JPS6247788A - Object discriminating device - Google Patents

Abstract

PURPOSE:To exactly discriminate an object by a simple constitution, by obtaining a video signal having the contents which have been brought to an image pickup by gradating an edge part of an object to be photographed, and comparing reference data which as been corrected, based on this video signal, and measured data. CONSTITUTION:A video signal VD corresponding to an image which has gradated an edge part of an image pickup object is obtained by a television camera 1. This video signal VD is applied to a deciding circuit 5 through a sample holding circuit 3 and an A/D converting circuit 4. This deciding circuit 5 compares a measured data which has been obtained from an object to be observed 2, with standard data which has been obtained from a reference object, and outputs a result of discrimination related to the object to be observed 2, based on a comparison output for showing a difference of both of them.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to an object identification device, and in particular, to an object identification device that determines the presence/absence, range, position, etc. of an object based on a video signal obtained from a television camera. It is something.

[Summary of the invention]

The present invention provides an object identification device that identifies the state of an observed object based on a video signal obtained by capturing an image of the observed object with a television camera. By obtaining the first detection data based on the reference object and comparing it with the second detection data obtained from the reference object to obtain the identification result for the observed object, it is possible to reliably detect the observed object with a simple configuration. It is designed to reliably identify misalignment of observed objects, presence or absence of missing items, mismatching of widths, etc.

[Conventional technology]

This type of object identification device can be used, for example, to inspect missing parts on printed circuit boards, to check for dropped or broken pins of injection molding products, to inspect missing parts to panel parts, to check the operation of a 7-segment display, etc. It is used to check whether an article is in a predetermined position (i.e., a designed position), such as checking for broken bits in the machine tool area.

In addition, this type of object identification device can be used to inspect the diameter or width of an article, such as inspecting the width of a pattern on a printed circuit board, inspecting the width of an IC internal pattern, sorting the size of fruits, detecting printing misalignment, etc. It is used when inspecting.

[Problem that the invention seeks to solve]

By the way, in this type of conventional object identification device, various devices are used to identify the video signal obtained from the object to be identified (i.e., the observed object) from among the video signals obtained from the television camera. is being done.

One method is to obtain the detection signal VDX (Fig. 14 (A)) representing the width W of the object based on the video signal obtained from the television camera, as shown in Fig. 14. There is a method in which signal processing is performed to make the rise and fall of the signal VDX as steep as possible, and this is compared with the reference signal VDR (FIG. 14(B)).

This conventional method is based on the idea that by matching the contour position of the object to be identified to the actual object as accurately as possible, it is possible to obtain identification results that are close to the truth (i.e., position detection with high accuracy). has been done.

In practice, as a method to satisfy such requirements, the video signal obtained from the television camera is compared with a predetermined threshold level, and when a video signal higher than the threshold level is obtained, the logic rHJ level is Efforts have been made to generate a detection signal VDX (FIG. 14(A)).

Meanwhile, the reference signal VDR is also generated in a similar manner. That is, a standard object having a width or contour according to the design standard is imaged with a television camera, the resulting video signal is compared with a predetermined threshold level, and the standard object shown in FIG. 14(B) is thus imaged. Reference signal VD for
Get R.

This reference signal VDR is converted into a digital signal and stored in a reference signal memory in advance, and when the object identification processing operation actually starts, the reference signal VDR is converted into a digital signal and is converted into a digital signal in synchronization with the identification processing.
is read out from the memory and compared with the detection signal VDX, and if they match, a determination signal of O level is obtained as shown in FIG. 14(C), thereby indicating a "pass" determination.

On the other hand, when the object to be identified is not at the predetermined position, the detection signal VDX is caused by the reference signal VDR (Fig. 15 (B)) rising to the logic rHJ, as shown in Fig. 15 (A). In the range W where it is, the O level corresponding to the logic rLJ is maintained, and as a result, as shown in FIG. 15(C),
When the signal level of the determination signal JUD falls from the O level in the negative direction beyond the threshold level SHO, it is possible to obtain an identification result with a determination of "fail".

Therefore, according to this method, two cases can be identified: when the object to be identified is at the reference signal position and when it is not.

