JPS58196405A - Method and device for aligning fixed focal point - Google Patents

Abstract

PURPOSE:To perform the alignment of a focal point automatically and efficiently, by photographing the filament of a lamp at the front and side by ITV cameras, computing the center of gravity and the slant amount of a coil, and driving a positioning driving part. CONSTITUTION:The front image, which is photographed by an ITV camera 13 is outputted to an image unit 36. The coordinates of the center of gravity of the front image is computed. An X stepping motor 27 and a Z stepping motor 22 are driven so that said coordinates are aligned with reference coordinates, and a lamp 4 is moved in the X direction and Y direction. A slant angle theta of the axial line of the front image is obtained, and a coil 3 is rotated by a stepping motor 24 by the angle theta. The coordinates of the center of gravity of the side image are computed by an image signal SB indicating the side of the coil 3, which is photographed by an ITV camera 14, and a rotary angle alpha of the coil 3 is obtained. The coil is moved and rotated by a Y stepping motor 29, the Z stepping motor and an alpha stepping motor 31, and the fixed focal point alignment of the coil 3 is completed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to a method and apparatus for automatically adjusting the focal point of a lamp that emits light onto a filament.

[Technical background of the invention and its problems]

In the assembly process of small halogen light bulbs for automobiles, there is a step in soldering the L-shaped 7-lunge (1) and the lamp base (2) shown in FIG. In this process, when soldering the 7 lunge (1) and the ramp paste (2), the 7 lunge (
Helical coil (3) with axis tV relative to the reference plane of 1)
It is necessary to perform positioning (tilt, up and down - left and right).

Therefore, in the past, as shown in Figure 2, a self-lit lamp (4) was sandwiched between lenses (5) on both the left and right sides.
, (5) 'e through the photocell (6) , (6
)=Th, and the amplifier (7), (differential amplifier (8) via the force). (1) Fixed focusing was manually performed by slightly moving the lamp pace (2) with respect to K. However, in this method, the coil (3) was
), In (6), the position is recognized indirectly based on the average amount of light received, and is not based on the direct position of the coil (3) itself. Therefore, the positioning accuracy is poor and automation is difficult. It is.

[Purpose of the invention]

The present invention has been made in consideration of the above-mentioned circumstances, and provides a fixed focus adjustment method and apparatus for automatically and efficiently adjusting the fixed focus position of a lamp that emits light onto a filament without human intervention. It is on offer.

[Summary of the invention]

After imaging the lamp filament and the 2ITV power camera Vc from the front and side, the video signal is converted into a binary value.
The center of gravity position, inclination, and content of the coil as seen from each side are calculated based on the binarized image data, and the positioning drive unit is designed to perform fixed focusing based on these calculation results. .

[Embodiments of the invention]

Hereinafter, the present invention will be described in detail based on embodiments with reference to the drawings. In the following description, the same parts as those used in the description of the background art are given the same symbols.

FIG. 3 shows the overall configuration of the fixed focusing device of this embodiment. It shows. This fixed focusing device binarizes the image signal output from the imaging unit αQ that images the front and side faces of the coil (3), and performs various arithmetic processing based on the binarized data. A calculation control unit Qυ that performs (not shown)
and an energizing section (not shown) that energizes the A7 lunge retainer (3).

The imaging unit Ql includes a first ITV camera, a second IT
It consists of a V camera I and camera controllers a and αe connected to these separately. The above first IT
The V camera (2) is held by the flange holding part.
Coil (3) of lamp (4) inserted into lunge (1)
It is arranged at the front t-image capturing position.

On the other hand, the second ITV camera α4 also has a lamp (4).
It is arranged at a position to image the side surface of the coil (3). Be friends, number 1. Second ITV camera (1 bottle).

I are placed in mutually orthogonal directions. FIG. 4 shows the configuration of the positioning drive section 0. As shown in FIG. In addition, as shown in this figure, the flange (1) is fixed to the lamp (4) tube which is inserted downward into the seven lamp holding parts. At this time, the flange (1) O projection piece (17
) , (I'l) , (17) The surface including the top surface is "
"Reference surface". The lamp head (2) inserted into this flange (1) has its electrode parts (2a), (2
b) 19 is held in the t chuck α barrel. This chuck (
II consists of a main body and a pair of clamping pieces (CA), one of which protrudes from the main body so that the position can be adjusted freely, and fixes the electrode parts (2a) and (2b), and CA. From these clamping pieces, coils (3) are passed through the wires.
It is supposed to be energized. This chuck bet is fixed on the Z table (b). This Z table (2)
is supported on the Z table support by two stepping motors so that it can move forward and backward in the Z direction (direction perpendicular to the reference plane) in FIG. 4. Furthermore, the Z table support is fixed to the rotating shaft of the θ stepping motor (c). The rotational axis (c) of this θ stepping motor (c) is parallel to the Y direction in FIG. It is set as follows. therefore,
The Z table (c) is rotatable about one axis (2) (
(Fig. 4).

