JPS62266403A - Position recognizing instrument for three-dimensional object - Google Patents

Abstract

PURPOSE:To enable a required part to be detected to be surely recognized by arranging the part located on a plane so that the projected images of the corners of the part are formed on the same plane and calculating the shadow information of the images by combining a plurality of pieces of the information. CONSTITUTION:The irradiation of light on a circuit part 3a from an oblique direction (4) forms shadows 20a-20d. Accordingly, an image pickup camera picks up and stores (10a-10d) the shadows 20a-20d, an electrode 30, a white marking 32 and the like. A density converting circuit is set so that the electrode 30 and the marking 32 are clipped to prescribed values. Then, a difference between the image signals A and B of memories 10a and 10b, respectively, is calculated (11a) for every signal of the same coordinates and difference signals E are outputted to an adding circuit 11c. Difference signals F between the image signals C and D of memories 10c and 10d, respectively, are outputted to the circuit 11c. Signals G are extracted as a region 36 with a zero density in the circuit 11c wherein the electrode 30 and the marking 32 are separated from each other. The signals G are binary-coded (12) to remove the unevenness of illumination, binary-coded signals H are inputted to an arithmetic circuit 16 and the center of gravity of the electrode 30 is arithmetically processed (16) whereby the presence or absence, positional deviation and the like of the part 3a can be recognized with a high accuracy.

Description

[Detailed description of the invention] [Purpose of the invention] (Industrial application field) This invention relates to components mounted on one circuit board.

The present invention relates to a three-dimensional object position recognition device that recognizes the shape and position of a detected object.

(Conventional technology) House 1! As the circuit boards used in equipment and OA equipment become smaller, for example, chipped circuit parts are used instead of circuit parts with lead wires.
The miniaturization of installed parts is progressing. Furthermore, as the mounting parts become smaller, the mounting density is also becoming higher.

Normally, components are mounted on a circuit board by temporarily mounting them on a board on which a predetermined wiring pattern is formed using an automatic mounting machine, and then soldering them using an automatic soldering machine. However, such automated mounting of circuit components often results in parts falling off, misaligned, or being misplaced. If this soldering is later discovered, a great deal of effort will be required to correct it. Therefore, it is necessary to detect and correct defects such as dropping of one component, misalignment, or wrong component before soldering.

Conventionally, defects in circuit boards have been detected mainly by visual inspection, but such a detection method requires obtaining a large number of defects, is not only inefficient, but also lacks reliability.

Therefore, conventionally, systems have been developed that perform this detection by capturing an image of a circuit board with an imaging camera 2, processing the image information, and detecting a dropout or a positional shift of a circuit component.

However, the detection systems developed so far only require wiring patterns and printed characters on a single circuit board.

From among a mix of circuit components of various types and shapes.

In order to extract and detect the required parts, the P1 degree, face credibility, etc. are not sufficient.

(Problems to be Solved by the Invention) This invention has been made in view of the situation that the above-mentioned three-dimensional object cannot be reliably recognized.
An object of the present invention is to provide a three-dimensional object position recognition device that can reliably recognize a desired target object to be detected.

[Structure of the invention]

(Means and operations for solving the problem) Illumination means arranged so that projected images of each corner of a three-dimensional object placed on a plane are formed on the same plane, and sequentially blinks the images; An imaging means that images the shadow of each corner formed by the illumination means from above, and a plurality of pieces of shadow information of each corner imaged by this imaging means are combined and calculated to determine the position of each corner. It has a calculation means for calculating Kg so that the required detected object can be reliably recognized.

(Example) Hereinafter, the present invention will be described based on an example with reference to the drawings.

FIG. 1 shows the configuration of a three-dimensional object position recognition device that recognizes parts attached to a circuit board, which is an embodiment of the present invention. As shown in this drawing, the 1-circuit part version (1) has a required wiring pattern (2) formed on its board surface, and multiple types of circuit components (3a) to (3d) are attached to the required parts. ing. Therefore, this recognition device is placed on the circuit board (1).

An optical lens (5) is attached to the front surface facing this circuit board (1).
) having an imaging camera (6) mounted thereon, and a circuit board (1) within the field of view of the imaging camera (6) with 9 on its side;
Four light irradiation devices (7a) that irradiate light onto a fixed area from diagonally above the four corners of the circuit board (1), (7
b), (7C).

(7d) is provided. Among these, light irradiation devices (7a), (7b) and light irradiation devices (7C), (7d
) They are arranged at positions facing each other. The above-mentioned imaging power) 5 (6>K means the above-mentioned light irradiation device (7a),
(7b), (7C).