However, according to this method, as shown in FIG. 16, even though it was possible to obtain the detection signal VDX with the same rise width W as the rise width W of the reference signal VDR (FIG. 16 (B)), the When the position of the object is slightly misaligned with respect to the reference signal VDR, the determination signal JUD rises or rises for a time period corresponding to the slight misalignment, as shown in FIG. 16(C). Generates an instantaneous change in signal level, such as a falling signal. These rising and falling signal levels become large enough to exceed the threshold level SHO because the detection signal VDX and the reference signal DR are processed so that they change sharply.
As a result, the final identification result is "fail".

However, in practical object identification devices, as long as the positional deviation does not exceed the allowable range, the result of identifying the presence or absence of an object is "pass".
It is desirable to judge that. However, depending on the conventional method, as described above with reference to FIG. 16, even if the positional deviation is within such an allowable range, a "fail" judgment result is issued, so there is a problem that it is difficult to use in practice.

In the above case, the problems of an object identification device that determines the presence or absence of an object to be identified have been described, but the same practical problems arise when determining the width or diameter of an object to be identified.

In this case, the object identification device detects the rising width W1 of the detection signal VDX for the observed object, as shown in FIG. 17(A).
If it matches the rising width WO of the reference signal VDH,
As shown in FIG. 17(C), the O level determination signal JU
D is obtained, thereby obtaining a judgment result of "pass".

On the other hand, as shown in FIG. 18, for example, since the diameter of the observed object is smaller than the reference object, the detection signal VDX
The rising width W1 of (FIG. 18(A)) is the reference signal VDR.
If the rising width WO is narrower than the rising width WO, the judgment signal JUD changes so that the open signal level of the time width corresponding to the difference in width has a falling waveform part, and this falling signal level becomes the threshold level. By exceeding SHO, a judgment result of "fail" can be obtained.

However, in this case as well, even if there is a difference in diameter in practice, if the difference is within the allowable range, it is necessary to mark it as "pass" rather than "fail." Therefore, in this case as well, there is a problem that it is difficult to use in practice.

Conventionally, as a method to solve such problems, the difference between the rising width of the detection signal VDX and the rising width of the reference signal VDR is counted using a reference clock signal, and when the counted value does not exceed the allowable range, A method has been proposed in which a test result is determined as "pass" and a "fail" is determined when the tolerance exceeds the tolerance range.

However, this method requires a pulse oscillation circuit with sufficient accuracy for practical use, and is relatively complicated, including a configuration for converting the detection signal VDX into a reference pulse number and a configuration for comparing this with the reference pulse number of the reference signal VDR. It is necessary to prepare a complicated configuration, and it is unavoidable that the overall configuration of the object identification device becomes complicated and expensive in practice.

The present invention has been made in consideration of the above points, and has a relatively simple configuration to provide a detection signal V that does not exceed a practically allowable range.
This paper attempts to propose an object identification device that can determine "pass" when DX is obtained.

[Means for solving problems]

In order to solve such problems, the present invention provides an object identification device that identifies the state of the observed object 2 based on the video signal VD obtained by insuring the observed object 2 with the television camera 1. The camera 1 obtains a video signal VD corresponding to an image with blurred edges of the object to be imaged, and includes a determination circuit 5 that receives detection data S2 formed based on the video signal VD. First detection data S2 obtained from observed object 2
The measurement data DAD consisting of is compared with the standard data DAR consisting of the second detection data S2 obtained from the reference object, and a comparison output C representing the difference between the measurement data DAD and the standard data DAR is generated.
Identification result S3 for observed object 2 based on OM
Try to get the following.

[Effect]

The television camera 1 is configured to obtain a video signal VD corresponding to an image with blurred edges, and by capturing an image of a reference object with the television camera 1 having such a configuration, Pi,-data DAR is formed.

Furthermore, measurement data DAD is obtained by lifting and wounding the object to be observed using the same television camera 1, and this measurement data DAD is compared with standard data DAR. This comparison is performed by obtaining a comparison output COM representing the difference between the measured data DAD and the standard data DAR.

Here, since the image of the edge part of the observed object 2 and the image of the edge part of the reference object are each blurred, the comparison output COM is the difference between the edge part of the observed object 2 and the edge part of the reference object. It will have an analog signal level of a magnitude corresponding to the difference in relative position.

As a result, the positional relationship of the observed object 2 with respect to the reference object is
With a simple configuration such as comparing with a threshold level, it is possible to obtain an object identification device that can reliably identify objects.

〔Example〕

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below in detail with reference to the drawings as an embodiment applied to inspecting the presence or absence of components mounted on a printed circuit board and the thickness of a wiring pattern.

