JPS63265104A - Object inspection - Google Patents

Abstract

PURPOSE:To correct the position deviation of an object to be observed highly accurately, by correcting the positions of monitoring regions from the correlation between the detected data obtained from the video signals of a master and the object to be observed using position deviation detecting markers, which are provided on an image screen. CONSTITUTION:Linear position-deviation detecting markers 7X and 7Y are provided on am image screen 5, which is obtained by imaging object 3 to be observed. First detected data INFN are obtained from a video signal parts corresponding to the position-deviation detecting markers 7X and 7Y in the image screen 5 with respect to a master. Second detected data INFD are obtained from a video signal parts corresponding to the position-deviation markers 7X and 7Y in the image screen 5 with respect to the object to be observed. The position deviation between the master and the object to be observed 3 is detected by obtaining the correlation between the first detected data INFN and the second detected data INFD. The relative position 6 with respect to the image screen 5 of the body to be observed 3 is corrected from the result of detection of the position deviation.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an object inspection device, and in particular, detects the presence or absence of an observed object based on a video signal obtained by imaging the observed object using a television camera. The present invention is suitable for application to an object inspection device that observes the state of an object to be observed, such as the presence or absence of defects.

[Summary of the invention]

The present invention recognizes the normality or abnormality of the state of an observed object at a monitoring point set on the imaging screen using an imaging screen defined based on a video signal of a television camera. In the object inspection device, the position of the monitoring area is corrected based on the correlation between the first and second detection information obtained from the video signals of the master and the observed object using a displacement detection marker set on the imaging screen. By doing so, the positional shift of the observed object can be corrected with high accuracy.

[Conventional technology]

This type of inspection device detects whether the signal level is within a predetermined range based on the signal level of the signal portion corresponding to a predetermined monitoring area among the video signals obtained by imaging the observed object with a television camera. A device has been proposed in which the presence or absence of an object and the presence or absence of defects in each part of the object are recognized by determining whether the object is inserted or not (Japanese Patent Application No. 148243/1982).

Using an inspection device with such a configuration, for example, when an assembly process for attaching keys to the surface of a keyboard board is completed, it may be inspected whether or not the attaching work has been carried out correctly.
For example, when inspecting whether or not the work was done correctly after completing the work process of pasting labels on the surface of the casing of heavy electrical equipment such as Tala, it is necessary to use a standard with a normal appearance. The brightness of a predetermined monitoring area set on the damaged screen when the television camera images the observed object (this is called the master), and the imaged screen when the observed object to be inspected is imaged. A method is adopted in which by comparing the brightness at corresponding positions above, if there is a difference between the two, it is determined that there is an abnormality in the observed object.

Incidentally, in the case of a keyboard, if there are keys that are not installed,
Since the brightness in the monitoring area is different from the brightness of the corresponding part in the master, if there is a key that is not installed for some reason during the key installation process, it can be reliably detected.

In addition, in the case of Tara, if the same label as the master label is not pasted at the same position, the brightness in the monitoring area set at the label position will be different from the master brightness, so it is necessary to reliably detect the abnormality. Can be done.

[Problem that the invention seeks to solve]

However, when attempting to inspect an object for anomalies using such a method, if the position of the monitoring area set on the observed object is even slightly misaligned with the position of the monitoring eM area set in the master, it is not correct. Since there is a risk that a determination result may not be obtained, if there is a positional shift in the position of the monitoring area set for the observed object, it is necessary to detect and correct it with as high precision as possible.

The present invention has been made in consideration of the above points, and uses a relatively simple method to reduce the positional deviation of the monitoring area set for the observed object with respect to the monitoring area set for the master. This paper attempts to propose an object inspection device that can detect objects with high accuracy.

