JPS636679A - Method for detecting tilt of wire - Google Patents

Abstract

PURPOSE:To check the 3-dimensional form of a wire with no contact by illuminating the wire with a ring-shaped illumination means and detecting a bright- dark pattern produced on the wire with a 2-dimensional sensor. CONSTITUTION:A subject 1 hooked by plural wires is photographed by a TV camera 3 via a lens system 2 and this pickup signal is processed by a picture processor 4. The ring light 5 is set vertical to the optical axis of the camera 3. A subject carrying device 6 is provided and controlled by a controller 7. Then the processor 4 detects the tilt of the wire from the photographed picture based on the bright-dark pattern of the wire. This detection result signal is given to the controller 7.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a method of inspecting an object using image processing, and in particular, detecting the inclination of a wire suitable for measuring or inspecting the three-dimensional shape of a wire with a mirror surface. Regarding the method.

[Prior Art] Various methods for detecting three-dimensional shapes have been studied in the past, and the content is described in 4 of the book "Automation of Appearance Inspection" published on November 20, 1980, edited by the Institute of Electrical Engineers of Japan. .Explained in Section 2.3 "Image input method for three-dimensional objects". Among these methods, there are two methods for determining the three-dimensional position along each of the plurality of wires: a method using a range finder and a binocular stereoscopic viewing method, but both methods have the following problems.

First, in the range finder method, each point on the object that is hit by the projected slit light is imaged by a TV camera from a different angle, but in order to do this, the object surface needs to have a diffuse reflection component. It is. However, since a metal wire with a specular reflective surface has almost no such component, no light enters the TV camera unless the wire is oriented in a certain direction, and therefore the point on which the light hits cannot be detected.

-On the other hand, in binocular stereopsis, a specific point of interest is
The distance to the object can be calculated based on where it appears on the camera screen, but for this purpose, that specific point must be found uniquely as one point on the scanning line of each screen. However, since the wire has the same shape continuously along the line, if the direction of the wire approaches the direction of the scanning line, there will be a large error in determining the corresponding points, and furthermore, if the direction of the wire coincides with the direction of the scanning line, it will theoretically not be detected. It becomes possible.

Other methods have similar problems, and so far there has been no effective method for detecting the three-dimensional shape of a wire.

[Problems to be solved by the invention] Conventional three-dimensional position detection methods require conditions such as whether the object has a diffuse reflection component or whether the object has corners, protrusions, or surface patterns. It was. There is a problem in that the three-dimensional shape of an object that does not meet either of these conditions cannot be detected.

SUMMARY OF THE INVENTION An object of the present invention is to provide a method for detecting the inclination of a linear object, particularly a wire, made of a mirror surface that does not meet the above-mentioned limiting conditions.

[Means for Solving the Problems] In order to achieve the above object, in the present invention, a two-dimensional sensor (hereinafter referred to as a camera) is used to image an object, and for illumination, a sensor perpendicular to the optical axis of the two-dimensional sensor is used. A light source (hereinafter referred to as a ring light) arranged in a circle centered on the optical axis is used.

First, a target wire is extracted from an image obtained by a two-dimensional sensor, and the two-dimensional position and direction of each point along the wire is determined. Next, the inclination of the wire with respect to the imaging plane is detected from the brightness distribution along the wire in the image.

[Function] When the diameter of the ring light is made larger than the wire to be inspected, it is possible to create a cone-shaped incident light beam that has a nearly constant angle with the camera optical axis at any point on the subject. 6 Therefore, if there is an element with a surface inclined by α/2 with respect to the plane perpendicular to the optical axis, strong reflected light will be directed toward the camera from that surface. As a result, it is captured as a point with high brightness in the image.

Also, wires can be considered to generally have a circular cross section.

