JP3054870B2 - Object recognition device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an object recognizing apparatus, and more particularly, to inspecting the presence or absence of an object to be recognized and the presence or absence of a defect based on a video signal obtained from a video camera. It is suitable to be applied to the case.

[Summary of the Invention]

The present invention relates to an object recognition apparatus using a television camera, wherein a monitoring point having brightness information that can be easily determined at the time of imaging of a recognition target based on video information obtained from a plurality of initial setting ranges forming an automatic setting section is provided. By judging and automatically initializing the monitoring area at the monitoring point,
The monitoring area can be initialized with a sufficiently high practical accuracy without imposing an excessive burden on the operator.

[Conventional technology]

Conventionally, as an object recognition device of this type, for example, in an injection molding machine, monitoring of occurrence of abnormal operation in an injection molding cycle, and inspection of a product in a manufacturing process of a printed wiring board, an image of an object to be recognized, that is, injection molding. In the case of a machine, images of the mold and its surroundings, and in the case of a printed wiring board, images of the printed wiring board, etc., are captured by a television camera and displayed as shown in FIG.
There has been proposed a method in which a video signal representing a PIC is obtained and an object is recognized based on whether a predetermined change has occurred in the video signal (Japanese Patent Laid-Open No. 60-39581).

In this case, predetermined positions P1, P2 in the display screen PIC represented by a video signal obtained from the television camera.
Are set to predetermined sizes, and by integrating video signals corresponding to the monitoring regions K1, K2,..., The monitoring regions K1, K2,.
The monitoring information regarding the luminance of the target is obtained, and when an unexpected change occurs in the monitoring information in each processing cycle, it is determined that an abnormality has occurred in the recognition target.

[Problems to be solved by the invention]

However, when trying to recognize the presence or absence of an object or the presence of a defect using such an object recognition device, the display screen
The coordinate position P where the monitoring area K1, K2 ... on the PIC is set
If the settings of 1 (x ₁ , y ₁ ), P2 (x ₂ , y ₂ ) are not appropriate, erroneous determination may occur in the recognition result.

For example, when inspecting a printed wiring board, as shown in FIG. 18, the television camera 2 is arranged so as to face the surface of the printed wiring board 1 to be recognized, and a predetermined camera is determined by the field of view FILD. The video signal VD is obtained by imaging the printed wiring board 1 positioned at the position. In such a case, when trying to determine whether or not a legitimate component is correctly mounted on the printed wiring board 1, in order to reduce the error rate of the determination result as much as possible, it is necessary to select " It is conceivable that the monitoring areas K1, K2,... Should be set in the video portion having brightness information that is easy to determine, that is, the brightest video portion or the darkest video portion.

However, if the monitoring areas K1, K2,... Are set in the brightest or darkest video area, and if an abnormality such as the missing of the part to be monitored occurs, the monitoring area
Since the brightness of K1, K2... Changes from the brightest brightness or the darkest brightness to the brightness of an insulating substrate having a normal brightness, it is easy to determine the difference in brightness. If the monitoring area K1, K2 ... is set in the video part with the brightness of, the luminance will only change from medium to medium even if the part is missing, so the judgment of the abnormality is incorrect. Is more likely to occur.

In the object recognition device applied to the injection molding machine monitoring device, as shown in FIG.
Together with the movable mold 5, the television camera 11 is mounted on a mount 10 fixed on the holder 9 of the solid mold 8 so that images of the movable mold 5 and its surroundings can be captured in the field of view FILD. The movable side mold 5 and its surroundings are imaged from obliquely above when mounted.

In this case, when setting the monitoring areas K1, K2,..., The video portion having the largest change in brightness in the relationship between the video before projection and the video after projection has “brightness information that is easy to determine. The video area "
If K1, K2,... Are set, the error rate of the determination can be reduced.

When monitoring whether or not the injection-molded product has been correctly ejected from the mold in the protruding process, the monitoring areas K1, K2,... Are set at positions where the brightness of the injection-molded product before projection is bright (or dark). In this case, if a change occurs in which the projected image becomes dark (or bright), the probability that the change can be correctly determined increases.

On the other hand, if the monitoring areas K1, K2,... Are set in an image portion having substantially the same brightness as the projected image, the brightness of the monitoring areas K1, K2,. Since the difference is small, the probability of erroneous determination increases.

As described above, in this type of object recognition apparatus, it is considered that whether or not the monitoring areas K1, K2,... Can be set at appropriate positions as much as possible is one of the conditions that significantly affects the error rate of the determination result. , A standard product that is conventionally monitored by an operator before the monitoring operation is started (this is called a master)
Are displayed on a monitor, and an image position considered to be optimal is selected while watching the image on the monitor, and the monitoring areas K1, K2,... Are manually set.