[Means for solving problems]

In order to solve this problem, in the present invention, a monitor is set at a predetermined monitoring point on the imaging screen 5 displayed based on the video signal VD obtained by imaging the object 3 to be observed by the television camera 13. In an object inspection device configured to detect whether the state of an observed object 3 in a monitoring area 6 is normal or abnormal by comparing a video signal portion corresponding to the monitoring area 6 with a corresponding video signal portion of a master. , linear positional deviation detection markers 7X, 7Y are set on the imaging screen 5, and the first detection is obtained from the video signal portion corresponding to the positional deviation detection markers 7X, 7Y on the imaging screen 5 for the master. By correlating the information INF with the second detection information INFo obtained from the video signal portion corresponding to the positional deviation detection markers 7X and 7Y on the imaging screen 5 regarding the observed object 3, the master and observed object 3 are detected. The positional deviation of the object 3 is detected, and the relative position of the monitoring area 6 of the observed object 3 with respect to the imaging screen 5 is corrected based on the positional deviation detection result.

[Effect]

Linear positional deviation detection marker 7 set on the imaging screen 5
When detection information INF and INF about the master and the observed object 3 are detected from the video signal portion corresponding to X and 7Y, these detection information INF and INF are linear positional deviation detection markers 7X and 7Y. It represents the positional shift in the direction along the extension direction of .

In this way, by being able to detect the positional deviation of the observed object 3 with respect to the master based on the luminance information on the imaging screen 5, the positional deviation can be reliably corrected.

〔Example〕

[1] First Embodiment An embodiment in which the present invention is applied to an object inspection device for checking the characters on a label attached to a cooler will be described in detail with reference to the drawings.

(1) Principle of positional deviation detection As shown in FIG. Check whether it is attached.

Therefore, an image around the label 3 of the cooler 1 is captured by a television camera, and the image of the label 3 is projected at a predetermined position on the imaging screen 5, as shown in FIG.

In this way, each position of the image projected on the imaging screen 5 is
It can be expressed by the XY coordinates set on the imaging screen 5, and the XY! A monitoring area 6 is set by specifying the position of the elevation, and a character pattern on the label 3 is created using a luminance signal corresponding to the video signal portion included in the monitoring area 6 of the video signal obtained from the television camera. It is designed to check the presence or absence of 4 or whether it is correct or incorrect.

In practice, a monitoring marker having a predetermined shape and size is assigned to the monitoring area 6, so that data regarding a portion of the video signal corresponding to the monitoring marker can be captured.

In this way, the brightness (for example, average brightness) of the monitoring area 6 on the label 3 as the observed object on the imaging screen 5 of the cooler 1
and the brightness of the monitoring area 6 on the imaging screen 5 obtained from the master, and depending on whether there is a difference of more than a predetermined value between the two, the character pattern 4 on the label 3 as the observed object is determined.
is the same as the master character pattern.

In order to detect whether the display position of the label 3 on the imaging screen 5 matches the master, a marker 7X for detecting positional deviation in the X direction and a marker 7Y for detecting positional deviation in the Y direction are provided on the imaging screen 5. ing.

As shown in FIG. 3, the X-direction positional deviation detection markers 7X include a predetermined number of pixel markers 7XI, 7X8, 7X, which are sequentially arranged in rows in the X direction on the imaging screen 5.
i, 7Xi, l, 7Xi, 8...7XII are formed in a straight line, and when set on the imaging screen 5, the luminance information of the corresponding pixel row can be extracted pixel by pixel. There is.

Similarly, the marker 7Y for detecting positional deviation in the Y direction is
They are formed in a straight line so as to correspond to pixel rows arranged sequentially in the direction, and thus the brightness information of each pixel on the imaging screen 5 can be specified.

The X-direction and Y-direction positional deviation detection markers 7X and 7Y are set at the corners of the label 3 so that a part thereof overlaps with the image of the label 3, and thus the X-direction and Y-direction positional deviation detection markers 7X and 7Y, the logic that expresses bright luminance from the part superimposed on label 3, for example, "
1" level pixel information can be obtained, and label 3 is a logical "
0'' level luminance information can be obtained.

In this way, when the marker 7X for detecting positional deviation in the X direction is set on the label 3M while the cooler 1 as the master is being imaged, as shown in FIG. 3(A),
Pixel marker 7X,...~7X.

As the part overlaps with label 3M, it is shown in Fig. 3 (
As shown in B), the detection information INF whose logic level switches from "0" to "1" at the pixel marker 7Xi, .
, can be obtained.

