JPH08213587A - Image reader - Google Patents

Abstract

PURPOSE: To secure the number of pixels which contribute to reading by dispensing with cumbersome packaging work and reliance on the size of an object to be read. CONSTITUTION: Image-pickup surfaces 111 and 121 of CCD line sensors 11 and 12 with different pitches among pixels are positioned in an image-forming surface, respectively, shunted by a beam splitter 10, and either one of the CCD line sensors 11 and 12 is selected by a switch SW depending on the size of a subject component. An image processor 17 performs image processing while recognizing which CCD line sensor is used.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image reading apparatus suitable for image recognition of a chip mounter, for example.

[0002]

2. Description of the Related Art Conventionally, electronic devices such as ICs and LSIs, chip-like chip parts, etc. are adsorbed by a head unit equipped with an adsorption nozzle and transferred onto a positioned printed circuit board. A chip mounter mounted on a predetermined position of a printed circuit board is known. Such chip mounters are
One-dimensional CCD (CCD line sensor) when transferring
An image of the component or the like is read by scanning upward, and the shape or the positional relationship between the chip component and the head unit is recognized. Here, the reason why the CCD line sensor is used for reading the image of the chip mounter is nothing but the shortening of the mounting time. This is because in the two-dimensional CCD sensor, it is necessary to temporarily stop the element or component that is the subject and capture an image, which causes a loss of time and cannot reduce the mounting time required in the chip mounter. .

By the way, the size and required positioning accuracy of electronic elements and chip parts are very different. For example, QFP (Quad Flat Pack) such as LSI
age) is as large as several tens of millimeters square, while some chip parts such as chip resistors and chip capacitors are as small as only 1 millimeter square.

[0004]

Here, for example, CC
The number of pixels of the D line sensor is set to 512, and the QF
It is assumed that the image reading angle of view is set to 50 mm square in order to read the P image. In this case, if the size of the QFP is 40 mm square, the number of pixels contributing to the reading of the QFP is about 400 pixels, which is necessary and sufficient for the reading accuracy. However, when a 1 mm square chip component is read in this setting state, the number of pixels contributing to the detection of the chip component is only about 10 pixels, which is insufficient for the reading accuracy and the mounting accuracy is low. Adversely affect. In this case, of course, this problem can be solved by resetting the angle of view for image reading by replacing the lens. However, the angle of view is set again according to the size of the part as the subject, and the mounting work is complicated.

Further, if a CCD line sensor having a high resolution, for example, a pixel having about 5000 pixels and a pitch between pixels is denser, this problem is apparently solved. Seem. However, as a matter required for the chip mounter, not only improvement of mounting accuracy but also reduction of mounting time is important as described above. Since a high-resolution CCD requires a lot of time to read because of the large number of pixels, for this reason, there is a circumstance that it is not desirable to use it in a chip mounter.

The present invention has been made in view of the above-mentioned circumstances, and an object of the present invention is not to complicate mounting work.
Moreover, without depending on the size of the object to be recognized,
An object of the present invention is to provide an image reading device capable of ensuring the number of pixels contributing to reading and performing highly accurate image recognition.

[0007]

In order to solve the above-mentioned problems, the invention according to claim 1 is a one-dimensional CCD in which an image pickup surface is located on an image forming surface of an object at the time of image pickup,
At least two or more one-dimensional C having different pitches between pixels
A CD and selection means for selecting one of the outputs of the two or more one-dimensional CCDs according to the size of a subject are provided, and the selection result of this selection means is used as an image signal. Is characterized by. According to a second aspect of the present invention, in the first aspect of the present invention, a beam splitter for shunting incident light from a subject is provided, and the at least two or more one-dimensional CCDs are separated by a beam splitter. The feature is that each of the incident light that is routed is incident. According to a third aspect of the present invention, in the second aspect of the invention, each of the at least two or more one-dimensional CCDs has a charge signal generated by a two-phase clock signal having an opposite phase relationship to each other. Is read out,
Filtering means for inputting one of the two-phase clock signals, and adding means for adding the charge signal read from the one-dimensional CCD selected by the selecting means and the output signal of the filtering means, The addition signal of the addition means is output as an image signal. In the invention according to claim 4, in the invention according to claim 1, the at least two or more one-dimensional CCDs are mounted, and an axis orthogonal to an optical axis of incident light from a subject is centered. And a rotation body that rotates the rotation body such that an appropriate one of the two or more one-dimensional CCDs is located on the image forming plane of the subject with respect to the size of the rotation body. And a dynamic urging means.