However, when trying to set the monitoring area manually, it is unavoidable that the setting operation becomes extremely complicated in practice.

For example, when the number of mount components on the printed wiring board is large, the operation of setting the monitoring area places an excessive burden on the operator.

In addition, in the case of an injection molding machine monitoring apparatus, it is often difficult to manually select an image position having the lowest determination error rate among the cases where the brightness of an injection molded product is at an intermediate level.

The present invention has been made in view of the above points, and will propose an object recognition device that can automatically and initially set a monitoring area to a position where practically sufficiently high determination accuracy can be obtained. It is assumed that.

[Means for solving the problem]

In order to solve such a problem, in the first invention,
The monitoring areas K1, K2,... Are set at predetermined monitoring positions on the display screen PIC represented by the video signal VD obtained by imaging the recognition target by the television camera 16, and the monitoring areas K1, K2,. In the object recognition device that determines whether or not the recognition target is normal by monitoring the change in the brightness information DWCH obtained from the display, when the master is displayed on the display screen PIC, a plurality of initial settings are displayed on the display screen PIC. Range SET
SEC11 to SEC34, SEC41 and SEC42 with automatic setting
In the state where the master is displayed on the display screen, the brightness information DWCH for each of the plurality of initial setting ranges SET included in the automatic setting sections SEC11 to SEC34, SEC41 and SEC2 is obtained, and the obtained brightness information DWCH is obtained. One or a plurality of representative initial setting ranges SET _PR of the initial setting range SET having brightness information that makes it easy to determine whether or not it is normal is determined, and the monitoring areas K1, K2,... Are set in the representative initial setting range SET _PR. Then, the determination processing for the recognition target is performed using the monitoring areas K1, K2,... Set for the master.

In the second invention, in addition to the first aspect, the representative initialization range SET monitored as the monitoring position sample points included in the _PR area K1, K2 respectively ...... sequentially setting the second
Are determined as the initial setting positions of the monitoring areas K1, K2,... Among the sample points having the brightness information which makes it easy to determine whether or not the second brightness information is normal. To do.

[Action]

In the first invention, the representative initial setting range SET _PR has brightness information that is most easily determined among the automatic setting sections SEC11 to SEC34, SEC41 and SEC42, and thus can reduce the error rate of the determination result, The monitoring area K1, K2 ...
The initial setting can be performed without burdening the operator.

Further, in the second invention, the monitoring areas K1, K2,... Can be set at the monitoring positions having the brightness information most easily determined in the representative initial setting range SET _PR , so that the error rate of the determination result is further reduced. It can be further reduced.

〔Example〕

 An embodiment of the present invention will be described below in detail with reference to the drawings.

[1] Overall Configuration of First Embodiment In FIG. 1, reference numeral 14 denotes an object recognition apparatus as a whole, which is applied to monitoring of the injection molding machine main body 15 described above with reference to FIG. The television camera 16 outputs a video signal VD obtained by imaging the movable type and its surroundings in the field of view FILD, and converts this video signal VD into video data DATA in the video data input circuit 18 under the control of the CPU 17. And store it in the recognition information memory 19.

The CPU 17 registers the register unit 23 according to command information input by the operator from the operation input unit 21 via the bus 22 based on the program stored in the program memory 20.
To perform data processing of the object recognition information.

In this embodiment, the CPU 17 receives the synchronization signal separated from the video signal VD by the synchronization separation circuit unit 25, and receives a video data input circuit 18 having an analog / digital conversion circuit configuration.
After the video signal VD is sampled, the video signal VD is converted into video data DATA and taken into a corresponding memory area of the recognition information memory 19.

The CPU 17 transfers the control signal CONT to and from the injection molding machine main body 15 via the control signal input / output circuit section 27, and in response to the execution of each step of the injection molding cycle by the injection molding machine main body 15, Corresponding data processing can be performed.

The video signal VD obtained from the television camera 16 is supplied to the monitor 32 via the synthesizing circuit 31, and thereby, as shown in FIG. 3, an image corresponding to the field of view FILD is displayed on the display screen PIC.

At the same time, the CPU 17 supplies monitoring area display data KDIS representing the monitoring areas K1, K2,... To the graphic display control section 33, thereby supplying the monitoring area display signal VDK to the synthesizing circuit section 31, and thereby the display screen of the monitor 32. Field of view FILD on PIC
Are superimposed and displayed in the monitoring areas K1, K2,.

In the case of this embodiment, the CPU 17 includes a register 23 (FIG. 2).
The mark shape and the monitoring position P1 (x) to be used in the monitoring areas K1, K2,... By the mark position specifying data D2 of the mark position specifying data register REG2 and the mark shape specifying data D1 of the mark shape specifying data register REG1. _1, y _1),
P2 (x _2, y ₂₎ data representing ...... monitor region display data KD
This is given to the graphic display control unit 33 as IS, and the graphic display control unit 33 generates a monitoring area display signal VDK composed of a video signal that increases the brightness of the display screen PIC at a timing corresponding to the monitoring area display data KDIS.