An A/D conversion circuit (8) is connected which performs A/D conversion of image information obtained by imaging the predetermined area under the light irradiation (7d) according to its brightness level. A density conversion circuit (9) is connected to the conversion circuit (8) for converting the density level of the image information that has been subjected to ~Φ conversion.

This density conversion circuit (9) has a first circuit that stores density-converted image information. Second. Third and fourth image memories (10
a), (10b), (10C), and (10d) are connected.

These images (10a), (10b), (IO
C), (10d) ji.

Light irradiation bag R (7a), (7b), (7C) respectively
, (7d) is a memory that stores image information captured under the light irradiation. These first. Second image memory (1
0a) and (10b), these image memories (101
), (10b) is connected to a first subtraction circuit (l1m) that calculates the difference between signals on the same coordinates, and on the other hand, third and fourth image memories (IOC),
(10d) are these image memories (IOCs), (10
A second subtraction circuit (llb) is connected to calculate the difference between signals on the same coordinates with respect to the image information stored in d). Furthermore, these first and second subtraction circuits (lla),
The output side of (llb) is the subtraction circuit (lla),
(llb) Connected to the input side of the addition circuit (IIC) that adds the subtraction results in K on the same coordinate. Therefore, the output side of the adder circuit (IIC) is as follows.

It is connected to the input side of a binarization circuit α2 in which a threshold value for binarization is set. Furthermore, the output side of this binarization circuit α2 integrates the binarized signals in one or more predetermined directions to obtain one or more pieces of projection image information, and this projection image information will be described later. While performing predetermined arithmetic processing, the light irradiation devices (7a), (7b), (7c
), (7d) f) k illumination switching error illumination FIA control via switching circuit (14) and imaging camera (6).

A/D conversion circuit (8), 1st. Second subtraction circuit (lla)
, (llb).

An arithmetic control circuit aQ is connected via a timing circuit a3 to control the operation timing of the addition circuit (11C), the binarization circuit α2, and the density conversion circuit (9).

The results of the arithmetic processing of the projection image information are displayed on a display device αη such as a printer connected to the arithmetic control circuit αe.

Next, we will discuss the operation of the position recognition system for this three-dimensional object.

First, a circuit board (1) or an imaging camera (6) and a pair of light irradiation devices! (7a), (7b), (7C),
(7d) and move the imaging camera (6) over a predetermined area of the circuit board (1).

That is, for example, as shown in FIG.
1) is positioned over the area including the circuit component (3) located approximately in the center. Then, according to a command from the arithmetic and control circuit α0, the light irradiation device (7a) is first turned on to irradiate the predetermined area with light. Then, the irradiated area is imaged by the imaging camera (6).Then, the image information obtained by this imaging is sent to the first image memo 9 (10a) via the Φ conversion circuit (8) and the density conversion circuit (9). Next, the light irradiation device (7a
), switch the lighting of (7b).

The light irradiation device (7b) irradiates the same area as the irradiation area of the light irradiation device (7a), and images this irradiation area in the same manner as described above. The image information obtained by this imaging is then stored in the second image memory (10b) via the ~Φ conversion circuit (8) and the density conversion circuit (9). below.

The image information when irradiated by the light irradiation device (7C) is stored in the third image memory, and the image information when irradiated by the light irradiation device (7d) is stored in the fourth image memory.

By the way, as mentioned above, the light irradiation devices (7a), (7b), (7C), (7
When irradiated from d), as shown in Fig. 2), the circuit component (3a) protruding from the circuit board (1) will cast a shadow on the side opposite to the light irradiation according to its protrusion height. (20a)
, (20b).

(20C) and (20d) are produced. Therefore, the imaging camera''') (6) can capture these shadows (20a) and (20b).
, (20c), and (20d), so the first. Second. Third. fourth image memory (10a),
(10b), (IOC), (10d) have these shadows (20a), (20b), (20c), (2
density-converted image information including od) is stored. By the way, one circuit component (3a) has electrodes (to) at both ends.
, has a circle, and .

At its periphery, there is a copper pattern Gl connected to the electrode (to).
) and a white marking C3LC33 for specifying the mounting position of the circuit component (3Jl). In this case, white! -King 03. eta, electrode (to), and C31 all appear white as images. Therefore, the electrode (to).

(7) White marking from part 02. (to) must be separated.

It is not possible to accurately recognize the circuit component (3a).