In the manufacturing process of printed circuit boards, there is a process in which a conductive wiring pattern is printed after making six holes in an insulating board as necessary, and then mounting chip components using the through holes or lands of the wiring pattern. Manufactured through.

In this manufacturing process, first, when a wiring pattern is printed on an insulating substrate, it is inspected to see whether the wiring pattern printed on the same day is formed according to a predetermined reference pattern. This inspection verifies that the width of the wiring pattern is within the allowable design range and that there are no missing or extra parts in the pattern.

On the other hand, after the chip components are mounted, it is inspected to see if there are any missing components on the printed circuit board, and if there are any misalignments in the mounting positions of the components.

The object identification device used for such inspection has the configuration shown in FIG. In FIG. 1, reference numeral 1 denotes a television camera, which images a printed circuit board 2 as an object to be identified, that is, an object to be observed.

The video signal VD obtained from the television camera 1 is
The sample output S1, which is sampled for each pixel in the sample hold circuit 3 and represents the brightness of each pixel, is converted into a digital signal in the analog/digital conversion circuit 4, and thus the detection data S2 obtained at the output terminal of the analog/digital conversion circuit 4 is supplied to the determination circuit 5.

The determination circuit 5 receives the base value sent from the reference value calculation circuit 6! #
The value multiplication signal 4 is given, and the determination circuit 5 determines whether the observed object 2 is "pass" or "fail" based on this reference value signal S4.
is determined, and the determination result output S3 is sent out.

A synchronizing signal Sll is applied to the television camera 1 from a synchronizing signal generating circuit 7, and a video signal VD is generated in synchronization with this synchronizing signal S11.

The synchronization signal generation circuit 7 also provides a synchronization signal 512 that is synchronized with the synchronization signal Sll to the timing control circuit 8, and outputs a sampling signal TI to the sample hold circuit 3, a sampling signal T2 to the analog/digital conversion circuit 4, and a timing signal 512 to the determination circuit 5. Generates signal T3.

In the case of the embodiment, the objects 2 to be observed are successively positioned one by one in the field of view of the television camera 1 by a conveyor device (not shown). The timing control circuit 8 receives an external timing signal TO representing the timing of the feeding operation from the conveyor means each time the observed object 2 is sequentially fed into the field of view of the television camera 1, and the timing control circuit 8 receives an external timing signal TO representing the timing of the feeding operation from the conveyor means, and the television camera 1 receives the external timing signal TO indicating the timing of the feeding operation. When the image is ready to be imaged, the sample and hold circuit 3, analog/digital conversion circuit 4,
The determination circuit 5 is operated.

The determination circuit 5 has a standard data memory 11, as shown in FIG. This standard data memory 11 stores detection data regarding standard printed circuit boards. Video signal data regarding the printed circuit board Fi12 is loaded in advance.

In this case, the reference printed circuit board 12 has dimensions, component placement, wiring pattern width, etc. according to the design specifications, and chip components 13 that become ICs and other circuit elements are mounted on the surface of the insulating substrate 14. Figure 8 (
As shown in B), a conductive wiring pattern 15 made of, for example, a copper plate is formed. Note that although a wiring pattern is also formed on the surface of the insulating substrate 14, it is omitted from the illustration in order to simplify the drawing.

Therefore, when inspecting missing parts or misalignment of components mounted on an insulating board, first, prior to the inspection, the surface of the insulating board 14 of the reference printed circuit board 12 is imaged by the television camera 1, and the video signal VD Detection data S2 based on this is taken into the determination circuit 5 (FIG. 2). In this standard data acquisition mode, the determination circuit 5 detects the detected data S
2 is stored in the standard data memory 11 via the standard data side terminal E1 of the switch circuit 21 in response to the write clock WT.

The detection data regarding the reference printed circuit board 12 stored in the standard data memory 11 is read out by the read clock RD when inspecting the printed circuit board 2 that is sent into the field of view of the television camera 1 via the conveyor device. The data is read out and sent to the comparison circuit 22 as reference data DAR.

In this inspection mode, detection data S obtained based on the video signal VD obtained from the television camera 1
2 are sequentially supplied to the comparison circuit 22 as measurement data DAD through the measurement data side terminal E2 of the switch circuit 21.

The comparison circuit 22 compares the measurement data DAD with the reference data DAR, and sends a comparison output COM representing the difference to the pass/fail judgment circuit 2.
Send to 3.

