JP3469490B2 - Component mounting equipment - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a component mounting apparatus used for mounting minute electronic components (hereinafter referred to as components) such as ICs, resistors and capacitors on a substrate.

[0002]

2. Description of the Related Art Conventionally, a component mounting apparatus is equipped with a suction nozzle in a mounting head unit, and the mounting head unit and a component recognition device move relatively to each other, so that the components picked up by the respective suction nozzles are removed. There is one that continuously recognizes the suction position. For example, regarding one component sucked by a head unit equipped with one nozzle, the suction position is recognized by a line sensor,
JP 61-287196, JP 03-110898
Etc. Further, in a rotary mounter in which a plurality of nozzles are arranged on the circumference, a line sensor is used to detect the posture of a component or the presence or absence of the component.
-202095, JP-A-63-202096, JP-A-6-2020
3-301600 etc.

[0003]

By the way, none of these known ones recognizes a plurality of parts continuously at the same time, and does not move parts relative to the recognition device to recognize the parts. Mounting efficiency is poor.

The present invention has been made in view of the above points, and it is possible to recognize a plurality of parts continuously and simultaneously.
An object of the present invention is to provide a component mounting apparatus which has a high mounting efficiency and which can easily control a recognition process by recognizing a plurality of components continuously and storing the component suction position data as continuous data.

[0005]

In order to solve the above-mentioned problems and to achieve the object, the present invention has the following constitution.

According to the invention described in claim 1, the " part supply unit
Equipped with a component mounting part to the board and multiple suction nozzles
The mounting head unit and the line sensor that moves relative to the mounting head unit ,
Image of the component sucked to the plurality of suction nozzles
Load the image data into the image memory and
Calculate the suction position of each part based on the image data
In each of the plurality of suction nozzles,
It has a component recognition device that recognizes the suction position of components, and calculates
Based on the picked up position of each part
Each component is preset on the board in the mounting part.
Is a component mounting device that is mounted at a different position , and the component recognition device is configured to attach the component according to the size of the component to be attached.
Small suction parts for each capture area set for each nozzle
The smaller the number of pixels, the shorter the image range in the relative movement direction.
In the image memory as intermittent image data that is set
A component mounting device characterized in that it is adapted to be taken in . ]. According to the invention of claim 2, "the part recognition
The device treats the intermittent image data as continuous data.
Is stored in the image memory.
The component mounting apparatus according to claim 1. ].

According to the present invention , since a plurality of components are continuously recognized at the same time, a plurality of components can be continuously picked up, continuously recognized, and continuously mounted, and the mounting efficiency is good.

Further, since a plurality of components are continuously recognized and the component suction position data is stored as continuous data,
The control of the recognition process is easier than when data is fetched and stored separately for each part.

[0009]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic configuration diagram of a component mounting apparatus,
2 is a plan view of the mounting head unit and the licensor unit, and FIG. 3 is a view seen from the direction of arrow A in FIG.

A carrier line 1 is provided in the center of the component mounting apparatus, and the carrier line 1 mounts a substrate 2 on the mounting portion 10.
It is designed to be transported to 0. A supply unit 3 is arranged on both sides of the transfer line 1, and the supply unit 3 supplies a component P by a feeder 4.

The component P of the supply unit 3 is the mounting head unit 6
And is moved by the operation of the horizontal direction moving means 7 and the vertical direction moving means 8 to be mounted on the substrate 2 of the mounting portion 100. The horizontal moving means 7 and the vertical moving means 8
Configures a component moving means Kl for moving the component P from the supply unit 3 to the mounting unit 1.

The mounting head unit 6 includes a head body 7
0 and the nozzle holder 7 provided on the head main body 70
1 and a plurality of suction nozzles 72 provided on the nozzle holder 71 so as to be able to move up and down.
At 2, the component P is sucked.

The head main body 70 comprises the shaft 9 of the lateral moving means 7.
The shaft 9 is rotated by a motor 10 so that the mounting head unit 6 moves laterally on the shaft 9. The bracket 11 of the lateral moving means 7 is attached to a shaft 12 of the vertical moving means 8, and the shaft 12 is rotated by a motor 13 so that the bracket 11 moves vertically on the shaft 12. ing.

The head body 7 of the mounting head unit 6
0 is attached to the shaft 9 of the lateral movement means 7, and this shaft 9
Is rotated by a motor 10 so that the mounting head unit 6 moves laterally on the shaft 9.

Further, the bracket 11 of the lateral moving means 7
Is attached to the shaft 12 of the vertical movement means 8, and the shaft 12 is rotated by a motor 13 so that the bracket 11 is moved vertically on the shaft 12.