[0008]

According to the first aspect of the present invention, one of the outputs of at least two one-dimensional CCDs having different pixel pitches is selected according to the size of the object, and the image signal is output. Used as. For example, when the object is large, the output of the one-dimensional CCD having a coarse pixel pitch is selected, while when the object is small, the output of the one-dimensional CCD having a fine pixel pitch is selected. The number of pixels on the subject can be secured by changing the inter-pixel pitch, that is, the reading pitch, without depending on the size of the subject. At this time, since the angle of view is constant, the work corresponding to the change in the size of the subject is unnecessary. According to the second aspect of the invention, the optical system from the beam splitter to the one-dimensional CCD is fixed, so the accuracy of the optical system is improved. According to the third aspect of the invention, since the charge signal of the one-dimensional CCD is read by the two-phase clock signals, the level difference between the charge signals is canceled by the addition with the signal obtained by filtering the clock signal. According to the invention described in claim 4, the rotating body is rotated by the urging of the rotating urging means, and the one-dimensional CCD suitable for the size of the object is positioned on the image forming surface of the object. Let For this reason, it is possible to capture an image without shunting the incident light from the subject, so that the amount of light from the subject is unlikely to be insufficient.

[0009]

Embodiments of the present invention will be described below with reference to the drawings.

1: First Embodiment First, a first embodiment of the present invention will be described. FIG.
FIG. 3 is a block diagram showing the configuration of this embodiment. As shown in this figure, the incident light from the subject enters through the lens L from the left side of the figure. In this figure, reference numeral 10 is a beam splitter, which splits incident light from a subject to the right and downward in the figure. Reference numeral 11 is a CCD line sensor for imaging the incident light shunted downward, and is of a small pixel / rough pixel pitch type. Code 12
Is a CCD line sensor that images the incident light shunted to the right, and is a multi-pixel / pixel-pitch dense type. For example, while the CCD line sensor 11 has 2048 pixels and its pixel pitch is 14 μm,
The line sensor 12 has 5000 pixels, and the pixel pitch is 7 μm. The CCD line sensor 1
The image pickup surfaces 11 ₁ and 12 ₁ of ₁ and 12 are provided and fixed so as to coincide with the image forming surface of the subject by the lens L.

On the other hand, reference numeral 13 is a clock generation circuit, which is based on the basic clock φ, and is provided with a start pulse ST, and clock signals for reading which are in opposite phase to each other as shown in FIGS. CK1 and CK2 are generated and supplied to the CCD line sensors 11 and 12, respectively. Here, each of the CCD line sensors 11 and 12 is
Each pixel of the photodiode array forming the light receiving portion is divided into an odd-numbered pixel and an even-numbered pixel, and has registers corresponding to these. As a result, the charge signal (image signal) corresponding to the odd-numbered pixel is read when the clock signal CK1 is “H”, while the charge signal corresponding to the even-numbered pixel is the clock signal CK1 “L”. , That is, when the clock signal CK2 having a phase opposite to this clock signal is "H". As a result, the image signal can be read at twice the normal speed. Then, the output of the CCD line sensor used for image reading is selected by switching with the switch SW. Here, a signal S for controlling the switching of this switch SW
EL is supplied from a chip mounter CM that supplies components. In other words, the chip mounter CM appropriately switches and selects either one of the CCD line sensors 11 and 12 used for imaging according to the size of the supplied component.

By the way, since the amount of light incident on each CCD line sensor decreases in accordance with the characteristics of the beam splitter 10, problems such as insufficient output level of the CCD line sensor are expected. In order to solve this, it is conceivable to increase the amount of illumination light to a part as a subject, or to lengthen the accumulation time of the CCD line sensor. However, increasing the amount of illumination light causes the light intensity to change drastically with time, and also shortens the life of the light that is the light source, while increasing the accumulation time requires time for reading, It goes against the reduction of mounting time. Therefore, in this embodiment, the following measures are taken. That is, the output voltage of the CCD line sensors 11 and 12 selected by the switch SW is amplified by the amplifier 14, and at this time, in order to prevent the occurrence of stripes in the image, the corrector 15 sets the odd number and the even number. The output corresponding to each pixel is corrected to the same level, then amplified again by the amplifier 16, and the amplification result is output to the image processing circuit 17 in the subsequent stage as an image signal. Since the image processing circuit 17 is generally known, its description is omitted.

