation (AREA)
Abstract
The invention provides a granite three-dimensional microstructure re-building method based on pixel statistics. The method comprises the following steps that the geometrical information of various minerals (quartz, feldspars and mica) in a granite surface image is counted; a digital rock sample with a two-dimensional or three-dimensional microstructure is constructed by utilizing a specific mineral growth mode and the growth probability distribution principle. For the three-dimensional condition, the mineral growth modes include three modes of the point, the line and the surface, and each growth mode has different growth positions. Proper probabilities are distributed for different growth positions. In order to reflect the mutual influence of same or different kinds of mineral group growth processes, two principles including an embezzlement forbidding principle and an embezzlement allowing principle are set. When the embezzlement forbidding principle is used, the merging of the same kind of mineral groups can be avoided, and the iteration convergence can be ensured. The method provided by the invention obtains the granite surface image, and is totally realized through computer programming. Compared with a CT (computed tomography) method or a machining method, the granite three-dimensional microstructure re-building method has the advantages of high precision, high efficiency, low manufacturing cost and wide application range.
Description
Technical field
The present invention relates to three-dimensional internal structural remodeling field, specifically, the present invention relates to the three-dimensional microscopical structure reconstructing method of a kind of grouan based on pixels statistics.
Background technology
Natural rock is by media such as mineral, hole, fluid, cementing matters spatially according to certain aligned transfer.Grouan is a kind of pyrogenic rock, and its mineral mainly comprise feldspar, quartz and mica.Although can be similar on a macroscopic scale and regard uniform rock material as, on meso-scale (0.1 ~ 10.0mm), also there is the heterogeneity of material, or even anisotropy.Heterogeneity on rock meso-scale has important impact to rock macromechanics behavior (strain and stress distribution and crack propagation path etc.).
Within period quite long in the past, be limited to technical conditions and research level, often rock material is considered as a kind of uniform material, the basic mechanical parameter bringing rock is thus difficult to the problem of standard, and has emerged in large numbers a large amount of more and more complicated constitutive models.In the constitutive model of some complexity, comprise a large amount of parameters, the physical significance of some parameter and how measuring all is difficult to set forth clear, therefore, is not easy to application.The root of the problems referred to above is mainly to ignore rock mesostructure.Since late 1990s; the importance of rock mesostructure is realized gradually; by considering the heterogeneity of unit intensive parameter in some numerical models; carry out the influence research (Chen Sha that more rock mesostructure responds its macroscopic view; Yue Zhongqi; Tan Guohuan. based on the geomaterial Three-Dimensional Numerical Analysis method of true microscopical structure. Chinese Journal of Rock Mechanics and Engineering, 2006,25 (10): 1951-1959; Yu Qinglei, Yang Tianhong, Zheng is superfine. the numerical analysis that rock mesostructure affects its evolution of deformation. and rock-soil mechanics, 2011,32 (11): 3468-3472).But the microscopical structure in these models is virtual, and the microscopical structure of any rock material is not alike, does not consider the space distribution rule of multi mineral composition in rock.This virtual microscopical structure to what extent embodies the nonuniformity of rock and unclear truly.No matter analyze any material, all adopt the practice of similar virtual microscopical structure to be also difficult to people is convinced.Even someone thinks, the analysis be based upon on virtual microscopical structure is difficult to the true mechanical property obtaining material, real microscopical structure must be considered when analyzing, just can obtain targetedly and the result of correctness (Yue Zhongqi. the method for ground mesoscopic medium space distribution numeral expression and relevant mechanics numerical analysis, application and progress. Chinese Journal of Rock Mechanics and Engineering, 2006,25 (5): 875-888).
The real microscopical structure of rock material is introduced from Digital image technology introduces rock mechanics in numerical model.In the rock surface image of shooting, different mineralogical composition can be distinguished due to the difference of color, by setting up least unit (pixel) in image and associating between unit in numerical model, and set up the gray scale of pixel and associating between unit mechanics parameter further, the thought considering the microscopical structure that rock is real in numerical model is achieved.But, because captured image is all two-dimentional, so mechanical analysis can only be carried out for two-dimensional digital rock specimens, can not meeting requirements on three-dimensional research needs.
In order to obtain the three-dimensional microscopical structure of rock, the method of normal employing have two kinds (Wang Yuanyin, Ma Shaopeng, Ma Qinwei etc. small-sized slicing-type three-dimensional structure reconstruction system. Chinese invention patent, publication No.: CN 102175175 A, date of publication: 2011.09.07): 1) CT scan method; 2) machine work method.CT scan method obtains the three-dimensional internal structure of whole sample to the mode that sample successively scans by X ray.But CT scan method equipment cost, takes up an area large, need professional's operation and maintenance, identical and color or the different component of structure cannot detect X ray absorptivity, in addition, conventional CT device to obtain the spatial resolution of image lower, be difficult to the meticulous detection carrying out labyrinth.Machine work method is by successively to grind or cut sample, and the mode of successively taking pictures obtains the image on sample many levels, and and then is spliced into the three-dimensional internal structure of sample.In the precision that the advantage of the method is mainly reflected in three-dimensionalreconstruction and resolution, in device miniaturization and cost degradation, also possesses certain advantage.But, the inferior position of the method is mainly reflected in: 1) damage sample, after the three-dimensional microscopical structure obtaining sample, sample disappears, the numerical result of the mechanical behavior of sample on testing machine and sample cannot be compared, that is, make troubles to the correctness of the numerical result investigating sample, be unsuitable for the situation not allowing to damage sample or do not allow to make sample; 2) equipment operating is complicated, and the process obtaining microscopical structure wastes time and energy, and the microvibration of mechanical system, the subtle change of illumination condition etc. all can have an impact to the precision of three-dimensionalreconstruction.Such as, the thickness of once grinding or cutting is 1mm, then for removing the sample thickness of 10cm, then need processing 100 times, each processing all may cause the position of sample that small change occurs, and leaves the vestige of processing on the surface of sample, and some mineral grains even may be made to come off in advance.The more important thing is, each employing capture apparatus is taken pictures all to be needed to ensure that the distance on camera lens and sample surface is constant, and in fact, this point is difficult to accomplish, because be difficult to control capture apparatus to move forward 1mm at every turn.In addition, 1mm thickness direction can only have a pixel, instead of multiple, this can affect the precision of three-dimensionalreconstruction, and the thickness as will be once ground reduces, be then unfavorable for the efficiency of three-dimensionalreconstruction.
