JPS6332667A - Object recognizing device - Google Patents

Abstract

PURPOSE:To stably recognize an object by setting evenly the sample holding output signals at a common control level in plural monitor areas for an object to be observed. CONSTITUTION:The video input signals VD received from plural monitor areas are samples held by main bodies SH1-16 of a sample holding circuit 23. Each of bodies SH1-16 has a simple constitution including a CR integration circuit, etc., and these bodies SH1-16 have variance of characteristics. Therefore the signals VD received from each monitor area of a standard object (master) are successively sampled held previously by the bodies SH1-16 and then fetched by a CPU via a gain control circuit GCON. Then the coefficient data PARA is calculated to that the each luminance data is equal to the fixed value and stored in a RAM. When the luminance data on each monitor area for the object to be observed is fetched by the CPU, the CPU feeds the corresponding data PARA back to the circuit GCON to control them at a common level. Thus the object is stably recognized.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an object recognition device, and in particular, detects the presence or absence of an object based on a video signal obtained by imaging an object to be recognized using a television camera. It is designed to observe the condition of objects, such as the presence or absence of defects.

[Summary of the invention]

The present invention relates to an object recognition device that observes the state of an object based on a video signal obtained by imaging an object to be observed using a television camera. By adjusting the level to a common adjustment level, stable object recognition results can be obtained.

[Conventional technology]

This type of object recognition device uses a video signal obtained by capturing an image of an observed object with a television camera.
Based on the signal level of a signal portion corresponding to a predetermined monitoring area, it is determined whether the signal level falls within a predetermined range, thereby recognizing the presence or absence of an object or the presence or absence of a defect in each part of the object. A method has been proposed (Japanese Patent Application No. 58-148243).

When attempting to determine the presence or absence of an object, the presence or absence of a defect, etc. using an object recognition device with such a configuration, -m is based on the video signal obtained when the television camera images the observed object. Compare the brightness of the monitoring area set at a predetermined position with the brightness of the corresponding monitoring area on the imaging screen when a standard object with a normal appearance (this is called the master) is imaged by the television camera. By doing so, a method is adopted in which if there is a difference between the two, it is determined that there is an abnormality in the observed object.

[Problem that the invention seeks to solve]

When trying to make such a determination, data representing the brightness of a monitoring area arbitrarily set on the imaging screen is
Like a sample hold circuit with an R integrator circuit configuration,
It is considered that an object recognition device with a simple configuration can be realized if brightness data of a monitoring area can be created using a configuration as simple as possible.

However, when trying to obtain luminance data of a monitoring area using a sampling hold circuit configured as a CR integration circuit,
Variations in the values of capacitor C and resistor R, changes over time,
There is a problem in that even if it is attempted to strictly compare the magnitude relationship of luminance data, it cannot be achieved because it is easily affected by variations in ibedance related to leakage current.

□This proposal was made in consideration of the above points, and a simple CR integration circuit is used to obtain luminance data from multiple arbitrarily set monitoring areas based on video signals obtained from television cameras. The purpose of this paper is to propose an object recognition device that can achieve stable judgment operations even when using a sampling and holding circuit.

[Means for solving problems]

In order to solve this problem, in the present invention, an arbitrarily set monitoring area A1 is provided in the imaging screen defined based on the video signal VD obtained by imaging the observed i-object 14 with the television camera 15. Object recognition configured to recognize whether the state of the observed object 14 in the monitoring area A1 to A16 is normal or abnormal by comparing the video signal portion corresponding to ~A16 with the corresponding video signal portion of the master. In the device, multiple monitoring areas A
A plurality of sample and hold means SHI that sample and hold video signal components obtained from 1 to A16, respectively;
5H16 and the plurality of sample hold means SHI~
i-word level adjustment means GCON which adjusts the signal level of the susceptible hold output signal VD and - signal sent out from 5H16, respectively, and each sample hold means SHI to 5H1.
When the sample hold output signal VDsH is sent from 6 to the signal level adjustment means GCON, the coefficient for adjusting the signal level of the sample hold output signal VD□ to the common signal level ≠ -ta PARA is set to the signal level. coefficient data supply means 22 for supplying to the adjustment means GCON sample-and-hold output signals VDs for a plurality of monitoring areas A1 to A16;

signal level to match the common adjustment level.