On the other hand, as shown in FIG. 3(C), the label 3D consisting of the observed object is the pixel marker 7X1. 7X, the logic level changes from "0" to "1" with the pixel marker 7Xi+* as the boundary, as shown in FIG. 3(D) from the X-direction positional deviation detection marker 7X. Detection information INF that switches to is obtained.

(detection information IN obtained from the master label 3M)
If the correlation information aCOR is obtained by, for example, performing an exclusive OR operation based on F, 1 and the detection information INFI obtained from the label 3D of the observed object, then the pixel markers 7Xl+I and 7x= in the central part are obtained. , , a mismatch section DIS occurs, and a match section AC appears on both sides of it.
CI and ACC2 result.

Thus, from the correlation information COR, the label 3D of the observed object
It can be determined that the label is shifted in the X direction from the master label 3M by the mismatch area DIS.

Therefore, when obtaining detection data from the monitoring area 6 for the label 3D as an observed object, if the position of the monitoring area 6 is moved as a whole in the Detection data can be obtained from the same position as the monitoring area 6 set on the label 3M.

Regarding the marker 7Y for detecting a positional deviation in the Y direction, it is possible to obtain the same correlation information as that described above (Fig. 3) for the marker 7X for detecting a positional deviation in the X direction. is misaligned in the Y direction from the label 3M serving as the master, this can be corrected.

If such correlation information COR is expressed as a general formula, it becomes the following formula. In equation (1), f (i) represents the brightness level that could be obtained from the i-th pixel marker 7X on the imaging screen 5 of the label 3M as the master, and g
(i) represents the brightness level that could be obtained from the i-th pixel marker 7X on the imaging screen 5 obtained from the label 3D as the observed object. Also, ΔX is the mismatch area DI
represents the number of pixels in S (Fig. 3(E)), thus (1)
The formula is, for the brightness level f (i) of the i-th pixel of the X-direction positional deviation detection marker 7X, the brightness level g (i + ΔX) obtained from the (i + ΔX)-th pixel marker of the label 3D as the observed object is This indicates that there is the largest correlation.

At this time, it is shown that luminance data with the highest correlation can be obtained from the pixel marker that is shifted by ΔX for all the pixel markers IX+ to 7XI of the X-direction positional shift detection marker 7X.

Therefore, if the amount of positional deviation ΔX that minimizes equation (1) is found, this can be determined as the amount of positional deviation ΔX of the observed object with respect to the master.

(2) Specific configuration The object inspection device 1) has the configuration shown in FIG.

That is, the labels 3 of the cooler 1 that are sequentially conveyed by the conveyor 12 are imaged by the television camera 13,
The video signal VD is input to the video data input section 14 and the timing control section 15, and is also supplied to the monitor section 17 via the superimpose circuit 16.

Conveyor controller 21 that drives and controls the conveyor 12
is adapted to give an inspection start signal SIGIM to the timing control section 15 every time the cooler 1 is positioned at the imaging position of the television camera 13.

The timing control unit 15 generates a clock signal synchronized with the horizontal and vertical synchronization signals of the video signal VD, counts the clock signal, and generates sampling pulses at predetermined intervals for the luminance signal of each line of the video signal VD. SMP 1 is applied to the analog/digital conversion circuit 22 of the video data input section 14, and the video signal VD is converted into pixel data DP pixel by pixel in the analog/digital conversion circuit 22, and the data is stored in the image memory 24 via the switching circuit 23. supply

The timing control section 15 controls the sampling pulse SMP.
1, the write address signal ADW is sent to the image memory 24 based on
give to

The switching circuit 23 has a master side switching output terminal M and a detection side switching output terminal, and outputs the pixel data DP to the master side switching output terminal M.
The pixel data DP is stored in the master image memory section 24M of the image memory 24 via the write address signal ADW.
The detected image is stored in the detected image memory section 24D according to the detected image.