In this embodiment, a plurality of mark data having different outer shapes and sizes are prepared as mark shapes of the monitoring areas K1, K2,..., And when the operator designates one of them through the operation input unit 21, the designation is made. It can be stored as data mark shape designation data D1.

The CPU 17 causes the monitoring data forming unit 35 to form brightness information in the monitoring areas K1, K2,... Based on the video data DATA loaded into the recognition information memory 19.

The recognition information memory 19 stores video data DATA obtained at a timing when the injection molding machine main body 15 is in a state before the mold is opened and before the ejection.
And a second frame memory 19B for taking in video data DATA obtained at a timing after the projection, and the CPU 17 comprises a first and a second memory 19B.
Of the video data DATA stored in the frame memories 19A and 19B, the mark position data D3 of the mark position data register REG3 of the register unit 23 (FIG. 2) and the mark shape designation data D1 of the mark shape designation data register REG1. Are determined by the shifter 35A and the integrator 35B of the monitoring data forming unit 35, and are integrated as monitoring data DWCH in the register unit 23 (the second unit).
In the figure, the post-molding pre-projection monitoring data register REG5 and the post-projection monitoring data register REG6 are stored as post-molding pre-projection monitoring data D5 and post-projection monitoring data D6, respectively.

Thus, when the injection molding machine body 15 performs a normal projecting operation, the value of the monitoring data DWCH at each monitoring point is determined by the determination reference value data D4 stored in the determination reference value data register REG4 (FIG. 2). When the injection-molded product remains without being pushed down, the value of the monitoring data DWCH does not change so as to cross the judgment level defined by the judgment reference value data D4. It can be determined whether or not the main body 15 has normally moved down the injection molded product.

Based on the monitoring data D5 and post-projection monitoring data D5 before and after the mold opening multiple times in the past as the determination reference value data D4, Is obtained by calculation and stored in the criterion value data register REG4 of the register unit 23.

[2] Initial setting of the monitoring area The CPU 17 displays the image of the standard product (that is, the master) on the display screen PIC in advance based on the brightness information processing operation principle shown in FIG. 4 and FIG. , The monitoring area Kj (j = 1, 2,...) Is automatically initialized to the optimum position.

That is, as shown in FIG. 4, the display screen PIC has 4 sections and 3 sections in the x and y directions, respectively, for a total of 4 ×
3 = 12 sections are automatically set and divided into grids into sections SEC11 to SEC34. Each of the automatic setting sections SEC11 to SEC34 is x
The initial setting ranges SET of 3 × 3 = 9 are arranged in a grid pattern in the direction and the y direction. As shown in FIG. 5, each initial setting range SET has 5 × 5 = 9 in the x direction and the y direction. Sample video data representing the luminance of 25 pixels is arranged.

In the monitoring area initial setting mode, CPU17 automatically sets SEC
11, SEC12 …… Initial setting range by integrating sample video data included in each initial setting range SET for each of the nine initial setting ranges SET included in SEC34.
Initial setting range brightness information, which is the integrated luminance data of SET, is obtained for the timing before the protrusion of the injection molding machine main body 15 and the timing after the protrusion, and the deviation of the initial setting range brightness information before and after the protrusion is maximized. (Therefore, it has the brightness information that is most easily determined.) Initial setting range SET
The initial setting range that represents each of the automatic setting sections SEC11, SEC12... SEC34, that is, the representative initial setting range SET
_PR (shown by hatching in FIG. 4).

Thus, the first stage of the brightness information processing is completed, and the CPU 1
Next, as the brightness information processing of the second stage, the optimal monitoring position having the brightness information that is most easily determined is determined from the representative initial setting range SET _PR .

That is, as shown in FIG. 5, the CPU 17 determines the positions (x = 5) of 5 × 5 = 25 pixels constituting the representative initial setting range SET _PR.
t, y = u), (x = t + 1, y = u) (x = t +
4, when y = u + 4) is set to a value obtained by integrating sample video data (3 × 3 = 9 pieces of sample video data in this embodiment) which fall within the range of the mark shape of the monitoring area Kj. The monitor area brightness information is formed by the operation, and the sample in which the deviation of the monitor area brightness information before and after the projection is maximized at each sample point (therefore, the sample has the brightness information that is most easily determined). The point is determined as the optimal setting position of the representative initial setting range SET _PR ,
This is stored in the mark position designation data register REG2 of the register unit 23 as the mark position designation data D2.

In the case of this embodiment, the first of the recognition information memories 19 (FIG. 1)
And the second frame memories 19A and 19B store sample video data DATA of each sample point before and after the projection with the intersection of the sampling point in the x direction and the sampling point in the y direction as an address. The first and second steps of the processing of the brightness information are executed using the data.