The lower end of the above Z table support @ is the X table @V
C is fixed. This X table) is a bus.

The chipping motor is supported by a Y table (c) so as to be able to move forward and backward in the X direction (direction perpendicular to the Y direction) in FIG. 4. This Y table) is a Y stepping motor @
It is supported by the Y table support body so that it can move forward and backward in the Y direction. This Y table support member is fixed on an α table (to) which is rotatable in the α direction in FIG. 4 by an α stepping motor (b). This α table (2)
The axis of the rotating shaft is set to be parallel to the Z direction in FIG. 4.

The above positioning drive unit a4t-constituting x, y, z, θ, α stepping motor @, 翰, @, (c), al) are as follows:
■Each is connected to the driver interface separately.

Meanwhile, Ml, the second ITV camera Q3. Q4)! 'j
:, selector switch e! It is connected to the image unit (the I image unit) which has the function of binarizing the video signal and storing data of logic "1" or rOJ in a fixed location corresponding to each pixel. These driver interfaces (to the image unit) are connected via the system bus@LTECP
U (Central Processing Un)
口; Central processing unit 1t) c4VC connected. This CPU wheel contains an arithmetic control program for fixed focus positioning, and an fcROM (Read Only
Memory) 7p where the wire and various variable data are stored.
ELA M (Ran-dom Access Me
% are connected via the system bus @. Further, a control interface (b) that outputs a switching control signal to the changeover switch (to) is connected to the system bus (b). However, there is a driver interface (goods) and a changeover switch (to).

Image unit c4. System bus @, CPU (to), R
OJl, RAM), control interface (b),
For example, arithmetic control unit 01) such as a microconbeater
It is fully configured. Although not shown, the binarized image in the image unit 1 is displayed on the TVVC monitor.

Next, an explanation of the operation of the above-mentioned fixed focusing device and %,
The fixed focusing method of this embodiment will be explained.

First, the flange (1) is fixed at a predetermined position on the flange holding section. Next, insert the lamp base (2) of the lamp (4) into the entire flange (1), and insert the clamping piece (to) of the chuck that protrudes from the other end of the flange, 12Ivr:
, grip the engineering υ. At this time, the axis of the coil (3) and the rotation axis of the θ stepping motor (c) are set to be perpendicular to VC. However, the clamping piece c! I
.

(to), electrode parts (2a), (2b) are fully valved, and the coil (
3) energizes and lights up by itself. Furthermore, the first ITV
If the camera Q3'Th is installed at a position where it can image the front of the coil (3) whose axis direction is perpendicular to the rotation axis (c) (imaging from the direction perpendicular to the axis of the coil (3)), The second ITV camera α4) t- is installed at the fill (in the axial direction of the coil (3)) so as to be able to image the side surface of the coil (3). At this time, the arithmetic control unit uses the first ITv camera (i3Vcz) under the same conditions as the above-mentioned friend as the reference coordinates (X10.ZF.) of the center of gravity stone 1 in the front image of the coil (3) located at the imaged foot focal position. a1) in the RAM). Keita, second ITV
The center of gravity in the side image f# of the coil (3) located at the foot focal position captured by the camera Q4) All second reference coordinates (Y
BO, Zllo) in the RAM- of the arithmetic control unit I. Next, we move on to image processing, but first, the CPU (
The switching signal SA is output from the control interface (2) to the changeover switch (to) in response to an instruction from the first IT
The frontal image (6) (see Figure 5) captured by the V camera a is captured by the camera controller Q! 9'! It is output to the image unit (to) as a video signal 8B via r.

