JPH0438193B2 - - Google Patents

Description

[Detailed description of the invention] [Technical field to which the invention pertains] The present invention relates to a two-dimensional solid-state image sensor, etc. used in an automatic visual inspection device for a moving imaged object (hereinafter referred to as a target object) as an object to be inspected. The present invention relates to an image input device using a two-dimensional scanning image sensor.

[Prior art and its problems]

In the following description of each figure, the same reference numerals indicate the same or corresponding parts. In addition, logic "High", "Low" and logic "1", "0" are simply written as "H", "L", "1", "0".

Conventionally, this type of image input device is
A system is known in which a TV camera is driven in a standard non-interlaced system, and instantaneous exposure is performed during the vertical retrace period of the TV camera using strobe light or a shutter to eliminate blurring of the image.

FIG. 9 shows an example of the configuration of an automatic visual inspection device using such a general image input device. , the signal processing device 2 uses the position detection device 5 to detect that the target object 7 moving on the belt conveyor 6 has entered the field of view of the image input device (hereinafter also simply referred to as a TV camera) 3.
The strobe firing signal 10 is sent to the strobe ignition device 1 to ignite the strobe light source 4 during the vertical retrace period immediately after the position detection signal 9 in the image input device 3. let As a result, an image of the target object 7 is instantaneously exposed on the imaging surface in the image input device 3 by the light from the strobe light source 4, and image information (image signal) with less image blur due to movement is captured by the image input device 3. Embedded,
Processed by the signal processing device 2.

FIG. 10 is a timing diagram of the operation of such a conventional image input device 3. 8 is a vertical synchronization signal of the TV camera 3, 9 is a position detection signal indicating that the target object 7 moving on the belt conveyor 6 has entered the field of view of the TV camera 3, and 10 is the strobe firing signal, which is used for position detection. This is the timing of instantaneous exposure by the strobe light source 4 performed within the vertical retrace period 8a immediately after the signal 9. Reference numeral 11 denotes an image read signal, which indicates a period during which image signals are read from the TV camera 3 after exposure.

In this case, the position detection signal 9 is normally emitted regardless of the image scanning cycle of the image input device 3, so if the exposure is performed at the same time as the rise of the position detection signal 9 or after a certain delay time, the vertical retrace line There is a high possibility that exposure will be performed at a time other than period 8a, and the image information will be divided and read out over two screens. Further, the image signal of the portion read out after exposure becomes a large signal and appears as a white line on the screen, resulting in problems such as an inability to obtain accurate image signals. Therefore, exposure is actually performed at the timing shown in FIG. 10 (within the vertical retrace period 8a).

As can be seen from the figure, in the conventional system, the position detection signal 9 is input asynchronously to the scanning of the TV camera 3. Moreover, since instantaneous exposure is performed within the next vertical retrace period 8a, the output time t0 of the position detection signal 9
An undefined delay time τ occurs between the exposure timing and the exposure timing (the output point of the strobe firing signal 10).

Generally, when standard non-interlaced scanning is performed, this delay time τ becomes undefined within the period of 16.7 msec of the vertical retrace signal, that is, the vertical synchronizing signal 8. Therefore, the moving speed of the belt conveyor 6 is now v
[m/sec] is 0 [m] ~ 1.67×10 ^-3 v [m]
Within this range, a deviation occurs between the detection position of the position detection device 5 and the position of the target object 7 during actual exposure. As a result, there are the following major problems.

If a positional shift occurs randomly during exposure between 10 and 16.7×10 ^-3 v [m], the detection accuracy of the apparatus will decrease because the illumination distribution will change depending on the location.

Positional deviations occur randomly during exposure between 20 and 16.7×10 ^-3 v [m], so when performing inspections such as pattern matching, it is necessary to perform position correction calculations over a wide range. The correction method and means become complicated, and the cost of the device increases.

3. If there is a random positional deviation of 0 to 16.7×10 ^-3 v [m] and the moving speed v of the target object is large, the target object will pass through the field of view of the TV camera with only the random positional deviation amount. In some cases, it has the fatal drawback of not being able to capture images.