The position of the monitoring area Kj (j = 1, 2,...) Is a point at the upper left corner of the mark shape (in FIG. 5, the sample point Qj _tu (x =
t, y = u)), and a value obtained by integrating video data of sample points falling within the range of the mark shape with reference to the upper left corner point is used as the monitoring area Kj (j = 1). , 2...) As the monitoring data DWCH, which is the monitoring area brightness information before and after the mold opening in the register unit 23 and the post-projection monitoring data register REG6 in the register unit 23 as the monitoring data DWCH. This is stored as D5 (referred to as first monitoring data) and post-projection monitoring data (referred to as second monitoring data) D6.

In this way, when the operator previously sets the monitoring position Qj _tu of the monitoring area Kj for the standard product (that is, the master) in advance, when the injection molding product is manufactured in each injection molding cycle, exactly the same image is displayed in the video data DATA.
Then, the CPU 17 integrates the monitoring area brightness information obtained by integrating the monitoring area Kj with reference to the monitoring position Qj _tu as appropriate first and second monitoring data D5 and D6 in the register unit 23. Can be captured.

After automatically setting the monitoring areas K1, K2,... For the master as described above, the object recognition processing is performed for each injection molding cycle of the injection molding machine main body 15 in accordance with the processing procedure shown in FIG. Execute

[3] Object Recognition Processing When the CPU 17 enters the object recognition processing operation from step SP1 in FIG. 6, in a state where the master is set in the injection molding machine main body 15, the operator at step SP2 operates the operation input section 21.
Waits for the input operation of the mark shape designation data D1 from, and when the input operation is performed, the input data is taken into the mark shape designation data register REG1.

Subsequently, the CPU 17 executes the recognition information memory 19 in step SP3.
The monitoring data forming unit 35 is based on the sample video data before projection after the mold opening captured in the first frame memory 19A and the sample video data after projection captured in the second frame memory 19B. , The integral data of the initial setting range SET in all the automatic setting sections SEC11 to SEC34 (FIG. 4) is obtained, and the initial setting range SET having the largest deviation of the integral data is determined as the representative initial setting range SET _PR .

Thus, one representative initial setting range SET _PR is determined for each of the twelve automatic setting sections SEC11 to SEC34 formed on the display screen PIC (FIG. 4), but the CPU 17 determines each representative initial setting range in the next step SP4. About the setting range SET _PR
Among them, the sample position Qj _tu where the deviation of the integrated data before and after the projection is maximized is determined as the optimum mark position.

As a result, the sample position Qj _tu obtained as the optimum mark position is, when the monitoring area Kj is set at all the sample positions included in each of the automatic setting sections SEC11 to SEC34, the integrated data before and after the protrusion is set. This means that the position has the largest deviation, and thus has the brightness information that makes it easy to determine the monitoring region Kj. Therefore, it means that the monitoring region Kj can be automatically set at the position where the sensitivity of the monitoring operation is the highest.

In this state, the CPU 17 proceeds to step SP5 and, for the monitoring area Kj set at the optimum mark position, stores the first monitoring data D5 after the mold opening and before-projection monitoring data register REG5.
(FIG. 2), after executing the processing to be taken in,
, A process of loading the second monitoring data D6 into the post-projection monitoring data register REG6.

In this way, the CPU 17 drives the injection molding machine body 15 on a trial basis to confirm that the normal operation of manufacturing a standard injection molded product (that is, a master) can be performed. First and second monitoring data D5 and D6 are stored in the register unit 23 as initial data.
It is made to take in.

Thus, the automatic initialization of the monitoring area Kj before the start of the injection molding cycle is completed, the CPU 17 moves to step SP7, and thereafter the injection molding machine main body 15 synchronizes with the operation of injection molding the injection molded products one by one in the injection molding cycle. Meanwhile, when the injection molding machine body 15 sequentially performs the mold opening operation and the protruding operation, a recognition process of whether or not the injection molded product has been normally pushed down is executed. At that time, when the image shifts with respect to the monitoring areas K1, K2,... Automatically set on the display screen PIC, the monitoring positions P1, P2,... Of the monitoring areas K1, K2,. A process for performing correction is performed.

That is, the CPU 17 determines the monitoring area Kj designated based on the mark position designation data D2 written in the mark position designation data register REG2 by the processing in steps SP3 and SP4.
With the monitoring position Pj _mn in FIG. 7 as the center, the monitoring position Pj
5 × 5 = 25 monitoring points Pj _xy (x = m−2, m−1... m + 2, y = n−) in the surrounding correction range AR including _mn
2, the n-1 ...... n + 2) , the sample data contained in the luminance data (i.e. monitoring area Kj when the monitoring region Kj from the first and second frame memories 19A and 19B move in each monitoring position Pj _xy Monitoring data forming unit 35
Obtained by the integral calculation in the deviation before and after the projecting protrusion of the operation result obtaining the greatest monitoring position Pj _xy.

