WO1993021594A1

WO1993021594A1 - Method of recognition of stereoscopic image

Info

Publication number: WO1993021594A1
Application number: PCT/KR1992/000012
Authority: WO
Inventors: Jae Moon Chung
Original assignee: Jae Moon Chung
Priority date: 1992-04-10
Filing date: 1992-04-10
Publication date: 1993-10-28

Abstract

This invention relates to a technical field of computer stereovision. To recognize a stereoscopic image, the present invention relates to a novel method of interpretation and description of stereoscopic profile data of a certain article obtained from two directions through sensors. To accomplish this object, the present invention postulates that the stereoscopic image is constituted in such a manner that a certain cut plane of the object article is expressed by a disk assumed as a specific portion of a specific sphere which is in turn assumed to exist inside this article. On the basis of this assumption, the present invention utilizes the characteristics of the sphere and disk, sets and interprets the relation of correspondence between the two profile data, and describes the object article in the form of an aggregate of disks defined completely in the space coordinates. The present invention can be applied to apparatus for recognizing, restoring and inspecting stereoscopic images. Particularly because only the data on the profile line of the object article is necessary for the input signal, pre-processing is easy. Therefore, the present invention has the advantage that its application is easy.

Description

Statement)

Recognition method of three-dimensional figure

Technical field

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a new method for recognizing stereoscopic contours of an object from two directions of the object through sensors to obtain stereoscopic contours of the object.

This is mainly related to Computer Stereo Vision, which is a research field of 3D image recognition through both eyes, and is written by BKRHorn, “Robot Bottom Vision” (Rob.ot. Vision, MIT Press According to, 1986), the biggest challenge in this field of research is to find the correspondence between a certain point in one video signal and the other's video signal. However, it has mainly relied on the interpretation of the brightness between a pair of video signals on the surface of the target object. Due to difficulties in interpretation, it has fundamental difficulties in practical application.

Disclosure of the invention

In the present invention, the understanding of the three-dimensional shape of an object through both human eyes Although the contour linear elephant information of the object is one of the elements in the elements for recognition and recognition, it is relatively easy to preprocess to extract it from an engineering point of view. We have come to note that is a relatively stable element for. Therefore, the present invention proposes a new geometrical method of searching for correspondence between a pair of signals of the stereoscopic contour linear elephant if¾ in order to recognize a three-dimensional image of an object. This method can be applied not only to a video recognition system (System) through a vision sensor, but also to any device that can provide contour linear elephant information for an object. As described above, a new interpretation method for the stereo contour linear elephant information and a new description method for expressing the three-dimensional image are also presented, so that the present invention is actually a useful methodology for recognizing the three-dimensional image. Show. The present invention provides a method of reconstructing a stereoscopic contour elephant ff¾ obtained from two directions of an object through a sensor for the recognition of the three-dimensional image. Each of which is defined as a particular part of a particular sphere (88) that can be assumed to be inside the object, as represented by a Disk (90). We start from the assumption that a three-dimensional figure is made.

Based on the above assumptions, we apply a Symmetric Axis Transform to each of the two contour linear ele- ments captured by each sensor coordinate. For the characteristics of the symmetry axis transformation, see "Evaluation of Shape Representation ¾ί (Criteria ior representations of shape.Human and Machine" ision, ed. According to the thesis, The application of the symmetry axis transformation to a contour linear elephant (10) as shown in Fig. 1 results in A contour having the contour (10) and two or more contact points such as (1 ² ) and (). The center of the maximum disk (Maximal Disk) (l ⁸ ) inside the linear elephant (creates a set of 1) Such a collection of center points is the symmetry axis (20) for the relevant contour obtained through the symmetry axis transformation, and was originally one of the means of expression (descriptor) for the plane figure. According to H. Blum, "A transform ior extracting new descriptor of shape.Models for perception of speech and visual form, ed.W. \ Vathen- Dunn, MIT press, 1967" Various methods are known for calculating this symmetry transformation, one of which is to be adapted to the application to signals through a digital sensor. Voronoi for points represented by a set of points on the contour The fast approximation method based on the Voronoi Diagram and the Delaunay Triangulation is based on DT 丄 ee. J. Computer Information sci., Vol.9, no.3, pp.219-142, 1980).