4 When the moving speed v is large, and the field of view is large enough to tolerate random positional deviations (the image of the target object can be captured within the field of view),
The resolution of the TV camera deteriorates, making it impossible to obtain the required inspection accuracy.

In reality, in addition to this positional deviation of 0 to 16.7×10 ^-3 v [m], positional deviation due to fluctuations in the moving speed v is added, resulting in an even worse condition.

In addition to this, there is also a method of creating a synchronization signal for a TV camera from the position detection signal 9 using a PLL circuit, etc.
A method is being considered in which a conveyance system such as a belt conveyor is driven in synchronization with a camera synchronization signal. However, in the former case, a position detection signal needs to be input at a constant cycle, it is necessary to detect the position by the transport system rather than by the target object, and the target object must be accurately positioned in the transport system. There is.

Also, in the latter case, the target object must be accurately positioned with respect to the conveyance system, and is not practical except in special cases. Furthermore, in either case, there is a drawback that the control circuit becomes complicated and the cost becomes high.

[Purpose of the invention]

This invention eliminates the above-mentioned problems, and when capturing an image of a high-speed moving object, even if the position detection signal is input asynchronously to the synchronization signal of the TV camera, the moving object can be accurately positioned on the screen. The purpose of the present invention is to provide, at a low price, an image input device that is capable of capturing images well and that can sequentially read out image signals corresponding to new exposure over one frame (one screen) from the starting point of the screen. That is.

As a document on this type of technology, there is Japanese Patent Publication No. 13905/1983 by the present applicant.

[Key points of the invention]

The key point of this invention is to make the delay time of the instantaneous exposure timing with respect to the position detection signal constant, and to
By performing post-exposure scanning sequentially from the starting point of one screen, images of high-speed moving objects can be captured on the screen with high positional accuracy, and the image signal is not divided between two frames, allowing normal images to be captured. This allows for reading. Furthermore, by shortening this delay time, even if the speed of the moving object varies, it is possible to capture the image of the moving object on the screen with high positional accuracy.

In other words, the gist of the present invention is a means (such as a strobe light source) that inputs a position detection signal that detects that a moving object to be imaged (such as a target object) has reached a predetermined position and instantaneously exposes the object to be imaged. The above exposure is performed by applying an exposure signal (such as a strobe firing signal) to the camera, and the image of the object to be imaged obtained by the above exposure is scanned through an imaging means (such as a MOS type two-dimensional solid-state image sensor). Starting point of screen scanning (starting point of vertical and horizontal scanning) via means (horizontal and vertical drive pulse circuits, horizontal and vertical shift registers, etc.)
In an image input device (such as a TV camera) that obtains an image signal by raster scanning from a source, the position detection signal is inputted to store the memory contents (horizontal and vertical shift register contents, etc.) that give the position of the current scanning point in the scanning means. ) and updating the storage contents to the storage contents of the starting point position (initialization circuit, horizontal (vertical) drive pulse circuit, etc.), and after updating the storage contents,
and means (such as a vertical drive pulse circuit) for outputting the exposure signal after a predetermined time (delay time) has elapsed from the input point of the position detection signal, and means for starting scanning of the scanning means after outputting the exposure signal. (such as an initialization circuit), and furthermore, the predetermined time is shorter than the time to finish scanning one screen (such as the period of a vertical synchronization signal).

[Embodiments of the invention]

The present invention will be explained below based on FIGS. 1 to 8. FIG. 1 is a block diagram showing the configuration of the first embodiment of the present invention, FIG. 2 is a time chart explaining the operation of FIG. 1, and FIG. 3 is an explanation of the initialization operation in FIG. 1. Figure 4 shows the second embodiment of the present invention.
FIG. 5 is a diagram illustrating the initialization operation of FIG. 4, FIG. 6 is a block diagram showing the configuration as a third embodiment of the present invention, and FIG. The figure is a diagram explaining the initialization operation of FIG. 6, and FIG. 8 is a MOS type two-dimensional solid-state image sensor (also referred to as an image sensor) 12 as an example of an image sensor used in the image input device 3 of the present invention. FIG.