Further, the bracket 11 is provided with a shaft 60, and the shaft 60 is rotated by a motor 61. The shaft 60 is provided with a line sensor unit 62, and the rotation of the shaft 60 causes the line sensor unit 62 to move on the shaft 60. The shaft 60, the motor 61, and the line sensor unit 62 constitute a sensor moving unit K2 that moves the line sensor in conjunction with the movement of the component P.

The line sensor unit 62 is composed of a line sensor 63 and mirrors 64 and 65. The mirror 65 is located below the component P sucked by the suction nozzle 72 of the mounting head unit 6, and an image of the component P is displayed. Is reflected from the mirror 65 via the mirror 64 and input to the line sensor 63.

In the process of moving the part P from the supply part 3 to the mounting part 100, the line sensor 63 is moved with respect to the part P to read the image of the part P and send it to the image processing part 20. Then, the relative position of the component P with respect to the mounting head unit 6 is detected based on the reading result of the line sensor 63, and the component P is mounted at the preset position based on the detection result.

That is, the parts P from the supply unit 3 to the mounting unit 100
In the process of moving to P, the line sensor 63 is moved with respect to the component P to read an image of the component P. Therefore, the line sensor 63 is used to move the component P from the supply unit 3 to the mounting unit 100. Can be recognized and the mounting time can be shortened.

For example, when the component P is located on the left side of the supply unit 3, it moves from the left side to the position on the substrate 2 of the mounting unit 100 along the paths al, bl, cl, dl, and el, and in the course of this movement, the line sensor. The unit 62 is moved with respect to the component P, the image is read, and the unit P is mounted at a predetermined position.

On the other hand, for example, when the component P is located on the right side of the supply unit 3, it moves from the right side to the position on the board 2 of the mounting unit 100 along the paths a2, b2, c2, d2, and el, and this movement process Then, the line sensor unit 62 is moved with respect to the part P to read an image, and is attached at a predetermined position.

FIG. 4 is a block diagram of the image processing section. The image information read by the line sensor 63 is input to the AD converter 21, and is input to the memory write controller 22 as digital information. On the other hand, a position signal of the component P and a fetching method designation command are input from the main body control device 26 to the fetching timing generator 23, and based on the command from the fetching timing generator 23, the memory writing control unit 22 determines the size of the component. The number of pixels written in the image memory 24 is changed according to the above.

The image memory 24 constitutes a storage means for storing a plurality of component image data taken in by the line sensor 63 as continuous data. Recognition unit 25
Has a calculation means for calculating the suction position of each component from the stored data, the recognition unit 25 recognizes the position and inclination of the component P based on the information stored in the image memory 24, and the mounting unit 100 The correction is made so that it can be mounted at a predetermined position.

In FIG . 4, an image from the line sensor 63 is displayed .
During the transmission of the image data signal to the memory 24, the part P
The number of pixels written in the image memory 24 is changed according to the size.
We are processing the image data so that
The data processing is the same as the captured image data in FIG.
Is done as follows. FIG. 5 shows an image capturing method using a line sensor. When capturing an image with the line sensor 63,
The number of pixels to be written in the image memory 24 is changed according to the size of the component P. That is, a memory for storing an image can be saved by capturing a rough image in a region having a low resolution and finely capturing an image in a region requiring a high resolution.

For example, in the example of FIG. 5, every one pixel and every one line in the capture area 101 where the component P is large, every pixel and every line in the capture area 102 where the component P is small, and the component P is large only in the central portion. 1 pixel in the capture area 103
Images are stored in the image memory 24 every other line.

As a result, the amount of memory required for all the pixels is 1/4 as compared with the case where the image is stored in the image memory 24 at each pixel and each line.

FIG. 6 shows a time chart of an intake area in which parts are large. In this example, the line sensor output b and the memory writing pulse d are intermittently output by the line sensor clock a, so that the image memory 2 is generated every other pixel.
4 is taken in every other line in the range Dl set by the output of the position pulse c indicating the position of the component P, so that an image of every one pixel and every one line is captured in the image memory 24. Take in.

FIG. 7 shows a time chart of the loading area where the parts are small. In this example, similarly, the line sensor output b generated by the line sensor clock a is continuously output as the memory writing pulse d only in the central portion of the line, so that the line capturing in which only the central portion is captured in the image memory 24 is performed. By capturing the line only in the range D2 set by the output of the position pulse c indicating the position of the component P, the image of each pixel and each line only in the central portion is captured in the image memory 24.

