A method for calculating uncertainty in visual measurement systems

Abstract

The present invention relates to the field of digital measurement technology, in particular to an uncertainty calculation method for a visual measurement system. First, a visual space is defined as a plurality of sections along a central axis of a visual measurement system and divided into grid nodes. Then, in a system measurement coordinate system, a standard device is used to evaluate the standard uncertainty component of the position of the grid node. Finally, according to the geometric relationship between the observation point and the adjacent sections and the grid node, the uncertainty of the observation point is predicted and calculated in combination with a set weight distribution and an interpolation algorithm. The calculation method is simple, efficient and highly reliable, and avoids the deviation between theory and practice caused by factors such as incomplete consideration of error sources caused by theoretical analysis.

Description

Technical Field

[0001] The invention relates to the technical field of digital measurement, and in particular to an uncertainty calculation method for a visual measurement system. Background Art

[0002] In the process of formulating the measurement plan, the evaluation of the measurement uncertainty of the measurement system is an extremely important link, which directly determines whether the measurement plan can meet the measurement requirements and whether the measurement data is reliable. Especially for aircraft assembly sites, the measurement scenarios are different, and the reliability of measurement data is extremely high. Each measurement task needs to be based on the systematic understanding of uncertainty.

[0003] At present, the research on uncertainty prediction mainly establishes error models through the analysis of measurement mechanisms. However, since there are many error sources in visual measurement systems, such as noise and distortion, in addition to the system's own integrated related factors, it is also affected by environmental factors. It is difficult to establish an accurate error theory error model for uncertainty assessment, and the correctness and robustness of the error theory model are difficult to guarantee. Therefore, it is particularly important to design an efficient and accurate measurement system uncertainty calculation method. Summary of the invention

[0004] Aiming at the problem that it is difficult to establish an accurate theoretical error model for uncertainty assessment in current measurement systems, and the correctness and robustness of the theoretical error model are difficult to ensure, the present invention proposes an uncertainty calculation method for a visual measurement system. Firstly, the visible space is defined as a plurality of sections along the central axis of the visual measurement system and divided into grid nodes; then, in the system measurement coordinate system, a standard device is used to evaluate the standard uncertainty component of the position of the grid node; finally, according to the geometric relationship between the observation point and the adjacent sections and the grid node, the uncertainty of the observation point is predicted and calculated in combination with the set weight distribution and interpolation algorithm. The calculation method is simple, efficient and highly reliable, and avoids the deviation between theory and practice caused by factors such as incomplete consideration of error sources caused by theoretical analysis.

[0005] The specific implementation contents of the present invention are as follows:

[0006] A method for calculating uncertainty of a visual measurement system is disclosed. First, the visible space is defined as multiple sections along the central axis of the visual measurement system and divided into grid nodes. Then, in the system measurement coordinate system, a standard device is used to evaluate the standard uncertainty component of the position of the grid node. Finally, according to the geometric relationship between the observation point and the adjacent sections and the grid nodes, the uncertainty of the observation point is predicted and calculated in combination with the set weight distribution and interpolation algorithm.

[0007] In order to better implement the present invention, the following steps are further specifically included:

[0008] Step S1: Cut the visible area into multiple sections S at equal intervals along the central axis of the visual measurement system. i , with the center axis and the cross section S i The intersection of the plane and the local coordinate system X are established as the origin. i O i Y i ;

[0009] Step S2: Set the plane local coordinate system X i O i Y i Dividing into a plurality of network nodes, setting standards at the network nodes, obtaining standard uncertainty, and calculating a mean value of the standard uncertainty;

[0010] Step S3: taking the central axis of the visual measurement system as the Z axis, establishing a measurement coordinate system O-XYZ, and constructing an interpolation function of the uncertainty mean and the z coordinate under the measurement coordinate system O-XYZ;

[0011] Step S4: Project the observation point P (x, y, z) along the central axis to two adjacent sections to obtain P i (x,y,z),P i+1 (x, y, z), and calculate the corresponding uncertainty u according to the inverse distance interpolation algorithm pi 、u p(i+1) ;

[0012] Step S5: According to the uncertainty u pi 、u p(i+1) , calculate the uncertainty u1 of the observation point P caused by measurement repeatability;

[0013] Step S6: Obtain the calibration uncertainty u2 of the standard instrument, synthesize the uncertainty u2 and the uncertainty u1 to obtain the synthesized standard uncertainty u c .