By the way, as shown in Fig. 1, the maximum disk (18) is connected to the contour line (10) at each point of contact (12), (14) with the center point ⁶ ) of the maximum disk and the contact line. Each tangent (22), (24) force with the contour at the point; each has the property of being mutually perpendicular. The straight line from the center point ⁶ ) of this maximum disk ⁶ ) depends on the nature of R as well as the bisecting point (28) of the connecting straight line (26) of each contact point (1 ² ) and (U). This contact point connecting straight line (26) Match.

Let us consider the related characteristics of the largest disk in the above symmetric axis transformation in relation to the three-dimensional image of an object assumed above as shown in FIG. It can be more explicit for the R-cone (96), but let's look at any morphological figure (94). When a stereoscopic image (94) is viewed with stereo sensors (100) and (102) like two eyes, a specific sphere (Sphere) is assumed to form a part of the stereoscopic image. Disk (90), which is a specific part of (88), can be represented as shown in Figure 2. By the way, the Identify 'of such a disk that exists in a given space coordinate field is the position of the center point of this disk in the space coordinate field, the orientation of the disk (Orientation ) And its size are possible. These are also possible depending on the position of the center point of the sphere to which the disk belongs, the size of the sphere, and the position of the center point of the disk within the sphere. Of course, in the latter case, if the position of the center point of the sphere is the same as that of the disk, it is necessary to separately apply the direction of the disk.

Here, when the three-dimensional image (94) is viewed by both sensors, the contour lines (10) and (30) connected to (100) and (102) at the sensor coordinates, and the symmetric axis transformation is applied to the image. It is assumed that each of the largest discs 8), ( ³⁸ ) is a shape connected to each sensor coordinate when the sphere (88) is viewed from each direction. That way the center points of the maximum disk (16), (36) corresponds to the center point ⁶⁾ in the sphere (88), each axis of symmetry (20), (40) shed points on ^6), ( The corresponding points (12), (14) and ( ³² ), ( ³⁴ ) on each contour line (10), (30) by sharing ³⁶ are the points ( ⁵² ) on the outer edge of the disc ( ⁹⁰ ). ), ( ⁵⁴ ) and (72), (74).

Also, the connecting straight lines (26), (46) between the corresponding points (1 ² ), () and ( ³² ), ( ³⁴ ) are located at the center point (76) of the disk (90), respectively. Points ( ⁵² ), ( ⁵⁴ ), and ( ⁷² ), ( ⁷⁴ ) can be made to correspond to straight lines (66), (86). Then, the bisecting points (28) and (48) of the corresponding point connecting straight lines (26) and (46) correspond to the center point (76) of the disk (90) when viewed from the direction of each sensor. Is possible. According to the relationship and characteristics of the assumed sphere, the disk, and the maximum disk for each contour linear elephant by the symmetrical axis transformation as described above, the two contour linear elephant signals are specifically interpreted as follows.

First, if the symmetry axis transformation is applied to both contour linear elephants, the point on the symmetry axis formed for each contour line, such as (16), and the point on the contour line due to the fact that the point is shared, Two groups of Units that unite three points, such as two corresponding points such as, (12) and (14), are formed for two contour shapes. In setting the correspondence between the units formed between the two groups formed in this way, the center point of a sphere is assumed from one direction, assuming the point on the axis of symmetry for each unit as described above. It is assumed that the point is connected to the sensor coordinates when viewed. Also, assuming that two corresponding points on the contour line of each unit, when a certain disk in a certain sphere assumed in the above is viewed from a certain direction, it is assumed to be both end points of the disk connected to sensor coordinates, The bisecting point of the connecting straight line of the corresponding point is the point connected to the sensor coordinates when the center point of the disk is viewed from one direction. At the end, the connecting line of the corresponding point on the contour line of each unit is drawn when the disk assumed above is viewed from one direction. It is considered to be related to the size of the disc because it is considered to be a connecting straight line between both end points on the disk outline that are viewed from each other while passing through the center of the disc.

The center of the sphere and the disk at m_h and the size of the disk, etc. are the special points or characteristics, and the characteristics of each sensor and the sensing environment, such as the two sensor coordinate fields and the spatial coordinate field. By taking into account and interpreting, a one-to-one (Oiie-to-One) correspondence between units is set.