The configuration and operation of the image sensor 12 will be explained in advance with reference to FIG. 27, 27aa, 27ba,...,
27ab, 27bb, ..., ~ are photodiodes,
35, 35aa, 35ba,..., 35ab, 35bb,
. . . - are vertical MOS switches provided corresponding to each photodiode, and one photodiode 27 and one MOS switch 35, which form a pair, constitute one pixel.

25, 25a, 25b, ... are output terminals (referred to as vertical shift register output terminals) of a vertical shift register 45 for switching the vertical MOS switch 35, and generally 25, 25a, 25b, . . .
a, 25b, . . . become "1"("H" level) in order, and turn on the MOS switch 35 (for one row) corresponding to each terminal 25 to perform vertical scanning.

33, 33a, 33b, ... are horizontal MOS switches provided corresponding to each vertical column of pixels,
Output terminals of the horizontal shift register 44 (referred to as horizontal shift register output terminals) 26, 26a, 26
26a, 26b, . . . become "1" in sequence during one horizontal scan, and the MOS switch 33 corresponding to each terminal 26 is turned on to perform horizontal scanning. In this way, each photodiode 27 is sequentially selected by the outputs of the horizontal and vertical MOS switches 33 and 35 that are turned ON, and therefore by the outputs of the horizontal and vertical shift register output terminals 26 and 25, and outputs the image signal 24. line (same symbol), and this output line 24
reverse biased by a power supply 37 connected to
The junction capacitance is charged. After that, when the photodiode 27 is irradiated with light while the connection between the photodiode 27 and the power supply 37 is cut off, the charged charges are discharged at a discharge rate according to the intensity of the light and the irradiation time. The next time the photodiode 27 is selected again, the discharged charge is replenished, and the amount of light irradiated by each photodiode 27 is sequentially read out as a current flowing through the load resistor 36 connected to the image signal output line 24, and an image signal is generated. It will be 24.

Horizontal and vertical shift registers 44 and 4 for turning on and off each MOS switch 33 and 35
5 can be operated by inputting horizontal and vertical input signals 19 and 22 as serial inputs, respectively, and two-phase horizontal and vertical shift clocks 17, 18 and 20, 21, respectively. For convenience of explanation, the signals 19, 22 and shift clocks 17, 18, 20, 21 are also referred to as drive pulses, and signal 19 and shift clocks 17, 18 are combined to form a horizontal drive pulse, signal 22.
and shift clocks 20 and 21 are also called a vertical drive pulse.

Returning to the explanation of the operation of the shift registers 44 and 45, first, when the vertical shift register 45 sets the vertical input signal 22 as a serial input to "1" and inputs one clock of the vertical shift clocks 20 and 21, the vertical shift register outputs the vertical shift register 45. Terminal 25a
becomes “1” and MOS switch 3 of the first row pixel
5aa, 35ab, ... are all turned ON. Next, the horizontal input signal 19 as a serial input of the horizontal shift register 44 is set to "1", and the horizontal shift clock 1 is set to "1".
When 7 and 18 are input for one clock, the horizontal shift register output terminal 26a becomes "1", and the first column
The MOS switch 33a is turned on, and the first
The column photodiodes 27aa are read out.
Further, when the horizontal input signal 19 of the horizontal shift register is set to "0" and the horizontal shift clocks 17 and 18 are input for one clock, the horizontal shift register output terminal 26a is set to "0", and therefore the MOS switch 3 of the first column is set to "0".
3a is OFF, horizontal shift register output terminal 26
b is “1”, therefore, the second row MOS switch 33b
is turned on, and the photodiode 27ab in the first row and second column is read out. Similarly, the horizontal shift clock 1 is turned on while the horizontal input signal 19 is set to "0".
When 7 and 18 are sequentially input, the photodiodes 27ac, 27ad, . . . in the first row are sequentially read out.
Photodiodes 27aa, 27ab,... in the first row
When all of 35aa, 35ab, . . . are OFF, and the vertical shift register output terminal 25b is “1”. Therefore, the vertical MOS switches 35ba, 35bb, . . . in the second row are OFF.
are all turned ON, and similarly, the signals of the photodiodes 27ba, 27bb, . . . in the second row can be sequentially read out. By sequentially reading out all the photodiodes 27 of one screen in the same manner, an image for one field (one screen) is obtained.