Deviation data resulting deviation is at a maximum and Natsuta mark position, if they match with the monitoring position Qj _tu which automatically initialized at step SP3 and SP4, the injection molding machine body 15 in this current injection molding cycle This means that the injection molding operation matches the brightness data of the monitoring position automatically set for the master, and this further enables the CPU 17 to capture information data that can obtain a practically correct recognition result. Means that

On the other hand, if the monitoring position Pj _xy where the deviation data is maximum is a position that does not match the automatically set monitoring position, this means that the image in the field of view FILD of the television camera 16 has been initialized. It means that it may be out of position.

However, even if such a displacement occurs, the monitoring area initially set to the monitoring position where the deviation data is maximum
It can be considered that the monitoring area Kj has moved so that the image portion of Kj has moved in the actual injection molding cycle while following it, so that there is no risk of causing false recognition in practical use. This means that high-precision identification information could be captured.

Further, in step SP7, the CPU 17 calculates the criterion value data D4 based on the post-mold-opening pre-projection monitoring data D5 and the post-projection monitoring data D6 according to equation (1).

Then, when the CPU 17 recognizes by the control signal CONT taken from the control signal input / output circuit 27 that the timing of the injection molding machine main body 15 has reached the timing before the ejection after the mold opening, the step SP
8, the first monitoring data D5 is fetched from the first frame memory 19A into the monitoring data register REG5 before protrusion after mold opening via the monitoring data forming unit 35, and then, at step SP9, the reference value is determined. It is determined whether the pre-projection monitoring data has a normal brightness level.

If a positive result is obtained in step SP9, the CPU 17 moves to step SP10 at the timing when the injection molding machine main body 15 performs the projecting operation, and the second monitoring data D6 is transmitted from the second frame memory 19B via the monitoring data forming unit 35. After the data is loaded into the post-projection monitoring data register REG6, step SP11
Is used to judge whether the second monitoring data D6 is normal or not.

If a positive result is obtained in step SP11, this means that the operation of the injection molding cycle of the injection molding machine main body 15 is normal, and at this time, the CPU 17 returns to step SP7 described above and performs a new injection molding. The calculation of the mark position and the determination reference value for the cycle is repeated.

On the other hand, if a negative result is obtained in step SP9 or SP11, this means that the above has occurred in the injection molding cycle of the injection molding machine body 15, and at this time, the CPU 1
7 waits for the operator to execute a process for removing the cause of the defect in step SP12, and when the process is completed, returns to step SP8 to repeat the fetching and judging process of new monitoring data D5 and D6. .

Here, the reason why the CPU 17 does not return from the step SP12 to the above-mentioned step SP7 is that correct monitoring data D5 and D6 cannot be obtained when a failure occurs, so that the monitoring position of the monitoring area Kj and the determination reference value data D4 are not changed. This means that it is used as it is in the next determination processing.

At the time of performing the calculation of step SP7 during the repetitive processing, the CPU 17 uses the monitoring data D5 and D6 newly acquired in steps SP8 and SP9, and thus monitors each time the injection molding machine body 15 executes one injection molding cycle. region
The monitoring position of Kj and the criterion value data D4 are updated.

According to the above configuration, when the monitoring area Kj is automatically initialized for the image on the display screen PIC, the display screen PI
C is divided into multiple automatic setting sections SEC11 to SEC34,
Initial setting range SE included in each automatic setting section SEC11 to SEC34
By determining the initial setting range in which the brightness information changes most when an abnormality occurs in T to the representative initial setting range SET _PR, and by setting the monitoring area Kj in the representative initial setting range SET _PR , The monitoring area can be automatically set in the video part of the monitoring target that has the largest response to the abnormal operation. Therefore, it is possible to realize an object recognition device that can execute the object recognition process with a sufficiently high determination accuracy in practical use, and can further reduce the burden on the operator because the setting operation by the operator is not required.

In addition to this, according to the above-described embodiment, the monitoring area Kj is set at the monitoring position where the response to the abnormal operation is largest among the sample information included in the representative initial setting range SET _PR. The monitoring area Kj can be set at a monitoring position where the reaction to the abnormal operation is larger, and an object recognition device with higher monitoring accuracy can be realized.

[4] Second Embodiment FIG. 8 shows a second embodiment, in which automatic setting sections SEC41 and SEC42 set in advance on the display screen PIC.
Is set to have an arbitrary size and shape at an arbitrary position on the display screen PIC.