The relationship between the angle difference β when the wire and the illumination light are projected onto a horizontal plane is shown below when the incident angle of the illumination light is α and the inclination of the wire is φ. Consider a sphere centered at point 0 on the central axis G of the wire and having the same diameter as the wire diameter. This sphere touches the surface of the ear at a circle H. From 0 to the direction of the camera optical axis.Here, let C be the vertically upward vector M and the sphere,
Let T be the intersection of the vertical plane containing G and C and the circle H. Otsu CO
T is equal to the wire inclination φ, and at T CT and circle H are orthogonal. Let P be a point on circle H that returns reflected light in the direction of the camera optical axis. The perpendicular N to the wire surface and the spherical surface at point P passes through O. From the specular reflection condition, the light incident vector L, N, and M are on the same plane, and the angle between M and L is α
, the angle COP formed by M and N is 1/2 of that, which is α/2.

Since there is a triangle CTP on the spherical surface, T is a right-angled triangle. When viewed from vertically above, CT is the direction of the wire axis, and CP is the direction of light incidence.

The angle at C is β. Regarding a right triangle on a spherical surface, the following equation (1) holds and β can be obtained.

Also,? When POT is δ, the following equation (2) holds true.si
nδ== sinβsin (α/2) ・
...(2) As you can understand from now on, the maximum value of φ is α/
2, it occurs when β and δ are 0, i.e. when P and T become the same point. Also, φ=O2β=le/2
When δ takes the maximum value α/2.

If the wire is arc-shaped, when viewed from above, the locus of the point P to which the reflected light returns will be elliptical, as shown in (c). Point P1, which is the end in the axial direction. The inclination of the wire at P2 is α/2. Therefore, if the wire inclination φ is small and φ<α/2, the reflected light from one of the light sources of the ring light will enter the camera, but if the wire inclination is large and φ>α/2, the light will be reflected from that part. The area where the light does not enter the camera, that is, the inclination of the wire is from O to α/2, is imaged as a brightly lit area.Also, since the inclination of the wire is usually not discontinuous, the edge points of the brightly lit area are The slope is considered to be α/2.

Based on the principle described above, an abnormal tilt (often a large tilt) of the wire under test can be detected by comparing it with the tilt pattern of a normal wire.

[Example] Hereinafter, an example of the present invention will be described with reference to FIG. An object 1 on which a plurality of wires are stretched is imaged by a TV camera 3 through a lens system 2, and the signal is processed by an image processing device 4.

Place the ring light 5 perpendicular to the camera optical axis. At this time, the angle between the light ray from the light source toward the object and the camera optical axis is α. There is a transport device 6 that sequentially transports the objects to be inspected, and is controlled by a control device 7. An inspection activation signal 8 is sent to the image processing device every time one cycle of transport is completed, and an inspection result signal 9 is returned to the control device when the inspection is completed. When a result of "control device r failure" is received, control is performed such as issuing an alarm and shutting down or ejecting the object to another route.

Note that the image processing device 4 and the control device 7 may be configured as one device using, for example, a microcomputer, and in that case, the signals 8 and 9 become data handled by a computer program.

Furthermore, the image processing device 4 has a standard pattern of brightness along the wire, which is used for comparison with that of the input image.

FIG. 2 shows the processing steps in this embodiment. First, in step 101, the object to be inspected is brought into the field of view of the camera. Upon receiving the signal from the control device, the image processing device inputs an image of the object in step 102 . In step 103, the position of the wire included in the target image is determined. Various methods can be considered for wire extraction, but the background differs for each application, and a method suitable for each will be adopted.1For example, if the background is a nearly homogeneous object with a diffuse component, the wire may be extremely Since there are bright and dark areas, a large spatial differential value is detected at the boundary in the image. Wire position candidates are detected by detecting pixels with large spatial differential values detected at intervals of the wire width and taking the midpoint.

Next, the position of the wire can be detected by removing isolated points among these candidate points and replenishing the broken points.

In step 104, the image value is read from the position in the image of each point along the wire determined in step 103, and the brightness distribution between both ends of the wire is determined, and this is set as Ii. where the subscript i is a number along the wire, 1
to N, and if there is a difference, divide the total length by the number of points in the standard pattern and interpolate so that N is the same as the number of points in the standard pattern, and then Ii
Look for patterns. In this pattern Ii, a large value is obtained if the wire inclination is less than or equal to α/2, and a small value is obtained if it is greater than or equal to α/2.

In step 105, the pattern I obtained in step 104 is
i and the reference pattern Ti are compared and verified.