With respect to the two units corresponding to the two groups thus set, the state for each disk assuming a three-dimensional image is derived through the following solution. First, the two bisecting points of the corresponding unit arrest straight line on the two contour lines of the corresponding unit are interpreted in consideration of the sensing environment to obtain the center point of the corresponding disk in the given spatial coordinate field. Calculate coordinate values. By interpreting the orientation (Orientation) of the corresponding point connecting straight line on the two contour lines of the corresponding units in the two sensor coordinates in consideration of the sensor environment, the orientation of the corresponding disk in the spatial coordinate field is determined. You will understand. Finally, the size of the corresponding disk can be determined by interpreting the length of the corresponding point connecting straight line of both units in consideration of the sensing environment. Reconstruct and describe the 3D figure of the target object in the form of a set of disks defined by the position and orientation of the center point in the space and coordinate space obtained by applying the Jith process to the overall correspondence unit. And a method for recognizing the three-dimensional image of the object.

Planar contour linear elephant by Symmetric Axis Transform The expressing means has the following characteristics.

• It is effective for extracting the symmetry axis of a simple contour curve without overlapping (not compounding).

• Since the definition of Local Symmetry is implied, it is possible to express faithfully even in the details of the shape.

• There is only one axis of symmetry for one shape. In other words, it has a unique reproducible (Uniquely Recoverable) characteristic. Since the present invention is basically based on the symmetry axis conversion, it can be said that the characteristics of the present invention have the same symmetry axis conversion characteristics of ΰϋι for a three-dimensional image.

BRIEF DESCRIPTION OF THE FIGURES

Fig. 1 is an illustration of Symmetric Axis Transform.

Figure ² is an explanatory diagram of a three-dimensional image reconstruction using the concept of a sphere and a disk.

FIG. 3 is an explanatory diagram showing the relationship between one sensor coordinate and a spatial coordinate field in the embodiment.

Fig. 4 is an explanatory diagram of the calculation of the disk orientation from the two sensor coordinate data.

Figure 5 shows the determined shape of the disc on the spatial coordinate field.

Fig. 6 shows the description of the shape of one disk on the spatial coordinate field. It is calculation explanatory drawing for restoring to the figure in which it is.

FIG. 7 is a configuration diagram of the three-dimensional ^ yg system.

* Explanation of sign

12, 1 (32, 34): Two corresponding points on contour 10 (30) sharing point 16 (36) on symmetry axis 20 (40)

18 (38): Maximum disk that touches the corresponding points 12, 14 (32, 34) at the same time

26 (46): Connected straight line of corresponding points 12, 14 (32, 34)

28 (48): Bisection point of the corresponding point connecting straight line 26 (46)

16 (36): The center point of the maximum disc 18 (38)

22 (24): tangent to the contour above point 12 (14)

9: An example of a solid figure

96: Example of a cone-

88 _: sphere

90: Disk

56: The center point of the sphere 88

100 (102): Left (right) sensor and sensor coordinates

52, 54 (72, 74): Both end points of the disc 90 corresponding to the points 12, 14 (32, 34) formed at the sensor one coordinate 100 (102).

66 (86): Connected straight line of points 52, 54 (72, 74), corresponding to 26 (46)

76: Disc 9 Q center point is the intersection of straight lines 66 and 86, corresponding to 28 and 48

104: Disk in X-z plane coordinates with diameter information and center at origin 1 10: Sensing space coordinate field

1 12: Stereo sensor and its coordinate field

1 14: Contour detection system (System)

1 16: Composite contour decomposition system

1 18: Pair, conversion system

120: Compatible unit search system

122: Supported unit analysis system

124: 3D image representation system

126: 3D figure synthesis system

128: Coordinate analysis system

130: Coordinate field of use space

132: 3D image restoration system

134: Recognition system for 3D objects

136: 3D image inspection system

140: Information on sensing environment

42: Use space coordinate information

144: Coordinate transformation relation between 110 and 130 † S¾

16: Configuration information of composite contour

In recognizing a three-dimensional image of an object from stereo contour linear elephant information, as one specific embodiment of the present invention, the following sensing environment is used. Check the application of.

First, consider a certain spatial coordinate field (World Coordinate) in which the left sensor coordinate (loo) is set to gi as shown in Fig. ³ .

