JPS58202671A - Auto focus method - Google Patents

Abstract

PURPOSE:To attain a proper focusing function regardless of the F value, etc. of an image pickup optical system, by using plural image contrast signals obtained from plural image pickup elements, applying a logical to these image contrast signals to obtain a focusing control signal and controllng the lens position with the focusing control signal. CONSTITUTION:A subject image is formed on an image pickup element 21 through an image pickup lens 11 and at the same time projected to two image pickup elements 31 and 41 via a beam splitter 12. The high-pass filters (HPFs) 22, 32 and 42 are connected to three image pickup elements 21, 31 and 41 respectively. The high frequency component amplitude of a video signal varies in response to the image contrast, and each amplitude has the maximum value when the correct focusing is obtained to each image pickup surface. The outputs of the HPFs are led to wave detectors 24, 34 and 44 via gates 23, 33 and 43 respectively, and outputs R, P and Q are obtained in response to maximum values. These outputs are used to operate a controller 50 and then to control the lens 11.

Description

[Detailed description of the invention] The present invention is mainly used as a television camera.

The present invention relates to an automatic focusing method for a camera having a photoelectric conversion image sensor. Here, automatic focusing refers to an optical adjustment function that automatically focuses an imaging optical system made up of lenses and reflecting mirrors without relying on the human eye.

Conventionally known techniques for automatic focusing means are basically classified as follows.

Fal Image sharpness detection type: Detects the sharpness of an image formed on the imaging surface or a position optically equivalent thereto.

The imaging optical system is adjusted based on this detection signal. The contrast of the video signal or the amplitude of the high frequency component of the video signal is used as a signal representing the sharpness of the image.

(b) Distance measurement type: The distance from the imaging optical system to the subject is detected by triangulation or other means, and the imaging optical system is adjusted based on this distance signal.

Of the above two methods, two of the methods employed in the apparatus of the present invention are the image sharpness detection type. The reason for this is (]) Since focusing means obtaining a clear image, it is most rational to detect the image clarity. (2) In the optical distance measuring type, the distance measuring accuracy may decrease for objects with a periodically repeating pattern. etc.

FIG. 1 shows an example of a conventionally known image clarity detection type television camera focusing device. Depending on the imaging lens 11, an image of the subject is formed on an imaging tube "□" or an imaging surface 21 (hereinafter referred to as an imaging device) of a solid-state imaging plate.

The video signal output from the image sensor 21 is passed through a high-pass filter 22 that passes only high-frequency signal components. The signal is inputted to the drive power source 29 via the detector 27° and the signal processor 28.

The drive mechanism 61 moves the lens 11 according to the output.

The operating principle of this device is as follows. That is, when the image is properly focused, the image clarity is maximized, and therefore the high frequency components of the video signal are maximized. Therefore, the high-pass filter 22
The output of the lens 11 is passed through a detector 27 to measure its amplitude, and the signal processor 28 judges whether the amplitude increases or decreases when the lens 11 is moved.
If you move it, it will focus.

The disadvantage of this method is that when the position of the lens 11 is significantly far from the correct in-focus position, or in other words, in a state of severe defocus, firstly the amplitude of the high-frequency components of the video signal becomes small, and secondly the amplitude of the high-frequency component of the video signal becomes small. When the lens 11 is moved a little, the amount of change in its amplitude also becomes small, and as a result, it becomes impossible to supply a defocus signal to the signal processor section. The second problem also occurs when the F number of the lens is large, that is, when the depth of focus is large. Additionally, with this method, if the subject or camera moves quickly, the focusing action may not be able to follow it.

The present invention obviates the above-mentioned drawbacks in the prior art.

It is an object of the present invention to provide a focusing device that can perform an appropriate focusing function regardless of the F value or defocus amount of an imaging optical system, and can quickly follow the movement of a subject or a camera.

In order to achieve this objective, the following novel principle is used in the focusing method of the present invention. That is, using a plurality of image sharpness signals obtained from a main image sensor and a single image sensor or a plurality of image sensors disposed at different positions on the optical path of the imaging optical system, and subjecting the signals to a certain logical judgment process. A focus control signal is generated by the lens, and the lens position is controlled by this signal.