[Effect]

Input video signal V obtained from television camera 15
A plurality of monitoring areas A1 to A16 are arbitrarily set on the imaging screen defined based on D, and each monitoring area A1 to A1
At the timing of 6, the sample hold circuit 5) 11-3
The video signal is sampled and held at H16.

Each sample-and-hold circuit SHI to 5H16 has a simple configuration, for example, a pole (simple CR integration circuit configuration), and therefore, the signal level of the sample-and-hold video signal VD, □ varies among the plurality of monitoring points A1 to A113. There is.

When the sample-and-hold video signal VDsH is sequentially sent out from each sample-and-hold circuit SHL to 5H16, the signal level adjustment means GCON is supplied with the corresponding coefficient data PARA from the coefficient data supply means 22, so that the signal level adjustment means GCOH After the signal level is adjusted, the signal level of the luminance data obtained from the signal level adjustment means GCON is adjusted to a common adjustment level for all sample and hold means SHI to 5H16.

In this way, even if the configuration of the sample and hold means itself is quite simple, the brightness data will be the same signal level if the brightness of each monitoring area A1 to A16 is the same. When the luminance data are compared with each other, the comparison result represents the difference in luminance of each monitoring area with high accuracy, and as a result, stable object recognition results can be obtained.

〔Example〕

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment in which the present invention is applied to an out-of-stock inspection device for a printed circuit board on which mounted components are mounted will be described in detail below with reference to the drawings.

(1) Overall configuration In FIG.
The upper mount component 14 is imaged from above by a television camera 15.

The XY child table 2 supports the printed circuit board 13 in the X direction and the Y direction.
By moving in the direction, the surface of the printed circuit board 13 is divided into, for example, 4×4=16 imaging areas, and by sequentially setting the most imaged area of each division in the field of view of the television camera 15, 16 divided imaging is performed. A video signal VD regarding the area is input to a video data input section 16 and a timing control section 17, and can also be supplied to a monitor section 18.

The XY child table controller 19 that drives and controls the XY child table 2 receives an inspection start signal SIG+ every time each divided imaging area of the printed circuit board 13 on which the mounted component 14 is mounted is positioned at the imaging position of the television camera 15.
, 4 are given to the timing control section 17.

The timing control unit 17 generates a clock signal synchronized with the horizontal and vertical synchronizing signals of the video signal VD, counts the clock signal, and forms position data at predetermined intervals for each line of the video signal VD. When the inspection start signal 31GIN is generated, the timing control unit 17 controls the central processing unit (C
Data regarding the mount component corresponding to the master is read from the PU) 22, and a sampling pulse SMP is generated at a timing representing the position of the mount component and sent to the sample hold circuit 23 of the video data input section 16.

In the case of this embodiment, the timing control section 17 displays a marker representing the corresponding monitoring area together with a cursor on the display screen of the monitor section 18 at the timing of sending out the sampling pulse SMP to notify the operator. is being done.

The CPU 22 is configured to be able to take in the data of the 16 monitoring areas Ah A2...-Al6 for each of the 16 divided imaging areas. As shown in Figure 2, the 16 sample and hold circuits SHI, SH2,
...has 5H16.

In FIG. 2, the sample-and-hold circuit 23 sequentially closes the input video signal VD using sampling pulses SMP as input-side switch circuits S W + l, SWI!
・・・・・・After sampling by the ON operation of 5WII& (for example, an analog switch), the output side switch circuit SWo5Woz...SWo,
.

(Similarly, it consists of an analog switch) Sample and hold video signal VD□
The luminance data of each monitoring area A1, A2, . It is done like this.

