CN104809362A - Method for preparing speckle pattern including heterogeneous deformation based on numerical calculation - Google Patents

Abstract

The invention provides a method for preparing a speckle pattern including heterogeneous deformation based on numerical calculation. The method comprises the following steps of acquiring strain of each unit and displacement of each node by utilizing a non-classical numerical calculation technology; building a correlation between a certain area of a unit or a surrounding node in a numerical model and a sub area in a digital image related method; mapping the position and grey value of each pixel in the sub area before deformation to a deformed sub area. The essence of the method disclosed by the invention lies in that a result of the non-classical numerical value calculation technology is taken as the result of the digital image related method so that the un-deformed speckle pattern can be mapped on a deformed grid. The method disclosed by the invention can implement all steps by utilizing computer programming and can be used for making speckle pattern including heterogeneous deformation or localized deformation at high precision and high efficiency without being limited by the type of load. The speckle pattern made by the method disclosed by the invention is mainly applied to the field of scientific researches, such as inspection on calculation precision of the digital image related method under the condition of heterogeneous deformation or even localized deformation.

Description

A kind ofly comprise inhomogeneous deformation speckle pattern method for making based on numerical evaluation

Technical field

The present invention relates to the strain measurement technique in engineering detecting, particularly a kind of high precision, high-level efficiency computer programming method for making comprising the speckle pattern of non-homogeneous and even Deformation Localization.

Background technology

Digital Image Correlation Method carries out relevant treatment, to realize the measurement of object strain field and displacement field to two width images (speckle pattern) of the body surface gathered before and after distortion.The method light path is simple, and measurement environment requires simple, can apply in the wild.Speckle can be that laser is formed, and also can be artificial speckle or some natural texture.Large to the galaxy motion in cosmic space, little of in the measurement of nano material distortion, all can see the shadow of the method application.

Speckle pattern is a kind of medium or carrier, and the distortion of object or motion are comprised on a series of speckle patterns of shooting.Digital Image Correlation Method obtains distortion and the characteristics of motion of object by related operation.The computational accuracy inspection of Digital Image Correlation Method be unable to do without the speckle pattern including distortion or motion.In the lot of documents delivered, the computational accuracy inspection of Digital Image Correlation Method generally operates in the following manner: first, to a certain amount of deflection of speckle pattern pre-add or amount of exercise; Then, then calculate by Digital Image Correlation Method, the degree differed by the deflection or amount of exercise comparing result of calculation and pre-add, evaluates the precision of Digital Image Correlation Method.In addition, in related operation process, often need some parameters, such as, the size, spacing etc. in subarea, these parameters also need to the influence research of result of calculation the speckle pattern including distortion or motion.It should be pointed out that and translation and rotation are comprised to the basic exercise of speckle pattern pre-add, stretching, compression are comprised to the basic deformation of speckle pattern pre-add or shears.These accentuation are all generally uniform, and this is suitable for computational accuracy inspection or the parameter sensitivity Journal of Sex Research of Digital Image Correlation Method under homogeneous deformation or rigid motion condition.But under loaded conditions, the distortion of object is always not uniform.For rock-soil material uniaxial compression, along with the increase of longitudinal strain, its deformation field changes from being uniformly distributed to non-uniform Distribution, until there is deformation localization phenomenon, after this just occurs macroscopic cracking.Shear deformation localization region is called shear zone.The distribution of shear zone internal strain is very uneven.Nearest research finds, outside the shear zone of sand material, the distribution of strain is also uneven (Wang Xuebin, Du Yazhi, Pan Yishan. the Digital Image Correlation Method observation of uniaxial compression damp sand sample local and overall volume strain. Geotechnical Engineering journal, 2014,36 (9): 1648-1656; Wang Xuebin, Du Yazhi, Pan Yishan, etc. based on Digital Image Correlation Method etc. different water cut sand sample shear zone observation under rate of strain. rock-soil mechanics, 2015,36 (3): 625-632).Strain field under non-homogeneous and even Deformation Localization condition and displacement field is measured exactly in order to adopt Digital Image Correlation Method, just must understand Digital Image Correlation Method computational accuracy with this understanding, and the acquisition of this precision be unable to do without the speckle pattern comprising non-homogeneous and even Deformation Localization.At present, the method for making also rare report of the speckle pattern of non-homogeneous and even Deformation Localization is comprised.Some simple functions of more existing method for making Main Basiss make the speckle pattern under inhomogeneous deformation conditions, do not have ubiquity and applicability widely.Some documents adopt the theory of shear zone deformation field answer to make the speckle pattern (Wang Xuebin comprising Deformation Localization, Du Yazhi, Pan Yishan. consider the comparison of the Digital Image Correlation Method of single order and Second Order Displacements gradient in shear zone is measured. engineering mechanics, 2013,30 (7): 282-287), be only applicable to simple shear condition, the speckle pattern comprising inhomogeneous deformation outside shear zone cannot be made.In fact, the theory answer of existing shear zone deformation field is very limited, due to the complex nature of the problem, even if to cut in pressure, under the relatively simple load-up condition such as direct shear, Biaxial Compression, the acquisition of the theory answer of shear zone internal strain field is also extremely difficult.