FIG. 2 shows an embodiment of the present invention. A subject image is formed by the imaging lens 11 onto the imaging device 21, and is also transmitted to the two imaging devices 31 and 4 via the beam splitter 12.
It is also projected onto 1. The amount of light incident on these image sensors is adjusted by the aperture 13 as necessary. The image sensor 21 is the main image sensor of a television camera or the like, and a part of the two-dimensional video signal generated therefrom is also used as an autofocus signal. In addition, two image sensors 31 and 4
Reference numeral 1 denotes a sub-imaging element for generating an automatic focusing signal, and a one-dimensional imaging element having a large number of minute solid-state photoelectric detection elements arranged in a straight line is useful for this purpose. When the subject is correctly focused on the main image sensor 21, the sub image sensor 31 is positioned closer to the lens 11 by a certain optical path length (hereinafter referred to as the front position);
It is located at a position far from the lens (hereinafter referred to as the rear position) by approximately the same optical path length as in the case of .

These three image sensors 21.31.41 are scanned synchronously. That is, when the main image sensor 21 scans one specific horizontal scanning line, the sub image sensors 31 and 41
The three video signals obtained at that time are adjusted to correspond to the same part of the object.

High-pass filters 22, 32, and 42 are connected to the three image sensors 21, 31, 4, respectively. The high frequency component amplitude of the video signal changes according to the image clarity, and each amplitude reaches its maximum value when each imaging plane is correctly focused. However, high frequency switching noise generated when driving the solid-state image sensor is removed by a filter or gate.

The outputs of the high-pass filters 22, 32 and 42 have a gain of −42
3° 33 and 43. This gate is used to extract only a specific part of one horizontal scanning line, and this allows the screen to be restricted horizontally and focus only on a specific subject within the nine screens. Become. Gate 1-23. The outputs of 33 and 43 are sent to the detector 24.
．． 34 and 44. The outputs of the detectors 24, 34 and 44 are signals corresponding to the peak value of the high frequency component amplitude of the video signal. The main image sensor 21 is a two-dimensional image sensor, and its peak value is output from the detector 24 for each horizontal scanning line. Only the book is extracted through the gate 25.

Here, the peak value obtained from the main image sensor 21 is P.

It is assumed that the peak values obtained from the sub-imaging elements 31 and 41 are Q and R. These signals are present at the input terminals 'rp, 'rq and Tr.
The signal is input to the controller 50 via. The output of the controller 50 appears at the terminal W, and its voltage is represented by W. This signal W drives the lens 1 through the drive mechanism 61. The position of the lens 11 is detected by a position detector 62 as required, and is input to the controller 50 via an input terminal Ts.

Note that the lens drive can be stopped using the manual switch 71 if necessary.

Next, the configuration and operation of the controller 50 will be explained. Wave height P,
Q and R are standardized as follows.

That is, the value of P when the subject is focused on the image sensor 21 is Po, the value of Q when the subject is focused on the image sensor 31 is Qo,
When the value of R when focused on the image sensor 41 is RO, it is adjusted so that Qo=Ro and Po=Qo. P, Q, and R are converted by the arithmetic operation unit 51 into the quantities r(P, Q, R) yo(P-Q)20(P-R)2, and then by the comparator 52 according to the discrimination condition F. (P, Q, R)=(P-Q)2+(p-R,)2:)
It is determined whether a (1) is satisfied. Here, a is a predetermined reference voltage, the value of which is greater than the noise level contained in the signals P, Q, and R, and
.

It is set to be smaller than QO and RO.

However, since the in-focus peak values Po, Qo and Ro change depending on the contrast of the subject, a relatively low contrast is desirable. Further, the value of a may be varied depending on the amplitude of the video signal.