The CPU 22 sequentially stores the data regarding the monitoring areas Al to A16 included in the 16 divided imaging areas as one page of data in the video data register 33C (FIG. 7) of the RAM 33, and stores the data for one page. When the data has been stored, an inspection end signal 5IGout is sent to the XY child table controller 19 via the control signal output unit 35, so that the next divided imaging area is positioned at the imaging position of the television camera 150. XY child table 2
to drive and control.

Sample and hold circuit main body SHI, SH2...
5H16 has a CR integration circuit consisting of a resistor R and a capacitor C, as shown in FIG.
WllSSWI□・・・・・・S W r I &,
The input video signal VD is integrated into the capacitor C through the resistor R by turning on at a timing corresponding to the area of the corresponding monitoring areas A1, A2, . . . A16. In this way, a voltage corresponding to a value obtained by integrating the luminance within the corresponding monitoring region over the entire area is held in the capacitor C.

The hold voltage of this capacitor C is the output side switch circuit S Wo+ , S Wo*--S Wo+ a is C
When they are sequentially closed under the control of the PU 22, they are sent to the gain adjustment circuit GCON as the sample-and-hold video signal VDsg, and then are discharged by closing the reset switch RST, thereby entering a state awaiting the next sampling-and-hold operation.

The CPU 22 controls the video data input section 16, the timing control section 17, and the monitor section 18 in accordance with the program stored in the program memory of the ROM 32 in response to the operation input from the operation switch section 31, as necessary.
Data processing in the monitoring area setting mode and inspection mode is executed using the register No. 3, and the processing results are output to the printer 34 as necessary.

(2) Pair inspection mode In the configuration above, the CPU 22 performs a pair test for each mount component (for example, an electrolytic capacitor, a diode, etc.) whose position and orientation are specified among the mount components 14 to be observed. Monitoring areas M1 and M2 are set, and the presence and orientation of the article is inspected based on the luminance data taken from the pair of monitoring areas M1 and M2 (this is called a pair inspection mode).

For example, as shown in FIG. 4, when an electrolytic capacitor CCN is mounted between wiring patterns PTNl and PTN2, a pair of monitoring areas Ml and M2 are placed on the image of the electrolytic capacitor CCN so that they face each other. Set to .

Here, on the upper surface of the container (made of aluminum, for example) of the electrolytic capacitor CCN, there is, for example, a black negative electrode marking part DI indicating the position where the negative electrode is provided, and another part D2
For example, they are painted separately with a coloring agent so that they can be distinguished from each other. One monitoring region M1 is set on this negative electrode label portion DI, and a monitoring region M2 is formed on the other portion D2 across the boundary D3. Here other part D
2 exhibits the ground color of aluminum, and thus the video camera 15 is installed within the negative electrode marking portion Dll and monitors the monitoring area M.
1, a video signal VD with low brightness is generated, whereas a video signal VD with high brightness is generated as the monitoring area M2 set in the other portion D2.

On the other hand, as shown in FIG. 5, when the mount component 14 is a diode CDI, the negative electrode label DI has a
Painted with silver colorant. As a result, the television camera 15 generates a high-luminance video signal VD from the monitoring area M1 of the negative electrode labeling part D1, and from the monitoring area M2 corresponding to the other part D2 representing the ground color of the container. A video signal VD of low brightness is obtained.

Thus, when the electrolytic capacitor CCN or the diode CDI having the position and orientation shown in FIG. 4 or 5 is mounted on the master, the CPU 22 calculates the luminance from the monitoring areas Ml and M2 of the mounted components on the master. In addition to obtaining data 13s+ and Bat, brightness data B□ and BTt with similar brightness for the observed object
is obtained from the monitoring areas M1 and M2, it is determined by the processing procedure shown in FIG. 6 that not only the mounting position of the electrolytic capacitor CCN or diode CDI but also its orientation is correct.