Summary of the invention
In order to the precision solving the existence of existing grouan three-dimensional microscopical structure reconstructing method is low, efficiency is low, price is high and the problem of damage sample, the invention provides the three-dimensional microscopical structure reconstructing method of a kind of grouan based on pixels statistics, Digital image technology is utilized to obtain the image of granite face, the regularity of distribution of statistics multi mineral composition, realize the three-dimensional microscopical structure of structure grouan accordingly, increase substantially precision and the efficiency of microscopical structure reconstruct, and significantly reduce cost.
For solving the problem, the invention provides three-dimensional microscopical structure reconstructing method (Fig. 1) of a kind of grouan based on pixels statistics, it is characterized in that, comprise: utilize Digital image technology to distinguish different minerals in the image of the granite face of shooting, obtain the statistical law of various mineralogical composition; According to above-mentioned statistical law, the mineral growth mode that utilization is specified and growth probability assignments principle there is the digital rock sample of two dimension or three-dimensional microscopical structure.
Further, wherein, the described Digital image technology that utilizes is distinguished different minerals in the image of the granite face of shooting, and the statistical law obtaining various mineralogical composition is further:
First, capture apparatus is utilized to obtain the image of granite face; Then, the rim detection in Digital image technology and partitioning algorithm is utilized to distinguish various mineralogical composition; Finally, the geological information of various mineralogical composition is added up, obtain the area percent of various mineral, arbitrary mineral group maximum radius and equivalent (equivalence) radius information.Wherein:
Described various mineralogical composition is mica, quartz and feldspar (Fig. 2).
One that described mineral group is the pixel composition that belongs to material of the same race isolated irregular collective body, the pixel of composition aggregate is connected with each other knot.Mode of connection has two kinds: line is connected to be connected with point.Line is connected and refers to two pixel share one line sections (Fig. 3-b), and point is connected and refers to that two pixels share a point (Fig. 3-a).
The area percent of described mineral is the ratio of overall pixel in the quantity sum of each pixel in certain mineral in an image and image.
The maximum radius of described mineral group is the distance of center to this center of certain mineral group's this mineral cluster centre of middle distance pixel farthest.Wherein, described mineral cluster centre rolled into a ball by mineral in the coordinate of each pixel try to achieve, specifically comprise: any one pixel is considered as a square, obtain coordinate and the area of any square center point, utilize planar graph geometric center formula to obtain the coordinate at the center of the complicated geometirc physique be made up of multiple square, be the Geometric center coordinates of these mineral group.
The equivalent radius that described arbitrary mineral are rolled into a ball is that the ratio of each elemental area sum and π in these mineral group extracts square root again.
Further, wherein, described according to above-mentioned statistical law, the digital rock sample that the mineral growth mode that utilization is specified and growth probability assignments principle there is two dimension or three-dimensional microscopical structure is further: first, need and Computing ability according to research, the size of reconstruct granite specimen is intended in setting, estimates the maximum pixel number intending holding in reconstruct sample; Afterwards, plan reconstruct sample is divided into some squares or cube cell element (Fig. 4), a cell element and a pixel or voxel (voxel) corresponding; Afterwards, by the one in 3 kinds of mineral as a setting, do not need reconstruct, select other two kinds of mineral to be reconstructed, in all cell elements respectively the some cell elements of Stochastic choice as the 1st, the seed (Fig. 5) of 2 kind of need reconstruct mineral; Afterwards, need the termination condition of statistical law and the reconstruct followed during given two kinds of mineral reconstruct, according to the growth rhythm of specifying, two kinds of mineral are grown near seed, until reconstruct terminates; Finally, reconstruction result is audited, if one or both current statistic results of mineral group reconstructed and the statistical law of setting have larger difference, current statistic result is suitably revised.Wherein: described square cell element is suitable for two-dimentional microscopical structure reconstruct situation, and cube cell element is suitable for three-dimensional situation.
The termination condition of described reconstruct comprises two kinds: single mineral group stops growing condition and sample integrative reconstruction termination condition.The single mineral group condition of stopping growing comprises: maximum radius or the equivalent radius of these mineral group have exceeded allowed scope; The termination condition of sample integrative reconstruction comprises: reach allow the maximum times of iteration, allow the maximum times of iteration not reach but iteration result area percent that is stable, two kinds of mineral groups exceeded the scope of permission.
Described growth rhythm comprises growth pattern and growth probability assignments principle, comprises two kinds: some growth and line growth (Fig. 6) for two-dimensional case growth pattern; 3 kinds are comprised for three-dimensional situation growth pattern: some growth, line growth and length (Fig. 7) of looking unfamiliar.Point growth refers to that the cell element as parent shares a point with the cell element grown; Line grows the cell element and the cell element share one line section grown that refer to as parent; Length of looking unfamiliar refers to that the cell element as parent shares a face with the cell element grown.Growth probability assignments principle refers to how between various growth pattern, to carry out probability assignments, point growth, line growth, look unfamiliar long between probability assignments can identical also can be different, in growth pattern of the same race, can be further divided into different growth orientation, they divide equally the probability of this growth pattern.A kind of simple and way efficiently can be adopted, to obtain around parent independently potential growth orientation number (Fig. 8), the growth of uniform distribution wherein probability.