The invention provides a kind of high precision, the high-level efficiency computer programming method for making that comprise the speckle pattern of non-homogeneous and even Deformation Localization.Essence of the present invention is: using the result of the result of non-classical numerical computation technology as Digital Image Correlation Method, realizes undeformed speckle pattern to be mapped on the grid after distortion.The method, owing to introducing non-classical numerical computation technology, no longer needs any theory about shear zone deformation field to answer, and utilizes the method to make and comprises speckle pattern that is non-homogeneous and even Deformation Localization under various simple load and combined load condition.

Summary of the invention

Under non-homogeneous and even Deformation Localization condition, in order to make the speckle pattern for the inspection of Digital Image Correlation Method computational accuracy and calculating parameter sensitivity study, the invention provides and a kind ofly comprise inhomogeneous deformation speckle pattern method for making based on numerical computation technology, the result of non-classical numerical computation technology is by the result as Digital Image Correlation Method, by to set up in numerical model unit or surround the certain area of node and associating of subarea in Digital Image Correlation Method, acquisition comprises speckle pattern that is non-homogeneous and even Deformation Localization.The present invention utilizes computer programming to realize Overall Steps, substantially increase precision and the efficiency of the speckle pattern making comprising non-homogeneous and even Deformation Localization, and, not by the restriction of load type, be applicable to simple load type (such as, stretch, compression and shearing etc.), be also applicable to combined load type (such as, draw cut and pressure is cut).

For solving the problem, the invention provides a kind of non-homogeneous speckle pattern method for making based on numerical computation technology, it is characterized in that, comprising:

Utilize non-classical numerical computation technology, obtain the strain of each unit and the displacement of each node;

To set up in numerical model unit or surround the certain area of node and associating of subarea in Digital Image Correlation Method;

The position of each pixel in the front subarea of distortion and gray-scale value are mapped on the subarea after distortion.

Described utilize non-classical numerical computation technology to obtain each unit strain and the displacement of each node be further:

In the yield function of the elastic and plastic strain softening constitutive model, introduce the second order term of Internal length and strain gradient, create two-dimension numerical calculation model, obtain the strain of each unit and the displacement of each node.

Further, wherein,

Described two-dimension numerical calculation model is made up of some square shaped cells, and be connected by node between unit, unit size equals subarea size.