In this image unit (to), the video signal 5Bf2 value 1
After exchanging love, a logic "1" or a logic "0" is stored in a predetermined location provided for each pixel. Based on the data stored in this image unit (to), the CPU (to)
The center of gravity coordinates (X, ZA) of the front image □□□ are calculated and temporarily stored in 几AM (to). Next, in the CPU (c), the coordinates (x, z,) are the first reference coordinates (X
,. , ZFO) to the X stepper motor) and Z stepper motor @ to match the drive signal 8C, 8
Output D=i and ramp (41) in the X and Z directions in Figure 4.
1 (block - in Figure 7). However, the fifth
The first reference coordinate (X70.”
When the center of gravity coordinates (X, Z, Then, the barycenter coordinates <X;, Z->* are determined again at
l, ) is the first reference coordinate (XyO.

Z. Lamp (4) as before to match the LIC? (block θ''7)). This operation is performed to remove the disturbance caused by the outer peripheral area (to the outer frame).Next, move the coil (3) in the image area (b).
) Find the inner frame (goods) circumscribing the front image O-bu (block ■). Next, in order to eliminate unevenness in brightness and unevenness in light emission at both ends of () in the front image, create a vertical line
−X4 (block II).

However, X, , X, are the X coordinate values of both the left and right ends of the inner frame (goods), respectively. Therefore, from the data of the front image (between the vertical line X and the vertical line Find (block (51)) Next, calculate the obtained linear equation and each vertical line
Father point coordinates with X4 (Xm,"a), (X1gZ4
), the inclination angle θ of the axis of the front image (b) (see Figure 5) is calculated using the following formula QvC, and the calculation control signal SE is output to the θ stepping motor (c), and the coil (3) t - The axis of the block (52) (resulting in the front image of the coil (3)) is parallel to the X direction of the image area (2). Next, CPU (to)
According to instructions from the control interface θη, a switching signal 8A'i is output to the switch (c), and a video signal 8B°i showing the side of the coil (3) imaged by the second factory TV camera (14) is outputted from the control interface θη. output to the camera controller α (to the image unit via 9) (block (53)).

In the image unit (to) that inputs the video signal SBI,
-6X・.

The video signals SB' and ll' are converted into binary values, and a logic "1" or logic "0" is stored in a predetermined location provided for each pixel. Based on the data stored in this image unit (to), the CPU (to) generates a side image (54) shown in FIG.
, (54) The barycentric coordinates (Y, Zs) of (two symmetrically displayed) are calculated and stored in the RAM. Next, in the CPU (to), the center of gravity coordinate (y, 8th) is the 8g20 reference coordinate (Yso).

Drive signal SF to Y stepping motor 1 and Z stepping motor 2 so as to match Z110').

Output SD and ramp in the Y and Z directions in Figure 4 (4)
(block ring). Therefore, the second reference coordinates (Yso.

2. ) and the center of gravity coordinates (Y8.Z, ) match, the side image (54) = (
54) k enclosing outer frame (56) t - set (to block)). Therefore, the outer frame (56) K Te again has the center of gravity coordinates (
Y'a, Zl, )? Find this centroid coordinate < Yl
a, Zl, ) are the second reference coordinates (Yso, ”
go ), do the same as before to match the ramp (
4) Move t'' (block 67)).

Next, a side image (54) within the image area (55).

(54) Inner frame circumscribing VC (57) t-find (in block)).

Next, the axis Z parallel to the horizontal line of the inner frame (57), 1F! : Find F using the formula ■ (block (58)).

Here Z6 and 2. indicate the Z-axis coordinate values of the upper horizontal line and lower horizontal line of the inner frame (57), respectively. Therefore, the intersection coordinate (Y
s, z, )-(Ya-Z?)-(Ya-Zy
), (Ys, z, )t-finding, axis y
Upper 1c side image (54), left and right width W1 of (54). W, is calculated by the following formula 〇 (block (59
) ).

Therefore, L coils (3
) rotation angle αf: Calculate the following formulas ■ and ■ (block (
60)).

α=to α・・・・・・・■Here, K is
Determine the constant value experimentally in advance. Note that the rotation direction is the left and right side images (54).

This is the direction in which the widths W1 and W of (54) are equal.

Therefore, the difference between the two is determined and the direction of rotation is completely determined based on whether the difference is positive or negative. Therefore, the drive signal SGiα step/
When the output is output to the stepping motor 6 and rotated by the angle α (block (61)), the focusing of the coil (3) is completed, but the coil (
When rotating the coil (3), the front side of the coil (3) may be slightly misaligned, so if high-precision constant focusing is required, change the position from the block ring to the block (on the front image again). 52) 'Repeat the operation at f (block (62)). However, when the fixed focusing is finally completed, the lamp base (2)? 7 lunge (1) VC soldering is solid.

