JP3650205B2 - Non-contact image measurement system and edge tracking measurement method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a non-contact image measurement system and an edge tracking measurement method for capturing an object to be measured by an imaging unit such as a CCD camera and extracting necessary measurement information by detecting a contour included in the image of the object to be measured. .
[0002]
[Prior art]
Conventionally, this type of non-contact image measurement system is used for measurement of a thin plate such as an IC lead frame, measurement of a wiring pattern, etc., which is difficult by contact measurement. When performing non-contact image measurement, after setting an object to be measured (workpiece) on a measurement table, an imaging device such as a CCD camera is moved to the location where the workpiece is to be measured, focus adjustment is performed, and the workpiece is displayed on the CRT display. An enlarged image of is displayed. Then, the location to be measured is indicated by a mouse cursor or window, the edge portion of the image is extracted based on the image processing technique, and a desired measurement value is obtained by arithmetic processing.
[0003]
[Problems to be solved by the invention]
However, in the above-described conventional non-contact image measurement system, when a specific edge portion of image information is to be continuously measured along the edge, the window for designating the measurement area must not be continuously moved along the edge. There is a problem that the operation is troublesome.
On the other hand, various edge tracking processing methods are known in the field of image processing. However, many of the known methods perform tracking while referring to neighboring pixels, and tracking takes time and is not suitable for non-contact measurement in which sampling is performed at an arbitrary interval. is there.
Furthermore, the conventional method has a problem that it is not suitable for edge measurement of a unit smaller than one pixel.
[0004]
The present invention has been made in order to solve such problems, and is excellent in operability that can flexibly perform edge tracking and extract necessary measurement information even for an object whose shape is unknown or changes. Another object of the present invention is to provide a non-contact measuring apparatus and an edge tracking measuring method.
[0005]
[Means for Solving the Problems]
An edge tracking measurement method according to the present invention includes a step for setting a window so that a part of the edge is included in a measured image including an edge to be tracked, and image information in the set window. Detecting a plurality of edge points, applying an approximate line to the plurality of edge points detected in this step, and overlapping a part of the current window along the approximate line applied in this step A step of setting a next new window as described above, and by sequentially repeating the detection of the edge point, the fitting of the straight line, and the generation of the new window based on the image information in the new window, Extract necessary measurement points in the window while sequentially moving the window along the edge of the measured image And wherein the go.
[0006]
In another edge tracking measurement method according to the present invention, a window is set so that a part of the edge is included in the measured image including the edge to be followed displayed on the display screen by imaging the measured object. And a step of detecting a plurality of edge points in the window from image information in the set window, a step of applying an approximate line to the plurality of edge points detected in this step, and a step of applying Setting a next new window so that a part of the area overlaps the current window along the approximate straight line, and the new window so that the new window generated in this step fits in the display screen. And changing the imaging position of the object to be measured in cooperation with the generation of a new window, the image in the new window While sequentially detecting the edge point based on the information, fitting the straight line, generating the new window, and changing the imaging position, the window is sequentially moved along the edge of the image to be measured. It is characterized in that necessary measurement points are extracted in the window.
[0007]
Further, the non-contact image measurement system according to the present invention includes an imaging unit that images the measurement target, a display unit that displays the image of the measurement target captured by the imaging unit, and the display unit that includes the measurement target. A window indicating a measurement region is displayed over the target image, and the window is moved along an edge included in the measurement target image to extract information necessary for measurement from the image information in the window. Control means, wherein the control means detects a plurality of edge points in the window from the image information in the window, applies an approximate straight line to the obtained plurality of edge points, and along the approximate straight line The next new window is generated so that a part of the area overlaps the current window.
[0008]
According to a more preferable aspect of the present invention, when an edge point is lost in the window, the window is rotated around the edge point detected last and the edge point detected last is continued. Detect edge points.
[0009]
Further, it is determined whether or not a predetermined measurement end point is included in the window, and when it is determined that the specified measurement end point is included in the window, edge tracking is ended.
[0010]
According to the edge tracking measurement method of the present invention, if the initial position of the window is first set, then the window moves autonomously along the edge and samples necessary edge points. Information can be obtained by a very simple operation.
[0011]
When measuring an object to be measured that does not fit in the display screen, it is possible to always fit the window in the display screen by coordinating the movement of the window and the imaging position of the object to be measured. For this reason, edge extraction and measurement point extraction can be performed without trouble even for a relatively large object to be measured.