That is, the first automatic setting section SEC41 is located at the substantially center position of the display screen PIC, as in the case of FIG.
It has a size including 12 default ranges SET having a size including 25 sample video data,
The initial setting ranges SET are arranged in a shape such as 4 × 3 in the x direction and the y direction.

The automatic setting section SEC42 is the first automatic setting section SEC41.
Is set at the lower right position of the display screen PIC, and has a size having eight initial setting ranges SET including 5 × 5 = 25 sample video data, and the initial setting range SET Are arranged in a shape such as 2 × 4 in the x direction and the y direction.

According to the configuration shown in FIG. 8, when there is a reaction portion of an abnormal occurrence intensively in a part of the display screen PIC as the image to be recognized, the automatic setting section is limited to the image portion having a large reaction. By setting SEC41 and SEC42, the amount of data to be processed can be remarkably reduced as compared with the case of setting an automatic setting section over the entire display screen PIC. Can be downsized.

[5] Other Embodiments (1) In the case of the embodiment shown in FIG. 4 or FIG. 8, when the representative initial setting range SET _PR is selected for each of the automatic setting sections SEC11 to SEC34 or SEC41 and SEC42, the representative initial setting range is set. Setting range SET _PR
FIG. 4 corresponds to FIG. 8, and as shown in FIG. 9 or FIG. 10, the brightness information of each of the automatic setting sections SEC11 to SEC34 or SEC41 and SEC42 is selected. When the deviation is equal to or greater than the predetermined determination level, not only one, but a plurality of representative initial setting ranges SET _PR are selected, and the monitoring area Kj is set for all the sample data included in the plurality of representative initial setting ranges SET _PR. Even if the determination of the optimal position to set is performed, the same effect as in the above case can be obtained.

(2) In the embodiment of the object recognition processing procedure shown in FIG. 6, after the representative initial setting range SET _PR is determined in step SP3, the optimum is selected from the sample data included in the representative initial setting range SET _PR in step SP4. The mark position was obtained, but instead of this, step SP
When the representative initial setting range SET _PR is determined in 3, a predetermined position (for example, a center position) determined in advance may be set as the optimum mark position.

In this way, when the representative initial setting range SET _PR is determined, the optimum mark position can be determined so as to be linked thereto, thereby simplifying the object recognition processing procedure.

However, in this case, in the processing procedure of the subsequent injection molding cycle, when the processing in step SP7 is executed, the video to be monitored is determined based on the optimum mark position determined in step SP3. A process of resetting the monitoring area is performed. At this time, the monitoring area can be moved so as to follow the video portion with the largest response in practice. Obtainable.

(3) In the case of the embodiment shown in FIGS. 4, 8, 9 and 10, the automatic setting sections SEC11 to SEC34, SEC41 and S
The case where the monitoring area Kj is individually moved for the EC42 has been described. However, when it is necessary to move one of the monitoring areas Kj, the monitoring area of the entire display screen PIC is changed according to this when the following method is used. They may be used to move them all at once.

The first method executes an operation for obtaining an optimum monitoring position for one monitoring region selected in advance among a plurality of monitoring regions Kj, and as a result, distances Δx and Δy in the x direction and the y direction.
If the calculation result that only moves the monitoring position is obtained, the position of the monitoring area other than the monitoring area on which the calculation of the movement amount is based is changed by the same amount, that is, the distances Δx and Δy
Just to move.

In this way, when the setting positions of the plurality of monitoring areas arranged on the display screen PIC are changed, the change calculation can be further simplified, and in particular, for example, as shown in FIGS. 8 and 10, It is suitable to be applied to the case where the automatic setting sections SEC41 and SEC42 are set in a part of the display screen PIC.

Further, as a second method, the movement amounts Δx and Δy are calculated by a function using the brightness information obtained from the monitoring area Kj set on the display screen PIC as a variable, and based on the movement amounts Δx and Δy. All the monitoring areas Kj are moved.

That is, deviations B1 (Δx, Δy) and B2 (Δ) of the brightness information before and after the projection of the monitoring area Kj (j = 1, 2,...)
x, Δy) When the following equation is detected, Is calculated, and the moving amounts Δx and Δy that maximize the luminance difference function B are obtained, and all the monitoring areas Kj (j = 1, 2,...) Are obtained by the moving amounts Δx and Δy.
…) To move the setting position all at once.

In this way, the moving distances Δx and Δy can be determined while taking care not to lose the luminance difference information of the monitoring area Kj on the display screen PIC.
As a whole, the monitoring area Kj can be set with sufficient accuracy for practical use.

(4) In the above embodiment, when integrating the sample video data included in the monitoring area Kj, all the sample video data are integrated under the common weighting coefficient. As shown, the weighting factor may be changed as necessary according to the position of the sample video data in the monitoring area Kj of the sample video data according to the feature of the recognition target.