The output X of the comparator 52 takes a high voltage level (hereinafter referred to as H) when the condition of equation (1) is satisfied; otherwise,
That is, when F(P,'Q,R)≦a, a low voltage level (hereinafter expressed as below) is assumed. This output X is the logical operation unit 5
3 is input.

On the other hand, Q and R are input to the subtracter 54, and the difference signal Q-
R is output. The signal Q-R is input to the comparator 55, and the discrimination condition Δ(Q, R) 21 Q-R1>b
It is determined whether (2) is satisfied. Here, b is a predetermined reference voltage, which is set to be larger than the noise contained in the signals Q and R. Since this value is much smaller than the peak values Qo and Ro, the condition of equation (2) is effectively Q\R
In other words, this is equivalent to determining that there is a difference between Q and R.

The output Y of the comparator 55 becomes H when the discrimination condition of equation (2) is satisfied, and conversely becomes L when (Q, R.)≦b, that is, Q=,R. is input.

Table 1 shows the relationship between the inputs X and Y of the logical operation unit 53 and the output Z thereto.

Table 1. As is clear from Table 1, the relationship between the input and output of the logical operation unit 53, only in the case of state ■, that is, F(P
, Q, R)=(P-Q)2+(P-R)2≦a,
and Z = only when Δ(Q, R)≦b, that is, Q−=R.
In other states r, n, and m, Z=H.

Further, the output Q-R of the subtracter 54 is output from the focus control signal generator 56.
is also input to generate a single focus control signal U=f(QR). The simplest form of the function f (Q-R) is the proportional relationship U=c (Q-R) (3)
It is. Here, C is a proportionality coefficient, and its value is adjusted so that the output voltage U becomes an appropriate value for driving the drive mechanism 61.

Note that in practice, in order to prevent the output voltage U from becoming an excessive value, the focus control signal generator 56 is provided with a saturation characteristic, and the IQ
In a range where -R1 is relatively small, the output U is proportional to Q-R according to equation (3), but as IQ-1al increases, it is preferable that the output saturates and becomes U-constant. The sign of U matches the sign of QR, and when U>0, the lens 11 is retracted and approaches the image pickup device 21, and when U<Oa, the lens 11 is extended and similarly moves away. If U=0, it stands still.

The output Z2 of the logical operator 53 and the output U of the focus control signal generator 56 are input to the changeover switch 57. In addition, the output (2) of the search signal generator 58 is also input to the changeover switch 57.

The search signal ■ is a constant positive or negative voltage.

Its polarity is controlled by a signal S generated by lens position sensor 62.

The changeover switch 57 operates as follows. In other words, the connection between the terminals changes depending on the signal level of input Z, and Z=H.
When , the output terminal Tw is connected to the input terminal of the focus control signal U, and as a result, W=U.

When Z=L, Tw is connected to the input terminal of the search signal V, and W=V.

Next, the operation of the focusing device of the present invention will be explained. First, the lens 11, its imaging point, and the three image sensors 21.31
The isometric positional relationship of and 41 is shown in FIG. When the lens 11 is extended to the front and reaches the position 11A in the figure, an object at close range will be imaged on the image sensor 21, and an object at infinity will be imaged at point M in the figure? 6° If the lens 11 moves to position IIB in the figure, the point at infinity will be on the image sensor 21.

The closest point 4 is imaged at point N in the figure. Therefore, regardless of the position of the lens 11, all objects ranging from close to infinity are generally imaged in the area from point M to N.

Assume now that the subject is correctly focused on the image sensor 21. At this time, the signal P takes a local maximum value, and Q=R(P
becomes. At this time, F(P, Q, R); (PQ)2+(P
-R)")a, that is, X=H, and Q-R=O, that is, Y=
L, and the state ■ in Table 1 is realized. Therefore,
The lens 11 remains stationary at the focused position.

Suppose that the imaging point shifts and focuses near the image sensor 31. At this time, Q > P > R (because 51).