Various methods can be considered for this verification, and the selection is made based on the significance of the inspection for each application.

Some of them are listed below.

(1) Find N,≠1 as an error value, and check whether this is within a predetermined tolerance.

(2) Find N]=1 as the error value. In this method, the number of points exceeding a certain threshold value E with an error for each point is determined, and the ratio to the total number of points is determined.

(3) Separate the brightness pattern Ii into dark areas and dark areas using a certain threshold value, find the boundary point number, compare this with the boundary point number obtained in advance from the reference pattern to find the amount of deviation, and calculate each Determine whether the boundary is within the permissible deviation amount defined for the boundary. At this time, the presence of an extra boundary or the absence of an expected boundary is detected as a serious defect.

In step 106, it is determined whether the errors in the various methods described above are within an acceptable range.

If it is within the allowable range, the wire shape is determined to be good in step 107, and the process moves to step 101 to return to the next cycle. If it is not acceptable, the wire shape is determined to be defective in step 108, and further action is taken in step 109 to detect the defect. For example, sounding an alarm and calling an operator, or activating an automatic exclusion device.
Furthermore, there are methods such as attaching marks to objects that clearly indicate that they are defective, but these can be determined according to the design of the larger production system.

In this embodiment, if the critical angle α/2 is set, which is effective for determining the quality of the wire shape, all that is required is to arrange the ring light 5 according to the double angle α, and the rest is done by the processing within the image processing device 4. Since it can be determined, it does not require a device with a complicated configuration and is easy to control.

Examples of ring lights that can be used include fluorescent ring lights, light emitting elements arranged in a ring shape, and optical fibers arranged in a circle at one end and illumination light input from the other end.

In the embodiment described above, only the magnitude of the inclination of the wire relative to a certain angle α/2 could be detected. There are two embodiments described below that detect a more detailed angular distribution.

FIG. 3 shows the configuration of the illumination system in the second embodiment. As shown in the figure, two ring lights 5-
1. Overlap 5-2 with the same axis. In this way, the angle of incidence of the illumination light on the object is α1. There are two ways α2.

Both of these two ring lights, or at least the ring light 5-2 having a large angle of incidence, can be switched on and off by the image processing device 4 or the control device 7.

The inspection procedure using two ring lights is as follows. First, the ring light 5-2 is turned on, a target image is captured, the wire position is detected, and the brightness along the wire is determined using the position. Wire inclination α2/2
It is now clear.

This brightness pattern is compared with the first reference pattern. If the error is outside the range, it can be determined that the product is defective at this stage. If the error is within the range, then the ring light 5-2 is turned off and the ring light 5-1 is turned on. From now on, the wire inclination α1/2 will become clear. Find the brightness pattern using the already determined wire position,
This brightness pattern is compared with the second reference pattern. If the error is outside the range, it is determined to be defective, and if it is within the range, it is determined to be good.

In this embodiment, the wire shape can be inspected in three angle ranges: from horizontal to α1/2, from α1/2 to α2/2, and over α2/2, so the wire shape can be inspected more finely than in the first embodiment. can be inspected. Also, as a way to use this ring light configuration, increase α2 by 1, for example 8.
When set to 0'', etc., the area within the 40° inclination of the wire will be imaged brightly, and in many applications, it can be expected that the entire area of the wire will be bright, and the wire position can be detected by simple binarization, etc. There is a strong point.

It is also possible to further advance the idea of the second embodiment and adopt a configuration in which three or more ring lights are stacked as many as necessary, allowing for more detailed inspection.

Further, the incident angle of each ring light may be changed depending on the product type to be inspected. Such a ring light support structure is easily realizable.

Furthermore, as shown in FIG. 4, the ring light may be automatically set in the direction of the camera optical axis by a drive mechanism so that these settings can be changed flexibly.

FIG. 5 schematically shows the structure of a ring light in the third embodiment. N light emitting elements 22-1, 22-2.
..., 22-N and individually wire 23-1.23-
2. . . , 23-N is performed up to the drive circuit 24. The drive circuit 24 is a circuit for individually turning on and off the drive current according to control signals from the image processing device 4 or the control device 7. Such a circuit can be easily constructed from a resistor, a current switch circuit, a control signal decoder, and the like.