[0012]
Further, according to the present invention, an approximate straight line is obtained from a plurality of edge points obtained in the window, and the next window position is determined along the approximate straight line. Therefore, an abnormal point is included in the sample value. However, it does not greatly affect the determination of the next window position. Since the position of the next window is determined so that a part of the area overlaps the previous window, the edge is not lost in the next window even when the edge changes sharply.
[0013]
In the unlikely event that the edge to be detected is steeper than expected and the edge point is lost in the window, the window is rotated around the last detected edge point, and the edge point following the edge point is detected. By setting the next new window, the detection of the edge point, the fitting of the straight line, and the generation of the new window can be continued without any trouble based on the image information in the new window. .
[0014]
If a measurement end point in the measured image is specified, when it is determined that the measurement end point is included in the window, a new window is not generated based on the image information in the window, Since extraction of necessary measurement points in the window ends, a series of processing can be automatically ended.
[0015]
Further, in the non-contact image measurement system of the present invention, the control means moves the window along the edge of the image to be measured and extracts information necessary for measurement, so that an object whose shape is unknown or changes is extracted. The necessary measurement information can be extracted efficiently.
[0016]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.
FIG. 1 is a perspective view showing the overall configuration of a non-contact image measurement system according to an embodiment of the present invention.
This system includes a non-contact image measurement type CMM 1, a computer system 2 that drives and controls the CMM 1, and executes necessary data processing, and a printer 3 that prints out measurement results. It is comprised by.
[0017]
The three-dimensional measuring machine 1 is configured as follows. That is, a measurement table 13 on which the workpiece 12 is placed is mounted on the gantry 11, and this measurement table 13 is driven in the Y-axis direction by a Y-axis drive mechanism (not shown). Support arms 14 and 15 extending upward are fixed to the center of both side edges of the gantry 11, and an X-axis guide 16 is fixed so as to connect both upper ends of the support arms 14 and 15. An imaging unit 17 is supported on the X-axis guide 16. The imaging unit 17 is driven along the X-axis guide 16 by an X-axis drive mechanism (not shown). A CCD camera 18 is mounted on the lower end of the imaging unit 17 so as to face the measurement table 13. The imaging unit 17 includes a Z-axis drive mechanism that moves the position of the CCD camera 18 in the Z-axis direction, in addition to a lighting device and a focusing mechanism (not shown).
[0018]
The computer system 2 includes a computer main body 21, a keyboard 22, a joystick box 23, a mouse 24, and a CRT display 25.
The computer main body 21 is configured, for example, as shown in FIG. That is, image information input from the CCD camera 18 is stored in the multi-value image memory 32 via an interface (hereinafter referred to as I / F) 31. The multi-value image information stored in the multi-value image memory 32 is displayed on the CRT display 25 via the display control unit 33. On the other hand, position information input from the mouse 24 is input to the CPU 35 via the I / F 34. In accordance with the program stored in the program memory 36, the CPU 35 generates data for displaying a rectangular area having a position, size, and orientation designated by the mouse 24 as a window, and stores the multi-value image information inside the window. The detection of edge points extracted from the multi-valued image memory 32, the process of fitting an approximate line to the detected edge points, and the process of generating the next window along the obtained approximate line are sequentially executed. The work memory 37 provides a work area for various processes in the CPU 35.
[0019]
Next, an edge tracking measurement procedure in the non-contact image measurement system configured as described above will be described.
FIG. 3 is a flowchart showing a processing procedure of the CPU 35 for the edge tracking measurement, and FIG. 4 is a diagram showing image information 41 showing a part of the workpiece 12 displayed on the CRT display 25 for explaining this processing. It is.
The image information 41 shown in FIG. 4 includes an edge 42 to be tracked. Therefore, first, the initial position of the rectangular window 43 indicating the measurement region is set so as to include a part of the edge 42 by operating the mouse 24 or the like (S1). In the window 43, for example, the four corners A, B, C, and D are set by clicking the mouse 24, or two points in the diagonal direction of the rectangle are specified, and then the rectangular area is dragged to an arbitrary angle. It is specified by operations such as tilting and moving. At this time, the direction to be traced along the edge 42 is also designated.