Embodiments (1) Next, embodiments of the present invention will be sequentially described. FIG. 1 is a block diagram showing the configuration of a first embodiment of the present invention, which shows the main parts in an image input device 3 used in an automatic visual inspection apparatus similar to the configuration example shown in FIG. This circuit consists of the MOS solid-state image sensor 12, its horizontal and vertical drive pulse circuits 13 and 14, which perform normal non-interlaced scanning, and pulse circuits 13, 14 using external position detection signals 9.
and an initialization circuit 151 for initializing the flash drive and transmitting the strobe firing signal 10.

If there is no position detection signal 9, the initialization circuit 151
Therefore, since the reset pulse 16 is not sent to the pulse circuits 13 and 14, the pulse circuits 13 and 14
A normal raster scan is performed on the image sensor 12. That is, as described in FIG. 8, the horizontal drive pulse circuit 13 outputs a horizontal input signal for setting the serial input section of the horizontal shift register 44 of the image sensor 12 to "1" for a period of one clock at the start of one horizontal scan. 19 and a two-phase horizontal shift clock 1 for shifting the horizontal shift register 44.
7 and 18 are output. Furthermore, the vertical drive pulse circuit 14 similarly outputs the vertical input signal 22 of "1" at the start of one vertical scan, and the two-phase vertical shift clock 21, 22 is output. The photodiode 27 of the image sensor 12 is controlled by this driving clock.
are sequentially read out, but if no exposure is performed, generally only the dark current component is read out, indicating "0"("L" level) of the image signal 24.

When the position detection signal 9 is input to the initialization circuit 151, the reset signal 16 is sent by the circuit 151 to the pulse circuits 13 and 14 immediately after the front edge of the position detection signal 9, and the scanning at that point is at an intermediate stage. Even if , the state returns to the initial condition (reading state from the starting point of one screen). After that, the pulse circuits 13 and 14 complete the normal preliminary scanning of one screen for reasons to be described later, and the initialization circuit 151 detects the one vertical scan end signal 23 from the vertical drive pulse circuit 14, and immediately after that, the strobe ignition signal 10
Send out.

The image sensor 12 is exposed to strobe light at the timing of outputting this signal 10, and the image signal output line 24 is exposed to light by normal one-screen raster scanning that is successively performed by the pulse circuits 13 and 14. image information (image signal 2
4) are sent out sequentially.

Figure 2 is a time chart showing this situation. Normal raster scanning is performed until the position detection signal 9 is input to the initialization circuit 151 at time t0. When the position detection signal 9 is input, the pulse circuits 13 and 14 are activated by the reset signal 16.
is reset and one vertical scan is newly started. That is, in a period 8a0, a vertical synchronizing signal 8 is output from a means not shown to the vertical drive pulse circuit 14, etc., and one screen is subsequently scanned in a period 8b0. Since no exposure has been performed at this point, the video signal generally consists of only dark current components. Thereafter, in response to the 1 vertical scan end signal 23 from the circuit 14, the strobe firing signal 10 is sent from the initialization circuit 151 during the vertical retrace period 8a1, and exposure is performed. Subsequently, by scanning one screen in the next period 8b1, the image information accumulated by exposure during the output period of the image read signal 11 is read out from the image sensor 12 as an image signal 24.

Therefore, the delay time τ from the input time t0 of the position detection signal 9 to exposure is constant at 16.7 msec. Furthermore, since scanning after exposure is performed from the starting point of one screen, image information can be read out within one field without being divided.

Next, in the above description, the reason why one screen worth of preliminary scanning is performed as in period 8b0 in FIG. 2 before exposure will be explained with reference to FIG. 25,2 in Figure 3
6 is the vertical and horizontal shift register output terminal 25, 25a, 25b, . . . , 26, 26a, 26 of each vertical and horizontal shift register 45, 44 described in FIG.
27aa, . . . correspond to the output signals b, . . . , and 27aa, .

