JPH042926B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an automatic focusing device suitable for use in, for example, a television camera.

For example, as a conventional automatic focusing method,
The CdS blur detection method and the ultrasonic echo method have been proposed, but the former has moving parts in the detection mechanism and is noisy, and the latter cannot be used when passing through glass, and it also produces smoke and holographic images. It also had drawbacks such as not being able to be used.

In view of the above, the present invention provides an automatic focusing device that has no moving parts in its detection mechanism, has a simple structure, can be used under various conditions, and has an excellent focusing function.

An embodiment of the present invention will be described in detail below with reference to FIGS. 1 to 8.

FIG. 1 shows the basic principle when an automatic focusing device according to the present invention is applied to a television camera. FIG. 1A shows a side view thereof, and FIG. 1B shows a front view thereof. In FIG. 1, 1 is a lens barrel in which a main lens is arranged, and 2 is a multi-lens consisting of a plurality of lenses arranged on the lens barrel, which is used as a focusing lens. This multi-lens 2 will be explained in detail later.

3 is a condenser lens provided behind the multi-lens 2, 4 is a reflector for reflecting the incident light transmitted from the multi-lens 2 through the condenser lens 3, 5 is an automatic aperture mechanism, and 6 is a reflector 4
7 is a reflector for reflecting the light focused by the projection lens 6; 8 is a reflector for directing the reflected light from the reflector 7 onto a target of an image pickup tube, for example, a vidicon 9; It is a right-angle prism for projecting images.

The structure of the multi-lens 2 consists of a plurality of rectangular lenses, for example, as shown in FIGS. 2A, B, and C, respectively. For example, a mask 11 having an aperture 10 four times the size of an image projected onto a vidicon target is provided corresponding to each lens. Various methods can be considered for providing the opening 10 in the mask 11. For example, the back surface of the lens 2 may be coated with a light-shielding material, leaving only a portion corresponding to the opening 10.

Further, each lens has a positional relationship such that an image is formed on the opening 10 of the mask 11 at a predetermined focal length, for example, infinity, 6 m, 3 m, 2 m, 1.5 m, and 1.2 m, and the light passing through each lens is They are separated by a separation plate 12 so that they do not influence each other. Note that to construct this multi-lens 2, a plurality of single lenses cut into rectangular shapes are used, but in principle, multiple single lenses can be used at positions where different focal lengths can be obtained. They may also be arranged in a relationship.

With this configuration, the light from the subject (not shown) is guided through the optical path to the target 9a of the bidicon as shown in FIG. Each image formed on the aperture 10 of the mask 11 corresponding to the incident light is projected onto the outside portion where the electron beam is normally scanned.

In the opening 10 of the mask 11, depending on the position of the subject, the image of the lens whose focal length matches the subject is formed, and the image of the lens whose focal length does not match the subject is not formed. As shown in FIG. 4, the image projected onto the target 9a also corresponds to the image formation state at each aperture 10, and the image of the lens with the same focal length is the clearest. Therefore, the output detected by the vidicon 9 also becomes maximum at the electrical output corresponding to the image of the lens with the correct focal length, and the high-frequency component or the maximum output component of this electrical output is processed by an electrical circuit as described later. By moving the lens barrel 1 back and forth, the focus is automatically adjusted.

Here, we will specifically explain examples such as how to project onto a target when using a 2/3'' Tri-Nicon, and how to set up a multi-lens structure when using a 14mm main camera lens. .

Normally 2/3″ tricon has a width of 1 on both sides of the target.
There is a masked part of mm, which is used to detect dark current, but since only one side of this part is sufficient, except for the mask on one side, this part is outside the effective screen of the target surface, but the electron beam is can be used as the part that has been successfully scanned, which in this case is 1 x 6.6 mm.

Now, assuming that you are using a 14mm to 84mm zoom lens and trying to focus on the center 1/4 of the screen, you will need 2.2mm (8.8/4) rather than the 8.8mm effective horizontal width of the 2/3″ Tri-Nikon. All you need to do is project at 1mm.This increases the sensitivity of this part of (2.2) ² , and in addition to the increased sensitivity due to the lack of a color filter, it increases by approximately 14.5mm.
The sensitivity increases by (2.2 ² × 3) times. Therefore, if the aperture area of the main lens of F1.8 is approximately 1600mm ² , it is 100mm ²
If a lens with an aperture of (5×20) is used, equivalent sensitivity can be obtained in this area.