In this embodiment, when the operator inputs a master image data import command from the operation switch unit 27 via the bus 26 while the television camera 13 is capturing an image of the cooler 1 as the master, the CPU 25 When the CPU 25 gives a switching command signal to the timing control section 15 via the bus 26, the timing control section 15 sends out a switching signal SW to the switching circuit 23, so that the analog/digital conversion circuit 22
The pixel data DP obtained at the output terminal of is taken into the master image memo 9 section 24M side via the master side switching output terminal M.

On the other hand, when a command signal for the inspection mode is input from the operation switch section 27, the CPU 25 controls the timing control section 1.
5, the switching circuit 23 is switched to the detection side switching output end side, and the cooler 1 as an object to be observed is switched to the television camera 13 by the conveyor controller 21.
The image memory unit 24 detects the pixel data DP by controlling the timing control unit 15 using the inspection start signal 5IGIN generated every time the camera is positioned at the imaging position.
Take it to the D side.

The CPU 25 controls the video data input section 14, the timing control section 15, and the monitor section 17 in accordance with the program stored in the program memory of the ROM 2B in response to the manual operation of the operation switch section 27, as necessary.
By controlling the object inspection device 1) as a whole using registers 9, the CPU 25 By outputting the inspection end signal 5IGout to the conveyor controller 21 via the bus 26 and the control signal output unit 30, the cooler 1 as a new object to be observed is brought to the imaging position of the television camera 13. being done.

As shown in FIG. 5, the RAM 29 has a cursor position register 29A, a monitoring area preset register 29B, a detection data register 29C, an evaluation data register 29D, a marker register for detecting positional deviation 29E, and a positional deviation data register 29F.

The cursor position register 29A controls the image capture screen 5 displayed on the monitor unit 17 based on the video signal VD.
You can set the monitoring area 6, the position, size,
The position of the cursor used when selecting the shape and setting the X-direction and Y-direction positional deviation detection markers 7X and 7Y is memorized moment by moment.

The monitoring area preset register 29B stores one label 3.
The operator can change the position of the monitoring area 6 set for the upper character pattern to a maximum of 16 points via the operation switch section 27.
When you input up to a point, store this.

When the cooler 1 as an object to be observed is positioned at the imaging position of the television camera 13, the detection data register 29C transfers the brightness data about the monitoring area 6 preset to the monitoring area preset register 29B to the CPU 2.
5, this detected data is stored.

The rating data register 29D is the master cooler 1.
is positioned at the imaging position of the television camera 13, the brightness data of the monitoring area 6 stored in the monitoring area preset register 29B is stored from the master image memory unit 24M.

The positional deviation detection marker register 29E
The image positions of the directional positional deviation detection markers 7X and 7Y are
When the operator makes an input using the operation switch unit 27, this is stored.

When the cooler 1 as a master or the cooler 1 as an observed object is positioned at the imaging position of the television camera 13, the positional deviation data register 29F stores the positional deviation detection data stored in the positional deviation detection marker register 29E. The luminance data of the markers 7X and 7Y is stored from the detected image memory section 24D.

Thus, the data stored in the RAM 29 is transferred to the CPU 25 as necessary when the operator specifies the data processing mode for the CPU 25 using the operation switch section 27.
It is read out by U25 and supplies the data necessary for each processing step.

At the same time, the data in the RAM 29 is supplied to the pattern generator 31 via the bus 26 as needed, and the pattern data DN is provided to the monitor unit 17 via the superimpose circuit 16, so that the currently set data can be set. A pattern representing a cursor, a monitoring area 6, and markers 7X and 7Y for detecting positional deviation in the X and Y directions is displayed in a superimposed manner on the image displayed on the display screen of the monitor unit 17 based on the video signal VD. let Thus, the operator sees the pattern displayed on the display screen of the monitor section 17 and issues an appropriate command signal via the operation switch section 27.
It is designed so that it can be input to the PU25.

In the above configuration, the CPU 25 executes inspection processing for the label 3D as an observed object according to the processing procedure shown in FIG.

That is, when the processing program is started in step SPI, the CPU 25 waits for a master data import command to be input by the operator from the operation switch section 27 in the next step SP2.

Here, after setting the cooler 1 as a master at the imaging position of the television camera 13, the operator inputs a master data import command to the CPU 25 via the operation switch section 27.