[0014] In order to better implement the present invention, further, the specific operation of calculating the mean value of the standard uncertainty in step S2 is:

[0015]

[0016] in, is the standard uncertainty mean, u i (x j ,y k ) is the standard uncertainty, j=1...m; k=1...n; i=1...N.

[0017] In order to better implement the present invention, further, the interpolation function in step S3 is:

[0018] .

[0019] In order to better implement the present invention, further, the corresponding uncertainty u is calculated according to the inverse distance interpolation algorithm in step S4. pi 、u p(i+1) The specific operations are:

[0020]

[0021] in, is the distance decay function, γ≥1.

[0022] In order to better implement the present invention, further, the specific operation of calculating the uncertainty u1 of the observation point P caused by measurement repeatability in step S5 is:

[0023]

[0024]

[0025]

[0026] Among them, λ i , i+1 is the weighted value of the uncertainty components of two adjacent sections, m z is the mean value obtained by the interpolation function.

[0027] In order to better implement the present invention, further, the synthetic standard uncertainty u in step S6 c for:

[0028] .

[0029] The present invention has the following beneficial effects:

[0030] (1) The uncertainty distribution law of the measurement system in the present invention is obtained based on the measured data of spatial sampling, and the data reliability is high, which can avoid the deviation between theory and practice caused by factors such as incomplete consideration of error sources caused by theoretical analysis.

[0031] (2) The present invention calculates the uncertainty at the measured object based on the geometric relationship between the measured data with limited uncertainty distribution and the measured object. The calculation method is simple and efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic flow chart of the uncertainty calculation method provided by the present invention.

[0033] Figure 2 It is a schematic diagram of the cross-section division of the visible area provided by the present invention.

[0034] Figure 3 It is a schematic diagram of grid division provided by the present invention.

[0035] Figure 4 It is a schematic diagram of the projection of the uncertainty observation point provided by the present invention.

[0036] Figure 5 It is a schematic diagram of calculating the uncertain component of the projection point provided by the present invention.

[0037] Figure 6 It is a schematic diagram of uncertainty interpolation calculation provided by the present invention.

[0038] Among them, 1. camera, 2. central axis, 3. cross section, 4. fitting point of interpolation function, 5. interpolation function curve. DETAILED DESCRIPTION

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. It should be understood that the described embodiments are only part of the embodiments of the present invention, not all of the embodiments, and therefore should not be regarded as limiting the scope of protection. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technical personnel in this field without making creative work are within the scope of protection of the present invention.

[0040] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "disposed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0041] Embodiment 1:

[0042] This embodiment proposes an uncertainty calculation method for a visual measurement system. First, the visible space is defined as multiple sections 3 along the central axis 2 of the visual measurement system and divided into grid nodes. Then, in the measurement coordinate system of the visual measurement system, a standard uncertainty component of the position of the grid node is evaluated using a standard device. Finally, based on the geometric relationship between the observation point and the adjacent sections 3 and the grid nodes, the uncertainty of the observation point is predicted and calculated in combination with an interpolation algorithm and a set weight.

[0043] The specific steps include:.

[0044] Step S1: Cut the visible area into multiple sections S at equal intervals along the central axis 2 of the visual measurement system i , establish a plane local coordinate system X with the intersection of the central axis 2 and the cross section 3 as the origin i O i Y i .

[0045] Step S2: Set the plane local coordinate system X i O i Y i The network is divided into a plurality of network nodes, standard instruments are set at the network nodes, standard uncertainty is obtained, and the uncertainty mean of the section 3 is calculated.

[0046] Step S3: Taking the central axis 2 of the visual measurement system as the Z axis, establishing a measurement coordinate system O-XYZ, and constructing an interpolation function of the uncertainty mean and the measurement distance in the visual space under the measurement coordinate system O-XYZ.

