JPH0629896B2 - Focus detection device - Google Patents

Description

The present invention relates to a focus detection device used in an automatic focus adjustment device suitable for a photographing device such as a video camera or a still camera.

Conventionally, various automatic focus adjustment devices have been proposed which detect a distance to a subject and automatically adjust the focus of an objective lens of a photographing device. These conventional examples are roughly classified into a so-called active method of detecting distance information by projecting a light projection pattern on a subject and detecting reflected light thereof, and a blur of a subject image or an optical path different from the subject image. 2 images are formed by two, and the so-called passive method for detecting the distance information by detecting the deviation amount of both (for example, US Pat. No. 4,185,191).
It is divided into

However, each of the former and latter methods has advantages and disadvantages, and it is difficult to expect highly accurate automatic focus adjustment operation for all subjects. For example, in the latter passive method, the focus adjustment operation is often unstable in the following cases.

(1) When the detection sensor cannot extract the contrast of the subject under low brightness.

(2) When the subject itself has low contrast and the detection sensor cannot extract its characteristics.

Under the above conditions, the detection sensor cannot accurately output the distance signal. Due to this, in a device that requires continuous operation such as a video camera, the taking lens is adjusted in response to an inaccurate detection signal, resulting in a decrease in focus stability. Will be invited. In order to eliminate the drawbacks in such a case, as a conventionally proposed method: a The focus adjustment operation is performed only when the previous determination result and the present determination result match.

b Two focus determination ranges (so-called dead band width) (wide range where stability is maintained and narrow range where focus adjustment accuracy is ensured)
Set and switch this at an appropriate timing.

There is a method such as. That is, in the former a, the distance signals before and after are detected, and if they are the same or almost the same signal, it is determined that they are the true object distance signals, and the automatic focusing operation is performed based on this signal. If they are different from each other, the pseudo focusing signal is canceled as the pseudo signal so that the automatic focusing operation is performed only based on the true signal.

In the latter method, when the contrast of the subject itself or the contrast signal is not sufficiently obtained, a wide dead band and then an accurate dead band are output so that unnecessary false signals are not output by alternately selecting the dead band. The narrow dead band that ensures the focus adjustment accuracy is switched alternately.

The above-mentioned methods have been conceived for the purpose of making the operation stability and the focus adjustment accuracy compatible with each other, but in the case of the passive type automatic focus adjustment device, they cannot cope with any condition of the subject.

The object of the present invention is to eliminate the drawbacks of the conventional example as described above, and as its characteristic configuration, a light receiving circuit for receiving a light flux from an object and a focus based on the light receiving circuit output. A focus detection circuit that outputs a focus signal indicating a state, a focus determination circuit that determines whether the focus signal has a value within a predetermined dead band width, a contrast state detection circuit that detects the contrast state of the subject, and A focus detection device is provided in which a adjusting circuit for changing the dead zone width according to the contrast state detected by the contrast state detection circuit is provided and the dead zone width is widened when the contrast state is low.

The contrast of the subject includes not only the contrast of the subject itself but also the case where the subject pattern is clear as described above, that is, even when the contrast is sufficient, it is not possible to obtain a sufficient contrast signal under low brightness. Used.

According to the above configuration, when the contrast signal is not sufficiently obtained, it is determined that the highly accurate focus adjustment information cannot be obtained, the wide dead band is selected, and the contrast signal is equal to or more than the predetermined value. If so, a narrow dead band is selected so as to respond to the focus adjustment information with high precision.

Therefore, due to the contrast signal itself, when the contrast is low, the sensitivity is poor, that is, the stability is emphasized,
If the automatic focus adjustment is not performed carelessly based on the false signal, and if the contrast exceeds the predetermined value, the responsiveness will be good and the automatic automatic focus adjustment will be performed accurately. Thus, the automatic focus adjustment operation becomes possible.