Here, the corrector 15 will be briefly described. As described above, the CCD line sensors 11 and 12 are both composed of two channels, but these channels do not necessarily have the same characteristics in the manufacturing process, but rather have different characteristics in many cases. For example, the transfer efficiency of a register that handles odd-numbered pixels of a CCD line sensor may be higher (or lower) than that of a register that handles even-numbered pixels. In this case,
Despite capturing an image of a subject with uniform brightness,
The obtained image signal has different levels between the odd-numbered pixels and the even-numbered pixels (see FIG. 3D). As a result, the image pickup screen has a striped pattern corresponding to the difference in level, which causes a problem that the reading accuracy is deteriorated. Note that the specifications of the CCD line sensor are generally determined in consideration of the difference in channel characteristics, dark current characteristics, and the like, and such a problem does not occur unless the output voltage of the CCD line sensor is amplified. . That is,
This problem can occur when amplifying the output voltage of the CCD line sensor.

In order to solve this, the compensator 15 has a structure shown in FIG.
As shown in FIG. _5, it is composed of an LPF 15 ₁ and an adder 15 ₂ . For the sake of explanation, the CCD line sensor 11,
12 assumes that uniform light is incident, and the actions of the switch SW and the amplifier 14 in the middle are omitted. First, consider the output signals of the CCD line sensors 11 and 12 supplied to one input end of the adder 15 ₂ . As shown in FIG. 3D, the output signals of the CCD line sensor 12 include the signals A, C, E, G, ... Corresponding to the odd-numbered pixels read by the clock signal CK1 and the clock signal CK.
In the signals B, D, F, H, ... Corresponding to the even-numbered pixels read by 2, there is a difference in level due to the difference in the characteristics of the two registers. Further, the output signal of the CCD line sensor is blunt because it is used near the maximum specification frequency.

Next, consider the output of the LPF 15 ₁ supplied to the other input terminal of the adder 15 ₂ . Adder 15 ₂
The waveform of the clock signal CK2 supplied to the other input terminal, as shown in FIG. 3 (c), dull by filtering action of LPF 15 _1. Here, by appropriately setting the cutoff frequency and the filter gain of the LPF 15 ₁ ,
The output waveform of the LPF 15 ₁ is transferred to the CCD line sensor 11,
The output waveforms of 12 are in the opposite phase relationship and are substantially the same.

Therefore, if the two signals are added by the adder 15 ₂ , as shown in FIG. 3 (e), it is possible to obtain a substantially uniform image signal with no level difference between the odd-numbered and even-numbered levels. That is, the level difference between the output signals of the CCD line sensors 11 and 12 is canceled by the clock signal synchronized with this. As a result, high-speed image reading can be performed with high accuracy without being affected by the difference in channel characteristics.

According to this embodiment, the CC used for imaging
D line sensor switches according to the size of parts
Is selected by For example, for a relatively large component such as QFP, while the CCD line sensor 11 is selected,
For relatively small parts such as chip resistors, the CCD line sensor 12 is selected. This eliminates the need to reset the angle of view according to the size of the component, and the number of pixels required for reading can be secured by the selected CCD line sensor, so that the image processing circuit 17 in the subsequent stage can be secured.
By recognizing the CCD line sensor currently selected by the signal SEL, it is possible to perform highly accurate image recognition.

2: Second Embodiment Next, a second embodiment of the present invention will be described. FIG.
FIG. 6 is a block diagram showing the configuration of the second embodiment. The second embodiment is different from the first embodiment in which incident light is shunted, and the CCD line sensor located on the image plane is selected according to the size of the component. In this figure, the same components as those of the first embodiment shown in FIG. 1 are designated by the same reference numerals, and the description thereof will be omitted. In this figure, reference numeral 20 is a rotation block, which rotates about an axis G orthogonal to the direction of incident light from a component as a subject, that is, the optical axis LG of the lens L, and mounts the CCD line sensors 11 and 12. To do. These image pickup surfaces 11 ₁ and 12 ₁ are set so as to be located equidistant from the axis G, respectively.

The rotating block 20 is urged clockwise by a coil-shaped spring SP wound around a shaft G in the figure, but its urging is restricted by a stopper 21R fixed to the frame F. Has been done. In this state, the image pickup surface 11 ₁ of the CCD line sensor 11 and the image forming surface of the subject coincide with each other. On the other hand, reference numeral 22
Is an electromagnetic solenoid, and when energized, its plunger 22 ₁ pushes the turning block 20 against the biasing force of the spring SP to turn counterclockwise around the axis G in the figure, but fixed to the frame F. Stopper 21L
Due to this, its rotation is restricted. Here, in a state where the electromagnetic solenoid 22 is energized and the rotation block 20 is rotated counterclockwise with the stopper 21L as a limit, the image pickup surface 12 ₁ of the CCD line sensor 12 and the image formation surface of the subject are aligned with each other. It has become.