That is, the CPU 22 first selects the monitoring area setting mode MOD.
In step SPI of EI, the printed circuit board consisting of the master is placed on the XY child table 2, and the electrolytic capacitor CCN (
After setting the monitoring areas M1 and M2 for the electrolytic capacitor CCN or diode CDI (Fig. 4) or the diode CDI (Fig. 5), the process moves to step SP2, where the video signal VD represents the brightness in the monitoring areas M1 and M2 on the electrolytic capacitor CCN or diode CDI.
The sample hold circuit 23 of the video data input section 16
After sample-holding, this is converted into luminance data B! in the analog/digital conversion circuit 24. I and B
St and fetched into the CPU 22 via the bus 21.

At this time, the CPU 22 stores the brightness data BS1 and Bo in the rating data register 33D of the RAM 33 (FIG. 7). Here, the brightness data B□ stored in the RAM 33
and B. are stored so that the data of the pair of monitoring areas M1 and M2 can be read out as a set.

In this way, 16 pages worth of evaluation data for the mount parts 14 having orientations included in the 16 divided imaging areas are stored in the RAM 33, and thus the monitoring area setting mode M
ODEI ends.

Next, the CPU 22 moves to step SP3 of the inspection mode. In this step SP3, the CPU 22 selects one printed circuit board to be inspected on the XY child table 2! 3! i, the brightness data Ba in the monitoring areas M1 and M2 for the 16 divided imaging areas,
and B7□ through the television camera 15, sample hold circuit 23, analog/digital conversion circuit 24, and bus 21,
The data is stored in the video data register 33C (FIG. 7).

Next, the CPU 22 moves to step SP4, and stores the pair of brightness data BS imported into the rating data register 33D (FIG. 7) of the RAM 33 in step SP2 described above.
+ and B. It is determined whether the relationship is 13g+>B, st. If a positive result is obtained in this step S24, this means that firstly, the electrolytic capacitor CCN is mounted in the opposite direction to that in FIG. 4, or secondly, the diode CDI is mounted as the mounting component 14 in the master This means that it is mounted in the same direction as in Figure 5.

At this time, the CPU 22 moves to the next step SP5. This step SP5 is performed by R in step SP3 described above.
The relationship between the data BTI and Bo taken into the video data register 33C (FIG. 7) of the AM33 is Btl>B.
If a positive result is obtained in the step of determining whether A2 is the case, it means that the state of the mount component 14 of the printed circuit board 13 to be observed is the same as that of the master mount component. Therefore, the CPU 22 determines that the mount component to be observed is acceptable, and in step S
Execute the process when it is determined to pass in P6.

Thereafter, the CPU 22, in the following step SP7, selects the brightness data Bs stored in the rating data register 33D.
+ and Bsz values are updated to new rating data. The processing in step SP7 is performed on the brightness data B! l and BSM
For the brightness data BTI and B7□ obtained from the printed circuit board 13 to be observed, for example, a simple average of the data obtained as a result of multiple inspections in the past is calculated, and a new rating brightness monitor B3I, B,! By storing the data in the evaluation data register 33D again, when inspecting the next printed circuit board 13 to be observed, the inspection conditions can be changed according to changes over time without having to import data from the master each time. Being able to go.

After completing this processing, the CPU 22 returns to step SP3 described above and executes the luminance data B □ and B , , (7) capture processing for the next observed printed circuit board 13, thereby performing the above-described operation again. repeat.

On the other hand, if a negative result is obtained in the above-mentioned step SP5, this means that the observed target printed circuit board 13
The relationship between the brightness data B7I and Byz obtained from Ba,
The CPU 22 indicates that <B7□ or Br+=Bt.
means that it has been determined. The reason for this determination is that firstly, the mount component 14 mounted on the observed target printed circuit board 13 is mounted in the opposite direction to the orientation of the corresponding mount component in the master (i.e., Bye<Btg ), or secondly, it is out of stock (i.e. Bye<Btzs or B1 depending on the background)
．． =B7 indicates that a component is incorrectly mounted, or thirdly, a non-polar component is incorrectly mounted.