Further, wherein,

Described yield function f has following form:

$f\left(\mathrm{\σ},\stackrel{\‾}{k}\right)=F\left(\mathrm{\σ}\right)-h\left(\stackrel{\‾}{k}\right)---\left(1\right)$

Wherein, F (σ) is yield stress index, for softening rule, for non local softening parameter, following form can be written as:

$\stackrel{\‾}{k}=k+c{\▿}^{2}k---\left(2\right)$

Wherein, k is local softening parameter, and c is the material parameter proportional with material internal length parameter, for Laplace operator, under two-dimensional condition, its expression formula is:

${\▿}^2=\frac{{\∂}^2}{\∂x^2}+\frac{{\∂}^2}{\∂y^2}---\left(3\right)$

Further, wherein,

The parameter relevant with length is introduced in described yield function, utilize finite element or finite difference grade numerical computation technology solve time, the width of localization band will be now uniquely determined, the phenomenon (i.e. Grid component) that the different result of calculation of grid is different will disappear, and uniqueness and the objective performance of numerical result are protected.

The described certain area setting up unit or encirclement node in numerical model with associating of subarea in Digital Image Correlation Method is further:

Obtain the displacement of subarea central point, or directly obtained by numerical computation technology, or be averaged acquisition by the displacement result of 4 nodes of 1 unit; Obtain the strain in subarea, or directly obtained by numerical computation technology, or be averaged acquisition by the strain result of some unit, or the displacement field directly obtained by numerical technique is obtained by centered Finite Difference Methods.

Further, wherein,

The concrete formula that the described displacement result by 4 nodes of 1 unit is averaged the displacement obtaining subarea central point is:

$u=\frac{1}{4}{\Σ}_{i=1}^4{u}_i,v=\frac{1}{4}{\Σ}_{i=1}^4{v}_i,---\left(4\right)$

Wherein, u, v are respectively the displacement of subarea central point in level and vertical direction.

Further, wherein,

The concrete formula that the described strain result by some unit is averaged the strain obtaining subarea is:

${\mathrm{\ϵ}}_x=\frac{1}{n}{\Σ}_{i=1}^n{\mathrm{\ϵ}}_{xi},{\mathrm{\ϵ}}_y=\frac{1}{n}{\Σ}_{i=1}^n{\mathrm{\ϵ}}_{yi},{\mathrm{\γ}}_{xy}=\frac{1}{n}{\Σ}_{i=1}^n{\mathrm{\γ}}_{xyi}---\left(5\right)$

Wherein, ε _x, ε _y, γ _xybe respectively the horizontal line strain in subarea, perpendicular line strain and shear strain, n is the number of unit around the central point of subarea.When the central point in subarea is positioned at the inside of computation model, n=4; When the central point in subarea is positioned on the border of computation model, n=1 or 2.The central point that n=1 represents subarea is taken at 4 jiaos of computation model, and in this special case, the unit around the central point of subarea only has 1, and the strain in subarea is the strain of this unit.

Further, wherein,

The described concrete formula obtaining strain field by centered Finite Difference Methods by displacement field is:

$\left\{\begin{array}{c}{\mathrm{\ϵ}}_x(i,j)=\frac{u(i,j+1)-u(i,j-1)}{2\mathrm{\Δ}}\\ {\mathrm{\ϵ}}_y(i,j)=\frac{v(i+1,j)-v(i-1,j)}{2\mathrm{\Δ}}\\ {\mathrm{\γ}}_{xy}(i,j)=\frac{v(i,j+1)-v(i,j-1)}{2\mathrm{\Δ}}+\frac{u(i+1,j)-u(i-1,j)}{2\mathrm{\Δ}}\end{array}\right.---\left(6\right)$

Wherein, Δ is the distance between node, and (i, j) represents the i-th row, jth row.

Described being mapped on the subarea after distortion in the position of each pixel in the front subarea of distortion and gray-scale value is further:

Obtain position and the gray-scale value of each pixel in the front subarea of distortion, obtain the position of each pixel in the rear subarea of distortion, the gray-scale value of each pixel in the front subarea of distortion is transformed on the subarea after distortion.

Further, wherein,

Described the concrete methods of realizing that the gray-scale value of each pixel in the front subarea of distortion transforms on the subarea after distortion can be classified as two classes: Direct mapping and back mapping.