At this time, the CPU 25 moves to the next step SP3 and controls the timing control unit 15 via the bus 26 to transfer the video signal VD for the master obtained from the television control camera 13 to the analog/digital conversion circuit 22 and the switching circuit. The image data of one frame is taken into the master image memory section 24M via the master side switching output terminal M of 23.

Next, the CPU 25 moves to step SP4, and uses the data of the positional deviation detection marker register 29E to select the X-direction and Y-direction positional deviation detection markers 7X and 7Y pixel data is stored in the positional deviation data register 29F.

Next, the CPU 25 moves to step SP5, and uses the data in the monitoring area preset register 29B to set the brightness data of the monitoring area 6 among the master image data taken into the master image memory section 24M as rating standard data in the rating data register 29D. Thus, the data import process for the label 3M as a master is completed.

Subsequently, the CPU 25 executes the inspection processing program. That is, the CPU 25 waits for the arrival of a detection data capture command in the subsequent step SP6.

Here, the operator causes the conveyor 12 to sequentially transport the coolers 1 as objects to be observed one by one to the imaging position of the television camera 13 via the operation switch section 27.

When the cooler 1 as one object to be observed is positioned at the imaging position of the television camera 13, the conveyor controller 21 sends an inspection start signal SIGIM to the timing control unit 15, and performs the next step SP7.
In , the label 3D of cooler 1 as the observed object (
In other words, the video signal VD regarding the object to be observed) is taken into the detected image memory section 24D.

Next, the CPU 25 moves to step SP8 and reads the RAM 29.
Using the position data of the X- and Y-direction positional deviation detection markers 7X and 7Y stored in the positional deviation detection marker register 29H of the The pixel data is read out from the detected image memory section 24D and stored in the RAM2.
9 is taken into the positional deviation data register 29F. In this way, a state is obtained in which the pixel data of the X-direction and Y-direction positional deviation detection markers 7X and 7Y obtained from the cooler 1 as the master and the observed object are loaded into the positional deviation data register 29F.

Subsequently, in step SP9, the CPU 25 reads out the luminance data corresponding to each pixel of the X-direction and Y-direction positional deviation detection markers 7X and 7Y stored in the positional deviation data register 29F, and calculates the brightness data corresponding to each pixel of the X-direction and Y-direction positional deviation detection markers 7X and 7Y, and calculates the luminance data as described above for equation (1). In this way, the positional deviations ΔX and ΔY are calculated based on the positional deviation data obtained for the master and the observed object.

Thus, the process step of obtaining positional deviation data representing the amount of positional deviation of the observed object of the cooler 1, ie, the label 3D, which is currently being imaged by the television camera 13 is completed.

Next, the CPU 25 moves to step 5PIO, reads the monitoring area position data from the monitoring area preset register 29B, and corrects it by the positional deviation amounts ΔX and ΔY using the positional deviation data stored in the positional deviation data register 29F. Detected image memory section 2 through bus 26
Timing control unit 15 as an address signal for 4D
Transfer to. The luminance data thus read out from the detected image memory section 24D is taken into the detected data register 29C.

Thereafter, the CPU 25 moves to step 5PII and compares the luminance data with the master luminance information stored in the evaluation data register 29D to determine the quality of the product.

The CPU 25 determines whether the judgment result is good or bad in the following step 5P12, and if it is defective, the operator performs defect processing to improve the defect in the label 3 of the cooler 1 currently being imaged by the television camera 13 in step 5P13. Wait until the above step S
Return to P6.

On the other hand, when it is determined in step 5P12 that the CPU 25 is not defective, the CPU 25 immediately returns to step SP6 described above.

In this way, the CPU 25 selects the cooler 1 as the next object to be observed.
The cooler 1 as a new object to be observed is set to the imaging position of the television camera 13, and a new inspection start signal 5IGIN is sent to the timing control unit. 15, reading the data of positional deviation detection markers 7X and 7Y for the new observed object;
The process of creating positional deviation data, correcting the monitoring area, capturing the brightness data of the monitoring area after correction, and determining the quality is repeated.