A cross section taken along the plane view A-A' is shown in the lower part of FIG. 5, and the light emitting element is bent toward the inside of the ring. The light flux from the light emitting element does not need to be a beam with sharp directivity, but rather it is desirable that it provides uniform light over the entire width of the object that is the field of view of the TV camera.

FIG. 6 shows the procedure for inspecting the wire shape using the ring light shown in FIG.

Bring the object into the field of view of the camera (Step 2o1), first turn on all the light emitting elements (Step 202), image the object, and extract the wire from the image (Step 203).
, from the extraction results, the position Ai(xi, y
i) and the direction θ are determined (step (204). The direction may be known.

Next, if there is an angle φ to be checked regarding the inclination of the wire, find the angle of incidence such that the reflected light enters the camera exactly when the inclination is φ. Of the incident angles, the angle θ with the camera optical axis is known, so find the angle θ' along the ring9 θ
Since the difference β between ' and θ can be calculated from φ and α using equation (1), it is possible to find θ' and determine a light emitting element that is closest to that direction. Since the light emitting elements can only be arranged discretely, φ' can also be determined from the angle θ'. By lighting only the determined light-emitting element and turning off the other elements, one point on the wire will light up (Step 2)
05). Find the brightness Ii along the wire (step 20
6) Compare this with the reference pattern Ti. As described above, there are various methods for this verification, but the method of comparing positions is particularly suitable for this embodiment.

If the matching error is within the range, it is determined whether to investigate other wire inclination angles (step 209).
. If it is checked, the process returns to step 205. If there is no need to investigate further, select the wire shape in step 210.
), the process returns to step 201 to proceed to the next target inspection. If the error is not within the range as a result of the verification, the wire shape is determined to be defective (step 211), and the defective product is dealt with (step 212).

This embodiment has the advantage that it is possible to detect various wire inclination angles with only one ring, and that the space required for the ring light is small. Alternatively, light may be transmitted from the light emitting element to the ring using optical fibers each having one end arranged in a ring shape. At this time, it is possible to make the ring smaller or set more angles.

[Effects of the Invention] As explained above, according to the present invention, the three-dimensional shape of a wire can be inspected in a non-contact manner by using one ordinary TV camera and a ring light that can be easily configured. This has the effect of enabling high-speed inspection with a simple device configuration. Additionally, there is no need for mechanical parts that move at high speed, allowing for long life, high reliability, and high-speed inspection.

[Brief explanation of drawings]

FIG. 1 is an apparatus configuration diagram of a first basic embodiment of the present invention;
FIG. 2 is a diagram showing the inspection procedure of the first embodiment, FIG. 3 is a configuration diagram of the illumination device in the second embodiment, and FIG. 4 is another configuration diagram of the illumination device in the second embodiment. Fig. 5 is a configuration diagram of the illumination device in the third embodiment, Fig. 6 is a diagram showing the inspection procedure in the third embodiment, and Fig. 7 explains the geometric relationship between the wire and the incident reflection of illumination light. This is a diagram. 1... Object to be inspected, 3... TV camera, 4... Image processing device, 5... Ring light, 7... Control device, 22... Light emitting element, 23... Light emitting element drive circuit. ,・'2) ,・゛I

Claims (1)

[Claims] 1. Ring-shaped illumination means that images the wire to be detected with a two-dimensional sensor and whose axis is the same as the optical axis of the secondary sensor,
A wire inclination detection method comprising: illuminating the wire and detecting the inclination of the wire with respect to a plane perpendicular to the optical axis based on a pattern of brightness and darkness occurring on the wire in the captured image. 2. The wire inclination according to claim 1, characterized in that a plurality of ring-shaped illumination means are provided on the optical axis of the two-dimensional sensor, and the ring-shaped illumination means are switched to illuminate the wire. Detection method. 3. The ring-shaped illumination means has a plurality of blinkable small light sources arranged along the ring of the illumination means,
3. The wire inclination detection method according to claim 1 or 2, characterized in that the small light sources are turned on all at once or sequentially.