In other words, if normal scanning is not completed at the input time t0 of the position detection signal 9 as shown in FIG. ”) (giving the scanning position at the time t0). Therefore, if you try to start scanning from the starting point of one screen in this state, when selecting the photodiode 27aa in the first row and first column, for example, as shown in FIG.
This is because 7af, 27da, and 27df are read out at the same time, and normal output cannot be obtained.

This method uses a MOS type two-dimensional solid-state image sensor 12,
In addition to TV cameras that use CCD type two-dimensional solid-state image sensors, etc., it can also be used for regular image pickup tube type TV cameras. In the case of an image pickup tube type TV camera, if the effects of dark current, extraneous light other than strobe light, and afterimages can be ignored, preliminary scanning of one screen before exposure is unnecessary.

In this embodiment, preliminary scanning of one screen before exposure is performed at a normal speed, but the clock frequency during this period is increased (in the case of an image pickup tube, the frequency of the deflection coil) is increased to a certain delay time τ in Fig. 2. can be shortened. Therefore, according to this embodiment, exposure is performed with a constant delay time τ with respect to the position detection signal that is input regardless of the vertical synchronization signal.
As long as the moving speed of the target object does not fluctuate, the image of the target object can be captured on the screen with high positional accuracy. Moreover, since the screen is sequentially scanned from the starting point after exposure, the screen of one field of the TV camera can be read out without being divided, and subsequent processing can be simplified.

Note that the delay time τ is 16.7 msec when the frequency of the vertical synchronizing signal 8 is set to a frequency equal to the commercial frequency, for example, 60 Hz.

Embodiment (2) FIG. 4 is a block diagram showing the configuration of a second embodiment of the main part of the present invention. This circuit includes a MOS type solid-state image sensor 12 and a horizontal and vertical drive pulse circuit 13.
In addition to the circuit that performs normal non-interlaced scanning consisting of
An initialization circuit 152 is provided which sequentially performs sending of the signal and initializing the pulse circuits 13 and 14. Furthermore, in addition to this, the horizontal and vertical shift clocks 1 are used as output drive pulses of the pulse circuits 13 and 14.
7, 18 and 20, 21 and the horizontal and vertical input signals 19 and 22 and the outputs 17 to 17 as initialization output drive pulses of the initialization circuit 152, respectively.
Shift clocks 17a, 18a and 20a, 21a corresponding to 22 and input signals 19a and 22
and a only during the initialization period of the image sensor 12,
During this period, the initialization output drive pulse 17
Horizontal and vertical drive pulse switching circuits 29 and 30 are provided for applying signals a to 22a to the image sensor 12.

When the position detection signal 9 is not input, the drive pulse switching signal 31 is not sent from the initialization circuit 152 to the pulse switching circuits 29 and 30. Furthermore, since the reset signal 16 is not sent from the initialization circuit 152 to the pulse circuits 13 and 14, the pulse circuit 1
3 and 14, drive pulses 17 to 22 are applied to the image sensor 12 via switching circuits 29 and 30, and normal raster scanning is performed on the image sensor 12.

When the position detection signal 9 is input to the initialization circuit 152, the circuit 152 sends a drive pulse switching signal 31 to the pulse switching circuits 29 and 30, and also sends a drive pulse switching signal 31 to the image sensor 12 for initialization via the circuits 29 and 30. The initialization horizontal shift clock 17a, 1 as the drive pulse, that is, the initialization horizontal drive pulse.
8a, initialization horizontal input signal 19a and initialization vertical shift clock 2 as initialization vertical drive pulse;
0a, 21a, and sends out an initialization vertical input signal 22a. In this way, when the initialization of the image sensor 12 is completed by sending out the initialization horizontal and vertical drive pulses 17a to 19a and 20a to 22a for a certain period of time,
The initialization circuit 152 sends out a strobe firing signal 10 and also sends out a reset signal 16 to the pulse circuits 13 and 14. Furthermore, initialization circuit 1
52 eliminates the drive pulse switching signal 31 sent to the pulse switching circuits 29 and 30, so that the horizontal and vertical drive pulses 17 to 22 from the pulse circuits 13 and 14 are applied to the image sensor 12. . Since the pulse circuits 13 and 14 have been reset by the initialization circuit 152, the drive pulses 17 to 22 sent out from the pulse circuits 13 and 14 are scanned from the starting point of one screen, and the strobe firing signal 10 is The image sensor 1 is exposed to light by
The image information stored in 2 can be continuously read out during one frame period after exposure without being divided into two screens as described above.