It is the process of nibbling background or other mineral that described two kinds of mineral grow near seed, follows two kinds of principles: forbid occupying principle and allowing to occupy principle.Forbid that occupying principle refers to only have the cell element as a setting around a parent just to have an opportunity to become mineral (Fig. 9-a, Fig. 9-c); Allow to occupy principle and refer to that any cell element around a parent all has an opportunity to become mineral (Fig. 9-a, Fig. 9-b).Forbid occupying principle if further do not limited, adjacent mineral group of the same race can be caused to be connected (Figure 10-a, Figure 10-b), thus the size making mineral roll into a ball is undergone mutation, be unfavorable for the statistics of reconstruction result, that also mineral may be caused to roll into a ball because the size of mineral group exceedes setting range stops growing.Therefore, be necessary to limit further the background cell element around those parents, the mineral of the same race around background cell element not except parent just can become mineral (Figure 10-a, Figure 10-c).
Described examination & verification reconstruction result refers to that the geological information of two kinds of mineral in the digital granite specimen after by reconstruct is compared with the reconstruct scope of setting.If reconstruction result and reconfiguration request coincide better, then representative reconstructs successfully, if it is bad to coincide, then needs to revise one or both reconstruction result.Enable during correction and allow to occupy principle, or change the probability assignments of different growth pattern.
The three-dimensional microscopical structure reconstructing method of a kind of grouan based on pixels statistics of the present invention, the reconstruct of grouan microscopical structure can be realized, except obtaining the image of granite face, all realized by computer programming, precision is high, efficiency is high, cost is low, applicability is wide, be not limited in laboratory, even do not need to make grouan sample, only need have and for the smooth granite face of shooting image, the grouan component of duty or the microscopical structure reconstruct of grouan history relic can be particularly suitable for.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the three-dimensional microscopical structure reconstructing method of a kind of grouan based on pixels statistics of the present invention.
Fig. 2 is that (1 for mica for the distribution plan of 3 kinds of mineral in granite face image; 2 is quartz; 3 is feldspar).
Fig. 3 is the schematic diagram of two kinds of connected modes of pixel, and Fig. 3-a is a succession figure; Fig. 3-b is line succession figure.4 is pixel; 5 is the pixel of certain mineral; 6 is the common point of mineral of the same race; 7 is the common edge of mineral of the same race.
Fig. 4 is divided into the sample of cell element and the schematic diagram of cell element numbering, and Fig. 4-a is three-dimensional sample; Fig. 4-b is two-dimentional sample.
Fig. 5 is the schematic diagram (for two dimension) of the seed distribution of two kinds of reconstruct mineral.
Fig. 6 is each growth orientation schematic diagram of two-dimentional cell element, and Fig. 6-a is 4 growth orientation of a growth; Fig. 6-b is 4 growth orientation of line growth, and 8 is two-dimensional growth orientation; 9 is two-dimentional parent.
Fig. 7 is the schematic diagram in each growth orientation of three-dimensional cell element, and Fig. 7-a is 8 growth orientation of a growth; 12 growth orientation of Fig. 7-b line growth; Fig. 7-c is 6 growth orientation of looking unfamiliar long, and 10 is three dimensional growth orientation; 11 is three-dimensional parent.
Fig. 8 is the schematic diagram (for two dimension) in the potential growth orientation of a parent or mineral group, and 12 is two-dimentional potential growth orientation.
Fig. 9 is the schematic diagram (for two dimension) allowing the effect of occupying principle and forbidding occupying principle, and Fig. 9-a is the current form of mineral group; Fig. 9-b is the possibility effect allowed after occupying principle application; Fig. 9-c is a possible effect of forbidding after occupying principle application.
Figure 10 forbids occupying the drawback of principle and the schematic diagram (for two dimension) of improvement effect, and Figure 10-a is the current form of mineral group; Figure 10-b is the merging of mineral of the same race group; Figure 10-c is the effect after limiting further the background cell element in potential growth orientation.
Figure 11 is the microscopical structure of the three-dimensional granite specimen utilizing the inventive method to obtain, and Figure 11-a is the microscopical structure on sample surface; Figure 11-b and Figure 11-c is the microscopical structure of some section of sample.
Embodiment
Below in conjunction with accompanying drawing, technical scheme of the present invention is described in detail.
Three-dimensional microscopical structure reconstructing method (Fig. 1) of a kind of grouan based on pixels statistics of the present invention, comprises step:
Step 1, utilizes Digital image technology to distinguish different minerals in the granite face image obtained, obtains the statistical law of various mineralogical composition;
This step is implemented specific as follows: first, utilizes capture apparatus to obtain the image of granite face, before capturing the image, one deck water or oil can be smeared at granite face, to make the mineral color of granite face vivid, be easy to distinguish, this is conducive to follow-up statistical work; Then, the rim detection in Digital image technology and partitioning algorithm is utilized to carry out science differentiation to various mineralogical composition; Finally, the geological information of various mineralogical composition is added up, obtain the area percent of various mineral, the maximum radius of arbitrary mineral group and equivalent (equivalence) radius.
Described various mineral comprise mica, quartz and feldspar.Three kinds of colors of mineral differences, usually, mica is black, quartzy gray, and feldspar is in white, and the gray-scale value of three from small to large (Fig. 2).