At this time, the CPU 22 moves to step SP8, determines that the observed target printed circuit board 13 has failed, and processes it, and then returns to step SP3 described above.

The above is the process when a positive result is obtained in the above-mentioned step SF3, but when the CPU 22 obtains a negative result in this step SP4, the electrolytic capacitor CCN is first mounted as a mounted component in the master as shown in FIG. Is it mounted in the same direction as (i.e. Bs
+<Bsz), or secondly, the diode CDI is mounted in the opposite direction to that in Figure 5 (i.e. Bs+
< B sz) or thirdly, whether a non-polar mount component is mounted (i.e. Bs+-Bs□
) indicates that it is in one of the following states.

At this time, the CPU 22 proceeds to step SP9 and determines whether the relationship between the brightness data B□ and Bo of the printed circuit board 13 to be observed is 13t+<I3tg.

If a positive result is obtained here, it means that the type and orientation of the mount component on the observed target printed circuit board 13 are the same as the master mount component, and therefore cp
U22 moves to the above-mentioned step SP6 and enters a processing procedure for executing the processing when the mount component of the printed circuit board 13 to be observed passes the test.

If a negative result is obtained in step SP9, this means that steps S and P regarding the master
This means that a judgment result different from the judgment No. 4 was obtained for the observed target printed circuit board 13. That is, first, the master has an electrolytic capacitor CC in the same direction as shown in Figure 4.
On the other hand, the electrolytic capacitor CCN is mounted in the reverse direction on the printed circuit board 13 to be observed, or secondly, the diode CDI is mounted on the master in the opposite direction to that shown in FIG. On the other hand, on the printed circuit board 13 to be observed, a diode CDI is mounted in the same direction as shown in FIG. □-B Is the T-mounted?
Or, fourthly, the mount component has fallen off from the observed printed circuit board 13 and is missing (that is, Bye>
This indicates that there is one failure condition: By□ or B〒,=Byz).

Therefore, at this time, the CPU 22 moves to the above-mentioned step SP8, and enters the process of rejecting the observed target printed circuit board 13.

As shown in FIG. 6, the CPU 22 compares and determines the brightness data BSI and BSt for the mounted components on the master with the corresponding brightness data E3y+ and l3vz for the mounted components on the observed printed circuit board 13. Accordingly, in the pair inspection mode, it is possible to reliably determine not only whether there is a mounted component on the printed m-board 13 to be observed, but also whether or not its orientation matches that of the master mounted component.

(3) Gain adjustment of sample hold video signal At step SP2 in FIG. The gain adjustment circuit GCON of the hold circuit 23 (Fig. 2) is connected to the sample and hold circuits SHI and SH2.
. . . By adjusting the signal level of the sample hold video signal VDsn taken in from each of the 5HI6 in the gain adjustment circuit GCON using the coefficient data PARA supplied from the CPU 22 via the bus 21, the monitoring N area Ml and the magnitude relationship between the brightness data Bit taken in from M2 and E3sz, and 13
The magnitude relationship between 'r+ and 811 is made to match the actual state of the master and the printed circuit board 13 to be observed.

Incidentally, when a CR integration circuit with a simple configuration as shown in Figure 3 is used as the sample and hold circuit main body SHI~5H16, variations in the value of capacitor C and the value of resistor R, changes over time, and peak current may occur. Associated impedance or analog switch circuit configuration switch circuit S W
Since there are variations in the switching characteristics of r r~5Wxh and SW o +~SWo+i or their resistance values, the input video signal VD based on the subject with the same brightness is sampled and held from the monitoring areas A1 to A16 to the sample and hold circuit main body SH.
Even if sample and hold is performed on I to 5H16, the signal levels of the sample and hold video signals VD3H sent out from each sample and hold circuit body 5) 11 to 5H16 may have different values.