FIG. 8 is a flow chart of the image processing section.

In step a, a capture method designation command is input, and it is determined whether or not every other pixel is captured (step b). In the case of capturing every other pixel, 1 pixel.
Each line is taken into the image memory 24 (step c).
Based on the information taken into the image memory 24, the part P
The position and the inclination of the component P are recognized (step d), the position and the inclination of the component P are corrected (step e), and the component P is mounted at a predetermined position.

On the other hand, if it is determined in step b that the pixel is not captured every other pixel, each pixel and each line are captured in the image memory 24 in accordance with the time chart of the capture region in which the component P is small in FIG. 7 ( Step f). Based on the information taken in the image memory 24, the position and inclination of the component P are recognized (step g), the position and inclination of the component P are corrected (step h), and the component P is mounted at a predetermined position.

As described above, since the plurality of components P are continuously recognized at the same time, the plurality of components P can be continuously picked up, continuously recognized, and continuously mounted, and the mounting efficiency is good. Further, since a plurality of components P are continuously recognized and the component suction position data is stored as continuous data, the control of the recognition process is easier than the case where the data is fetched and stored separately for each component P. .

In this way, it is necessary to extract parts from the image captured in the image memory 24. Here, a case where a boundary of a part is converted into a chain code using a method called boundary tracking will be described.

It is necessary to extract parts from the image captured in the image memory. Here, a case where a boundary of a part is converted into a chain code using a method called boundary tracking will be described.

Here, the chain code represents, as shown in FIG. 9, a direction code of 0 to 7 indicating in which direction a pixel adjacent to the central pixel * is in a 3 × 3 area. A line image is represented by a sequence of start points and direction codes.

In the example of the line image shown in FIG.

[0037]

It is expressed as follows. [Boundary Tracking Process] 1) Detection Mode In this detection mode, for example, in order to find the part K in FIG. 11, the image is scanned from the upper left to the right. Here, when the part K is found, if the found pixel PO is not already tracked, the scan is interrupted and the mode is changed to the tracking mode.

If not found, the scan from left to right is moved down by one pixel and the same processing is repeated.

When the scan reaches the lower right corner of the screen, the process ends.

2) When the part K is found in the tracking mode [a] detection mode, the coordinates (x, y) of the found pixel PO are stored. Also, the same point 2
A tracking prohibition mark is attached to the pixel PO so as not to be tracked once.

[B] Whether or not there is a component around the detection pixel PO is examined in the order around the detection pixel PO (1 to 4) as shown in FIG. If not found, this indicates that the part K has only one pixel, in which case the tracking mode is ended.

If found, the coordinates of the first found pixel Pi (i = 1) are stored, and the tracking prohibition mark is also attached to the pixel Pl.

[C] The tracking direction from the previous detection pixel Pi-1 to Pi is stored as a chain code for the eight neighboring pixels centered on the detection pixel Pi, and FIG. 13 to FIG. As shown in FIG. 20, it is sequentially checked whether or not there is a component, and the first pixel found is set to Pi + 1. Further, in this case, a tracking prohibition mark is attached so that the same pixel is not tracked twice. 13 to 20, the symbol ◯ represents the pixel Pi and the symbol Δ represents the pixel Pi-1. The numbers represent the order in which the pixels around the pixel Pi are checked. Also,
The arrow indicates the tracking direction. The chain code in the tracking direction is as shown in FIG. [D] The first found pixel is set as a new detected pixel,
The processing of [c] is repeated. By doing this until the detection pixel Pi matches the tracking start pixel PO and the next detection pixel Pi + 1 also matches the detection pixel Pl next to the tracking start pixel, the contour of the component K is detected and converted into a chain code. It [E] When the tracking is completed, the mode is returned to the detection mode of 1) and the scanning is restarted. In this case, as shown in FIG. 22, since the scan enters the inside of the component K that has just been detected, the component detection is skipped until it goes outside the component K (or enters the hole). Next, a method for obtaining various feature quantities from the chain code obtained by this boundary tracking processing will be described below. Here, the feature amount is a basic item required in image processing. For example, the center of gravity of the target image can be obtained by the first moment, and the inclination of the target image can be obtained by the second moment. [1) 0th / first / second moment] For example, consider the case where a part as shown in FIG. 23 is traced. The moment feature amount is, for all points inside the traced part, 0th moment ... Pixel number first moment ... Accumulated value of X and Y (ΣX, ΣY) Secondary moment X ² Y ² accumulated The value (ΣX ² ΣY ² ) is obtained by calculating the integrated value (ΣXY) of XY. From this, as shown in FIG. 23, the moment feature amount is calculated by obtaining the right end point (○) and the left end point (●) of the contour and calculating the above values for all points in between. can do. A point (Δ) at which the right end point and the left end point coincide can be calculated by the same process. As shown in FIG. 23, in the example of the point P ₁ (x ₁ , y ₁ ) at the left end and the point P ₂ (x ₂ , y ₂ ) at the right end,