On the other hand, when the images of the six lenses are placed on the target plane most effectively, the mask provided on the back of the lens becomes as shown in Figure 2, and its opening (4
x 3 mm) and the area ratio of the target surface is 1:4, the distance between each lens is 5 mm, and the image is projected onto the target as shown in FIG. Therefore, 4
A lens that makes a mm wide screen equivalent to a 2.2 mm wide screen with a 14 mm lens is 14 x 4/2.2 = 25.45 mm. Therefore, if a lens with f=26 mm is used as a unit lens of a multi-lens, the amount that each lens projects in front of the infinity lens will be as shown in the following table. In the table, x is the amount that each lens moves forward relative to the infinity lens, and D
is the distance to the imaging point, that is, the focal length.

x(mm) D(m) 0 ∝ 0.11 6.1 0.23 2.9 0.35 1.9 0.46 1.5 0.58 1.2 Multi-lens is created by stacking each lens set in this way through the separation plate 12 as shown in Figure 2. configured.

The above is for the case where the main camera lens is 14 mm, in which case each lens making up the multi-lens 2 has a resolution of approximately 34 horizontal scanning lines (300/8.8) and approximately 26 lines per field (525/8.8). 2×1/6×3/
5) There is a scanning line that can be used to detect . Also, when the main camera lens is at the 6x end, the resolution is reduced to about 5 horizontal scanning lines (34/6) and about 4 scanning lines per field (26/6), which can be used for detection, but it is difficult to judge distance. You can get enough output without any problems.

FIG. 5 shows an embodiment of the present invention, in which 20 is a main lens, 21 is a focusing lens barrel on which the main lens 20 is arranged, 22 is a gear that engages with the lens barrel 21, 23 is linked with the lens barrel 21, and the lens barrel 2
1 is a potentiometer for varying the setting value according to each predetermined focal length to be moved; 24 is a gear 2;
This is a focusing motor that can be rotated in forward and reverse directions. By rotating the motor 24 forward and backward as necessary, the lens barrel 21 is moved back and forth to adjust the focus.

25 is a zoom lens barrel (not shown) in which a zoom lens is arranged, 26 is a gear that engages with the lens barrel 25, and 27
is a zoom potentiometer that is linked to the lens barrel 25, and this potentiometer 27 adjusts the sampling hold time width, which will be described later, according to the rotation angle of the lens barrel 25.

28 is a focusing lens using a multi-lens as described above, 29 is a reflecting mirror, and the light incident through the focusing lens 28 for focus adjustment is reflected by the reflecting mirror 2.
9 and is introduced into the vidicon 30 via a projection lens, a reflecting mirror, and a right-angle prism (not shown) as shown in FIG. Incident light for imaging is also introduced into the vidicon 30 through the main lens 20. Therefore, a photoelectrically converted video signal S _V and a focus signal S _F as shown in FIG. 6 are outputted to the output side of the vidicon 30.

Note that FIG. 6 schematically shows each signal included in 1H, and the focal signal is outside the effective screen of the vidicon 30 in the range obtained by normal scanning of the electron beam, for example, in the horizontal blanking period or vertical blanking period. It is preferable to insert it in the corresponding part,
In this embodiment, a case is shown in which the insertion is performed within the horizontal blanking period. Further, in FIG. 6, _HD is a horizontal synchronizing signal, and S _G is a gate signal to be described later, and only the focus signal S _F is derived from this gate signal S _G.

31 is an amplifier for amplifying the imaging output from the vidicon 30, that is, a video signal; 32 is a mixing circuit for adding a composite synchronization signal supplied from the synchronization signal generator 34 to the video signal; The output is sent out from the output terminal 33 as a composite video signal.

35 is an amplifier; 36 is a switching circuit for removing this video signal from the video signal containing the focus signal supplied from the amplifier 35 using a switching signal obtained by inverting the blanking signal from the synchronization signal generator 34; A counter 38 for counting horizontal scanning lines in response to the horizontal synchronization signal from the synchronization signal generator 34 removes noise included in the synchronization part from the output from the switch circuit 36 in response to the output from the counter 37. A gate circuit 39 is for deriving only the focal signal, and 39 is a high frequency device for extracting only high frequency components from the output of the gate circuit 38. As this high frequency device, for example, a differential circuit is used. . The reason why the high-frequency filter 39 is used here is that the high-frequency component increases at the focused point.
It takes advantage of its large differential output and high contrast. 40 is an integration circuit for removing high-frequency noise components from the output of the high-frequency filter 39 and improving the S/N ratio of the focal signal; 41 is a peak detection circuit for the output of the integration circuit 40 to obtain a DC signal; 42 is a waveform width generator for varying the width of the sampling pulse according to the setting value of the zoom potentiometer 27 and the output of the counter 37. The field of view changes depending on whether the main lens 20 is a zoom lens (telephoto or wide-angle), but since the focusing lens 28 is not a zoom lens, the viewing angle is fixed. The viewing angle is 1/3 to 1/4. Therefore, the same effect can be obtained by adjusting the sampling and holding time width using the waveform width generator 42.