Next, the operation of the initialization circuit 152 will be explained in more detail. FIG. 5 is a diagram for explaining the operation of initializing the image sensor 12 by the circuit 152.
The configuration of a to z in the figure is the same as that in FIG. 3. However, in this figure, the photodiodes 27 selected by the vertical and horizontal shift register output terminals 25 and 26, which are respectively "1", are shown as "1", and those that are not selected are shown as blank.

After inputting the position detection signal 9, the initialization circuit 152 sends out the drive pulse switching signal 31 to
Each initialization drive pulse 17a to 22a of the circuit 152
is added to the image sensor 12. 17a,
18a is an initialization horizontal shift clock, which may be common to horizontal shift clocks 17 and 18. 19a is an initialization horizontal input signal, first set the signal 19a to "1"
Initialize horizontal shift clock 17a while maintaining
Continue adding 18a. Therefore, all "1"s are stored in the horizontal shift register 44, and all "1s" are stored in the horizontal shift register output terminals 26. Next, the initialization vertical input signal 22a is applied as "1" during one clock's worth of initialization vertical shift clocks 20a and 21a. This state is shown in Figure 5a.
All of the photodiodes 27 in the row are read (“1”), that is, are in a charging state. Next, the initialization vertical input signal 22a is set to "0" and one clock is added to the initialization vertical shift clocks 20a and 21a. This state is shown in FIG. 5b, where all the photodiodes 27 in the second row are in the read ("1") state. Thereafter, b...y and all photodiodes 27 are charged in the same manner. Next, initialize horizontal input signal 19
a is set to "0", and the initialization horizontal shift clocks 17a and 18a are applied until all "1"s in the horizontal shift register 44 are swept out, all photodiodes 27 are placed in a non-reading state, and the initialization of the image sensor 12 is completed. (Figure z). Thereafter, as described above, the horizontal and vertical drive pulse circuits 13 and 14 are initialized and exposed, and then image signals are read out by normal scanning.

According to this second embodiment, the delay time τ (not shown) from the input point of the position detection signal 9 to the exposure is the time to set all the horizontal shift register output terminals 26 to "1", and the vertical shift register output terminal 25 This is the sum of the time for sequentially setting "1" to "1" and the time for setting all the horizontal shift register output terminals 26 to "0" again, and is constant. In normal scanning, one horizontal scan is completed (horizontal shift register outputs 26a, 26
The vertical shift register 45 is shifted by one step by the vertical shift clocks 20 and 21 each time the shift is completed by sequentially setting "1" to "1".
Since a can be output continuously, the delay time τ can be shortened. If the initialization horizontal shift clocks 17a, 18a are set to the commercial frequency, and the shift clock of the vertical shift register 45 (initialization vertical shift clocks 20a, 21a are set to the same frequency as the horizontal shift clocks 17a, 18a), if the number of vertical and horizontal pixels is approximately the same, there will be a delay. The time τ is about 3 horizontal scanning times (200 μsec), which can be significantly shortened compared to 16.7 msec in the first embodiment.

Therefore, even if there are fluctuations in the moving speed of the target object, it is possible to capture images with higher positional accuracy on the screen. Note that when the frequencies of the horizontal and vertical shift clocks are the same, the horizontal and vertical shift register output terminals 26,
A method in which all 25 are set to "1" at the same time and then all set to "0" without shifting can reduce the time required for initialization to about 2 horizontal scanning hours, and can further reduce the delay time τ. .