F(P,Q,R)==(P-Q)2+(P-R)2>
a, that is, X=H, and Q=R'-<0, that is, Y=H, and the state (2) in Table 1 is realized. Therefore, Z=H, and the lens 11 moves to the right in FIG. 3 and stops after correctly focusing on the image sensor 21. Conversely, if the in-focus point deviates to the vicinity of the image sensor 41, it becomes U(O), and the focus point is returned to the correct in-focus position.

Next, suppose that the imaging point has shifted significantly to the left and has come near point M in the figure. This corresponds to the lens 11 being fully extended and the subject being at infinity.

At this time, the images produced by the three image sensors 21, 31 and 41 are all extremely blurred, and only very small high frequency components can be obtained from the two image sensors. That is, P=Q=R
,=Q.

F(P, Q, R)=(P-Q)2+(P-R)2<
a, that is, X=L, and Q-R, ==Q, that is, Y=
L Tonary 2 State ■ in Table 1 is realized. Therefore, the lens 11 is driven by the output (2) of the search signal generator 58. However, since the signal V is a constant positive or negative value, if the polarity of the signal (2) at that time is positive, the lens 11 will be pulled toward the right in FIG. 3, and will finally reach the infinity end. reach At this position position detector 6
2 is activated and sends a signal to the search signal generator 58, which inverts the polarity of its output. In response to this, the drive mechanism 61 begins to rotate in reverse, and the lens 11 is extended to the left in FIG. 3. When it reaches the close end, a signal is sent from the position sensor 62 again and the direction of lens movement is reversed. In reality, the imaging point approaches the image sensor 31 or 41 during such movement, and at this time F (P, Q, R) mi (P-Q) 20 (P
-R) 2) Since the two conditions of a, Q -R\0 are satisfied, the state T in Table 1 is realized, and Z=H, W=U
Then, automatic focusing control is activated and the lens 11 is pulled into the correct focusing position.

Here, if the contrast of the subject is low, palm 7□'
11(7) Focal point 4 degrees / 3 large aperture: '・, ν is 27
Regardless of the position, F(P, Q, R) = (P-
Q)2+(P-l(), 2.=Q, P=Q. In such cases, it always becomes W-■ and the lens 1
1 repeats forwards and backwards endlessly. To avoid this, when the lens 11 almost reaches the in-focus position, turn off the switch 71 manually to stop the lens 11.

Note that the imaging point is largely shifted from the image sensors 21, 31 and 41, resulting in P(P, Q, R)mi(P-Q)2+(P-
R) Even if 2<a, if IQ-R1>b, Z=H,
When W=U, the autofocus function is activated. This corresponds to state (■) in Table 1.

FIG. 4 shows another embodiment of the invention. In the embodiment described above, two sub-imaging elements were used.

In this embodiment, a single sub-image sensor 31A is used. Sub-imaging device 3], A is driven in the optical axis direction by a drive device 37, and is positioned at a front position 31. A and the rear position 41A. Let Q be the peak value of the high frequency component of the video signal generated at the front position, and R be the peak value of the high frequency component of the video signal generated at the rear position, these are Q and 'r in the above embodiment.
Compatible with JR. The peak value of the high frequency component of the video signal generated from the main image sensor 21 is P, as in the previous embodiment. That is, the image clarity signals obtained from the two fixed sub-imaging devices in the above embodiment can be improved by sequentially moving a single sub-imaging device to different positions on the axis of the imaging optical system. 9 is obtained as a time-series image clarity signal.

A signal obtained from the main image sensor 21 is input to an input terminal Tp of the controller 50. The signal obtained from the sub-imaging device 31A is transmitted to the changeover switch 3 which operates in synchronization with the drive device 37.
6, the signal at the front position is input to the input terminal Tq, and the signal at the rear position is input to the input terminal Tr.

The subsequent signal processing is the same as in the previous embodiment.

The following modifications can be made to the embodiments described above.

(1) Image sensor This device requires a plurality of sub image sensors that generate video signals at positions in front and behind the main image sensor. Alternatively, a single sub-imaging element may be moved spatially to generate forward signals and backward signals alternately in time. In this case, the two sub-imaging elements are arranged at different optical path lengths from the lens, but it is most preferable to arrange the main imaging element exactly in the center thereof. If arranged like this, Q=R,
When the following conditions are met, the main image sensor automatically comes to the correct focus position. If the optical position of the main image sensor is shifted from the midpoint of the two sub image sensors.