One that described mineral group is the pixel composition that belongs to mineral of the same race isolated irregular collective body, it is different.In a mineral group, each pixel should be connected to each other.On image, connected mode comprises two kinds (Fig. 3-a and Fig. 3-b): line is connected and is connected with point, and line is connected and refers to two pixel share one line sections (Fig. 3-b), and point is connected refers to that two pixels share a point (Fig. 3-a).For a pixel, generally, its line is connected, the connected each own 4 kinds of situations of point.Line is connected, has the concrete situation in 4 kinds, upper and lower, left and right, upper left, upper right, lower-left, the concrete situation in 4 kinds, bottom right are connected with for point.If a pixel is positioned on the border of image, then no longer comprise some concrete situation.Described rim detection and partitioning algorithm are by Matlab software simulating.
The maximum radius of described arbitrary mineral group describes the out to out of these mineral group, should be determined, and mineral cluster centre is tried to achieve by planar graph geometric center formula by the distance of these mineral group middle distance mineral cluster centre pixel farthest to this mineral cluster centre.Arbitrary pixel can be considered as a square, like this, the aggregation in irregular shape be made up of some pixels can be considered the planar graph be made up of some squares.Any foursquare area and its center point coordinate are easy to determine, like this, the geometric center of planar graph can be determined, this center is the geometric center of mineral group.
The equivalent radius that described arbitrary mineral are rolled into a ball is that the ratio of all elemental area sums and π in these mineral group extracts square root again.That is, mineral group in irregular shape is visualized as the circle of an area equivalent, this radius of a circle is equivalent radius.The ratio of equivalent radius and maximum radius describes the form that mineral are rolled into a ball, if this ratio and 1 close, then represent the form relatively circle of mineral group, if the form of mineral group is in oval, then this ratio will be less than 1.
The area percent of certain mineral described is the number percent that the quantity of all pixels of this kind of mineral in an image accounts for the quantity of all pixels in image.This number percent is higher, then the ratio represented in grouan shared by this kind of mineral is larger.When being reconstructed the three-dimensional microscopical structure of grouan according to the statistics of two dimension, if it is then very effective area percent to be extended to percent by volume, but, this relation is also not easy to obtain, not final conclusion, this is similar to two-dimentional factor of porosity and three-dimensional apertures porosity in soil mechanics and differs comparatively large, neither one uniform expression.For this reason, the present invention, by carrying out the mode of 3 mutually perpendicular direction sections to 3-dimensional digital granite specimen rock core, spot-check the area percent of every aspect.
Step 2, according to above-mentioned statistical law, the mineral growth mode that utilization is specified and growth probability assignments principle construction have the digital rock sample of two dimension or three-dimensional microscopical structure;
This step is implemented specific as follows: first, needs and Computing ability according to research, and the size of reconstruct granite specimen is intended in setting, according to the conversion relation between pixel and length measuring unit, and the maximum pixel number can held in estimation sample; Afterwards, plan reconstruct sample is divided into the equal square of some sizes (Two-Dimensional Reconstruction) or cube (three-dimensionalreconstruction) cell element and numbers (Fig. 4-a and Fig. 4-b), a cell element and a pixel or voxel (voxel) corresponding; Afterwards, by the one cell element as a setting in 3 kinds of mineral, do not need reconstruct, such as, the feldspar cell element as a setting that content is maximum can be selected, and only mica and quartz two kinds of mineral are reconstructed, between 1 to cell element total amount, the some unduplicated random numbers of Stochastic choice are as mica and quartz mineral original position (seed) (Fig. 5); Afterwards, according to actual count result, the condition that the statistical law need followed when setting often kind of mineral reconstruct and reconstruct terminate, specifies growth pattern and the growth probability assignments principle of two kinds of mineral, makes them grow near seed, until reconstruct terminates; Finally, reconstruction result is audited, if having larger difference for the statistics of Contemporary Digital rock sample and the statistical law of setting, suitably need revise current statistic result, until satisfied.
The simple and easy step that conversion relation between described pixel and length measuring unit is determined is: before taking pictures to granite face, place at granite face or fix a steel ruler, on the image obtained, determine within 1 centimetre, to include how many pixels, like this, 1 pixel is equivalent to how many centimetres and can determines.
Described Computing ability depends on the performance of hardware and software.On the microcomputer of common configuration, the granite specimen that reconstruct has up to a million cell elements can realize.But such sample is generally difficult to carry out to be out of shape on the computing machine of common configuration, destructive process simulation.The microcomputer considering common configuration generally can only carry out mechanical process simulation to the computation model being no more than Unit 400,000, so, the size of tissue specimen intending reconstruct is unsuitable excessive, if want reconstruct one to comprise the cube granite specimen of about 400,000 unit, then on 3 orthogonal directions, the number of unit is greatly about about 74, according to document (Wang Xuebin, Du Yazhi, Pan Yishan. 2012. based on DIC thick-Strain Distribution of Uniaxial Compression sand sample of fine searching method and the experimental study [J] of strain gradient. Geotechnical Engineering journal, 34 (11): 2050-2057), the about corresponding 0.11mm of 1 pixel, like this, then the length of side of cube granite specimen is approximately 8.14mm.Such sample seems too small in actual applications, so, can select to reconstruct grouan microscopical structure on the microcomputer of common configuration, and on high configuration server, carry out the mechanical process simulation of the suitable sample in 3 directions.
Described between 1 to cell element total amount, the some unduplicated random numbers of Stochastic choice can by Matlab software simulating, if allow the random number selected to repeat, then by the seed that causes above by kind sub-covering below.
The position of described seed is the original position of mineral growth, and when there being new background cell element to be transformed into mica or quartz, seed and newly-generated mineral all will by as parents, and new mineral produce by near parent.Along with the increase of mineral size, the chance being gone out mineral by seed growth will reduce.