[0020]
When the initial position of the window 43 is set, the CPU 35 next detects a plurality of edge points 44 from the multi-value image information in the window 43 (S2). FIG. 5 shows details of this sampling. The sampling interval Δ of the edge point is set in the work memory 37 in advance. The CPU 35 first sets the x coordinate from the start point A (xa, ya) to the end point B (xb, yb) as cos θ [where θ is the inclination of the window 43. The multi-value image information at the address indicated by the x and y coordinates is extracted from the multi-value image memory 32 while changing the y coordinate by sin θ. An appropriate threshold level is set from the obtained multi-valued point sequence data, and a point where the threshold level and the point sequence data intersect is sampled as an edge point. Next, the starting point and the ending point are moved by Δ · sin θ and Δ · cos θ, respectively, and similar sampling is executed. When the above processing is continuously performed to the start point C (xc, yc) and the end point D (xd, yd), sampling of the plurality of edge points 44 at the preset interval Δ is completed.
[0021]
Next, the CPU 35 applies an approximate straight line to the obtained sampling values of the plurality of edge points 44 by, for example, the least square method (S3).
Now, as shown in FIG. 6, if the approximate line L is obtained from the sampling value of the edge point 44 obtained by the window 43, the CPU 35 determines the next window 43 'along the approximate line L. (S4).
For this reason, first, a perpendicular line is drawn from the edge point 43a in the moving direction of the window 43 obtained in the current window 43 to the approximate straight line L, and along the approximate straight line L from the intersection of the normal line and the approximate straight line L. H · m / 100 opposite to the moving direction of the window 43 (where H is the height of the window, m is a preset overlap rate (%)) and the point P1 from the point P1 A point P2 separated by H in the movement direction is obtained. Next, points that are separated from the approximate line L by W / 2 (W is the width of the window) on the straight lines orthogonal to the approximate line L at points P1 and P2, respectively, at the four corners of the new window 43 '. Points A ′, B ′, C ′, and D ′ are assumed. Thereby, the next window 43 'is determined.
[0022]
When the next window 43 'is determined, the window 43 is sequentially moved while sampling the edge points in the window 43' and fitting the approximate straight line as described above. When all the edges to be tracked are tracked, the process is terminated (S5).
[0023]
When the initial position of the window 43 is set, the mouse 24 or the like is operated to set a measurement end point on the edge 42, or a series of values is directly input as a point on the work coordinate system. Can be automatically terminated at the measurement end point.
FIG. 7 is a flowchart showing a processing procedure of the CPU 35 for the end of the measurement, and FIG. 8 is a diagram showing image information 41 showing a part of the work 12 displayed on the CRT display 25 for explaining this processing. 9 and 10 are diagrams for explaining the end determination process.
For simplification of description, as shown in FIG. 9, the window coordinate systems m and n are defined with reference to the center of the window 43, and the origin is Ow.
[0024]
First, the CPU 35 calculates a vector E representing the relative position between the center Ow of the window 43 and the end point E in the work coordinate system x, y (S11). Next, the vector E is rotated on the work memory 37 so that, for example, the inclination between the work coordinate system and the window coordinate system becomes 0 ° (S12). As shown in FIG. 10, when the vector components Ex and Ey of the rotated vector E ′ are smaller than ½ of the width W and the height H of the window 43 ′, that is, Ex <W / 2, Ey <H If it is / 2 (S13), it is determined that the end condition is satisfied, a new window 43 is not generated, and when necessary measurement points in the window 43 are extracted, a series of processing ends. On the other hand, if the vector components Ex and Ey of the rotated vector E ′ do not satisfy one of the conditions Ex <W / 2 and Ey <H / 2 as shown in FIG. 10 (S13), the process ends. The process described above is repeated for the newly generated window 43 until it is determined that the condition is satisfied (S11 to S13).
When the work 12 has a closed shape like a circle, the measurement start point and the end point coincide. For this reason, when the first window 43 is generated, the above-described termination condition is not determined, and the determination may be performed when the window 43 is generated for the second time or later. This is because the measurement end point deviates from the window 43 due to the movement of the window 43 at the second time.
[0025]
As described above, according to the system of the present embodiment, if the initial position of the window is set first, then the window moves autonomously along the edge and samples necessary edge points. There is an advantage that necessary information on the information can be obtained by an extremely simple operation.
Although it is possible to determine the next window position by curve approximation, in this case, if an abnormal point is included in the sampling point, the edge may be lost in the next window depending on that point. Conceivable. In this point, in this system, an approximate straight line is obtained from a plurality of edge points obtained in the window by the least square method, and the next window position is determined along this approximate straight line. Even if included, it does not significantly affect the determination of the next window position.
[0026]
Further, in this system, since the next window is determined so as to overlap the previous window at a predetermined overlap rate m, for example, as shown in FIG. 11, even when the edge changes sharply, the next window Never miss an edge in the window. The overlap ratio m may be set to an arbitrary value in consideration of the edge steepness and the edge tracking efficiency. The illustrated example shows an example in which the duplication rate is about 20%.