The pass/fail judgment circuit 23 is given the reference value signal S4 generated in the reference value calculation circuit 6 (FIG. 1), and judges whether the signal level of the comparison output COM is within the permissible range, and determines whether or not the signal level of the comparison output COM is within the permissible range. When the signal level of the comparison output COM is within the range, the currently observed printed circuit board 2 is judged as "passing", whereas when the signal level of the comparison output COM exceeds the allowable range, it is judged as "fail". being done.

In addition to the above configuration, the object identification device shown in FIG. A signal corresponding to an image with blurred edges is obtained as a video signal VD.

In the case of this embodiment, the television camera 1 has an image pickup tube as the image pickup device main body 32, and as shown by the solid line in FIG. The light is focused on the light receiving surface 34, so that the captured image IMO of the printed circuit board is formed on the light receiving surface 34, as shown in FIG. 4(B).

The blurring device 31 inserted in the front surface of the imaging optical system 33 having such a configuration has a blurring lens 35, and by combining this lens 35 with the imaging optical system 33, the broken line in FIG. As shown in , substantially the imaging optical system 3
3 closer to the light-receiving surface 34, thereby moving the focal position to the rear of the light-receiving surface 34. As a result, the image IMO of the printed circuit board that should be formed on the light receiving surface 34 is blurred, and the blurred image IMI (fourth image) on the light receiving surface 34 is blurred.
(B)).

In the above configuration, when inspecting the printed circuit board on which the mounted components 13 are mounted on the insulating substrate 14 for missing parts or misalignment of each mounted component, the operator first uses the television camera 1 to inspect the reference printed circuit board. The substrate 12 (FIG. 3(A)) is imaged (FIG. 1). At this time, the switch circuit 21 of the determination circuit 5 is switched to the state connected to the standard data side contact E1, and the video signal V
Detection data S2 corresponding to D is stored in the standard data memory 11 for each pixel (FIG. 2).

At the same time, the detection data S2 is sequentially taken into the reference value calculation circuit 6 (FIG. 1), and a threshold level is calculated for each pixel. In this embodiment, the reference value calculation circuit 6 requires a value of, for example, 25% or 50% of the imported standard data as the reference value signal S4 representing the threshold level for determining whether the product is acceptable or not. The operator can select according to his/her needs.

Once the standard data has been captured in this way, the printed circuit board 2 (Fig. 1) to be observed is sequentially set in the field of view of the television camera 1 using a conveyor device, and the television camera 1 is set at the set timing. john camera
Detection data S corresponding to the video signal VD sent from
2 is taken into the determination circuit 5.

At this time, the switch circuit 21 of the determination circuit 5 is connected to the identification data side contact E2 (FIG. 2), and thus the detection signal S2 is sequentially sent to the comparison circuit 22 for each pixel as measured data DA.
It will be supplied as D.

On the other hand, the data in the standard data memory 11 is sequentially read out at the timing when pixels of measurement data arrive at the comparison circuit 22, and is supplied to the comparison circuit 22 as reference data DAR. At this time, the comparison circuit 22 compares the position of the mounted component on the printed circuit board 2 and the position of the mounted component 13 on the reference printed circuit board 12 (FIG. 3(A)) stored in the standard data memory 11. As a result, the comparison output COM of the comparison circuit 22 is determined as shown in FIG. - As shown in FIG. 9(C), the circuit operates in response to changes in the analog signal level.

First, the measurement data DAD (6th
(A)) is not misaligned with respect to the reference signal DAR (FIG. 6(B)), the signal level of the comparison output COM becomes approximately 0 level. Therefore, the pass/fail judgment circuit 23
As shown in FIG. 6(C), the analog signal level of the comparison output COM is the threshold level SH and SH
Based on the fact that it is between t, a determination signal S3 indicating "pass" is output.

In addition, in Fig. 6 (A), 13X is the observed object '2
It represents the position when the edge portion of the mounted component above is converted into measurement data DAD. In addition, in FIG. 6(B), 13S
represents the position when the edge of the corresponding mounted component on the reference printed circuit board 12 (FIG. 3) is converted into the reference signal DAR. The same applies to other drawings.