Therefore, as shown in FIG. 3, the density conversion circuit (9) is set so that the electrode (7) and the white marking (■9 country) are clipped to a constant value. In addition, the shadow (20a),
(20b), (20C), (20d) and copper bap
- (oI) and the surfaces □□□ other than the electrodes ■ and (to) of the circuit component (3a) undergo concentration conversion in a direct proportional relationship.

That is, one image signal is 1. For example, when the input density level is equal to or higher than α, all the output values are constant at α, for example.

When the input value is α unrefined, the density conversion circuit (9) performs conversion processing so that the output value and the input value become the same value. As shown in FIG. 2, the image signals A and B stored in the first image memo 9J (10a) and (10b) are called to the subtraction circuit (lla), and each signal at the same coordinates is The difference between them is calculated, and a signal E indicating the difference is further output to the addition circuit (IIC). On the other hand, 3rd. Fourth image memory (IOC), (10d) Also calculates the difference between the stored image signals C and D by outputting them to the subtraction circuit (llb), and adds a signal F indicating the difference. Output to the circuit (IIC). Thus, FIG. 4 shows the image represented by signal E, and FIG. 5 shows the image represented by signal F. Here, the shaded part (b)
, (to) shows the electrode COG, (7) and the white marking cormorant, oa extracted from other parts by the subtraction process,
Since the density levels are clipped to the same value, the density is zero. Thus, the adder circuit (lie) K is
Signals E and FK are added and signal G is output. This signal G is, as shown in FIG.
) and white marking o21. oa is separated and extracted as a beveled edge region (to) with zero density. However,
When the signal G is binarized by a binarization circuit (L) with a threshold value of concentration 1, for example, a binarized signal H is obtained in which one electrode ω has a logical value "l". In particular, in this case, A /D Since the image information is density-converted by the density conversion circuit (9), the light illumination devices (7a), (7b), (7C), (71
Is it KJ illumination? , noise can be removed by the directionality of the parts.

Next, in the control circuit Jt1 (l) to which this signal H is input, as shown in FIG.
Center of gravity (" t Yc ) w (xcy )'c )
By processing the .

Presence/absence of circuit component (3a), which is the object to be detected, 1 positional deviation.

It is possible to recognize the inclination θ at high speed and with high accuracy.

In the above embodiment, four light irradiation devices were used to selectively irradiate a predetermined area, but this light irradiation device
Any shadow may be used as long as it casts a shadow evenly around the target part, and the number is not particularly limited to four.

Furthermore, in the embodiment described above, the image information obtained from the imaging camera is once stored in the first image. Second. Third. Although the image information is stored in the fourth image memory, the image information obtained later may be processed in real time without being stored in the first image memory. Also.

The density conversion circuit (9) may convert values above a certain level into fixed values, or may perform variable length value conversion processing into fixed values below a certain level.

Although the above embodiment describes the recognition of circuit components on a circuit board, the present invention can also be applied to the detection of other objects to be detected.

〔Effect of the invention〕

The projection image of each corner of a three-dimensional object placed on a plane is formed on the same plane by means of illumination means, and the shadow information of each corner obtained by sequentially blinking the images is obtained. Since the position of each corner is calculated by performing a combination calculation on a plurality of corners, the target object to be detected can be reliably recognized.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the configuration of a three-dimensional object position recognition device that recognizes circuit components attached to a circuit board, which is an embodiment of the present invention, and FIG. Figure 3 is a graph showing density conversion, Figures 4 to 6 are explanatory diagrams of image processing, and Figure 7 is circuit components. FIG. 3 is a diagram showing a recognition method. (3a) to (3d): Circuit components (three-dimensional objects), (6
): Imaging camera 2 (imaging means). (71), (7b), (7C), (7d): Light irradiation device (illumination means). αls: Arithmetic control circuit (arithmetic means). Agent Patent Attorney Nori Chika Yudo Kikuo Takehana! TO Night Figure 1 C2Db, 20c, 20d) Figure 3 4s 4 Figure 11゜(. Figure 6 Figure 5 Figure 7

Claims (1)

[Claims] An illumination means arranged so that projected images of each corner of a three-dimensional object placed on a plane are formed on the same plane and sequentially blinks the projected image, and a shadow of each corner formed by the illumination means The present invention is characterized by comprising: an imaging means for imaging from the top surface; and an arithmetic means for calculating the position of each corner by calculating a combination of a plurality of pieces of shadow information of each corner imaged by the imaging means. Three-dimensional object position recognition device.