Set the X axis in the same direction parallel to the U axis and the Y axis in the same direction parallel to the V axis with respect to the two-dimensional plane U-V coordinate field of the sensor coordinates (100). Let's say you are pointing towards the Infinity point in the Minus direction. Also, let's assume that the sensor (102) with the same characteristics is located at a position rotated counterclockwise around the Y axis 6 degrees from the sensor (100). Therefore, as can be seen in Fig. 3, the plane of the sensor (102) coordinate is parallel to the U-axis with respect to the u-V coordinate field and in the same direction. Let's assume that the sensor (10 ² ) points to the point of infinity in the minus direction of the axis.

Here, when the angle between the two sensors is rotated by のため ° due to calibration of the sensing environment, one point on the appropriate space with IftJ which is thought to be the Y axis as the center of rotation of the space coordinate field For each point, the corresponding point in the U-V coordinate field of each sensor locator 檫 (100), (102) is set as the origin of each U-V sensor coordinate, and a point in this space is defined as X- Set to the origin of the Y-Z axis. Then, it can be seen that the JT and ^ axes have rotated counterclockwise by about Θ degrees about the Y axis with respect to the X and Z axes, respectively. In addition, the sensor characteristics have uniform characteristics to the sensor coordinate plane for convenience of calculation, and the value corresponding to the focal length (Focal Length) of each sensor is infinite and U-V and X (or: T )-Suppose the magnification between Y is 1. Let us examine the desirable conditions for establishing a one-to-one correspondence between units between two groups for two contour linear elephants in such a sensing environment.

• In relation to the center point of a hypothetical sphere that is a feature point, comparing the V-axis values on both sensor coordinates with respect to the point on the symmetry axis between the two groups, it is basically determined whether they are the same. investigate.

• Compare the V-axis value on both sensor coordinates to the bisecting point of the corresponding point connecting straight line between the units, in relation to the center point of the assumed disk, which is another feature point. Investigate whether they are the same in principle.

• Investigate whether the length of the straight line connecting the corresponding points between the two groups is essentially the same, in relation to the remaining feature, the assumed disk size. If there is not one-to-one unity relationship between the two groups that satisfy all three conditions outside of the above three desirable conditions, and there are multiple correspondences in this sensing environment, both groups will be related to the sensor system calibration. The u-axis values of the points on the symmetry axis between the units in the two sensor coordinates are examined, and the values are compared to form a unit that is relatively close to one-to-one correspondence. That is, if the two sensors are focused on the object, the U-axis values on both sensor coordinates of the assumed center point of a certain sphere do not differ greatly. To this end, it is necessary to position the target object as close as possible to the Y-axis of the spatial coordinate field. The interpretation is as follows, taking into account the sensing environment given to both corresponding units between the two groups in the one-to-one correspondence defined in the above-described correspondence setting process.

First, consider the diagram in Fig. ^{3 to} calculate the position of the center point of the disk in the spatial coordinate field. O A point PO, y, ^z ) in the given spatial coordinate field is determined by the coordinates of the left and right sensors (PO, y, ^z ). Assuming that Pr (Ur, V _r ) is connected to (100) and (10 ² ), respectively, point ρ is connected so that it is projected on the left sensor coordinate (100), and it is connected to the right sensor coordinate (102). Are connected so that they are projected in a state of being rotated clockwise by 0 degrees about the Y axis, and if 《is not zero,

I «{, vi, wi, 1)

(Μ _Γ , υ _Γ , ΐϋ _Γ , 1)

(2)

Can be expressed as In equations (1) and (2), f and w _r are dummy variables. Assuming that the value f corresponding to the focal length of one sensor is infinite, each coordinate value in the space coordinate field of the point rc, y, z) in space is

X = Ui

_{V = vi (or) ν τ} (3) ui · cos θ- u T

sin O Can be represented by '

Therefore, the bisecting point of the corresponding point connecting straight line on the contour of the unity of the left sensor coordinate (100) in both units in the corresponding relation, for example, the U and V axis values of the point (28) are (u,, ) in to the corresponding point of the right sensor coordinates (10 ^2), eg if its corresponding disk by substituting computation U point (48), the V-axis value (M _r, in the t _r) equation (3) For example, the coordinates of the center point of, for example, the point (76) in this spatial coordinate field can be found.

Also, taking into account the length of the two corresponding point connecting straight lines in both corresponding units, the size of the corresponding disk, that is, the diameter d can be calculated, for example, by averaging the two values.