To adjust the gain of the electronic circuit connected to two sub-imaging elements so that the gain of the electronic circuit connected to the sub-imaging element which is close to the main imaging element is lowered, and conversely the gain of the electronic circuit connected to the sub-imaging element further away from the main imaging element is increased. Accordingly, it is possible to electrically achieve Q=R at the focal point of the main image sensor.

(2) Logic processor Functions shown in the above examples.

F(P, Q, R) Yo(P-Q)2+(P-R)2(
8) may be considered to represent the degree of mismatch between the signals Q and R of the two sub-image sensors and the signal P of the main image sensor. When the FQ value is large, the imaging point of the lens is at the correct focal point, that is, near the main image sensor. At this time, more precise focusing is achieved by controlling the position of the lens using the focus control signal U. Conversely, a small value of F means that the imaging position of the lens is far away from the main image sensor. At this time, the lens is first driven by the search signal V and then precisely controlled in position by the control signal U.

In this way, the position of the lens imaging point is indicated by the value of the function F(P, Q, R), and the lens drive signal is selected based on the result.

In addition to the above-mentioned function F (P, Q, R), various forms can be considered as the function that indicates the in-focus state of the lens. For example, a quadratic function may be used with A and B being arbitrary constants.

Another function is more convenient to know the lens position.

F (P, Q, R,): αp + βQ + γR (10
)1' may be used. Here α, β and γ are constants. α
Adjust the gain of the circuit so that -β-γ-1.

F(P, Q, R)=P+Q-111).

In the above embodiment, an example is shown in which a one-dimensional solid-state image sensor is used as the sub-image sensor. Various other techniques are known as image sharpness detection means, and these known techniques may be applied to the sub-imaging device. For example, a method in which a slit I or a spatial filter is provided in front of a photoelectric conversion element and the filter is scanned.

A method that utilizes the property that the resistance of the photoelectric element takes an extreme value at nine focal points due to the nonlinear response characteristic of the photoelectric element may be applied to the sub-imaging element of the present invention. Furthermore, the subject may be irradiated with light from a light emitting diode, and scattered light from the subject may be received.

A feature of the present invention is that image sharpness signals obtained from the main image sensor and the sub image sensor are used as logic inputs.

This means that the position of the lens is controlled by the output obtained by performing certain logical operations on these inputs. According to this method, the lens is always drawn into the correct focus position regardless of the distance, contrast, or movement of the subject. Further, a part of the main image sensor can also be used as the sub image sensor. This is due to the following reasons. Generally, there is some margin around the photosensitive surface of an image sensor. This extra portion has a function as an image sensor, but is not scanned, or even if scanned, does not appear on the image receiving surface as an image to be actually viewed. Since this extra portion exists on the top, bottom, left and right sides of the screen, this portion can be used as a sub-imaging device. In this case, of course, light shielding is required to prevent the main screen from being projected onto the sub-imaging element portion. Furthermore, if the part to be used is a part that is not scanned in the normal state, it is necessary to provide the function as a sub-imaging element by separately scanning that part using the retrace period.

FIG. 5 shows an embodiment in which the periphery of the main image sensor is also used as a sub-image sensor. beam brick 1
A part of the light separated by the mirror 2 passes through the sub-optical system 122 of the reflecting mirror 12L and forms an image on the first sub-imaging element section 31, which is the outer side of the main image-sensing element 21.

This sub-image sensor section 31 is optically arranged at a forward position compared to the main image sensor 21. The other part of the light separated by the beam subricter 12 passes through a reflecting mirror 123 and a sub-optical system 124, and then passes through a second part of the light separated from the sub-image sensor part 31 at the outer edge of the main image sensor 21. Sub image sensor section 41
image on top. This sub image sensor section 41 is the main image sensor 21
It is optically located at a rearward position compared to the Since the parts other than those mentioned above are the same as those of the previously described embodiments, the explanation of the operation will be omitted.