43, the output of the integrating circuit 40 is connected to the waveform width generator 42
44 is a comparator for comparing the output of the peak detector 41 and the output of the sampling hold circuit 43, and this comparator 41 is a peak detector. 41
At the point where the output of the sampling hold circuit 43 becomes larger than the output of the sampling hold circuit 43, the output level changes from logic "1" to "0", for example. That is, comparator 4
1 generates a pulse output at the maximum output point of each focus signal. A monostable multivibrator 45 generates a signal that rises in response to the vertical synchronization signal from the synchronization signal generator 34 and falls in response to the output of the comparator 41. Therefore, monostable multivibrator 45
The falling point of the output of represents the best focus position. Reference numeral 46 denotes a mirror integration circuit for generating a signal having a slope portion in a predetermined range from the point where the output of the comparator 41 occurs, that is, a portion corresponding to the focal length range to which the output belongs.

47 is a monostable multivibrator for generating a signal that rises in response to the vertical synchronization signal from the synchronization signal generator 34 and falls with a time constant set by the potentiometer 23; this monostable multivibrator 47 The time constant is set by a potentiometer 23 interlocked with the lens barrel 21 so as to vary according to a preset focal length. In other words, to change the focal length from 0 m to infinity, the time constant of the monostable multivibrator 47 must be 16.6.
mS (1V) to 0mS, and divide this 1V period into 9 sections as shown in Figure 7, respectively, 1m, 1.5m, 2m, 3m, 5m, 10mS.
It corresponds to the focal lengths m and ∝, and the output of the monostable multivibrator 47 is obtained according to each of these focal lengths. Therefore, the falling point of the output of the monostable multivibrator 47 represents the current position of the lens barrel 21, that is, the focal length.

48 is a differentiating circuit for generating a narrow pulse in response to the output of the monostable multivibrator 47; 49 is a comparator for comparing the output of the integrating circuit 46 and the output of the differentiating circuit 48; This comparator 49 generates a negative signal when the pulse from the differentiating circuit 48 is within the logic "0" level range of the output of the integrating circuit 46, and generates a positive signal when the pulse is within the logic "1" level range. occurs, and logic “1”
When the output signal is in the transition range from "0" to "0", that is, in the slope portion, the output signal is set to zero. Reference numeral 50 denotes an integrator that integrates the output of the comparator 49 and drives and controls the focusing motor 24 via an amplifier 51 in accordance with the output level of the comparator 49. This integrator 50 also serves as a phase correction circuit to prevent hunting in the servo loop.

Reference numeral 52 denotes an aperture control circuit for generating an aperture control signal in response to the output of the peak detector 41, which drives and controls the aperture motor 53 according to the amount of incident light, and adjusts focus signal detection sensitivity by blooming of the image pickup tube. This prevents deterioration of the S/N ratio of the focal signal component when the image becomes dark. This focal signal has almost no error due to shedding, and since the difference in frequency components is important,
As shown in FIG. 5, it is sufficient for the diaphragm plate to have a structure in which one plate 54 moves either vertically or horizontally in the incident light path for focusing.

55 is a delay circuit for delaying the output of the monostable multivibrator 45 related to the focus signal by a predetermined time so that the output falls within the effective screen; 56 is a switch 5;
A summing circuit for adding the output supplied from the delay circuit 55 via 7 and the composite video signal from the mixing circuit 32, 58 an amplifier, and 59 a view finder, which are focused by such a configuration. The position can be displayed on the monitor screen.

Next, the operation of the automatic focusing device according to the present invention will be explained in the sixth section.
This will be explained with reference to FIGS. 8 through 8.

The focus adjustment light from the subject that enters through the focusing lens 28 is converted into an electric signal by the vidicon 30 together with the photographing light that enters through the main lens 20, and the video signal S _V is converted as shown in FIG. Focus signal S _F inserted during horizontal blanking period
It is expressed as When these signals are viewed as a 1V waveform, they are represented as a signal _S1 as shown in FIG. 7A, where _VD is a vertical synchronizing signal.