Further, wherein,

Described Direct mapping is by circulating successively to each pixel in the front subarea of distortion, and obtain its position after deformation in subarea, concrete formula is:

$\left\{\begin{array}{c}{x}^{\′}=x+u+\frac{\∂u}{\∂x}\mathrm{\Δ}x+\frac{\∂u}{\∂y}\mathrm{\Δ}y\\ y^{\′}=y+v+\frac{\∂v}{\∂x}\mathrm{\Δ}x+\frac{\∂v}{\∂y}\mathrm{\Δ}y\end{array}\right.---\left(7\right)$

Wherein, Q (x, y) is the coordinate of arbitrary pixel center point in the front subarea of distortion, Q'(x', y') be the coordinate in this point after deformation subarea, u, v are respectively subarea center point P (x ₀, y ₀) displacement (conversion relation between the metric system and pixel need be obtained, and then convert this metric system displacement to pixel displacement) in level and vertical direction, u=x' ₀– x ₀, v=y' ₀– y ₀, Δ x, Δ y be respectively distortion front subarea in arbitrary pixel center point Q (x, y) to subarea center point P (x ₀, y ₀) horizontal and vertical distance (Fig. 1-a and Fig. 1-b).Various partial derivative describes the distortion in subarea, if the position calculated is positioned on the central point of a certain pixel just, obtain the gray-scale value of this pixel, in most cases, the position calculated will be dropped in sub-pixel location, now, interpolation is carried out in reply gray scale field, is transformed on the subarea after distortion by the gray-scale value of each pixel in the front subarea of distortion.

Further, wherein,

Described back mapping be by after distortion in subarea each pixel circulate successively, obtain its position before being deformed in subarea.The formula of back mapping is:

$\left\{\begin{array}{c}x=x^{\′}-u-\frac{\∂u}{\∂x}\mathrm{\Δ}x-\frac{\∂u}{\∂y}\mathrm{\Δ}y\\ y=y^{\′}-v-\frac{\∂v}{\∂x}\mathrm{\Δ}x-\frac{\∂v}{\∂y}\mathrm{\Δ}y\end{array}\right.---\left(8\right)$

Wherein, Q'(x', y') for being out of shape the coordinate of arbitrary pixel center point in rear subarea, Q (x, y) is the coordinate in this point before being deformed subarea.

Further, wherein said back mapping before proceeding, needs the coordinate calculating 4 angle points in the rear subarea of distortion.

Of the present inventionly a kind ofly comprise non-homogeneous speckle pattern method for making based on numerical computation technology, the result of non-classical numerical computation technology is by the result as Digital Image Correlation Method, by to set up in numerical model unit or surround the certain area of node and associating of subarea in Digital Image Correlation Method, acquisition comprises speckle pattern that is non-homogeneous and even Deformation Localization.The present invention utilizes computer programming to realize Overall Steps, substantially increase precision and the efficiency of the speckle pattern making comprising non-homogeneous and even Deformation Localization, and, not by the restriction of load type, be applicable to simple load type (such as, stretch, compression and shearing etc.), be also applicable to combined load type (such as, draw cut and pressure is cut).

Accompanying drawing explanation

Fig. 1-a is the schematic diagram in the front subarea of distortion in Digital Image Correlation Method, and Fig. 1-b is the schematic diagram in subarea after distortion;

In figure, 1 is the image in Digital Image Correlation Method before distortion, and 2 is a subarea before distortion in image, P (x ₀, y ₀) for being out of shape the central point in front subarea, Q (x, y) is the central point of arbitrary pixel in the front subarea of distortion, 3 is the image in Digital Image Correlation Method after distortion, and 4 is a subarea after distortion in image, P'(x' ₀, y' ₀) be the central point in subarea after distortion, Q'(x', y') be any point in subarea after distortion, Δ x, Δ y to be respectively in the front subarea of distortion arbitrary pixel center point Q (x, y) to subarea center point P (x ₀, y ₀) horizontal and vertical distance, u, v are respectively subarea center point P (x ₀, y ₀) displacement in level and vertical direction;