[0047] Step S4: Project the observation point P along the central axis 2 to two adjacent sections 3 to obtain a projection point P i , projection point P i+1 , and calculate the corresponding uncertainty prediction value u according to the inverse distance interpolation algorithm pi 、u p(i+1) .

[0048] Step S5: Predict the value u according to the uncertainty pi 、u p(i+1) , calculate the uncertainty component u1 of the observation point P caused by measurement repeatability.

[0049] Step S6: Calculate the comprehensive measurement uncertainty u at the observation point based on the calibration result of the standard instrument and the uncertainty u2 of the standard instrument. c .

[0050] Working principle: This embodiment first defines the visible space as multiple sections 3 along the central axis 2 of the system and divides it into grid nodes. In the system measurement coordinate system, a measured standard device is used to evaluate the standard uncertainty component of the grid node position. Finally, for any point in the space, based on the geometric relationship between the point and the adjacent sections 3 and the grid nodes, a reasonable weight distribution and interpolation algorithm are used to predict and calculate the uncertainty of the point. This method establishes the uncertainty distribution law in the measurement space based on the measured data, has high reliability, can quickly calculate the uncertainty based on the geometric relationship between the measured object and the system, and the calculation method is simple.

[0051] Embodiment 2:

[0052] This embodiment is based on the above embodiment 1. Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 As shown, a specific embodiment is used for description.

[0053] Specifically include the following steps.

[0054] Step S1, first, Figure 1 As shown, the visual interval is cut into multiple sections S at equal intervals along the central axis 2 of the visual measurement system. i (i=1...N), with the center axis 2 and the cross section S i The intersection of the two points is the origin of the local coordinate system X i O i Y i ;

[0055] Step S2, such as Figure 2 As shown, the local coordinate system X i O i Y i The plane is meshed, and at the mesh node P i (x j ,y k ) (j=1…m; j=1…n) Set the standard device measured by the higher precision measuring equipment, and use the visual measurement system to measure point P i (x j ,y k ) Repeat the measurement of the standard instrument at the place where the standard uncertainty introduced by the measurement repeatability is recorded as u i (x j ,y k ), for all points P on the plane i (x j ,y k ) corresponds to the standard uncertainty u i (x j ,y k ), take the mean uncertainty:

[0056] .

[0057] Step S3, define the measurement coordinate system O-XYZ, the Z axis coincides with the central axis 2 of the measurement system, and construct the uncertainty mean of each section 3 in the measurement coordinate system O-XYZ Interpolation function with z coordinate , so that the mean value of the cross section 3 at any point P (x, y, z) in space can be obtained by the interpolation function To obtain.

[0058] Step S3, the uncertain mean of the plane where the observation point P (x, y, z) is located is calculated by interpolation of the mean of the known section 3. The selection of the interpolation function and method in the actual operation process is related to the node data.

[0059] Step S4, such as Figure 4 As shown, the observation point P (x, y, z) is projected along the central axis 2 to the two adjacent sections 3, that is, Figure 2 Section S i and section S i+1 , and obtain two projection points P respectively i (x,y,z),P i+1 (x, y, z), the uncertainty caused by measurement repeatability at the projection point is u pi 、u p（i+1） ;like Figure 5 As shown, the uncertainty distribution data based on the plane where the two points are located is calculated using the inverse distance interpolation algorithm. The specific calculation process is expressed by the following formula:

[0060]

[0061] Similarly, we can calculate u p（i+1） .

[0062] in is the distance attenuation function, γ≥1. If γ=1, the uncertainty value of the point is positively correlated with the inverse of the distance to the node.

[0063] Step S4, the uncertainty of the point observation point P (x, y, z) caused by the measurement repeatability is recorded as u1, and the value of the uncertainty recorded as u1 is calculated from the two adjacent sections 3, that is, Figure 2 Section S i and section S i+1 The measurement uncertainty component u pi 、u p(i+1) and its weight λ i , i+1 The calculation process of the uncertainty u1 is expressed by the following formula:

[0064]

[0065] where λ i and i+1 For two adjacent sections 3, that is Figure 2 Section S i and section S i+1 The weighted value of the uncertainty component of is obtained based on the mean value of uncertainty. The specific calculation process is expressed by the following formula:

[0066]

[0067]

[0068] Step S5 , the uncertainty of the observation point P (x, y, z) caused by measurement repeatability is a function of the uncertainty mean of each section 3 and the local coordinates of the section 3 .