The signal S ₁ thus converted into an electric signal by the vidicon 30 and amplified by the amplifier 31 is added with a composite synchronizing signal by the mixing circuit 32 and sent out from the output terminal 33 as a composite video signal.

On the other hand, the signal _S1 is supplied to a switch circuit 36 through an amplifier 35, where the video signal component is removed. Switch circuit 36 from which video signal components have been removed
The output of the counter 37 is supplied to the gate circuit 38, which receives the gate signal S _G from the counter 37 as shown in FIG.
As a result, only the focal signal component _S2 as shown in FIG. 7B is derived, and the noise component included in the synchronous part is removed. The output of the gate circuit 38 is a high frequency amplifier 39
A signal _S3 containing only high-frequency components as shown in FIG. 7C is derived. The signal S ₃ is further integrated by an integrating circuit 40 in the next stage to remove high-frequency noise components, and is supplied to a peak detector 41 and a sampling hold circuit 43 as a signal S ₄ as shown in FIG. 7D, respectively.

The peak detector 41 performs peak detection on the signal S ₄ and generates on its output side a signal S ₅ as shown in FIG. 7E, in which the maximum point of the signal S ₄ is peak-held. Further, the sampling and holding circuit 43 samples and holds the signal _S4 using the sampling pulse from the waveform width generator 42, and generates a step-like signal _S6 as shown in FIG. 7F corresponding to the amplitude level of the signal _S4 . . Note that the sampling hold time width of the sampling hold circuit 43 is the same as that of the zoom lens barrel 25.
The waveform width generator 42 varies the width according to the position of the image frame, that is, the size of the image frame. That is, the main lens 20
While the zoom lens can change the viewing angle, the viewing angle of the focusing lens 28 is fixed, so the main lens 20
When using telephoto, the viewing angle of the central 1/3 to 1/4 of the image cannot be achieved unless the frame becomes narrower. Therefore, the setting value of the potentiometer 27 is varied according to the rotation angle of the lens barrel 25 to change the pulse width of the sampling pulse from the waveform width generator 42, thereby changing the sampling hold time width of the sampling hold circuit 43. Thus, no matter where the zoom lens barrel 25 is located, only a signal having a pulse width of 1/3 to 1/4 of the image frame and the center is used as a focus signal. FIG. 8 shows an example in which the sampling and hold time width is varied by the waveform width generator 42, and shows cases where the focal length is, for example, 14 mm and 40 mm. In FIG. 8, W ₁ is the screen vertical width of the focus signal, W ₂ is the vertical width of the detection signal, that is, the sampling pulse, A ₁ ,
If A ₂ is an image frame with a focal length of 14 mm and 40 mm, respectively, then by setting the sampling hold time widths to SH ₁ and SH ₂ when the focal length is 14 mm and 40 mm, respectively, the image frame will be fixed regardless of the position of the lens barrel. A signal with a width of 1/3 to 1/4 of the center can be used as a focal signal.

The signal S ₆ from the sampling hold circuit 43 and the signal S ₅ from the peak detector 41 thus obtained are supplied to a comparator 44 and compared. Then, at the time when the signal _S5 becomes larger than the signal _S6 , that is, at the time when the focus signal component becomes maximum, the comparator 44 generates a signal _S7 as shown in FIG. 7G. This signal _S7 is supplied to the monostable multivibrator 45, and as shown in FIG.
The monostable multivibrator 45, which is in the energized state, is deenergized by the vertical synchronizing signal V _D from the multivibrator 4, and the signal S ₈ is generated on its output side. The falling point of this signal _S8 represents the focal length of the lens that is focused on the subject, and FIG. 7 shows that the focal length is 2 m.

The signal S ₈ from the monostable multivibrator 45 is integrated by an integrating circuit 46 and waveform-shaped into a signal S ₉ having a slope portion at a portion corresponding to the focused focal length as shown in FIG. 7I. The signal is supplied to a comparator 49, where it is compared with a signal representing the current position of the focusing lens barrel 21. That is, the monostable multivibrator 47 whose time constant is varied by the potentiometer 23 according to the movement of the lens barrel 21 rises in response to the vertical synchronization signal V _D from the synchronization signal generator 34, and corresponds to the moving distance of the lens barrel 21. A signal _S10 as shown in FIG. 7J, which falls with a time constant, is generated. Therefore, the falling point of this signal _S10 is at lens barrel 2.
1, and FIG. 7 shows that the lens barrel 21 is at a focal length of 5 m.