Therefore, in the case of the configuration shown in FIG. 2, the monitoring areas Al to A16
After sequentially sampling and holding the input video signals VD based on objects with the same brightness from the sample and hold circuit bodies SHI to 5H16, this is converted into brightness data in the analog/digital conversion circuit 24 via the gain adjustment circuit GCON, and then The coefficient data PARA is fetched into the CPU 22 and is stored in the coefficient data register 33E of the RAM 33 in advance before the CPU 22 calculates the coefficient data PARA so that the value of the data becomes a predetermined value and performs the processing shown in FIG.

Then, when performing the processing in FIG. 6, the monitoring areas A1 to A16
Regarding the brightness data B31, BSZ or B?B based on the sample hold video signal VDss? □CPU
22, the CPU 22 reads out the corresponding coefficient data PARA from the coefficient data register 33E and feeds it back to the gain adjustment circuit GCON via the bus 21.

By doing this, the coefficient data register 3 of the RAM 33
3E includes sample and hold circuit main bodies SH1, SH2,
... Variations that occur in the sample-and-hold video signal VDio based on differences in the circuit configuration around 5H16 are captured in advance as coefficient data PARA.

The CPU 22 then controls the sample and hold circuit main body SHI.
, SH2 . You can level up.

Thus, when comparing the magnitude relationship based on the luminance data taken in from the pair of monitoring regions M1 and M2 set on the mount components shown in FIGS. 4 and 5, the sample and hold circuit bodies SH1, SH2, . . . ...5H
It is possible to effectively avoid the possibility of making an erroneous determination based on differences in circuit configurations around 16.

Incidentally, regarding the monitoring areas M1 and M2, the brightness data B! obtained from the master! The magnitude relationship between I and BSZ is Bs+
When <Bst (or Bs+>Bsz), brightness data BTI and BT obtained from the observed target printed circuit board 13
The magnitude relationship of t is B y+ < Brz (or BTt
> Byz), it is necessary to pass the test, and when the opposite is the case, it is judged to be a fail. However, the brightness data is B$. Or B, iB,
The signal level of the sample and hold circuit body SHI, SH
2...5 If it varies based on the configuration around H16, there is a risk that the determination result will become unstable.

According to the above configuration, this problem can be effectively solved.

(4) Other embodiments (a) In the above embodiment, the coefficient data P
Although an embodiment has been described in which the signal level (therefore, the gain) of the sample-and-hold video signal VDsH, which is an analog value, is controlled in an analog manner using ARA, instead of this, digital conversion is performed in the analog/digital conversion circuit 24. The same effect as described above can be obtained even if the value of the digital data is controlled by performing a multiplication operation on the luminance data taken into the CPU 22 via the rear bus 21 using the coefficient data PARA.

(bl In the above embodiment, the data in the rating data register 33D is updated (step SP? in FIG. 6).
), the past simple average value of the brightness data B A similar effect can be obtained.

(C) In the above embodiment, a pair of monitoring regions M1 and M2 are set for one observed object to determine the presence or absence and orientation of the observed object. The number of monitoring areas set for the observation object is not limited to one pair, and the same effect as described above can be obtained even if three or more monitoring areas are set as necessary.

(d) In the above-described embodiment, the configuration in which the signal levels of the sample-and-hold video signals VD, , obtained from the sample-and-hold circuit bodies SHI to 5H16 are aligned to a common adjustment level is applied to the pair test mode. The present invention is not limited to this, but can be broadly applied to capturing luminance data from a video signal VD, thereby making it possible to evaluate luminance data with high stability.

(e) In the above embodiment, a case was described in which no tolerance range was set when comparing the luminance data B, iaS2, and BaiBo, but instead of this,
A permissible range may be provided.

That is, when comparing master luminance data 13t+ and E3tt in step SP9 in FIG. 6, instead of determining B□<B, □, 13tt+
It is also possible to judge whether or not ∆ is within the acceptable range (∆ is an acceptable range).
If the brightness of the video signal VD changes slightly because the mounting condition of item 4 has changed to a practically acceptable extent, it is possible to make a pass or fail judgment in accordance with the actual situation. .