At the end, the remaining element, the orientation of the disc in the spatial coordinate field, is calculated as follows: 4 As shown in Fig. 4, the direction of the connecting line of the corresponding point on the contour of the unit on the left sensor coordinate in this coordinate field (100) in both corresponding units is the line of the connecting line of the corresponding point on the right unit. If the direction in the right sensor coordinate field (102) is taken as r, = at + (4) r _a = a'l + b't, it can be said that & = 6 'in this sensing environment. Then, the direction r of the relevant disk is obtained as follows for the relevant spatial coordinate field by vector synthesis of 7 and. r = at + b + ck (5) 4 In equation (5), α = sb = cosa

sm a-oos ^-cos hi

c =

81Ώ.Θ where o :,? Is a-tan- ^{l Pfl (t?)} - ^P " ^(u)

_{= tan} . ₁ ^^ l

P _r2 (") -P _rl (") where i fc in equations (), (5) and 4 and the unit vector of the X, Y, Z and X 'axes of this spatial coordinate field, respectively.

The calculation method for calculating the position, orientation, and size, that is, the diameter, of the center point in the corresponding spatial coordinate system for a certain disk assumed to be a part of an object through the interpretation of J¾_ is the sensing environment of this embodiment. Was sought. If these processes are applied to the whole correspondence unit of both contour line groups, it can be calculated for the whole set of disks representing the three-dimensional image.

One disk determined in this way, for example, the disk (90) of order i among the disks constituting a certain object (94) can be represented on the corresponding space coordinate field as shown in FIG. The expression for this can be expressed as the expression in ( ⁶ ).

X {yi Z {d {a {b {ί 6) In expression (6), _yi , is the corresponding spatial coordinate value of the center point of this disk, is the diameter of this disk and α, ·, b α are the corresponding spatial coordinate fields X, Y, and z-axis directions, respectively. It has information about the orientation of this disc in the amount of vector per unit.

If a certain three-dimensional figure is composed of a set of n specific disks, it can be represented by the following expression (7).

\ (7) Xi yi Zi di, also, 'c,

However, in such an expression,, 'regarding the position of the center point of the disc and ひ & ,, = 1 ~ regarding the direction of the disc are relative values to the corresponding spatial coordinate field which are not absolute values. It is a value that changes relative to the change in the spatial coordinate field, but the relative value of this disc to one object is constant.

The corresponding spatial coordinates on the outline of one disk (90), which is considered to be one part of the three-dimensional image surface for reconstruction of the three-dimensional image from the expression of one disk constituting the three-dimensional image as in Expression (6) The coordinate value meter 界 in the field is as follows. First, consider a disk (104) whose diameter is as shown in Fig. 6 (a) and whose center point is on the X-Z plane at the origin.

Orientation of the disc is labeled same as r _a and the Y-axis direction. The coordinate values on the outline of the disk (104) can be expressed as in equation (8). di wide

— — · Cos ς y = 0.0 (8) z = ― · sin で In equation (8), ζ is a value from 0 to 360 degrees.

And should the current orientation r _a is translated as r in 6 (b) view of the Formula ⁽⁹⁾ by expression (6).

From Figure 6 (b)

6;

a = cos

a, '

β = cos

Ci

7 = cos

i,-i + a ^ where if a, = 6, = 0 then a = 0 °, = 90. and 7 = 0 °, if a _t = c, = 0 then a = 0 ° = 0 ° and 7 = 90 °, if

°. Der ¹ & shed

From J ^ JLh, the coordinate value (Xo Vo, Z ₀ ) in the spatial coordinate field on the outer line considering the rotation Jl and the movement T of the disk (104) can be calculated as follows.

0 = IR + T (10) _{_{O = (x 0, y D}} , z 0j 1) , the equation (8) from J = (a ;, y, z , 1), T = (Xi from expression (6), the formula (10), 2 1)

It is. However, 7 '= 90. -7,

Industrial Applicability In the above embodiment, the realization of the present invention for one specified sensing environment (Sensing Condition) has been described. However, it goes without saying that the application of the present invention utilizing the characteristics of a sphere and a disk is not limited to this embodiment.

That is, in order to set the one-to-one correspondence between the units between the two contour groups using the characteristics of the sphere and the disk, and to interpret both units set to be in the correspondence described above, In relation to the characteristics of the two sensors and the relationship between the coordinates of the two sensors and the spatial coordinate field, etc. An appropriate calibration is required to set the environment. The method in each case for such a calibration can generally be adequately documented in the extensive literature. Therefore, the spirit of the present invention can be realized for all sensors that can obtain stereo contour information together with the execution of such a calibration, and for all the sensors and their systems.