By the way, in the case of FIG. 11, when an image portion having a particularly high luminance on the display screen at the center point is to be detected by the monitoring area Kj, the weighting coefficient is set to “1” for the surrounding sample data. At the same time, a weighting coefficient “8” is set for the sample data at the center point.
In this way, the monitoring area Kj can be set as the optimum position of the monitoring area Kj at the position where the luminance at the center position is large.

(5) In the above embodiment, the television camera
A description will be given of a case in which sample video data is taken in based on the 16 video signals VD, and the data stored and held in the recognition information memory 19 is formed with deviation data in the monitoring data forming unit 35 which becomes the shifter 35A and the integrator 35B. However, instead of this, as shown in FIG. 12, after the video signal VD of the television camera 16 is converted into digital data by the analog / digital conversion circuit 41, the video signal VD is transmitted to the integration circuit 44 comprising the addition circuit 42 and the memory 43. Supply and memory 43
By incorporating the integrated data stored in the CPU 17 into the CPU 17 via the bus 22, sample data of the monitoring area may be obtained in a real-time manner.

In the case of the embodiment shown in FIG. 12, the synchronization signal SYNC included in the video signal VD is separated by the synchronization separation circuit 45 and supplied to the timing control circuit 46, and at the same time, the initialization range SET (FIG. , FIG. 8, FIG. 9 and FIG. 10) or the timing signal TM for capturing the monitoring area Kj (sample 2 data in FIG. 5 and FIG. 7).
The signal is supplied to the timing control circuit 46 via the circuit 22.

When the timing signal TM is supplied, the timing control circuit 46 supplies drive signals to the analog / digital conversion circuit 41, the addition circuit 42, and the memory 43 while synchronizing with the synchronization signal SYNC. Integration data for the initial setting range SET and the monitoring area Kj is accumulated.

In the configuration of FIG. 12, the initial setting range SET or the monitoring area Kj can be shifted in the horizontal direction by shifting the timing of the conversion start operation in the analog / digital conversion circuit 41 by ± a few clocks. , The initial setting range SET and the monitoring area Kj can be shifted in the vertical direction.

According to the configuration of FIG. 12, when the brightness information is captured from the video signal VD of the television camera 16 as necessary, the brightness information can be captured in real time. The recognition process can be executed with good responsiveness.

Also, instead of the configuration of FIG. 12, as shown in FIG. 13, a data transmission circuit gate circuit 50 from the analog / digital change circuit 41 to the addition circuit 42 is inserted, and this gate circuit 50 is connected to the CP.
By controlling the opening and closing by the gate signal formed in the timing control circuit 46 based on the timing signal TM supplied from U17, video data corresponding to the initial setting range SET and the monitoring area Kj is taken into the integration circuit 44. In this case, the same effect as in the case of FIG. 12 can be obtained.

In the embodiment of FIGS. 12 and 13, as described above with reference to FIG. 8 or FIG. 10, when the automatic setting sections SEC41 and SEC42 having arbitrary sizes and shapes are set at arbitrary positions, Automatic setting section SEC41, S for each successive frame
You may make it take in data of EC42 one block at a time.

(6) In the case of the embodiment of FIG. 5, the representative initial setting range
In setting the monitoring area Kj at the optimum monitoring position in SET _PR , the representative initial setting range SET _PR is configured to obtain integral value data from all 5 × 5 = 25 sample video data in the x and y directions. However, instead of this, for example, as shown in FIGS. 14, 15, and 16, the number may be reduced as necessary, and in this case, the same effect as in the above-described case is obtained. And the calculation time can be further reduced.

In the case of FIG. 14, as the representative initial setting range SET _PR ,
A range of 5 × 5 is set in the x direction and the y direction with the monitoring position Qj _tu as the center. The monitoring point Qj _tu and the monitoring point Qj are data to be captured in order to determine the optimum monitoring position.
Monitoring data is obtained from eight monitoring points arranged in the x and y directions through _tu and eight monitoring points arranged in an oblique direction through the monitoring point Qj _tu . In this way, the number of data to be captured is increased from 5 × 5 = 25 to 1 while securing the width of the representative initial setting range SET _PR as the range of the monitoring data to be captured to 5 × 5.
+ 8 + 8 = 17, and the calculation time for determining the optimum monitoring position can be shortened accordingly.

In the case of FIG. 15 is limited to the range of 3 × 3 around the monitoring point Qj _tu Representative initial setting range SET _PR. In this case, in the case of FIG. 5, the number of monitoring points from which data is to be fetched is 5 × 5 = 25.
The number can be reduced to 3 = 9, and thus the operation time can be further reduced.

In the case of FIG. 16, compared with the case of FIG.
Four monitoring points in the oblique direction with Qj _tu as the center are deleted from the monitoring points from which data is to be taken. In this case, the calculation time can be further reduced.