According to the above configuration, when the position of the label 3D of the cooler 1 as the observed object is shifted in the X direction or the Y direction with respect to the position of the label 3M of the cooler 1 as the master on the imaging screen, this Positional deviation detection markers 7X and 7Y
Label 3 of Label 3D as an observed object is detected based on pixel data obtained from
The amount of positional deviation with respect to M can be easily detected.

By thus correcting the position of the monitoring area 6 based on the detection result, it is possible to determine with high accuracy whether or not there is an abnormality in the label 3.

[2] Second Embodiment FIG. 7 shows another embodiment of the present invention. As shown by assigning the same reference numerals to corresponding parts as in FIG. 4, the video data input section 14 is a sample and hold circuit. 41 and analog/
It has a digital conversion circuit 42 and a timing control section 15.
Sampling pulses SMP2 and SMP sent out from
3, the luminance information of the monitoring area 6 (FIG. 1) is held in the sample and hold circuit 41, converted into digital data in the analog/digital conversion circuit 42, and then stored in the evaluation data register 29D of the RAM 29 or detected. The data is taken into the data register 29G.

In this way, in the case of FIG. 7, the brightness information about the monitoring area 6 set in the monitoring area preset register 29B of the RAM 29 can be loaded into the RAM 29 separately from the data of the positional deviation detection markers 7X and 7Y. It has a different feature from the configuration shown in FIG. 4 in that it is configured differently.

In the configuration shown in FIG. 7, the CPU 25 executes the inspection process according to the processing procedure shown in FIG. 8, in which parts corresponding to those in FIG. 6 are denoted by the same reference numerals.

In FIG. 8, compared to the case of FIG. 6, a processing step for correcting the position of the monitoring area 6 is performed between the processing step SP9 for creating positional deviation data and the processing step 5P10 for capturing the luminance data of the monitoring area 6. The difference is that 5P14 is executed.

In the processing procedure of FIG. 8, the CPU 25 performs step SP
After importing the master image data in step 3, step S
When capturing marker data for positional deviation detection in P4, the video signal VD obtained from the television camera 13 is captured into the master image memory section 24M via the analog/digital conversion circuit 22 and the switching circuit 23. At this time, the data taken into the master image memory section 24M are the markers 7X and 7 for detecting positional deviation in the X direction and Y direction.
It is only necessary to import pixel data DP for detecting positional deviation regarding Y, and in this embodiment, the pixel data DP
is 1-bit digital information, and this 1-bit digital information is taken into the master image memory unit 24M for the entire imaging screen 5.

Subsequently, the CPU 25 causes the sample and hold circuit 41 to sample and hold the video signal VD when taking in the evaluation data in step SP5. Here, the video signal VD sampled and held by the sample and hold circuit 41 is analog luminance information of pixels included in the monitoring area 6, and is an analog signal.

Thus, the CPU 25 is the analog/digital conversion circuit 4.
2, the sample-and-hold output of the sample-and-hold circuit 41 is converted into data of a plurality of predetermined bits (for example, 5 bits), and the data is stored in the evaluation data register 29D.

Thereafter, in step SP7, the CPU 25 converts the video signal VD into 2-bit pixel data DP as positional deviation detection data from the observed object in the analog/digital conversion circuit 22, and then imports it into the detected image memory section 24D via the switching circuit 23. , In the next step SP8, the pixel data of the X-direction and Y-direction positional deviation detection markers 7X and 7Y of the data in the detected image memory section 24D is taken into the positional deviation data register 29F, and then in the next step SP9, the pixel data of the markers 7X and 7Y for detecting positional deviation in the X direction and Y direction are taken into the positional deviation data register 29F. The correlation with the positional deviation data is calculated to determine the positional deviation amounts ΔX and ΔY.

However, in this embodiment, in the following step 5P14, the CPU 25 uses the positional deviation amounts ΔX and ΔY to correct the monitoring area position data read from the monitoring area preset register 29B by the positional deviation amounts ΔX and ΔY, and adjusts the timing. It is sent to the control unit 15. In this way, the timing control unit 15 adjusts the positional deviation amount ΔX detected by the X-direction and Y-direction positional deviation detection markers 7X and 7Y.
The monitoring area 6 is moved to a position parallelly moved by ΔY in the X and Y directions, and the analog luminance information of each pixel included in the corrected monitoring area 6 is sampled and held by the sample hold circuit 4.
1 to hold the sample.