[0027]
However, even if an appropriate overlap ratio is set, an edge that is steeper than expected may be lost. In this case, as shown in FIG. 12, it is determined that the shape of the workpiece 12 is discontinuous in the direction along the approximate straight line, and the vicinity of the last edge point Pn obtained in the window 43 is searched. Thus, the lost edge 42 can be searched for.
FIG. 13 is a flowchart showing a processing procedure of the CPU 35 for this edge tracking measurement, and FIG. 14 is a diagram for explaining this processing.
[0028]
The CPU 35 detects the edge point from the multi-value image information in the window 43 (S2), and if the edge point is not detected (S21), the brightness (for example, reflection intensity) is examined in the vicinity of the edge point Pn ( In S22, it is determined whether the shape of the workpiece 12 is an acute angle as shown in FIG. 14A or an obtuse angle as shown in FIG. 14B (S23). Here, the position for checking the brightness is determined by the angle and the distance with the final edge detection point Pn as a reference. As shown in FIG. 12, since the window 43 when the edge is lost is set perpendicular to the approximate straight line, the direction of the continuous edge is estimated based on that direction. The point Pn + 1 at which no edge is obtained is a position separated from the position immediately before losing sight of the continuous edge by the sampling interval Δ, and therefore there is little inconsistency with the actual shape of the workpiece 12, and the position for checking the brightness It is considered as optimal.
[0029]
When the shape of the workpiece 12 is an acute angle, as shown in FIG. 15A, the window 43 is rotated by −90 ° about the edge point Pn (S24), and the multivalued image information in the window 43 is used. Edge point Pn + 1 is detected (S25). If the edge point Pn + 1 is detected (S26), the above-described straight line fitting process is performed on the assumption that the missing edge 42 is found. On the other hand, if the edge point Pn + 1 is not detected again (S26), the window 43 is rotated by −45 ° around the edge point Pn (S27) as shown in FIG. An edge point Pn + 1 is detected from the image information (S28). If the edge point Pn + 1 is detected (S29), the above-described straight line fitting process is performed on the assumption that the missing edge 42 is found. On the other hand, if the edge point Pn + 1 is not detected again (S29), the continuation of the measurement is abandoned, and the user is warned via the CRT display 25, for example.
Hereinafter, even when the shape of the workpiece 12 is an obtuse angle as shown in FIG. 14B, the rotation angle is changed to, for example, + 90 ° and + 45 °, and the same processing is performed (S30 to S33).
[0030]
Further, when measuring the workpiece 12 that does not fit on the screen of the CRT display 25, the movement of the window 43 and the imaging position of the workpiece 12 are coordinated so that the window 43 is always within the display screen. be able to. For this reason, edge extraction and measurement point extraction can be performed without any trouble even for a relatively large workpiece 12.
FIG. 16 is a flowchart showing the processing procedure of the CPU 35 for the edge tracking measurement, and FIG. 17 is a diagram for explaining this processing.
[0031]
The CPU 35 determines the next window 43 ′ along the approximate line (S4), calculates a vector A representing the relative position between the center Oa of the screen 41 and the center Ow ′ of the window 43 ′ (S41), It is determined whether or not the window 43 'fits on the screen 41 (S42). If the window 43 'does not fit on the screen 41, a vector that internally divides the distance from the center Ob of the next screen 41' to the boundary FF 'along the approximate straight line at a predetermined rate (for example, 50%). After calculating B (S43), a vector BA (movement amount) representing the relative position between the center Oa of the screen 41 and the center Ob of the screen 41 'is calculated (S44), and the direction of the vector BA is as follows: In the opposite direction, the stage is moved by B-A (S45).
Since the above-described ratio designation is realized by software stored in the program memory 36, it is possible to easily change the handling as a variable.
[0032]
【The invention's effect】
As described above, according to the present invention, the window moves along the edge of the image to be measured to extract information necessary for measurement, and thus is necessary even for an object whose shape is unknown or changes. There is an effect that information can be extracted efficiently.
[0033]
In addition, according to the present invention, it is not necessary to create a program according to the measurement target, and a series of processing in edge tracking measurement can be automatically performed only by initial setting, and the operation can be acquired in a short time. There is an effect that can be done.
[Brief description of the drawings]
FIG. 1 is a perspective view showing a configuration of a non-contact image measurement system according to an embodiment of the present invention.
FIG. 2 is a block diagram of a computer main body in the system.
FIG. 3 is a flowchart of edge tracking processing in the system.