(7) In the above-described embodiment, as described above with reference to FIG. 5, the case where the sample point at the upper left corner among the sample points included in the monitor region Kj is used as the monitor point of the monitor region Kj has been described. However, other sample points may be used. In short, one of the sample points included in the monitoring area Kj may be used.

(8) In the above-described embodiment, the monitoring position where the deviation of the video information is maximum is determined as the optimum monitoring point. However, in practice, the monitoring point becomes the television camera 16 or the object to be measured, and is formed by injection molding. Since it varies due to product vibration, misalignment, etc., for example, an average value of the data of the optimum monitoring point obtained in the past multiple injection molding cycles is calculated,
The monitoring position represented by the average value may be set as the corrected monitoring position.

When the average value of the optimum monitoring position is obtained, the monitoring position represented by the average value corresponds to the center of the monitoring position that is varied by the measurement.

〔The invention's effect〕

As described above, according to the present invention, based on a video signal obtained by imaging a master, a representative having brightness information that makes it easy to determine whether or not each of the automatic setting sections including a plurality of initial setting ranges is normal or not. By selecting the initial setting range and setting the monitoring area at one of the monitoring points included in the representative initial setting range, it is possible to obtain a highly accurate determination result without imposing an excessive burden on the operator. it can.

In this way, by setting a monitoring area by selecting a monitoring point having brightness information that makes it easy to determine whether the monitoring point is normal among monitoring points included in the representative initial setting range,
Further, a determination result with higher accuracy can be obtained.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of an object recognizing device according to the present invention, FIG. 2 is a schematic diagram showing a detailed configuration of a register section, and FIG. 3 is set for a video signal display screen. Schematic diagram for explanation of monitoring area, Fig. 4 is display screen PI
Initial setting range SET formed in C, representative initial setting range SET _PR
And a schematic diagram for explaining the automatic setting sections SEC11 to SEC34,
FIG. 5 is a schematic diagram for explaining a method of setting an optimum monitoring point in the representative initial setting range SET _{PR, and} FIG. 6 is a diagram of FIG.
7 is a flowchart showing the object recognition processing procedure of the CPU 17, and FIG.
The figure is a schematic diagram for explaining the correction range AR of the monitoring position Pj _xy ,
FIG. 8 is a schematic diagram showing another method of setting the representative initial setting range, FIGS. 9 and 10 are schematic diagrams showing other embodiments of FIGS. 4 and 8, and FIG. FIGS. 12 and 13 are schematic diagrams for explaining a weighting calculation method at the time of integration processing of the monitoring area Kj.
FIG. 14 is a block diagram showing another embodiment of FIG. 1, and FIGS.
FIG. 16 is a schematic diagram showing another embodiment of FIG. 5 for the representative initial setting range, and FIGS. 17 to 19 are schematic diagrams showing a conventional configuration. 14 ... object recognition device, 15 ... injection molding machine body, 16 ... television camera, 17 ... CPU, 19 ... recognition information memory, 19A, 19B ... frame memory, 23 ... register section,
35 ... Monitoring data forming unit.

Claims (2)

(57) [Claims] 1. A monitoring area is set at a predetermined monitoring position on a display screen represented by a video signal obtained by capturing an image of an object to be recognized by a television camera, and a change in brightness information obtained from the monitoring area. In the object recognition device to determine whether the recognition target is normal by monitoring the, in the state where the master is displayed on the display screen, on the display screen, having a plurality of initial setting range, an automatic setting section In the state where the master is displayed on the display screen, brightness information is obtained for each of the plurality of initial setting ranges included in the automatic setting section, and it is determined whether the brightness information is normal among the obtained brightness information. Determining one or more of the initial setting ranges having easy brightness information as a representative initial setting range, setting the monitoring area in the representative initial setting range, Object recognition apparatus, characterized in that the determination process for the recognition target by using the monitoring area set for. 2. A monitoring area is set at a predetermined monitoring position on a display screen represented by a video signal obtained by imaging a recognition target by a television camera, and a change in brightness information obtained from the monitoring area. In the object recognition device to determine whether the recognition target is normal by monitoring the, in the state where the master is displayed on the display screen, on the display screen, having a plurality of initial setting range, an automatic setting section Setting, and in a state where the master is displayed on the display screen, first brightness information is obtained for each of the plurality of initial setting ranges included in the automatic setting section, and whether or not the first brightness information is normal is determined. One or a plurality of representative initial setting ranges of the initial setting range having brightness information that makes it easy to determine whether or not the sample included in the representative initial setting range Is set as the monitoring position, and the monitoring areas are sequentially set to obtain the second brightness information, and among the sample points, the second brightness information has brightness information which makes it easy to determine whether or not the second brightness information is normal. An object recognition device, comprising: determining a sample point as an initial setting position of the monitoring area; and performing a determination process on the recognition target using the monitoring area set for the master.