The termination condition of described reconstruct comprises two kinds: single mineral roll into a ball the condition that dormant condition and sample integrative reconstruction terminate.Single mineral growth to a certain extent after, once meet its condition that stops growing, by not regrowth.All mineral groups grow into a certain degree, once meet respective stop condition, sample integrative reconstruction condition will inevitably be made to be reached.Even if some mineral do not grow into maturity, the condition that sample integrative reconstruction terminates also may be reached.
The condition that described sample integrative reconstruction terminates specifically comprises: the maximum iteration time 1) reaching permission, and a computation cycles is exactly an iteration; 2) result of iteration has reached stable, although not yet reach the maximum iteration time of permission, such as, the result of front and back successive ignition no longer makes background cell element be transformed into mica or quartz; 3) area percent of the mineral of mica and quartz has exceeded allowed scope in multiple different aspect.As long as above-mentioned 3 conditions meet any one, sample integrative reconstruction just terminates.
Described unit mineral are rolled into a ball dormant condition and are specifically comprised: 1) maximum radius of these mineral group has exceeded allowed scope; 2) equivalent radius of these mineral group has exceeded allowed scope.In order to the impact on reconstruction result that the efficiency improving reconstruct is brought with the difference avoiding mineral to roll into a ball succession, in single mineral group growth course, sub-iteration is not set.Once overall iteration corresponds to a secondary growth of all mineral group, in once overall iterative process, some mineral group will obtain once substantial growth (mineral group size increases), and the size that some mineral are rolled into a ball may not change (restriction of rolling into a ball by other mineral and cannot grow).
The growth pattern of described mineral and growth probability assignments principle directly determine form and the reconstruct efficiency of each mineral group.For two-dimensional case, mineral group growth pattern comprises two kinds: some growth and line growth; For three-dimensional case, the growth pattern of mineral comprises 3 kinds: some growth, line growth and length of looking unfamiliar.Point growth refers to that the cell element as growth shares a point with the mineral grown.For two-dimensional case, a most multipotency of the cell element as parent above grows the mineral (Fig. 6-a) made new advances in 4 orientation (upper left, lower-left, upper right, bottom right); For three-dimensional case, a most multipotency of the cell element as parent above grows in 8 orientation (front upper left, front upper right, rear upper left, rear upper right, front bottom left, front bottom right, rear bottom left, rear bottom right) mineral (Fig. 7-a) made new advances.Line grows the cell element and the mineral share one line section grown that refer to as parent.For two-dimensional case, a most multipotency of the cell element as parent above grows the mineral (Fig. 6-b) made new advances 4 orientation (upper and lower, left and right); For three-dimensional case, a most multipotency of the cell element as parent grows the mineral (Fig. 7-b) made new advances in 12 orientation, and this is because 1 space cell element has 12 seamed edges.Length of looking unfamiliar is the exclusive growth pattern of three-dimensional case, and a most multipotency of the cell element as parent grows the mineral (Fig. 7-c) made new advances in 6 orientation, and this is because 1 space cell element has 6 faces.
Described growth orientation, for two-dimensional case, amounts to 8 at most; For three-dimensional case, amount to 26 at most.If a cell element as parent is positioned on the surface of granite specimen, then some growth orientation can be positioned at outside sample, so effective growth orientation is not so much.In addition, some growth orientation can occupy by parent, or to be occupied by other parents closed on, so effective growth orientation also can reduce, especially when parent develops into a certain degree or parent is surrounded by other parents time.
Described growth probability assignments principle comprises and being uniformly distributed and uneven distribution principle.One is simple and the efficient practice is (Fig. 8): the potential growth orientation of all cell elements as parent can be pooled together, reject the orientation overlapped, only obtain independently potential growth orientation number, and each potential growth orientation is numbered, then between 1 to potential growth orientation number, extract random number by being uniformly distributed principle, in an iteration, the numbered positions only corresponding to random number allows the growth of mineral.In order to construct grouan microscopical structure in different poses and with different expressions, different growth probability can be distributed in different growth patterns.Such as, for two-dimensional case, a probability for growth is appointed as 1/4, and the probability that line grows is appointed as 3/4, the growth of postulated point, line respectively has 4 to grow orientation, like this, the growth probability of any one in 4 orientation of some growth is 1/16, and the growth probability of any one in 4 orientation of line growth is 3/16, so, parent is more prone to by the growth of line growth pattern, and this will change the form of mineral group significantly.Equally, can a certain or some growth orientation on improve growth probability, like this, mineral group growth will have predominant direction.
Described mineral growth course near seed is not an isolated process, and will inevitably be subject to the impact of other mineral group, both can be mineral group of the same race, also can be xenogenesis mineral group.The growth course of certain mineral is in fact occupy, nibble or replace the process of background cell element (or matrix) or other mineral.At the initial stage of mineral growth, influencing each other between usual mineral group is less.But after mineral group develops into certain size, of the same race or xenogenesis mineral group will very closely or come in contact, and this is using occupied for some the potential growth orientation caused as the cell element of parent.Now, the growth pattern of mineral need be further limited, and has two kinds of corresponding principles can for selecting.One occupies principle for forbidding, another kind occupies principle for allowing.Forbid that occupying principle refers to that the background cell element around a parent just can have an opportunity to become mineral, that is, a background cell element can only have 1 chance to become mineral.Allow to occupy principle and refer to that any cell element around a parent all has an opportunity to become mineral, that is, the mineral attribute having become the cell element of mineral can change again.
There is continuous adjustment (Fig. 9-a and Fig. 9-b) in an iterative process in the described interface allowing the use of occupying principle that xenogenesis mineral can be made to roll into a ball, like this, iterative process is difficult to convergence, xenogenesis mineral can be caused to roll into a ball invade, certain mineral group even can be caused to be divided into some parts or part disappearance, also may to cause the merging that mineral of the same race are rolled into a ball.The appearance of above-mentioned phenomenon is unfavorable for the statistics of reconstruction result, and the preservation of the reconstruction result be unfavorable for, is also unfavorable for the convergence of iterative process.By contrast, forbid that the result of use of occupying principle generally can better (Fig. 9-a and Fig. 9-c).