Thus, in the imaging screen 5, if the position of the label 3D as the observed object is shifted compared to the position of the label 3M as the master, the position shift ΔX
After correcting the position of the monitoring area 6 by an amount corresponding to ΔY and ΔY, the brightness information is captured in step 5pio.

As a result, the positional information obtained from the label 3M as the master and the positional information obtained from the label 3D as the observed object are almost the same, so that errors due to positional deviation will not occur in the result of the determination operation in step 5PII. It can be done.

According to the above configuration, the effects described above with respect to FIGS. 4 and 6 can be obtained, and in addition, according to the configurations of FIGS. 7 and 8, the quality of the observed object can be determined. Since the analog luminance information taken in for this purpose is limited to the monitoring area 6 rather than the entire imaging screen 5, the time to take in the determination data to the RAM 29 can be further shortened, and the inspection time can be shortened as a result of this division.

[3] Other embodiments (1) In the above-mentioned embodiment, the markers 7X and 7Y for detecting positional deviation in the X and Y directions, which extend linearly in the X and Y directions, are used to detect the object to be observed. label 3d
The positional deviation from the label 3M as a master is detected in two directions, the X direction and the Y direction.
The extending direction and the number of positional deviation detection markers are not limited to these, and may be changed in various ways.

For example, as shown in FIG. 9, when a circular object 41M is imaged as a master, when detecting the positional deviation of the observed object 41D with respect to this object 41M, three positional deviation detection markers 42A, 42B, 42C are used. is set at a predetermined position on the circumference in an oblique direction so as to cross the periphery of the object 41M14LD.

In this way, detection information in three directions can be obtained as positional deviation information, and thus the direction and amount of positional deviation of the observed object 41D can be easily detected.

(2) In the above-mentioned embodiment, an example was described in which a marker extending in a straight line is used as a positional deviation detection marker, but the present invention is not limited to this. For example, as shown in FIG. 10, a position extending in an arc shape is used. A shift detection marker 49 may also be used.

In FIG. 10, reference numeral 45 is an image of a gear, and a monitoring area 47 is set at each tooth tip 46, thereby detecting whether or not the tooth tip 46 is damaged.

In this case, the periphery of the center hole 47 of the gear 45 has a center 0
Three positional deviation detection markers 48A, 4
8B and 48C are provided, and the gear 45 can be positioned at the center o position based on the positional deviation information obtained from these three positional deviation detection markers 48A to 48C.

In addition, when the gear 45 is positioned at the center O, the arcuate positional deviation detection marker 49 is located at the tooth tip 4.
It can be set so as to sequentially traverse along the 6 columns.

According to the configuration shown in FIG. 10, the arcuate positional deviation detection marker 49 can detect detection data representing the positions of successively adjacent tooth tips 46 pixel by pixel. If the tooth tip 46 is misaligned in the rotational direction with respect to the position of the tooth of the master gear, the amount of misalignment in the rotational direction can be detected.

Therefore, if the monitoring area 47 as a whole is moved in the rotational direction to eliminate the positional deviation based on the detection output of the arcuate positional deviation detection marker 49, if there is a chipped tooth with respect to the master tooth tip, it will be removed. can be reliably detected without being affected by a 1-position shift.

(3) In the above embodiment, a marker with a width of one pixel was used as the positional deviation detection marker, but the width of the positional deviation detection marker is not limited to this. In short, it is fine as long as it is linear as a whole.

(4) In the above embodiment, as shown in FIG.
The observed object displayed on the imaging screen clearly has a white level or black level, and therefore the pixels included in the positional deviation detection marker clearly fall from the logical "0" level corresponding to the boundary position of the observed object. The embodiment has been described in which the present invention is applied when it is possible to obtain detection information that switches to the logical "1" level. However, when the boundary of the observed object on the imaging screen includes a gray level, the boundary position It may become unclear.