Is the

As shown in [Equation 1],

[0046]

[Formula 1]

(In the case of the second moment, the coordinates x and y are calculated by the relative coordinates from the center of gravity of the part.)
The value for the point between P ₁ and P ₂ can be obtained by calculating Therefore, while following the chain code, check whether the point of interest is the left end point / right end point,
By performing the corresponding calculation, the moment feature quantity of the part can be obtained.

At the point (Δ) where the left end point and the right end point coincide,
Perform the following calculations.

[0049]

[2) Area AREA] Area is Area = M0
It can be represented by 0. [3) Center of gravity XG, YG] The center of gravity XG, YG can be expressed by XG = M10 / M00 YG = M01 / M00. [4) Spindle inclination θ]

The inclination θ of the main axis is

It can be represented by [Formula 2].

[0052]

[Formula 2]

[0053]

As described above, according to the first aspect of the present invention, since a plurality of components are continuously recognized at the same time, a plurality of components can be continuously sucked, continuously recognized, and continuously mounted, and the mounting efficiency is good.

Since a plurality of parts are continuously recognized and the component suction position data is stored as continuous data,
The control of the recognition process is easier than when data is fetched and stored separately for each part.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a component mounting apparatus.

FIG. 2 is a plan view of a mounting head unit and a licensor unit.

3 is a view seen from the direction of arrow A in FIG.

FIG. 4 is a block diagram of an image processing unit.

FIG. 5 is a diagram showing an image capturing method with a line sensor.

FIG. 6 is a diagram showing a time chart of an uptake area in which parts are large.

FIG. 7 is a diagram showing a time chart of an intake area in which a component is small.

FIG. 8 is a flowchart of an image processing unit.

FIG. 9 is a diagram showing direction codes for a central pixel *.

FIG. 10 is a diagram showing a line image by a sequence of start points and direction codes.

FIG. 11 is a diagram of scanning an image from the upper left to the right in order to find a part in the detection mode.

FIG. 12 shows the pixels around the detection pixel PO in order (1 to 4)
FIG.

FIG. 13 is a diagram for checking whether or not there is a part in order according to the chain code.

FIG. 14 is a diagram for checking whether or not there are parts in order according to a chain code.

FIG. 15 is a diagram for checking whether or not there is a part in order according to a chain code.

FIG. 16 is a diagram for checking whether or not there is a part in order according to the chain code.

FIG. 17 is a diagram for checking whether or not there are parts in order according to the chain code.

FIG. 18 is a diagram for checking whether or not there are parts in order according to the chain code.

FIG. 19 is a diagram for checking whether or not there is a part in order according to the chain code.

FIG. 20 is a diagram for checking whether or not there is a part in order according to the chain code.

FIG. 21 is a diagram showing a tracking direction.

FIG. 22 is a diagram showing a case where component detection is skipped.

FIG. 23 is a diagram illustrating a method of obtaining a feature amount.

[Explanation of symbols]

6 Mounting head unit 63 line sensor 20 Image processing unit 22 Memory write controller 24 image memory 72 Suction nozzle P parts

Claims (2)

(57) [Claims] 1. A component supply unit and a component mounting unit for a substrate.
And mounting head unit equipped with multiple suction nozzles
And a line sensor that moves relative to the mounting head unit, stores image data of each component sucked by each of the plurality of suction nozzles in the supply unit in an image memory.
Capture, each of the above based on the captured image data
By calculating the suction position of the part,
Recognize the suction position of each of the above-mentioned components sucked on each
And a component recognition device that
Based on the position, attach each of the parts to the mounting part.
In the component mounting device, the component mounting device is mounted at a preset position on the board , and the component recognition device is configured to match the size of the component to be sucked.
Correspondingly, for each capture area set for each suction nozzle
, The smaller the suction component, the larger the number of pixels and the relative movement direction
As an intermittent image data in which the image range of is set short
A component mounting apparatus, which is adapted to be loaded into the image memory . 2. The component recognizing device is configured to detect the intermittent image data.
Data in the image memory as continuous data
The component mounting device according to claim 1, wherein
Place