On the other hand, if the position of the mounted component of the observed object 2 is slightly shifted from the position of the corresponding mounted component on the reference printed wiring board 12, the As shown in Figures 7 (A) to (C), the rising and falling parts of the measurement data DAD and reference data DAR change smoothly because the edges of the image are blurred. . Thus, the comparison output COM of the comparison circuit 22
corresponds to the difference between the reference data DAR and the measured data DAD, and exhibits a change similar to when the DC signal level is slightly varied around the 0 level.

The peak value of this comparison output COM represents the difference between the position of the edge portion of the mounted component of the observed object 2 and the position of the edge portion of the reference mounted component. And this comparison output CO
The analog signal level of M is the threshold level SH,
and SHL, the pass/fail judgment circuit 23 sends out a judgment output S3 indicating "pass".

As the positional deviation further increases, the peak value of the comparison output COM increases correspondingly to the positional deviation, as shown in FIG. 8(C), and this increases the threshold levels SHH and SHL. It becomes a state of exceeding. At this time, the pass/fail judgment circuit 23 sends out a judgment output S3 whose content is "fail".

In this way, the analog signal level of the comparison output COM of the comparison circuit 23 has a large fluctuation range from the O level, corresponding to the positional deviation between the edge part of the mounted component of the observed object 2 and the edge part of the reference mounted component. Therefore, if the threshold levels SH, l, and SHL are selected to predetermined values as necessary, even if a position shift occurs, the comparison output COM
The signal level is the threshold level 5H)l and SH
As long as the range between L is not exceeded, it is possible to realize an object identification device that can accept this and determine "pass".

On the other hand, if the mounted component of the observed object 2 is out of stock, as shown in FIG. 9, the measurement data DAD becomes almost O level, so the analog signal level of the comparison output COM is set to the lower limit threshold. It will be in a state that greatly exceeds the level SHL. At this time, the pass/fail judgment circuit 23 sends out a judgment output S3 whose content is ``fail J.'' In this way, missing items can be reliably detected.

6 to 9 have described the object identification operation for identifying misalignment and missing parts of mounted components on a printed circuit board. However, when identifying wiring patterns, the determination circuit 5 operates as shown in FIG. 1O. -Performs a determination operation as shown in FIG.

In this case, before measuring the object to be observed 2, the operator first images the wiring pattern (FIG. 3(B)) of the reference printed circuit board 12 with the television camera l, and obtains standard data representing the video signal VD. Switch circuit 21 (
(FIG. 7) is stored in the standard data memory 11.

Then, the printed circuit board to be observed object 2 is sequentially positioned in the field of view of the television camera 1 using a conveyor device, and the detection data S2 representing the video signal VD representing the wiring pattern of the printed circuit board obtained at the timing of the positioning is switched. Measurement data DAD via circuit 21
It is given to the comparator circuit 22 as .

As a result, if the width W of the wiring pattern of the observed object 2 is equal to the width WR of the wiring pattern of the reference printed circuit board, the analog signal level of the comparison output COM will be at the threshold level, as shown in FIG. 10(C). By placing SHO and SH between them, the pass/fail judgment circuit 23 sends out a judgment output S3 of "pass".

On the other hand, when the width W of the wiring pattern of the observed object 2 is narrower than the width WR of the reference pattern (Fig. 11 (A) and (B)), the analog signal level of the comparison output COM is The value corresponds to the difference between the rising position or falling position of DAD and the rising position or falling position of reference data DAR. If the value of this comparison output COM exceeds the threshold level SHL, the pass/fail judgment circuit 23
sends out a "fail" determination result output S3.

However, in this case, if the positional deviation between the rising position or falling position of the measurement data DAD and the rising position or falling position of the reference data is not so large, the comparative output COM
The signal level does not exceed the threshold level SHL. At this time, the pass/fail judgment circuit 23 determines that the result is within the allowable range and sends out a "pass" judgment output S3.

On the other hand, when the pattern width W of the wiring pattern of the observed object 2 is wider than the width WR of the reference pattern, the analog signal level of the comparison output COM is as shown in FIG. A comparison output CO having an analog signal level corresponding to the difference between the rising position or falling position of the reference data DAR and the rising position or falling position of the reference data DAR.
M is obtained, and if this signal level is higher than the threshold level 5H11, the pass/fail judgment circuit 23 sends out a judgment output S3 indicating "fail".

However, in the state shown in FIG. 12, if the difference between the rising position or falling position of the measurement data DAD and the rising position or falling position of the reference data DAR is small, the comparison output CO
The analog signal level of M is the threshold level 5H1
It will not exceed 1. At this time, the pass/fail judgment circuit 23 sends out a "pass" judgment output S3.