On the other hand, reference numeral 23 is a drive circuit for driving the electromagnetic solenoid 22, and drives the electromagnetic solenoid 22 when the signal SEL supplied from the chip mounter CM becomes "H" level. In detail, when the chip mounter CM supplies a relatively large component such as QFP, the signal SEL becomes “L”, while when the chip mounter CM supplies a relatively small component such as a chip resistor, the signal SEL becomes “L”. SE
L becomes "H". As a result, in the former case, the plunger 22 ₁ does not operate, and the rotation block 20 stops at the stopper 21R.
CCD with a small number of pixels and a coarse pitch
While the line sensor 11 is selected, in the latter case, the plunger 22 ₁ is activated and the rotary block 20 is stopped by the stopper 21.
When biased to L, the multi-pixel / fine-pitch CCD line sensor 12 is selected.

Therefore, also in the second embodiment, it is not necessary to reset the angle of view according to the size of the parts, and the number of pixels required for reading can be secured by the selected CCD line sensor. Therefore, the image processing circuit 17 in the subsequent stage can recognize the selected CCD line sensor based on the signal SEL to perform highly accurate image recognition. Moreover, this second embodiment is similar to the first embodiment.
As compared with the embodiment, since the incident light is not shunted, it is possible to perform the image pickup while keeping the incident light amount sufficiently. Therefore, the amplifier 14 and the compensator 15 in the first embodiment
Can be eliminated.

Although the number of CCD line sensors is "2" in each of the above embodiments, the present invention is not limited to this. For example, as “3” or more, the pixel pitches may be different and the CCD line sensor used for imaging may be selected according to the size of the components supplied by the chip mounter.

[0023]

As described above, according to the present invention, the following effects are obtained. It is possible to perform high-accuracy image recognition by ensuring the number of pixels that contribute to reading without complicated mounting work and without depending on the size of an object to be recognized (claim Item 1). Since the optical system is fixed, the accuracy of the optical system can be improved (claim 2). The level difference between the charge signals read by the signals of the two-phase clocks is canceled by addition with the signal obtained by filtering the clock signals, so that even when the output level of the one-dimensional CCD is amplified, a substantially uniform image is obtained. A signal can be obtained (claim 3). Since the image pickup can be performed without shunting the incident light from the subject, the shortage of the light amount is less likely to occur from the subject, and an additional circuit is not provided in the output stage of the one-dimensional CCD (claim 4).

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of a first embodiment according to the present invention.

FIG. 2 is a block diagram showing a configuration of a corrector in the same embodiment.

FIG. 3 is a waveform diagram for explaining the operation of each unit in the embodiment.

FIG. 4 is a block diagram showing a configuration of a second embodiment according to the present invention.

[Explanation of symbols]

10 ... Beam splitter, 11, 12 ... CCD line sensor (one-dimensional CCD), 11 ₁ , 12 ₁ ... Imaging surface, 15 ₁ ... LPF (filtering means), 15 ₂ ... Adder (adding means) , 20 ... Rotating block (rotating body), 22 ... Electromagnetic solenoid (rotating biasing means), SW ... Switch (selecting means), CK1, CK2 ... Clock signal, G ... Axis (with incident light) Are orthogonal axes)

Claims (4)

[Claims] 1. A one-dimensional CCD whose imaging surface is located on an image-forming surface of a subject at the time of imaging, and at least two or more one-dimensional CCDs having different pitches between pixels, and, according to the size of the subject, The two or more one-dimensional CCDs
An image reading apparatus, comprising: a selecting unit that selects any one of the outputs of 1. and using the selection result of the selecting unit as an image signal. 2. A beam splitter for shunting incident light from an object, wherein the at least two or more one-dimensional CCDs respectively enter the incident light shunted by the beam splitter. 1. The image reading device described in 1. 3. The at least two or more one-dimensional CCDs
Of each of which the charge signal is read by two-phase clock signals that are in opposite phase to each other, and a filtering means that receives one of the two-phase clock signals as input, and a primary signal selected by the selecting means. 3. The image reading apparatus according to claim 2, further comprising an adding unit that adds the charge signal read from the original CCD and the output signal of the filtering unit, and outputs the added signal of the adding unit as an image signal. apparatus. 4. The at least two or more one-dimensional CCDs
And a rotating body that rotates about an axis orthogonal to the optical axis of incident light from the subject, and the two or more one-dimensional CCDs for the size of the subject.
The image reading apparatus according to claim 1, further comprising: a rotating biasing unit that rotates the rotating body so that an appropriate one of them is located on an image forming surface of the subject.