The signal _S10 is supplied to the differentiating circuit 48, which generates a narrow pulse signal _S11 as shown in FIG.
9'', where the signal from the integrating circuit 46
Compare with S ₉ . The comparator 49 changes the signal S ₁₁ to the signal S ₉
Since the level is within the logic "0" level range, _a negative signal _S12 as shown in FIG. It is converted into a DC signal _S13 . This negative DC signal _S13 is amplified by the amplifier 51 to a level sufficient to drive the motor 24, and this amplified drive signal causes the motor 24 to rotate the focusing lens barrel 21 to a focal length of 2 m via the gear 22. move it. Accordingly, the output _of the monostable multivibrator 47 also changes, and as shown in _FIG . As shown in FIG. 7S, the signal from the differentiating circuit 48
When S ₁₁ enters the slope portion of the signal S ₉ from the integrating circuit 46, the output of the comparator 49 becomes zero as shown in FIG. 7T, and the output of the integrating circuit 50 also becomes zero as shown in FIG. 7U. As a result, the drive of the motor 24 is suppressed, and the lens barrel 21 is stopped at a focused position, that is, at a focal length of 2 m.

On the other hand, the current position of the focusing lens barrel 21 is focal length 1
m, the output side of the monostable multivibrator 47 has a seventh lens corresponding to its focal length.
A signal _S10 as shown in Figure N is generated. Corresponding to this signal _S10 , a pulse signal _S11 as shown in FIG. 7O is generated on the output side of the differentiating circuit 48.
This signal S ₁₁ is the logic “1” output of the integrating circuit 46.
Since the level is within the range, a positive signal S ₁₂ as shown in FIG. 7P is generated at the output side of 49.
This signal S ₁₂ is integrated by an integrating circuit 50, converted into a positive DC signal S ₁₃ as shown in FIG. 7Q, amplified by an amplifier 51, and supplied to the motor 24 as a drive signal. Then, the motor 24 rotates the focusing lens barrel 21 in the opposite direction, and stops the lens barrel 21 at a focal length of 2 m in the same manner as described above, thereby completing the focus adjustment.

The result of focus adjustment in this manner can be visually confirmed by closing the switch 57 and looking through the viewfinder 59.

As described above, according to the automatic focusing device according to the present invention, image blurring on the focal plane of the same image is focused as focus information outside the effective screen of the image pickup tube using multiple lenses, so a special sensor is not required. Focus control is possible with a simple structure without the need for

Furthermore, since there are no moving parts in the detection mechanism, there is no noise unlike conventional devices, and the present device can also be used for smoke and holographic images, providing versatility.

In the above-described embodiments, the present device is applied to a television camera, but the present invention is not limited thereto, and can be similarly applied to, for example, an electro-optical camera.

Furthermore, although the above embodiments have been described using a vidicon as a detection device, other detection devices such as a CCD may also be used.

[Brief explanation of drawings]

FIG. 1 is a side view and front view showing the basic principle of this invention, FIG. 2 is a front view, side view, and rear view showing the structure of a multi-lens according to this invention, and FIG.
5 and 4 are schematic diagrams for explaining the operation of FIG. 1, FIG. 5 is a circuit configuration diagram showing an embodiment of the present invention, and FIGS. FIG. 3 is a signal waveform diagram for explaining the operation. 2 is a multi-lens, 4 and 7 are reflecting mirrors, 6 is a projection lens, 8 is a right angle prism, 9 and 30 are vidicon, 20 is a main lens, 21 is a focusing lens barrel, and 28 is a focusing lens.

Claims (1)

[Claims] 1. A main lens whose focus is adjustable; a main lens moving means for moving the main lens to adjust the focus; and a unit onto which a subject image obtained through the main lens is projected. In an automatic focusing device for a television camera having an image sensor, a plurality of fixed focus lenses having different focal lengths are provided near the main lens, and a plurality of subject images are obtained through the plurality of fixed focus lenses. are configured to project the images directly onto a portion outside the effective screen of the image sensor at respective predetermined positions without going through the main lens, and the fixed focus lens included in the output signal of the image sensor A circuit is provided for detecting a focus information signal having the most high frequency components from among focus information signals corresponding to a plurality of subject images projected by the 1, and the focus information signal having the most high frequency components detected by the circuit is provided. An automatic focusing device for a television camera configured to adjust the focus of the main lens by driving the main lens moving means in correspondence with two fixed focus lenses.