In this way, regarding the width of the wiring pattern of observed object 2,
If the difference between the width of the observed object and the reference width is small and within the allowable range, it is possible to realize an object identification device that can output a "pass" determination result.

According to the embodiment described above, when observing the positional shift or width difference of the observed object, the television camera 1
By blurring the edge portions of the object to be imaged, it is possible to convert the difference in position of the edge portions of the object to be observed into an analog signal level. Thereby, analog signal level comparison means can be used as means for determining the difference. Therefore, it is possible to realize an object identification device that can reliably identify objects with a simple configuration that does not require the use of a complicated and large-scale configuration such as a means for counting using reference pulses as in the conventional case. obtain.

In the case of the embodiment shown in FIG. 1, an example in which an image pickup tube is used as the imaging device main body 32 has been described, but even if a solid-state imaging device is used as the imaging device main body 32 instead of this,
The same effects as in the above case can be obtained.

In the case of the embodiment shown in FIG. 4, as a blurring device, the blurring lens 35 has an effect of moving the focal position of the imaging optical system 33 to the rear of the light receiving surface 34, but instead of this, the blurring lens 35 is used as a blurring device. As shown in (A), by producing an effect of substantially moving the imaging optical system 33 in a direction away from the light receiving surface 34, the image IMO of the printed circuit board on the light receiving surface 34 is blurred and a blurred image IM2 is created. may be caused to occur.

In the embodiment shown in FIG. 1, the optical focus position is changed and controlled by the blurring device 31 in order to obtain the video signal VD based on the blurred image. It may be configured as follows.

In FIG. 13, the imaging device main body 32 is an imaging tube;
The voltage applied to this focus coil 41 can be changed by a focus coil voltage setting variable resistor 42.

In this way, by controlling the degree of convergence of the scanning beam on the target of the image pickup tube by the variable resistor 42, the diameter of the scanning beam landing on the target can be blurred, and thus the configuration shown in FIG. Video signal V obtained from television camera 1
Similarly to D, it is possible to obtain a video signal VD similar to that obtained by capturing an image with the object blurred.

〔Effect of the invention〕

As described above, according to the present invention, a video signal VD is obtained from the television camera 1, and the video signal VD is captured with the edges of the object blurred, and reference data corrected based on this video signal and measured data are compared. By doing this, the analog signal level can be adjusted according to the degree to which the position of the edge part deviates from the position of the edge part of the reference object, such as in the case of positional deviation of the observed object, missing parts, mismatched widths, etc. A comparison output COM can be obtained, and thus an object identification device that can reliably identify whether or not it deviates from a predetermined tolerance range can be easily obtained with a construction that is extremely simple in practice.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the object identification device according to the present invention, FIG. 2 is a block diagram showing its determination circuit, FIG. 3 is a route map showing the configuration of a reference object, and FIGS. The figure is a route map to explain the blurring operation of the television camera, Figures 6 to 9 are signal waveform diagrams to explain the operation to identify misaligned objects or missing items, and Figures 10 to 12.
The figure is a signal waveform diagram for explaining the operation of observing the width of the observed object, and FIG. 13 is a block diagram showing another embodiment of the present invention.
FIGS. 14 to 19 are signal waveform diagrams for explaining the conventional technology. 1...Television camera, 2...Object to be observed, 5...Judgment circuit, 11...
Standard data memory, 21...Switch circuit, 2
3... Pass/fail judgment circuit, 31... Blur device, 32... Imaging device main body, 33...
・Imaging optical system, 34... Light receiving surface, 35...
...Blur lens. /r Figure 3 (A) (B) Figure 5 @ Figure 12, Figure II Figure 13 Figure 16 Figure 19

Claims (1)

[Scope of Claims] An object identification device that identifies the state of the observed object based on a video signal obtained by imaging the observed object with the television camera, It has a judgment circuit that obtains a video signal corresponding to an image with blurred edges of the object and receives detection data formed based on this video signal, and the judgment circuit is configured to obtain a first detection data obtained from the observed object. Comparing the measurement data consisting of the detection data with standard data consisting of the second detection data obtained from the reference object, and identifying the observed object based on the comparison output representing the difference between the measurement data and the standard data. An object identification device characterized by obtaining results.