Described forbidding occupies principle if further do not limited, adjacent mineral group of the same race can be caused to be connected (Figure 10-a and Figure 10-b), thus the size making mineral roll into a ball is undergone mutation, be unfavorable for the statistics of reconstruction result, that also mineral may be caused to roll into a ball because the size of mineral group exceedes setting range stops growing.Therefore, be necessary to limit further the background cell element around those parents, the mineral of the same race around background cell element not except parent just can become mineral (Figure 10-a and Figure 10-c).
Described examination & verification reconstruction result is realized by the area percent of two kinds of reconstruct mineral in selective examination different aspects, is necessary.Influence each other this is because exist in mineral growth process, these mineral made are difficult to growth, and some mineral growth cross Sheng, or iterations has been finished but has not yet reached convergence.So need the geological information of the mineral after reconstructing two kinds compared with the reconstruct scope preset, if reconstruction result and reconfiguration request coincide better, then representative reconstructs successfully, does not need to revise reconstruction result.If it is bad to coincide, then needs to revise one or both reconstruction result, usually the change of another kind of reconstruction result will be caused naturally to a kind of correction of reconstruction result.All re-construct with sample microscopical structure and compare, efficiency is undoubtedly revised to reconstruction result higher, and simple and easy to do, and the concrete practice comprises: 1) change the probability assignments scheme between different growth pattern; 2) some position some mineral seed of random arrangement again in sample, such as, at background cell element place; 3) enable and allow to occupy principle; 4) random some mineral group of rejecting certain mineral, is converted into background cell element or another kind of mineral group (when not causing mineral of the same race to roll into a ball merging).
The microscopical structure of the three-dimensional granite specimen utilizing the inventive method to obtain is shown in Figure 11-a, Figure 11-b, Figure 11-c.
Claims (5)
1. based on the three-dimensional microscopical structure reconstructing method of grouan of pixels statistics, it is characterized in that, comprising: utilize Digital image technology to distinguish different minerals in the image of the granite face of shooting, obtain the statistical law of various mineralogical composition; According to above-mentioned statistical law, the mineral growth mode that utilization is specified and growth probability assignments principle there is the digital rock sample of two dimension or three-dimensional microscopical structure.
2. the three-dimensional microscopical structure reconstructing method of the grouan based on pixels statistics according to claim 1, it is characterized in that, the described Digital image technology that utilizes is distinguished different minerals in the image of the granite face of shooting, the statistical law obtaining 3 kinds of mineralogical compositions is further: first, utilizes capture apparatus to obtain the image of granite face; Then, the rim detection in Digital image technology and partitioning algorithm is utilized to distinguish various mineralogical composition; Finally, the geological information of various mineralogical composition is added up, the maximum radius of the area percent of acquisition 3 kinds of mineral, arbitrary mineral group and equivalent (equivalence) radius information.
3. the three-dimensional microscopical structure reconstructing method of the grouan based on pixels statistics according to claim 2, it is characterized in that, described 3 kinds of mineralogical compositions are mica, quartz and feldspar; One that described mineral group is the pixel composition that belongs to mineral of the same race isolated irregular collective body, the pixel of composition aggregate is connected with each other knot, mode of connection has two kinds: line is connected to be connected with point, line is connected and refers to interior two the pixel share one line sections of mineral group, and point is connected and refers to that interior two pixels of mineral group share a point; The area percent of described mineral is the ratio of overall pixel in the quantity sum of each pixel in certain mineral in an image and image; The maximum radius of described mineral group is the distance of center to this center of certain mineral group's this mineral cluster centre of middle distance pixel farthest; Wherein, described mineral cluster centre rolled into a ball by mineral in the coordinate of each pixel try to achieve, specifically comprise: any one pixel is considered as a square, obtain coordinate and the area of any square center point, utilize planar graph geometric center formula to obtain the coordinate at the center of the complicated geometirc physique be made up of multiple square, be the Geometric center coordinates of these mineral group; The equivalent radius that described arbitrary mineral are rolled into a ball is that the ratio of each elemental area sum and π in these mineral group extracts square root again.
4. the three-dimensional microscopical structure reconstructing method of the grouan based on pixels statistics according to claim 1, it is characterized in that, described according to above-mentioned statistical law, the digital rock sample that the mineral growth mode that utilization is specified and growth probability assignments principle there is two dimension or three-dimensional microscopical structure is further: first, need and Computing ability according to research, the size of reconstruct granite specimen is intended in setting, estimates the maximum pixel number intending holding in reconstruct sample; Afterwards, plan reconstruct sample is divided into some squares or cube cell element, a cell element and a pixel or voxel (voxel) corresponding; Afterwards, by the one in 3 kinds of mineral as a setting, select other two kinds of mineral to be reconstructed, in all cell elements respectively the some cell elements of Stochastic choice as the 1st, the seed of 2 kind of need reconstruct mineral; Afterwards, need the termination condition of statistical law and the reconstruct followed during given two kinds of mineral reconstruct, according to the growth rhythm of specifying, two kinds of mineral are grown near seed, until reconstruct terminates; Finally, reconstruction result is audited, if one or both current statistic results of mineral group reconstructed and the statistical law of setting have larger difference, current statistic result is suitably revised.