Therefore, not only when the moving speed of the target object is constant, but also when the moving speed changes greatly, since the time from the input of the position detection signal to the exposure is short, the distance the object moves during that time is small and the object can be positioned easily. Less is. For example, if the movement speed is 0~
If the speed is 2m/sec, the moving distance during 200μsec is 0~
It is 0.4mm, which is a very small value.

In this method, it is not always necessary to put "1" into the horizontal and vertical shift registers 44 and 45 to charge all photodiodes in advance, and the influence of external light other than strobe light, dark current afterimages, etc. If the deterioration in the quality of the image signal caused by this can be ignored, the same effect can be obtained by simply shifting the maximum number of clocks required to sweep out the "1"s remaining in the shift register from the previous scan.

Embodiment (3) FIG. 6 is a block diagram showing the configuration of a third embodiment of the main part of the present invention, in which the initialization circuit 153 is different from the second embodiment, and the initialization of the image sensor 12 is Each drive pulse for each is different. In addition, this initialization drive pulse (initialization horizontal/vertical shift clocks 17b, 18b and initialization horizontal/vertical input signal 1)
9b) is commonly used as an initialization horizontal drive pulse and an initialization vertical drive pulse.

In the third embodiment, unlike the first and second embodiments, the image sensor 12 is always initialized without the position detection signal 9 being input. That is, the initialization circuit 153 always sends the drive pulse switching signal 31 to the horizontal and vertical drive pulse switching circuits 29 and 30, and uses the initialization drive pulses 17b to 19b from the initialization circuit 153 to horizontally and vertically shift the image sensor 12. It is added to registers 44 and 45.

When the initialization circuit 153 receives the position detection signal 9, it sends out different initialization drive pulses 17b to 19b for a certain period of time. Thereafter, the initialization circuit 153 sends out a strobe firing signal 10 to cause the image sensor 12 to perform exposure, and horizontally and
Reset signal 1 to vertical drive pulse circuits 13 and 14
Send 6.

Furthermore, the initialization circuit 153 eliminates the drive pulse switching signal 31 until the one vertical scan end signal 23 is sent from the vertical drive pulse circuit 14, so that the normal raster scan for the image sensor 12 is changed to the starting point of one screen. It is executed more. When the scanning of one screen, that is, the reading of the image information of one screen is completed, the drive pulse switching signal 31 is sent from the initialization circuit 153, and the initialization drive pulses 17b to 19b from the initialization circuit 153 are applied to the image sensor 12 again. .

FIG. 7 is a diagram illustrating the operation of this initial circuit 153, and is similar to FIG. 5. In this embodiment, when the position detection signal 9 is not input, the serial input signals of the horizontal and vertical shift registers 44 and 45 are always sent to the image sensor 12 via the horizontal and vertical drive pulse switching circuits 29 and 30, that is, as shown in FIG. The common initialization horizontal/vertical serial input signal 19b corresponding to the horizontal/vertical input signals 19, 22 is set to "1", and the horizontal/vertical shift clock 1 in FIG.
A common initialization horizontal/vertical shift clock 17b, 18b corresponding to 7, 18, 20, 21 is given to each shift register 44, 45, and all of the horizontal and vertical shift register output terminals 26, 25 are set to "1". As a result, all photodiodes 27 are kept in a charged state (FIG. 7a). When the position detection signal 9 is input, the common serial input signal (initialization horizontal/vertical input signal 19b) to the horizontal and vertical shift registers 44, 45 is set to "0", and the common shift clock (initialization horizontal/vertical input signal 19b) is set to "0". Vertical shift clocks 17b and 18b are input, and in order, the display shifts to Fig. 7a,
All the photodiodes 27 are set to a non-selected state as in b...m...z.

After this, exposure is performed, and image information is then read out by standard scanning.

With this method, image signals cannot be obtained all the time, so it is necessary to switch to standard scanning for focusing, etc., but the delay time τ from the input of the position detection signal 9 to exposure can be further reduced to about 1 horizontal It can be the time of scanning.

Furthermore, it goes without saying that the delay time .tau. can be shortened by making the clock frequency higher than the standard clock frequency.