(f) In the above embodiment, when setting the pair of monitoring areas M1 and M2 for the master, it is not judged whether the settings are appropriate or not. In step sP4 of FIG. 6, 13s
By determining whether t+Δ>B38 or B, Otori-Δ>BSt, a predetermined tolerance range Δ is taken into consideration, and -Δ< B sz −B□<Δ
If the following conditions hold true, it may be determined that the settings are inappropriate.

Incidentally, if this condition holds, the pair of monitoring areas M
This indicates that the difference between the brightness BS1 and 13st in BS1 and M2 is insufficient. Therefore, in such a case, the operator may be notified using a separate alarm device such as a buzzer or display, and the operator may be required to reset the monitoring areas M1 and/or M2, or adjust the camera position, lighting method, etc. You may also let them do so.

(g) In the above, an embodiment has been described in which the present invention is applied to inspecting the presence or absence of mounted components mounted on a printed circuit board, but the present invention is not limited to this, and can be applied to inspecting the condition of an article. Applicable to a wide range of cases.

〔Effect of the invention〕

As described above, according to the present invention, the signal level of the sample-and-hold video signal sampled and held from a plurality of monitoring areas is adjusted using coefficient data so as to be equal to a common adjustment level for the plurality of monitoring areas. Even if a sample and hold circuit with an extremely simple configuration is used, the determination operation for determining the state of the observed object can be further stabilized without increasing the determination error.

Incidentally, due to deterioration or variations in parts, it is important to note that if brightness data of different signal levels are input even though objects with the same brightness are being imaged, the error in the judgment results will increase. Although it is unavoidable, according to the present invention, this error can be prevented from occurring.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing an embodiment of the object recognition device according to the present invention, Fig. 2 is a connection diagram showing the detailed configuration of its sample and hold circuit, and Fig. 3 is a more detailed diagram of the sample and hold circuit shown in Fig. 2. Connection diagrams showing the configuration, Figures 4 and 5
The figure is a plan view showing the mounted parts to be inspected, FIG. 6 is a flowchart showing the processing procedure in the bare detection mode by the CPU 22 in FIG. 1, and FIG. 7 is the RAM 33 in FIG. 1.
It is a route map showing the structure of. 11...Object U2! ! Device, 12...
XY child table 13...Printed circuit board, 14.
... Mount parts, 15 ... Television camera, 16 ... Video data input section, 22
...CPU, 33...RAM. 23...Sample hold circuit, 24...
・・Analog/digital conversion circuit, CCN・・・・
・Electrolytic capacitor, CDI...Diode, A
1 to A16, Ml, M2...Monitoring area, Dl.
...Negative electrode label part, D2...Other parts, PT
NISPTN2...Wiring pattern, SH1~5
H16・・・Sample hold circuit main body, GCN
...Gain adjustment circuit.

Claims (2)

[Claims] (1) Of the imaging screen defined based on the video signal obtained by imaging the observed object with a television camera, the video signal portion corresponding to the arbitrarily set monitoring area is transferred to the corresponding master. The object recognition device is configured to recognize whether the state of the observed object in the monitoring area is normal or abnormal by comparing the video signal component with the video signal component obtained from the plurality of monitoring areas. a plurality of sample and hold means for respectively holding samples; a signal level adjustment means for adjusting the signal level of the sample and hold output signals respectively sent from the plurality of sample and hold means; and a signal level adjustment means for each of the sample and hold means. When the above sample hold output signal is sent out,
coefficient data supply means for supplying coefficient data for adjusting the signal level of the sample-hold output signal to a common adjustment level to the signal level adjustment means, the signal of the sample-hold output signal for the plurality of monitoring areas; An object recognition device characterized in that the level is adjusted to the common adjustment level. (2) An object having approximately the same brightness is placed at the position of the plurality of monitoring areas and is imaged by the television camera, and the coefficient data is converted into the coefficient data based on the signal level of the sample hold output signal obtained from the sample hold means. The object recognition device according to claim 1, wherein the object recognition device is adapted to be input into a data supply means.