[0069] Step S6, record the uncertainty caused by the accurate calibration of the standard as u2, combine it with the repeatability measurement uncertainty, and calculate the combined standard uncertainty of the measurement system at the point observation point P (x, y, z) as u c , the specific calculation process is expressed by the following formula:

[0070]

[0071] The mean uncertainty of each section is obtained by evaluating the standard instrument ,like Figure 6 As shown, the uncertainty mean of each section 3 is used as a known point for function fitting, which can be a polynomial function, exponential function, power function, or any other function in which the residual is acceptable to the user. The function is expressed as , the uncertainty mean of any section 3 in the measurement coordinate system The interpolation function To obtain.

[0072] The other parts of this embodiment are the same as those of the above-mentioned embodiment 1, and thus will not be described in detail.

[0073] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any simple modification or equivalent change made to the above embodiment based on the technical essence of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for calculating uncertainty of a visual measurement system, characterized in that: First, the visible space is defined as a plurality of sections (3) along the central axis (2) of the visual measurement system and divided into grid nodes; then, in the measurement coordinate system of the visual measurement system, the standard uncertainty component of the position of the grid node is evaluated using a standard instrument; finally, based on the geometric relationship between the observation point and the adjacent sections (3) and the grid nodes, the uncertainty of the observation point is predicted and calculated in combination with an interpolation algorithm and a set weight; The specific steps include: Step S1: Cut the visible area into multiple sections S at equal intervals along the central axis of the visual measurement system (2) i , a plane local coordinate system X is established with the intersection of the central axis (2) and the cross section (3) as the origin i O i Y i ; Step S2: Set the plane local coordinate system X i O i Y i Dividing into a plurality of network nodes, setting a standard device at the network node, obtaining a standard uncertainty, and calculating a mean value of the uncertainty of the cross section (3); Step S3: taking the central axis (2) of the visual measurement system as the Z axis, establishing a measurement coordinate system O-XYZ, and constructing an interpolation function of the uncertainty mean and the measurement distance in the visual space under the measurement coordinate system O-XYZ; Step S4: Project the observation point P along the central axis (2) to two adjacent sections (3) to obtain a projection point P i , projection point P i+1 , and calculate the corresponding uncertainty prediction value u according to the inverse distance interpolation algorithm pi 、u p(i+1) ; Step S5: Predict the value u according to the uncertainty pi 、u p(i+1) , calculate the uncertainty component u1 of the observation point P caused by measurement repeatability; Step S6: Calculate the comprehensive measurement uncertainty u at the observation point based on the calibration result of the standard instrument and the uncertainty u2 of the standard instrument. c .

2. The uncertainty calculation method for a visual measurement system according to claim 1, characterized in that: The specific operation of calculating the uncertainty mean of the cross section (3) in step S2 is: in, is the standard uncertainty mean, u i (x j ,y k ) is the standard uncertainty, j=1...m; k=1...n; i=1...N.

3. The uncertainty calculation method for a visual measurement system according to claim 2, characterized in that: The interpolation function in step S3 is: 。 4. The uncertainty calculation method for a visual measurement system according to claim 2, characterized in that: The corresponding uncertainty u is calculated according to the inverse distance interpolation algorithm in step S4. pi 、u p(i+1) The specific operations are: in, is the distance decay function, γ≥1.

5. The uncertainty calculation method for a visual measurement system according to claim 4, characterized in that: The specific operation of calculating the uncertainty u1 of the observation point P caused by measurement repeatability in step S5 is: Among them, λ i , i+1 is the weighted value of the uncertainty component of two adjacent sections (3), is the interpolation function Solve for the mean.

6. The uncertainty calculation method for a visual measurement system according to claim 5, characterized in that: The comprehensive measurement uncertainty u in step S6 c for: Among them, u1 is the uncertainty of the observation point P caused by measurement repeatability, and u2 is the uncertainty of the standard instrument.

Images