5. the three-dimensional microscopical structure reconstructing method of the grouan based on pixels statistics according to claim 4, is characterized in that, described square cell element is suitable for two-dimentional microscopical structure reconstruct situation, and cube cell element is suitable for three-dimensional situation; The termination condition of described reconstruct comprises two kinds: single mineral group stops growing condition and sample integrative reconstruction termination condition, and the single mineral group condition of stopping growing comprises: maximum radius or the equivalent radius of these mineral group have exceeded allowed scope; Sample integrative reconstruction termination condition comprises: reach allow the maximum times of iteration, allow the maximum times of iteration not reach but iteration result area percent that is stable, two kinds of mineral groups exceeded the scope of permission; Described growth rhythm comprises growth pattern and growth probability assignments principle, comprises two kinds for two-dimensional case growth pattern: some growth and line growth; 3 kinds are comprised for three-dimensional situation growth pattern: some growth, line growth and length of looking unfamiliar; Described growth refers to that the cell element as parent shares a point with the cell element grown; Described line grows the cell element and the cell element share one line section grown that refer to as parent; The described long cell element face shared with the cell element grown referred to as parent of looking unfamiliar; Described growth probability assignments principle refers to how between various growth pattern, to carry out probability assignments, point growth, line growth, look unfamiliar long between probability assignments can identical also can be different, in growth pattern of the same race, can be further divided into different growth orientation, they divide equally the probability of this growth pattern; It is the process of nibbling background or other mineral that described two kinds of mineral grow near seed, follows two kinds of principles: forbid occupying principle and allowing to occupy principle; Describedly forbid that occupying principle refers to only have the cell element as a setting around a parent just to have an opportunity to become mineral; Describedly allow to occupy principle and refer to that any cell element around a parent all has an opportunity to become mineral; Described forbidding occupies principle if further do not limited, adjacent mineral group of the same race can be caused to be connected, thus the size making mineral roll into a ball is undergone mutation, be unfavorable for the statistics of reconstruction result, that also mineral may be caused to roll into a ball because the size of mineral group exceedes setting range stops growing, therefore, be necessary to limit further the background cell element around those parents, the mineral of the same race around background cell element not except parent just can become mineral; Described examination & verification reconstruction result refers to that the geological information of two kinds of mineral in the digital granite specimen after by reconstruct is compared with the reconstruct scope of setting, if reconstruction result and reconfiguration request coincide better, then representative reconstructs successfully, if it is bad to coincide, then need to revise one or both reconstruction result, enable during correction and allow to occupy principle, or change the probability assignments of different growth pattern.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510258725.5A CN104849276B (en) | 2015-05-20 | 2015-05-20 | A kind of granite three-dimensional microscopical structure reconstructing method based on pixels statistics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510258725.5A CN104849276B (en) | 2015-05-20 | 2015-05-20 | A kind of granite three-dimensional microscopical structure reconstructing method based on pixels statistics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104849276A true CN104849276A (en) | 2015-08-19 |
| CN104849276B CN104849276B (en) | 2018-07-03 |
Family
ID=53849082
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510258725.5A Expired - Fee Related CN104849276B (en) | 2015-05-20 | 2015-05-20 | A kind of granite three-dimensional microscopical structure reconstructing method based on pixels statistics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104849276B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106093035A (en) * | 2016-05-30 | 2016-11-09 | 武汉大学 | A kind of microspur video image identification method that soil body develops |
| CN108776979A (en) * | 2018-07-04 | 2018-11-09 | 安图实验仪器(郑州)有限公司 | Utilize the method for image measurement plating medium bacterial strain diameter |
| CN110487794A (en) * | 2019-09-20 | 2019-11-22 | 鞍钢集团矿业有限公司 | A kind of homogeneity recognition methods of the rock strength based on Digital image technology |
| CN111553878A (en) * | 2020-03-23 | 2020-08-18 | 四川公路工程咨询监理有限公司 | Method for detecting paving uniformity of asphalt pavement mixture based on binocular vision |
| CN112683915A (en) * | 2020-12-11 | 2021-04-20 | 中国文化遗产研究院 | Method for rapidly measuring macroscopic engineering parameters of deteriorated stone cultural relics |
| CN113284089A (en) * | 2021-04-20 | 2021-08-20 | 深圳大学 | Crack generation method based on generator, storage medium and terminal equipment |
| CN114419135A (en) * | 2022-03-29 | 2022-04-29 | 武汉楚精灵医疗科技有限公司 | Pancreas marker size quantification method, pancreas marker size quantification device, pancreas marker size quantification terminal and readable storage medium |
| CN114862783A (en) * | 2022-04-27 | 2022-08-05 | 天津大学 | Method and equipment for representing rock structural surface coincidence degree |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6516080B1 (en) * | 2000-04-05 | 2003-02-04 | The Board Of Trustees Of The Leland Stanford Junior University | Numerical method of estimating physical properties of three-dimensional porous media |
| US20110004448A1 (en) * | 2009-07-01 | 2011-01-06 | Schlumberger Technology Corporation | Method to quantify discrete pore shapes, volumes, and surface areas using confocal profilometry |
| CN102236737A (en) * | 2011-07-14 | 2011-11-09 | 西安交通大学 | Method for reconstructing micro structure finite element of multiphase material based on sequence image |