The positioning method and the positioning device according to this embodiment of the present invention are as follows: Since the coil (3), which is self-lit (a'a, a3), is directly imaged and the positioning drive 11Qa is used to perform positioning based on the images gl from these different directions, positioning can be performed quickly and with high precision. It can be automatically determined with precision.

In the above embodiment, the center of gravity coordinates of the front image (in the front image) are determined twice, once within the image area (goods) and once within the outer frame (), but if there is no problem with positioning accuracy, the coordinates of the center of gravity of the front image
The operation of determining the coordinates of the center of gravity in (c) is omitted.Similarly, the linear equation Z= a
The process of cutting out the vertical lines X, , X4vcx when calculating 2 can also be omitted. Furthermore, it is also possible to calculate the inclination angle θ of the axis of the front image to θ−tan a using the coefficient a of the linear equation Z = a X + b. moreover,
Side image processing may be performed first instead of using the fixed focusing procedure shown in FIG. 7. Furthermore, as another example,
Directly below the axis of the coil (3) (lamp (4)
) optical axis direction) and the rectangular shape shown directly to the side.
The advantage tV is that the complicated arithmetic operations shown in FIG. 6 are no longer necessary, and the rotation angle α can be directly determined.

〔Effect of the invention〕

The fixed focusing method and device of the present invention capture images of a spiral coil of a lamp in a lit state from mutually orthogonal directions, and detect positional deviations from the focused coil based on the captured images. It is a device that quickly and accurately adjusts a fixed focus based on the position of the foot, making it possible to automate the fixed focusing work that previously relied on manual labor, greatly reducing the burden on workers. In addition to this, production efficiency also increases.

[Brief explanation of drawings]

FIG. 1 is a diagram showing a lamp attached to a flange, FIG. 2 is a diagram showing a conventional focusing method, and FIG. 3 is an electric circuit diagram of a focusing device according to an embodiment of the present invention. FIG. 4 is a block diagram of the positioning drive section of the fixed focusing device shown in FIG. FIG. 7 is a flowchart showing a foot focusing method according to an embodiment of the present invention. (1): 7 rank, (3): “Il, uI:
Imaging unit, (11): Arithmetic control unit, Qri: Positioning drive unit, αri: First ITv camera, (14): gg 2
(7) ITV camera. Agent Patent attorney Noriyuki Chika (and 1 other person) 2 figures $1 figure 13 figure ¥te figure

Claims (8)

[Claims] (1) In a foot focusing method in which a lamp having a spiral coil is attached to a flange and a fixed focal point is determined, the above-mentioned flange? a method of holding the lamp in a fixed position, a method of holding the lamp in the above-mentioned -leg position and inserting it into the flange, a method of lighting the lamp inserted in the seven lunges, and a method of lighting the lamp inserted in the seven lunges. A method for capturing an image of a coil, converting it into a binary image, and determining the amount of positional deviation of the coil from the fixed focus position based on the binary image, and determining the fixed focus position of the coil of the nine lighting lamps based on the amount of positional deviation. A foot focusing method comprising: (2) The fixed focusing method according to claim 1, characterized in that the coil of the lamp is imaged 20,000 times orthogonally to each other. (3) One imaging direction is the axial direction of the coil, and the other imaging direction is perpendicular to the axial line of the coil and perpendicular to the base axis of the lamp. How to focus. (4) The fixed focusing method according to claim 1, characterized in that the amount of positional shift is determined at the center of gravity of the binarized image. (5) The fixed focusing method according to claim 1, wherein the amount of positional shift is determined by calculating the inclination angle vc of the axis of the binarized image. (6) The foot focusing method according to claim 1, wherein the inclination angle of the axis is determined using the least squares method. (7) A leg focusing device that positions a 4VC fixed focal point when a lamp with a spiral coil is attached to a flange includes a 7-lamp holding part that fixes the flange at one leg position, and a 1kV chuck that holds the lamp. a positioning drive unit that supports the lamp in a freely movable position; an energizing unit that energizes the coil of the lamp to turn it on; an imaging unit that takes an image of the lit coil; The output signals are binarized and stored for each pixel, and based on these stored data, the positional deviation of the fixed focus position of the illuminated coil is calculated and sent to the positioning drive section. 1. A fixed focus adjustment barrel, comprising: a calculation control unit that outputs a drive signal for correcting positional deviation. (8) The imaging unit is an ITV camera.1-*Fixed focusing according to claim 7, wherein the feature is 1-*! "Device.