In addition, a block diagram as shown in FIG. 7 can be considered as one configuration example for realizing such a method. First, a shape signal for a certain object in the sensing spatial coordinate field (110) is received through a stereo sensor (112), and a contour line is detected for each signal. The complex contours are decomposed into simple forms (116), and each is sequentially sent to the symmetric axis conversion system (118). A corresponding unit search (120), a corresponding unit interpretation (122), and a three-dimensional figure based on the interpreted information are expressed (124).

The three-dimensional representation of each of the decomposed parts is synthesized (126) with reference to the composition relation information (146) between the composite contours from the composite contour decomposition system (116) and expressed. This way representable the sensing space coordinate field from the three-dimensional figure (110) and use space coordinates field (1 ³ 0) coordinates related interpretation † (128) the coordinate transformation information (144) with each use for purposes by between Available. The fields of use that can be generally considered include a wide range, such as reconstruction of three-dimensional figures (132), recognition of three-dimensional figures (134), and inspection of three-dimensional figures (136).

As described above, the method according to the present invention can be applied to stereoscopic image recognition. It requires only simple preprocessing for the sensor input signal because only the contour linear elephant information for the input signal is required.Also, detailed representation of the target object is possible, and only one It has a unique output (Unique), and the basic units (Primitives) for expressing the three-dimensional image are simple and can be sufficiently expressed by synthesizing it. It has important advantages for application to recognition systems.

† This method can be sufficiently achieved by known techniques.

Claims

1 · Recognition of the three-dimensional object from the stereo contour information obtained from the two directions of the object through the sensor through the sensor. Something is assumed to exist inside the object! Assuming that a cubic representation of the object has been made, as represented by the disk '(Disk), assumed as a particular part of the sphere, (Sphere), two contour linear elephants Each of them applies a Symmetric Axis Transform, and as a result, two points on the contour line due to という share the point with a point on the Symmetric Axis formed for each contour linear elephant. Two groups of Units that unite three points, such as two corresponding points, are formed for the two contour linear elephants; b) Using the characteristics of the sphere and the disk assumed above, A one-to-one (1: 1) correspondence between units is set; c) The characteristics of the sphere and the disk, assumed above, for both units between the two groups in such a set correspondence. In the spatial coordinate field set for each disk by using and interpreting Recognizing the state; d) Reconstructing the three-dimensional shape of the target object so that it becomes a collection of disks whose state is defined by the size, the position and orientation of the center point in the spatial coordinate field obtained in this way. Constitute; A method characterized by including such a process.

In setting the one-to-one correspondence between units between the two groups in Section 2.1, a) The point on the symmetry axis for each unit is viewed from a certain direction with respect to the center point of the sphere assumed above. B) Assuming two corresponding points on the contour line for each unit, the disk tied to the sensor coordinates when one disk is viewed from one direction And the bisecting point of the line connecting the corresponding points is regarded as the point connected to the sensor coordinates when the center point of the disk is viewed from one direction; c) The contour for each unit Assuming that the connecting straight line of the two corresponding points on the line is the connecting straight line between the two end points of the disk viewed from each other while passing through the center of the disk when the disk is viewed from one direction, assuming the above, See that it is related to the size, etc .; d) J¾_h The characteristics of each sensor and the coordinates of the two sensors and the spatial coordinate field using the characteristics of the sphere and the disk, with the characteristic points and characteristics such as the center point of the disk inside the disk and the size of the disk, etc. Interpretation taking into account the sensing environment (Sensing Condition) related to the relationship with

3 · Correspondence between the two groups set as corresponding in paragraph 1 In calculating the state of each corresponding disk from the information on the two units to be used, a) the two bisecting points of the corresponding point connecting straight line on each contour line of the corresponding units are considered in consideration of the sensing environment. By interpreting, the coordinate value of the center point of the relevant disk in the corresponding spatial coordinate field is known; b) The orientation of the corresponding point connecting straight line on each contour line of the corresponding unit in the sensor coordinate field of each sensor (Orientation) ) Is taken into account in consideration of the sensing environment to determine the orientation of the relevant disc in the corresponding spatial coordinate field; c) The length of the two corresponding points connecting straight line of both units in the corresponding relationship is considered in consideration of the sensing environment. Interpretation to determine the size of the disc; a method that includes such a process