That is, in this case, the video signal VD obtained from the television camera 13 has a white level L at the boundary position as shown in FIG. A distance of 66 pixels is required to fall from to the black level Ll, and therefore a detection signal whose positional shift amount is unclear is generated by the number of pixels between this distance Δd.

In such a case, in the configuration shown in FIG. 4, for example, among the luminance data taken into the master image memory section 24M and the detected image memory section 24D, the positional deviation data is extracted from the pixels included in the positional deviation detection marker. Register 29F
Data on the white level Lw and the black level L are extracted based on the detection data taken in, and the threshold levels L, yM are calculated as the average values, and the threshold level LTH is used as the detection signal representing the video signal VD. By comparing them, detection information I N F M about the master and detection information INFD about the object to be detected may be obtained as shown in FIG. 1 (B).

In this way, it is possible to reliably detect positional deviations even for objects to be detected that include gray levels.

(5) In the above-described embodiments, the present invention was applied to the case of inspecting labels attached to data and the tips of gear teeth, but the present invention is not limited to this and can be applied to various objects to be observed. .

(6) In the above embodiment, the position of the monitoring area is corrected when the image of the observed object is misaligned with respect to the master image. Alternatively, by changing the relative position of the object to be observed which is imaged by this television camera 13, the same effect as in the above case can be obtained even if the imaging screen of the video signal VD is corrected as a whole. Can be done.

〔Effect of the invention〕

As described above, according to the present invention, a linear positional deviation detection marker is set on the imaging screen, and the position is determined based on the correlation of detection information obtained based on the luminance information of pixels included in the positional deviation detection marker. By detecting the deviation,
It is possible to detect the positional shift of the observed object on the image with respect to the master with high accuracy.

[Brief explanation of the drawing]

FIG. 1 is a route map showing an imaging screen in the first embodiment;
Fig. 2 is a side view showing the data of the observed object, Fig. 3 is a signal diagram for explaining the detection information obtained from the positional deviation detection marker, and Fig. 4 is the object inspection device in the first embodiment. Systematic connection diagram showing specific configuration, Figure 5 is RAM
6 is a flowchart showing the object inspection processing procedure, FIG. 7 is a systematic connection diagram showing the second embodiment of the object inspection device, and FIG. 8 is the object inspection processing procedure. A flowchart, FIG. 9 is a route map showing the arrangement of positional deviation detection markers when the observed object is circular, and FIG. 10 is a route map showing the arrangement of positional deviation detection markers when the observed object is a gear. Figure 1) is a signal waveform diagram for explaining signal processing in the case of an observed object including gray levels. 3...Label, 4...Character pattern,
5... Imaging screen, 6... Monitoring area, 7
X, 7Y... Marker for positional deviation detection, 1)...
...Object inspection device, 13...Television camera, 14...Video data input section, 15.
...timing control section, 25 ... CPU
.

Claims (3)

[Claims] (1) Of the image capture screen displayed based on the video signal obtained by imaging the object to be observed with a television camera, the video signal portion corresponding to the monitoring area set at a predetermined monitoring point is transferred to the master. In an object inspection device configured to detect whether the state of the observed object in the monitoring area is normal or abnormal by comparing it with a corresponding video signal portion, a linear positional deviation is detected on the imaging screen. A detection marker is set, and the first detection information obtained from the video signal portion corresponding to the positional deviation detection marker in the imaged screen for the master and the positional deviation in the imaged screen for the observed object are set. The positional deviation of the master and the observed object is detected by correlating with the second detection information obtained from the video signal portion corresponding to the detection marker, and the positional deviation of the observed object is detected based on the positional deviation detection result. An object inspection device characterized in that the relative position of the monitoring area with respect to the imaging screen is corrected. (2) Based on the positional deviation detection results, the display position of the monitoring area on the imaging screen for the observed object is adjusted;
The object inspection device according to claim 1, wherein the object inspection device corrects the positional deviation by moving the object to be observed. (3) Based on the positional deviation detection result, the positional deviation is corrected by moving the display position of the imaging screen for the observed object with respect to the monitoring area. The object inspection device described in .