Fig. 2 is the schematic diagram calculation of the embodiment of the present invention one;

In figure, 5 is mathematical calculation model, and 6 is a unit in mathematical calculation model, its size and subarea measure-alike, 7 is the node of unit in numerical model, the displacement of the arrow representation node of Nodes, and 8 is the central point of unit in numerical model;

Fig. 3-a is grid chart before the distortion of numerical model, and Fig. 3-b and Fig. 3-c is different speckle pattern;

When Fig. 4-a is longitudinal strain=0.025 numerical model distortion after grid chart, when Fig. 4-b is longitudinal strain=0.0375 numerical model distortion after grid chart, when Fig. 4-c is longitudinal strain=0.05 numerical model distortion after grid chart, when Fig. 4-d is longitudinal strain=0.0625 numerical model distortion after grid chart;

Fig. 5-a, Fig. 5-b, Fig. 5-c, Fig. 5-d are respectively the result figure on the grid that to be transformed to by Fig. 3-b in Fig. 4-a, Fig. 4-b, Fig. 4-c, Fig. 4-d;

Fig. 6-a, Fig. 6-b, Fig. 6-c, Fig. 6-d are respectively the result figure on the grid that to be transformed to by Fig. 3-c in Fig. 4-a, Fig. 4-b, Fig. 4-c, Fig. 4-d;

Fig. 7 is the schematic diagram calculation of the embodiment of the present invention two;

In figure, 9 is the certain area surrounding node in mathematical calculation model, its size and subarea measure-alike.

Embodiment

Below in conjunction with accompanying drawing, technical scheme of the present invention is described in detail.

Two kinds of specific embodiments are adopted to be described, the difference of two schemes is that the front subarea of distortion 2 in Digital Image Correlation Method (its ultimate principle is shown in Fig. 1-a and Fig. 1-b) is different from the corresponding region in numerical model 5, a unit 6 in the corresponding numerical model in the front subarea of distortion in embodiment one in Digital Image Correlation Method 2, be out of shape the center point P (x in front subarea 2, y) central point 8 of unit 6 in corresponding numerical model, the certain area 9 of the encirclement node in the corresponding numerical model in the front subarea of the distortion in embodiment two in Digital Image Correlation Method 2, be out of shape the center point P (x in front subarea 2, y) node 7 of unit 6 in corresponding numerical model.

As shown in Figure 2, be the schematic diagram calculation of the embodiment of the present invention one, comprise step:

Step 1: utilize non-classical numerical computation technology, obtains the strain of each unit and the displacement of each node.

This step is implemented specific as follows: first, the second order term of Internal length and strain gradient is introduced in the yield function (formula (1-3)) of the elastic and plastic strain softening constitutive model, to avoid the mesh dependence of numerical result, guarantee objectivity and the uniqueness of numerical result; Then, create two-dimension numerical calculation model 5 (Fig. 3-a), the square shaped cells length of side is 5mm, parameter in the boundary condition of given computation model, loading environment and constitutive model, obtains the strain of each unit 6 and the displacement (Fig. 4-a, Fig. 4-b, Fig. 4-c and Fig. 4-d) of each node 7.

Step 2: set up associating of unit and subarea in Digital Image Correlation Method in numerical model.

This step is specifically implemented as follows: first, and according to the displacement of node 7 each in numerical model 5, obtain the displacement of each unit 6 central point 8 according to formula (4), the strain of each unit is obtained by the first step; Then, the second best in quality speckle pattern that selection and calculation moulded dimension is consistent, as the speckle pattern 1 before distortion, speckle pattern can according to Some Related Algorithms manual manufacture, also can the rock or other materials surface with native texture be taken pictures, by intercepting a part as speckle pattern (Fig. 3-b and Fig. 3-c), the size in subarea 2 equals the size of unit 6, is 11 × 11 pixels; Finally, the central point 8 of unit is appointed as the center point P (x being out of shape front subarea 2, y), the size of unit 6 is appointed as the size of being out of shape front subarea 2, the front subarea 2 center point P (x of distortion is appointed as in the displacement of unit center point 8, y) displacement, the strain of being out of shape front subarea 2 is appointed as in the strain of unit 6.