〔Effect of the invention〕

As is clear from the above explanation, according to the present invention, after the position detection signal is input, the TV at that point is immediately detected.
In order to initialize the memory data etc. that give the scanning position in the camera and perform instantaneous exposure with a certain delay time,
Even when capturing images using the position detection signal of the target object that is input asynchronously with the synchronization signal of the TV camera,
As long as the moving speed of the target object is constant, an image of the target object can be captured within the screen with high positional accuracy.

Furthermore, by reducing the delay time from inputting the position detection signal to performing instantaneous exposure, it is possible to capture an image with high positional accuracy even when the moving speed of the target object varies. Therefore, there is no variation in the illumination distribution due to positional errors, and the detection accuracy of the device can be maximized.

In this way, it is possible to eliminate the fatal drawback that the object protrudes from the field of view and cannot be captured as an image, and even if the object is magnified and imaged, it does not protrude from the field of view, so the image sensor, etc. Highly accurate automatic visual inspection using resolution effectively becomes possible.

Furthermore, there is no need to perform a large amount of positional deviation correction calculations in a pattern recognition calculation section of an automatic visual inspection device, etc., and costs can be reduced.

There is a certain effect.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of the first embodiment of the present invention, FIG. 2 is a time chart explaining the operation of FIG. 1, and FIG. 3 is an explanation of the initialization operation in FIG. 1. 4 is a block diagram showing the configuration of the second embodiment of the present invention, FIG. 5 is a diagram explaining the initialization operation of FIG. 4, and FIG. 6 is a block diagram showing the configuration of the second embodiment of the present invention. A block diagram showing the configuration as an example, FIG. 7 is a diagram explaining the initialization operation of FIG. FIG. 10 is a time chart illustrating the operation of a conventional image input device. 1... Strobe ignition device, 2... Signal processing device, 3... Image input device (TV camera), 4...
Strobe light source, 5...Position detection device, 6...Belt conveyor, 7...Target object, 8...Vertical synchronization signal, 8a1...Vertical blanking period, τ...Delay time,
9...Position detection signal, 10...Strobe firing signal, 11...Image reading signal, 12...MOS type 2
dimensional solid-state image sensor (image sensor), 13...horizontal drive pulse circuit, 14...vertical drive pulse circuit, 1
51, 152, 153...Initialization circuit, 16...
Reset signal, 17, 18...Horizontal shift clock, 17a, 18a...Initialization horizontal shift clock, 17b, 18b...Initialization horizontal/vertical shift clock, 19...Horizontal input signal, 19a...Initialization horizontal input signal, 19b ...Initialization horizontal/vertical input signal, 20, 21...Vertical shift clock,
20a, 21a...initialization vertical shift clock,
22... Vertical input signal, 22a... Initialization vertical input signal, 23... 1 vertical scanning end signal, 24...
Image signal, 25, 25a, 25b... Vertical shift register output terminal, 26, 26a, 26b... Horizontal shift register output terminal, 27, 27aa, 2
7ba..., 27ab, 27bb...,...Photodiode, 29...Horizontal drive pulse switching circuit, 3
0... Vertical drive pulse switching circuit, 31... Drive pulse switching signal, 44... Horizontal shift register, 4
5...Vertical shift register.

Claims (1)

[Scope of Claims] 1. A position detection signal for detecting that a moving object to be imaged has reached a predetermined position is input, and an exposure signal is applied to a means for instantaneously exposing the object to be imaged to perform the exposure. , an image input device that obtains an image signal by raster scanning an image of the object to be imaged obtained by the exposure through an imaging means from a starting point of screen scanning through a scanning means, scanning position memory updating means for inputting and clearing memory contents giving the position of the current scanning point in the scanning means, and updating the memory contents to the memory contents of the position of the starting point; An image input device comprising: means for outputting the exposure signal after a predetermined time has elapsed from the time when the detection signal is input; and means for causing the scanning means to start scanning after outputting the exposure signal. 2. The image input device according to claim 1, wherein the predetermined time is shorter than the time required to finish scanning one screen.