FIG. 6 shows another embodiment in which the outer peripheral portion of the main image sensor is used as a sub image sensor.

The imaging optical system in this example is a zoom lens. This zoom lens is composed of a variable magnification optical system 111 and a relay lens system 112, and a beam splitter 12 is arranged in between. The light separated by the beam splitter 12 passes through the reflective surface 21° sub optical system 122 and then passes through the sub image sensor section 31 that forms the outer periphery of the main image sensor 21.
image is formed. Mechanism 1 for moving the constituent lenses of the sub-optical system 122
25, the image formed on the sub-imaging element section 31 reciprocates between the front imaging state and the rear imaging state.

That is, in the case of this example, the control of the optical position of the sub-imaging element section is as follows:
This is done by controlling the positions of the constituent lenses of the sub optical system 122. Therefore, the position of the sub-imaging element is isometrically represented by the position of the lens.

Since the points other than those mentioned above are the same as those of the previously described embodiments, their explanations will be omitted.

It should be noted that any function other than the function P (P, Q, R) listed in the example above is included in the present invention as long as it represents a difference between P and Q or P and R. Applicable.

In the above embodiment, an example was shown in which the focusing state of the imaging optical system is controlled by moving the lens.

However, it is obvious that the image sensor side may be moved instead of moving the lens. This is because the focusing state of the optical system is determined by the relative distance between the lens and the image sensor, so movement of the lens is equivalent to movement of the image element. The mechanism for making the position of the image pickup element movable is the same in principle as the mechanism for making the lens movable or the mechanism for making the sub image pickup element movable in FIG.

As explained above, the focusing method of the present invention can be used for two types of subjects.
It is characterized by its ability to perform highly accurate focusing that is less affected by position, brightness, contrast, and movement, making it useful as a focusing device for television cameras.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an example of an automatic focusing device according to the prior art, FIG. 2 is a block diagram showing an embodiment of the present invention,
FIG. 3 is a schematic diagram of an optical system used to explain the operating principle of the apparatus of the present invention, and FIGS. 4, 5, and 6 are block diagrams showing other embodiments of the present invention. 11: Imaging lens, 21: Main image sensor, 31.31A
, 41: Sub-image sensor, 22, 32, 42: High-pass filter, 5o: Tsukuda Goki, 51. : Arithmetic operation unit, ・2゛52,55: Comparator. 53: Logical operation unit, 54: Subtractor, 56: Focus control signal generator, 57: Changeover switch, 58: Search signal generator, 61: Drive mechanism. Figure 9 Figure 2 L J Figure 3 Figure 5 Figure 6

Claims (1)

[Scope of Claims] 1) A photoelectric conversion type main image sensor and at least one photoelectric conversion type sub image sensor arranged in a predetermined positional relationship on the optical path of the imaging optical system, and images projected onto these image sensors. A means for converting each image sharpness into an electrical signal, and a predetermined calculation for each image sharpness signal to determine the focusing state of the imaging optical system, and to adjust the imaging optical system and imaging surface according to the determination result. and means for generating a control signal for adjusting the positional relationship of the automatic focusing system. 2. In the automatic focusing device according to claim 1, the main image sensor and the sub image sensor are used as an input signal for determining a focus state by a predetermined calculation of the control means for adjusting the positional relationship of the imaging optical system. An automatic focusing method characterized by using a level comparison function between each image definition signal obtained from the sub-imaging device and a level comparison function between a plurality of image definition signals obtained from the sub-imaging device. 3) In the automatic focusing device according to claim 1 or 2, when the control signal generated from the imaging optical system positional relationship adjustment control means is within the 9 focus control range, Switches and sends a focus control signal calculated from the image clarity signal, and a search signal that forcibly moves the positional relationship between the imaging optical system and the imaging surface from the current positional relationship when the focus is outside the focus control range. An automatic focusing system that is characterized by the ability to