Step 3: the gray-scale value of each pixel in the front subarea of distortion is mapped on the subarea after distortion.

The embodiment of this step is as follows: first, calculates the position of 4 angle points in the rear subarea of distortion according to the displacement of each node 7 of unit in numerical model 5; Then, the gray-scale value of each pixel Q (x, y) in the subarea 2 in image 1 before distortion is transformed on the subarea 4 in the rear image 3 of distortion.

The specific implementation method of said process has two kinds: Direct mapping and back mapping.Direct mapping is each pixel Q (x in order in the front subarea of scanning distortion 2, y), then its position after deformation in subarea 4 is calculated according to formula (7), by interpolation, the gray-scale value of each pixel Q (x, y) in the subarea 2 in image 1 before distortion is transformed on the subarea 4 in the rear image 3 of distortion.Now, the number of the pixel in unit length need be obtained, the conversion relation namely between the metric system and pixel.If the position calculated is positioned on the central point of a certain pixel just, obtain the gray-scale value of this pixel, in most cases, the position calculated will be dropped in sub-pixel location, and now, interpolation is carried out in reply gray scale field.Back mapping is each pixel Q'(x' in subarea 4 after scanning distortion in order, y'), its position before being deformed in subarea 2 is calculated according to formula (8), by interpolation by each pixel Q (x in the subarea 2 in image 1 before distortion, y) gray-scale value transforms on the subarea 4 in the rear image 3 of distortion, and result is as shown in Fig. 5-a, Fig. 5-b, Fig. 5-c, Fig. 5-d, Fig. 6-a, Fig. 6-b, Fig. 6-c, Fig. 6-d.

As shown in Figure 7, be the schematic diagram calculation of the embodiment of the present invention two, comprise step:

Step 1: identical with the step 1 of embodiment one.

Step 2: set up in numerical model associating of the certain area that surrounds node and subarea in Digital Image Correlation Method.

This step is specifically implemented as follows: first, and according to the encirclement element number of any node and the strain of each unit, obtain the strain in the region 9 surrounding any node according to formula (5), the displacement of each node is obtained by the first step; Secondly, the second best in quality speckle pattern that selection and calculation moulded dimension is consistent, as the speckle pattern 1 before distortion; Finally, the position of node 7 is appointed as the center point P (x being out of shape front subarea 2, y), the size of surrounding the region 9 of node is appointed as the size in subarea 2, subarea 2 center point P (x is appointed as in the displacement of node 7, y) displacement, the strain in subarea 2 is appointed as in the strain of surrounding the region 9 of any node.The another kind of method obtaining the strain in the region surrounding any node carries out central difference according to formula (6) to the displacement field of node, like this, just need not average to the strain of the some unit surrounding any node.

Step 3: identical with the step 3 of embodiment one.

Should be understood that, because the unit number of surrounding node on border may be 1 or 2, like this, the size of surrounding the region of node on model boundary may for surrounding the half or 1/4th of the size in the region of model internal node, and this can make mapping process become complicated a little.In order to avoid above-mentioned defect, in mapping process, can consider only to operate the region that those surround the node of model inside, the node on model boundary not operated.Like this, the speckle pattern after the distortion after mapping is by a circle less of the speckle pattern before distortion.This processing mode also has another kind of advantage: centered Finite Difference Methods is easily implemented, otherwise, carry out central difference again after needing the positional displacement interpolation to the node on model boundary.

By contrast, embodiment one is most preferred embodiment, can ensure that numerical model is intactly mapped to speckle pattern, and it is convenient to implement.