| CN102760309A (en) * | 2012-05-30 | 2012-10-31 | 合肥工业大学 | Bituminous pavement test piece X-ray CT (computed tomography) image-based aggregate mesoscopic solid model reconstruction method |
| CN103679650A (en) * | 2013-11-26 | 2014-03-26 | 四川大学 | Core three-dimension image repairing method |
| CN103698803A (en) * | 2012-09-27 | 2014-04-02 | 中国石油天然气股份有限公司 | Rock pore structure characterization method and device |
-
2015
- 2015-05-20 CN CN201510258725.5A patent/CN104849276B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6516080B1 (en) * | 2000-04-05 | 2003-02-04 | The Board Of Trustees Of The Leland Stanford Junior University | Numerical method of estimating physical properties of three-dimensional porous media |
| US20110004448A1 (en) * | 2009-07-01 | 2011-01-06 | Schlumberger Technology Corporation | Method to quantify discrete pore shapes, volumes, and surface areas using confocal profilometry |
| CN102236737A (en) * | 2011-07-14 | 2011-11-09 | 西安交通大学 | Method for reconstructing micro structure finite element of multiphase material based on sequence image |
| CN102760309A (en) * | 2012-05-30 | 2012-10-31 | 合肥工业大学 | Bituminous pavement test piece X-ray CT (computed tomography) image-based aggregate mesoscopic solid model reconstruction method |
| CN103698803A (en) * | 2012-09-27 | 2014-04-02 | 中国石油天然气股份有限公司 | Rock pore structure characterization method and device |
| CN103679650A (en) * | 2013-11-26 | 2014-03-26 | 四川大学 | Core three-dimension image repairing method |
Non-Patent Citations (2)
| Title |
|---|
| 张挺 等: "基于二维图像和多点统计方法的多孔介质三维重构研究", 《中国科学技术大学学报》 * |
| 陈沙 等: "基于真实细观结构的岩土工程材料三维数值分析方法", 《岩石力学与工程学报》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106093035A (en) * | 2016-05-30 | 2016-11-09 | 武汉大学 | A kind of microspur video image identification method that soil body develops |
| CN106093035B (en) * | 2016-05-30 | 2018-10-26 | 武汉大学 | A kind of microspur video image identification method that the soil body develops |
| CN108776979A (en) * | 2018-07-04 | 2018-11-09 | 安图实验仪器(郑州)有限公司 | Utilize the method for image measurement plating medium bacterial strain diameter |
| CN110487794A (en) * | 2019-09-20 | 2019-11-22 | 鞍钢集团矿业有限公司 | A kind of homogeneity recognition methods of the rock strength based on Digital image technology |
| CN110487794B (en) * | 2019-09-20 | 2021-10-22 | 鞍钢集团矿业有限公司 | Rock strength homogeneity degree identification method based on digital image technology |
| CN111553878A (en) * | 2020-03-23 | 2020-08-18 | 四川公路工程咨询监理有限公司 | Method for detecting paving uniformity of asphalt pavement mixture based on binocular vision |
| CN112683915A (en) * | 2020-12-11 | 2021-04-20 | 中国文化遗产研究院 | Method for rapidly measuring macroscopic engineering parameters of deteriorated stone cultural relics |
| CN113284089A (en) * | 2021-04-20 | 2021-08-20 | 深圳大学 | Crack generation method based on generator, storage medium and terminal equipment |
| CN114419135A (en) * | 2022-03-29 | 2022-04-29 | 武汉楚精灵医疗科技有限公司 | Pancreas marker size quantification method, pancreas marker size quantification device, pancreas marker size quantification terminal and readable storage medium |
| CN114419135B (en) * | 2022-03-29 | 2022-06-28 | 武汉楚精灵医疗科技有限公司 | Pancreas marker size quantification method and device, terminal and readable storage medium |
| CN114862783A (en) * | 2022-04-27 | 2022-08-05 | 天津大学 | Method and equipment for representing rock structural surface coincidence degree |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104849276B (en) | 2018-07-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104849276A (en) | Granite three-dimensional microstructure re-building method based on pixel statistics | |
| CN109087396B (en) | Mesostructure reconstruction method based on concrete CT image pixel characteristics | |
| CN108280290B (en) | Concrete aggregate numerical model reconstruction method | |
| CN102768699B (en) | Method based on CT image Accurate Reconstruction dissimilar materials microcosmic FEM mesh | |
| CN104331922B (en) | A kind of Soil Microstructure three-dimensional rebuilding method | |
| CN108305286A (en) | Multi-view stereo vision foot type method for three-dimensional measurement, system and medium based on color coding | |
| DE102016103870A1 (en) | Systems and methods for soft shading in 3D rendering | |
| CN104809756B (en) | Asphalt mixture gap spatial construction method based on X-ray CT images | |
| Barraud | The use of watershed segmentation and GIS software for textural analysis of thin sections | |
| CN102222323A (en) | Histogram statistic extension and gradient filtering-based method for enhancing infrared image details | |
| CN103473805A (en) | Method for measuring size of three-dimensional reconstruction liver model on basis of improved region growing algorithm | |
| CN106096135A (en) | A kind of grating granule library file numerical generation method based on AIMS equipment | |
| CN102779354B (en) | Three-dimensional reconstruction method for traditional Chinese medicine inspection information surface based on photometric stereo technology | |
| CN109887073B (en) | Method and device for building three-dimensional digital model of rock core | |
| CN105095581B (en) | A kind of casting shrinking hole defect image generation method | |
| CN108986024A (en) | A kind of regularly arranged processing method of laser point cloud based on grid | |
| CN110047079A (en) | A kind of optimum segmentation scale selection method based on objects similarity | |
| CN108038881B (en) | A kind of calculation method using directional light evaluation rock structural face fluctuating quantity | |
| CN114565658A (en) | Pore size calculation method and device based on CT technology | |
| CN108053415A (en) | Based on the bionical profile testing method for improving non-classical receptive field | |
| CN107590858A (en) | Medical sample methods of exhibiting and computer equipment, storage medium based on AR technologies | |
| CN110148208A (en) | A kind of pharynx nasalis radiotherapy teaching mode construction method based on Chinese Digital Human | |
| CN109191571A (en) | A method of gather materials using 3D printing technique preparation mechanical test standard | |
| CN109829879A (en) | The detection method and device of vascular bundle | |
| CN208922290U (en) | Potato image collecting device based on RGB-D camera |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180703 Termination date: 20200520 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |