阅读说明：本技术 机器视觉系统 (machine vision system ) 是由 埃萨·莱卡斯 于 2018-01-11 设计创作，主要内容包括：公开了一种能够测量绝对尺度的差异的机器视觉系统。在该机器视觉系统中,使用诸如坐标测量机之类的参考测量设备来准确地测量参考物体。在测量中,测量确定的虚拟点的集合。参考测量被用于计算确定的坐标矩阵与参考测量结果之间的偏差矩阵,其随后被用于根据机器视觉测量来计算绝对尺度值。(A machine vision system capable of measuring differences in absolute scale is disclosed. In the machine vision system, a reference measuring device such as a coordinate measuring machine is used to accurately measure a reference object. In the measurement, a set of determined virtual points is measured. The reference measurement is used to calculate a deviation matrix between the determined coordinate matrix and the reference measurement, which is then used to calculate an absolute scale value from the machine vision measurement.)

1. A controller for a machine vision system, comprising:

at least one processor (101);

At least one memory (102); and

At least one data communication connection (106);

Wherein the controller (100) is configured to receive a deviation matrix, wherein the deviation matrix is based on measurements of a reference object using a reference measurement device; and

Wherein the processor (101) is configured to:

Measuring a reference object and at least one target object using a machine vision system;

Calculating a difference matrix representing a difference between the target object and the reference object; and

An absolute scale result of at least one target object is calculated based on the measured target object, the corresponding difference matrix and the bias difference matrix.

2. The controller of claim 1, wherein the controller (100) is further configured to receive a determined coordinate matrix, wherein the determined coordinate matrix comprises coordinates of at least one virtual point.

3. The controller of claim 1 or 2, wherein the controller (100) is further configured to receive the determined coordinate matrix.

4. The controller of claim 3, wherein the processor (101) is configured to calculate the absolute scale result of the at least one target object as a sum of the deviation matrix, the difference matrix and the determined coordinate matrix.

5. A machine vision system, comprising:

The controller (100) of any of the preceding claims 1-4; and

At least two cameras (103a-103c) connected to the controller.

6. The machine vision system of claim 5, wherein the machine vision system further comprises a coordinate measuring machine (106).

7. A method for measuring an object, comprising:

Measuring a reference object and at least one target object using a machine vision system (300, 301);

Calculating a difference matrix (302) representing a difference between the target object and the reference object;

An absolute scale result (303) of the at least one target object is calculated based on the target object under test, the corresponding difference matrix and a deviation matrix, wherein the deviation matrix is based on measurements of the reference object using a reference measurement device.

8. The method of claim 6, wherein the method further comprises receiving a deviation matrix.

9. the method of claim 7 or 8, wherein the method further comprises receiving a determined coordinate matrix, wherein the determined coordinate matrix comprises coordinates of at least one virtual point.

10. The method of claim 7, 8 or 9, wherein the method further comprises receiving the determined coordinate matrix.

11. The method of claim 10, wherein the method further comprises calculating the absolute scale result of the at least one target object as a sum of the bias matrix, the difference matrix, and the determined coordinate matrix.

12. A computer program comprising computer program code, wherein the computer program code is configured to perform the method of any of claims 7-11 when executed by a computing device.

Technical Field

The following disclosure relates to machine vision systems. In particular, the present application relates to measuring machine vision systems for use in quality control or other similar tasks requiring measurement of an object.

background

Computer controlled machine vision systems are used in a variety of applications. A typical application is the quality control of objects manufactured in the manufacturing industry. By imaging the manufactured object using one or more cameras, it is possible to measure various properties of the object. The measurement may involve measuring the entire object or some selected characteristic of the object. Thus, depending on the selected features, the measurements may be one-dimensional, two-dimensional or three-dimensional, or may even be performed in a combination of dimensions. In addition to size and shape, it is also possible to measure other characteristics such as color, roughness, or other such features. The measurements of machine vision systems are typically made by comparing manufactured objects to model objects. The results obtained generally give the relative difference of the measured object and the model object used.

For measuring the three-dimensional coordinates only two cameras are required, since the three-dimensional coordinates can be calculated from two-dimensional images, provided that the point to be measured is visible in both images. However, the number of cameras is generally large. This is because a large number of cameras increases the coverage and accuracy of the measurement. The cameras are typically positioned so that they can see all, or at least as many, of the features of the object under test. Accordingly, not all cameras typically see the feature being measured. In addition to increasing the number of cameras, many other concepts are known to improve measurement accuracy, such as precision calibration and image processing algorithms. Furthermore, it is possible to plan the camera position for a specific object, or to use a more accurate camera or specific lighting to improve the quality of the image acquired from the desired feature.

Measuring machine vision systems are particularly excellent because they can similarly identify measured features under different conditions. Therefore, when measuring an object, features such as edges and holes will be similarly detected even if conditions change. Therefore, it is possible to accurately measure even a slight change in the position or shape of the object. Although the measurements derived from the acquired images are accurate, they cannot be compared with measurements made with other measurement tools, such as coordinate measuring machines. This is because it is difficult to measure, for example, the absolute position of the edge using conventional machine vision systems and methods. While it is possible to accurately measure the relative change in size, position, or other change of an object under test, it is difficult to measure the same change in absolute scale, rather than the relative difference.

In conventional solutions, these measurements are sometimes supplemented by accurately measuring the position of the measured object or by placing the measured object in a measuring fixture so that the exact position is known. When the position is exactly known, it is possible to measure at least some absolute measurements of the object to be measured. One approach is to make a reference object ("gold object") that is manufactured as accurately as possible to meet the nominal dimensions of the object.

however, these methods can be problematic if different types of objects or a large number of objects need to be measured. If the object to be measured needs to be accurately positioned before the measurement can be made, the measurement will be slow. Accordingly, if different types of objects need to be measured, then different types of jigs or other positioning means may also need to be changed between measurements. All of these mechanical methods are expensive and subject to wear.

Disclosure of Invention

A machine vision system capable of measuring differences in absolute scale is disclosed. In machine vision systems, reference measurement devices, such as coordinate measuring machines, are used to accurately measure reference objects. In the measurement, a set of points determined from drawing or CAD data is measured. The reference measurement is used to calculate a deviation matrix between the determined coordinate matrix and the reference measurement, which is then used to calculate an absolute scale value from the machine vision measurement.

In one aspect, a controller for a machine vision system is disclosed. The controller includes at least one processor, at least one memory, and at least one data communication connection. The controller is configured to receive a deviation matrix, wherein the deviation matrix is based on measurements of a reference object using a reference measurement device. The processor is configured to measure a reference object and at least one target object using the machine vision system, calculate a difference matrix representing a difference between the target object and the reference object, and calculate an absolute scale result for the at least one target object based on the measured target object, the corresponding difference matrix, and the deviation matrix.

in an embodiment, the controller is further configured to receive a determined coordinate matrix, wherein the determined coordinate matrix comprises coordinates of at least one virtual point. In another embodiment, the controller is further configured to receive the determined coordinate matrix. In another embodiment, the processor is configured to calculate the absolute scale result of the at least one target object as a sum of the deviation matrix, the difference matrix, and the determined coordinate matrix.

in another aspect, a machine vision system is disclosed. The machine vision system includes a controller as described above and at least two cameras connected to the controller. In an embodiment, the machine vision system further comprises a coordinate measuring machine.

In one aspect, a method for measuring an object is disclosed. A method for measuring objects includes measuring a reference object and at least one target object using a machine vision system; calculating a difference matrix representing a difference between the target object and the reference object; and calculating an absolute scale result of the at least one target object based on the measured target object, the corresponding difference matrix and the deviation matrix.

In an embodiment, the method further comprises receiving a deviation matrix, wherein the deviation matrix is based on measurements of the reference object using the reference measurement device. In another embodiment, the method further comprises receiving a determined coordinate matrix, wherein the determined coordinate matrix comprises coordinates of at least one virtual point. In another embodiment, the method further comprises receiving the determined coordinate matrix. In an embodiment, the method further comprises calculating the absolute scale result of the at least one target object as a sum of the deviation matrix, the difference matrix and the determined coordinate matrix.

in one aspect, the above method is implemented as a computer program. The computer program comprises computer program code, wherein the computer program code is configured to perform the method as described above when executed by a computing device.

benefits of the described embodiments include the possibility to use absolute measurements using machine vision systems. When absolute measurements are required, the disclosed arrangements and methods provide a fast and cost-effective way of performing absolute measurements, since not all measured objects need to be measured using a coordinate measuring machine.

Drawings

The accompanying drawings, which are included to provide a further understanding of the machine vision system and are incorporated in and constitute a part of this specification, illustrate embodiments and together with the description help to explain the principles of the machine vision system. In the drawings:

Figure 1 is an example of a machine vision system,

Figure 2 is an example of a method of a machine vision system,

Fig. 3 is an example of a method of a machine vision system.

Detailed Description

Reference will now be made in detail to embodiments, examples of which are illustrated in the accompanying drawings.

In fig. 1, a block diagram illustrating an example of a controller 100 for a machine vision system is shown. The controller 100 includes at least one processor 101 and at least one memory 102. The at least one processor 101 is configured to execute the computer program and the at least one memory is configured to store the computer program comprising the computer program code and data required by the computer program. Typically, the controller is a general purpose computer, but may also be a specially manufactured controller. The controller 100 includes several connectors so that the controller 100 can connect to external devices and data communication networks using data communication connections 106. In addition, it is possible to include typical computing devices such as displays, keyboards, and other generally known devices.

The controller 100 is configured to control a machine vision system. The machine vision system includes a plurality of cameras 103a-103 c. The processor of the controller is configured to acquire images of the object 104 to be measured using the cameras 103a-103 c. Even though the figure illustrates the use of three cameras, the number of cameras will not be limited to three, and a greater number of cameras will typically be used.

The controller 100 of the figure that controls the machine vision system is also configured to receive measurements from an external measurement device (e.g., coordinate measuring machine 105). Coordinate measuring machine 105 includes a probe 106 that is used to measure object 104. In the following description, the use of a coordinate measuring machine will be explained in more detail; however, the object 104 in fig. 1 should be understood to be the same object. Thus, measurements will be made first by using machine vision equipment and then using a coordinate measuring machine, and vice versa. In other words, it is not assumed that two measurements are taken of the object at the same time, but the results of the two measurements are stored in the memory 102 of the controller 100.

the object 104 need not be a specially manufactured model object. The object 104 may be a generic or pre-manufactured object. The advantage of this arrangement is therefore that no special and more accurately manufactured model objects are required.

In the example of fig. 1, a coordinate measuring machine 105 is coupled to the controller 100. However, this is not essential. The coordinate measuring machine need not be located at the same site as the controller 100, nor need it be directly connected to the controller 100. However, it is desirable that the measurement results obtained with the coordinate measuring machine 105 can be brought to the controller 100. If there is no direct connection, a media carrier (such as a memory stick, memory card, optical disc, conventional magnetic disc, etc.) may be used to obtain the measurement. Accordingly, the controller 100 is connected to additional computers using a local area network connection. The local area network connection may be used for controlling and configuring the controller 100 and for general data exchange. For example, a local area network connection may be used to receive data relating to the object to be measured.

In fig. 2, an example method for performing initial measurements is shown. In this method, three-dimensional virtual points are used. As explained above, different cameras see the measured object at different positions. Thus, the point measured by the camera system is not the true exact position of the point being measured, but represents the point seen by the camera system comprising at least two cameras. This so-called virtual point is offset from the real coordinates of the measured object. However, if the object to be measured moves, the three-dimensional virtual point and its offset will move accordingly. If the object under measurement does not move and different parts of the object are measured, the movement of the virtual point will correspond to a change in the object under measurement. These may be due to, for example, thermal expansion. If the measured object is changed into a similar object, the change of the three-dimensional virtual point corresponds to the change between the measured objects.

The virtual point can be freely selected; however, the coordinate values are typically selected to correspond to the feature to be measured. Virtual three-dimensional point (X, Y, Z) and virtual two-dimensional (X) viewable using a camera₀，y₀，x₁，y₁，…x_n，y_n) The coordinate pairs correspond, where n is the number of cameras. To determine the position of the virtual point, the point to be measured can be searched and detected using conventional methods.

The method of fig. 2 begins by determining a three-dimensional virtual point (step 200). This determination will be performed by using a reference object. A virtual point is a 3D point bound to a feature of the reference part to be measured. A virtual 3D point is typically (but not exclusively) a point close to, for example, the center point of a hole. The reference object may be a specific model object or may simply be any manufactured object. The determined three-dimensional virtual points form a determined coordinate matrix N, the matrix size being m x 3, where m is the number of three-dimensional virtual points. It is also possible to use several reference parts and calculate the matrix E in equation (1) to include the average values.

After the three-dimensional virtual points are determined, the corresponding points are measured from the reference object using a coordinate measuring machine or other reference measuring device, step 201. From the measurements, a result matrix M is formed₀. Finally, a deviation matrix E is determined in step 202. The deviation matrix is determined according to equation (1)

E＝M₀-N equation (1)

The method of fig. 2 may be considered part of a calibration process for a machine vision system; however, it may also be interpreted as an initial measurement of the measurement method described below with respect to fig. 3. It should be noted that even the reference measurement system cannot directly measure certain features, such as the hole center point, because the hole center point is not a physical object. The position of the centre point of the hole has to be determined indirectly by measuring the hole edges and assuming e.g. that the hole has an accurate circular shape.

in fig. 3, an example method for measuring an object is shown. The measurement process uses the determined three-dimensional virtual points. The method is initiated by measuring a reference object measured using a reference measurement device using a camera system (step 300). Measurement provision result matrix V₀. This step may also be implemented into the method discussed above with respect to fig. 2. Since it does not depend on the rest of the method of fig. 2, it can be performed at any stage.

After the reference object has been measured, other objects may be measured, step 301. For each object, a result matrix V is obtained_n. Result matrix V_nIs a common result in machine vision measurements. The resulting matrix obtained and the resulting matrix of the reference object are adapted to each other, for example by using the least squares method, so that the difference matrix D between the measured object and the reference object can be calculated according to equation 2_nStep 302.

D_n＝V_n–V₀Equation (2)

In calculating a difference matrix D_nthereafter, a matrix M representing the absolute scale measurement of the measured object n may be calculated by using equation 3_nStep 303.

M_n＝D_n+ E + N equation (3)

The resulting matrix M obtained_nIs comparable to the absolute scale results given by a reference measuring device, such as a coordinate measuring machine. In the above description, the object to be measured may be referred to as a target object, and the reference may be one of the target objects; however, the reference may also be a specific reference object. Depending on the situation, we can also compute more directly

D_n＝V_n-N equation (4) gives

M_n＝V_n+ E equation (5)

The above mentioned method may be implemented as computer software comprising computer program code executed in a computing device capable of communicating with external devices. When the software is executed in a computing device, it is configured to perform the inventive method described above. The software is embodied on a computer-readable medium so that it may be provided to a computing device, such as the controller 100 of fig. 1.

As mentioned above, the components of the exemplary embodiments can include computer-readable media or memory for holding instructions programmed according to the teachings of the inventions and for holding data structures, tables, records, and/or other data described herein. Computer-readable media may include any suitable media that participates in providing instructions to a processor for execution. Common forms of computer-readable media may include, for example, a floppy disk, a flexible disk, hard disk, magnetic tape, any other suitable magnetic medium, a CD-ROM, CD R, CD + -RW, DVD-RAM, DVD + -RW, DVD + -R, HD DVD-R, HD DVD-RW, HD DVD-RAM, Blu-ray disk, any other suitable optical medium, a RAM, a PROM, an EPROM, a flash-EPROM, any other suitable memory chip or cartridge, a carrier wave, or any other suitable medium from which a computer can read.

It is obvious to a person skilled in the art that with the advancement of technology, the basic idea of a machine vision system may be implemented in various ways. Thus, the machine vision system and its embodiments are not limited to the above examples; instead, they may vary within the scope of the claims.