Of the present inventionly a kind ofly comprise non-homogeneous speckle pattern method for making based on numerical computation technology, using the result of non-classical numerical computation technology as the result of Digital Image Correlation Method, by to set up in numerical model unit or surround the certain area of node and associating of subarea in Digital Image Correlation Method, obtain and comprise speckle pattern that is non-homogeneous and even Deformation Localization, the main application of this speckle pattern is field of scientific study: under non-homogeneous and even Deformation Localization condition, the computational accuracy inspection of Digital Image Correlation Method and calculating parameter sensitivity study, in addition, also can be used for the objects such as decoration or image display.The present invention utilizes computer programming to realize Overall Steps, substantially increases precision and the efficiency of the speckle pattern making comprising non-homogeneous and even Deformation Localization, and, not by the restriction of load type, be widely used.

Claims (9)

1. based on the speckle pattern method for making comprising inhomogeneous deformation of numerical computation technology, it is characterized in that, comprising: utilize non-classical numerical computation technology, obtain the strain of each unit and the displacement of each node; To set up in numerical model unit or surround the certain area of node and associating of subarea in Digital Image Correlation Method; The position of each pixel in the front subarea of distortion and gray-scale value are mapped on the subarea after distortion.

2. according to claim 1ly a kind ofly comprise inhomogeneous deformation speckle pattern method for making based on numerical computation technology, it is characterized in that, describedly utilize non-classical numerical computation technology, obtain the strain of each unit and the displacement of each node, be further divided into: the second order term introducing Internal length and strain gradient in the yield function of the elastic and plastic strain softening constitutive model, create two-dimension numerical calculation model, obtain the strain of each unit and the displacement of each node.

3. two-dimension numerical calculation model according to claim 2 is made up of some square shaped cells, and be connected by node between unit, unit size equals subarea size.

4. Internal length according to claim 2 determines size and the energy dissipation of localization band, and the Grid component of the numerical result utilizing finite element or finite difference to obtain will disappear, and uniqueness and objective performance are protected.

5. according to claim 1ly a kind ofly comprise inhomogeneous deformation speckle pattern method for making based on numerical computation technology, it is characterized in that, describedly to set up in numerical model unit or surround the certain area of node and associating of subarea in Digital Image Correlation Method, be further divided into: the displacement obtaining subarea central point, or directly obtained by numerical computation technology, or be averaged acquisition by the displacement result of 4 nodes of 1 unit; Obtain the strain in subarea, or directly obtained by numerical computation technology, or be averaged acquisition by the strain result of some unit, or the displacement field directly obtained by numerical technique is obtained by centered Finite Difference Methods.

6. according to claim 1ly a kind ofly comprise inhomogeneous deformation speckle pattern method for making based on numerical computation technology, it is characterized in that, described the position of each pixel and gray-scale value in the front subarea of distortion are mapped on the subarea after distortion, be further divided into: the position and the gray-scale value that obtain each pixel in the front subarea of distortion, obtain the position of each pixel in the rear subarea of distortion, the gray-scale value of each pixel in the front subarea of distortion is transformed on the subarea after distortion.

7. according to claim 7 the concrete methods of realizing that the gray-scale value of each pixel in the front subarea of distortion transforms on the subarea after distortion can be classified as two classes: Direct mapping and back mapping.

8. Direct mapping according to claim 8, by circulating successively to each pixel in the front subarea of distortion, obtain its position after deformation in subarea, by interpolation, the gray-scale value of each pixel in the front subarea of distortion is transformed on the subarea after distortion, if the position calculated is positioned on the central point of a certain pixel just, obtain the gray-scale value of this pixel, in most cases, the position calculated will be dropped in sub-pixel location, and now, interpolation is carried out in reply gray scale field.

9. back mapping according to claim 8, by circulating successively to each pixel in subarea after distortion, obtain its position before being deformed in subarea, by interpolation, the gray-scale value of each pixel in the front subarea of distortion is transformed on the subarea after distortion, before carrying out back mapping, need the coordinate obtaining 4 angle points in the rear subarea of distortion.