FIG. 4 is a diagram showing a display screen in the system.
FIG. 5 is a diagram for explaining edge point detection in a window in the system.
FIG. 6 is a diagram for explaining a procedure for determining a next window position in the system.
FIG. 7 is a flowchart of end processing in the system.
FIG. 8 is a diagram showing a display screen in the system.
FIG. 9 is a diagram for explaining an end point determination procedure in the system;
FIG. 10 is a diagram for explaining end point determination in the system;
FIG. 11 is a view showing a moving locus of a window in the system.
FIG. 12 is a diagram for explaining edge point detection in a window in the system.
FIG. 13 is a flowchart of edge tracking processing in the system.
FIG. 14 is a diagram for explaining a workpiece shape determination procedure in the system;
FIG. 15 is a diagram for explaining the procedure for determining the next window position in the system;
FIG. 16 is a flowchart of stage movement processing in the system.
FIG. 17 is a diagram for explaining a procedure for determining the next screen position in the system.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... CMM, 2 ... Computer system, 3 ... Printer, 11 ... Mount, 12 ... Workpiece, 13 ... Measurement table, 14, 15 ... Support arm, 16 ... X-axis guide, 17 ... Imaging unit, 18 ... CCD Camera, 21 ... Computer body, 22 ... Keyboard, 23 ... Joystick box, 24 ... Mouse, 25 ... CRT display, 31,34 ... Interface, 32 ... Multi-valued image memory, 33 ... Display control unit, 35 ... CPU, 36 ... Program memory 37 ... Work memory 41, 41 '... Image information 42 ... Edge, 43, 43', 43 "... Window, 44 ... Edge point.

Claims (8)

Setting a window to include a portion of the edge in the measured image including the edge to be followed;
Detecting a plurality of edge points in the window from image information in the set window;
Fitting an approximate line to a plurality of edge points detected in this step;
Setting the next new window so that some area overlaps the current window along the approximate line fitted in this step,
The window is sequentially moved along the edge of the image to be measured by sequentially detecting the edge point, fitting the straight line, and generating the new window based on the image information in the new window. An edge tracking measurement method, wherein necessary measurement points are extracted in the window. Setting a window to include a part of the edge in the image to be measured including the edge to be followed displayed on the display screen by imaging the object to be measured;
Detecting a plurality of edge points in the window from image information in the set window;
Fitting an approximate line to a plurality of edge points detected in this step;
Setting the next new window so that some area overlaps the current window along the approximation line fitted in this step;
Changing the imaging position of the measurement object in cooperation with the generation of the new window so that the new window generated in this step fits in the display screen,
Based on the image information in the new window, the detection of the edge point, the fitting of the straight line, the generation of the new window, and the change of the imaging position are sequentially repeated, thereby making the window an edge of the measured image. An edge tracking measurement method, wherein necessary measurement points are extracted in the window while sequentially moving along the window. The step of detecting the edge point continues to the last detected edge point by rotating the window around the last detected edge point when an edge point is lost in the set window. The edge tracking measurement method according to claim 1, wherein the edge tracking measurement method is a step of detecting an edge point. Designating a measurement end point in the measurement object;
Determining whether a specified measurement end point is included in the window;
The edge tracking according to claim 1, further comprising a step of ending edge tracking when it is determined that the specified measurement end point is included in the window. Measuring method. Imaging means for imaging a measurement object;
Display means for displaying an image of the measurement object imaged by the imaging means;
A window indicating a measurement region is displayed on the display unit so as to overlap the measurement target image, and the window is moved along an edge included in the measurement target image, and measurement is performed from image information in the window. And a control means for extracting information necessary for
The control means detects a plurality of edge points in the window from the image information in the window, applies an approximate straight line to the obtained plurality of edge points, and a part of the area along the approximate straight line is currently A non-contact image measurement system for generating the next new window so as to overlap the window. The control means moves the position of the imaging means with respect to the measurement target in cooperation with the generation of the new window so that the generated new window fits within the display screen of the display means. The non-contact image measurement system according to claim 5, wherein the non-contact image measurement system is provided. The control means is configured to detect an edge point following the last detected edge point by rotating the window around the last detected edge point when an edge point is lost in the window. The non-contact image measurement system according to claim 5, wherein the non-contact image measurement system is provided. The control means determines whether or not the measurement end point designated in advance is included in the window,
8. The non-contact image measurement system according to claim 5, wherein, when it is determined that the designated measurement end point is included in the window, edge tracking is ended. 9. .

Images