Specific embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows a principle diagram of an optical system of an automatic focusing device of a form suitable for carrying out the present invention. In FIG.
Is a lens element, and 2 is a small lens group for re-imaging the image of the pupil plane of the lens element. In FIG. Things are used. Reference numerals 3 to 6 denote the respective pixel elements of the light receiving element formed of, for example, a CCD element, the pixel elements 3 and 4 correspond to the small lens 2-1 and the pixel elements 5 and 6 correspond to 2-2. The pixel element 3 includes the small lens 2-1.
The aperture 8 part through and the pixel element 4 stares at the aperture 7 part through the small lens 2-1.

Similarly, the pixel element 5 looks at the aperture 8 portion through the small lens 2-2, and the pixel element 6 looks at the aperture 7 portion through the small lens 2-2. As mentioned above,
The small lens and the sensor are arranged in more than 20 rows in the small lens. If 2-1 is the n-th small lens and 2-2 is the n-1-th small lens, for the sake of explanation, the aperture is opened through the n-th lens. the pixel elements stare the An7 pixel elements stare the 8 Bn, the pixel elements stare the Apachiya 8 through n-1 th lens pixel elements stare the a _n-1 Apachiya 7, B _n-1.

The F surface corresponds to the focal plane of the taking lens 1. That is, assuming a video camera, the image pickup surface position (equivalent position) of the image pickup tube becomes F. Therefore, when the lens 1 changes its position in the left-right direction, the object distance at which the image is correctly formed on the F-plane is different, and when the image of the object at the in-focus distance is captured as the output of the light-receiving element array, An = Bn, While A _n-1 = B _n-1 , when capturing an image of an object that is not in focus, An = B _n , n ≠ m.

In FIG. 1, assuming that the distance R1 is the focusing distance of the lens 1, the images at 9 points enter the An, Bn pixel elements (3,4).
The image of the point enters the A _n-1 , B _n-1 pixel element (A _n-1 , B _n-1 ). From the opposite viewpoint, the light rays 11 and 12 intersect at the point 10, and the light rays 13 and 14 intersect at the point 9.

FIG. 2 is a diagram showing the output of the light receiving element array described in FIG. 1 with the position of the light receiving element on the horizontal axis. In this figure, the number of pixel elements of the light receiving element is 1
The case where the positions of Bn and Bn are deviated is shown. By detecting this deviation, out-of-focus can be detected.

FIG. 3 is an overall configuration diagram for implementing the present invention by applying the optical system shown in FIG. 1 to an actual automatic focus adjusting device for a video camera. In FIG. 3, 10 is a photographing lens system. A so-called front lens group of a focus adjustment system, 15 is composed of two groups of a variable power system and a correction system, and is a lens group for changing the focal length of the photographing lens, 16 is a half prism, 17
Is a total reflection mirror, 18 is an AF lens group, and 19 is a first
A sensor unit including the small lens array and the light receiving element array described in the figure, 20 is an aperture meter, 21 is an aperture blade, and 22
Is a lens of the image forming system, 23 is a solid-state image pickup plate that forms a video signal as an image sensor, 24 is an A / D converter that makes the signal obtained from the sensor a digital signal, and 25 is this information and focal length information, Focusing according to focusing distance information,
CPU for determining out of focus, 26 is a motor drive circuit, 27 is an AF for changing the position of the lens group 1.
A motor, 28 is an encoder for detecting the positional information of the lens group 1, and 29 is an encoder device for detecting the focal length information of the taking lens.

The CPU 25 is equipped with an arithmetic circuit that calculates the output of the AD converter 24 and the focus signal as distance information.
In the arithmetic circuit, for example, Hf calculated by Hf = Σ {| An-B _{n + 1} |-| A _{n + 1} -Bn |} is calculated, and when Hf is almost zero, it is determined to be in focus. .

Here, in order to calculate the deviation and the direction of the An curve and the Bn curve in FIG. 2 (that is, the degree of out-of-focus and the distinction between the front bin and the rear bin), the CPU further performs the second arithmetic processing, that is, H _pf =? {| A _{n + b} -B _{n + 1} |-| A _{n + b + 1} -B _n |}, and H _pf is calculated by varying from k = -a to a. Here, the numerical value of a is determined by the number of the small lenses described above.

FIG. 4 shows a so-called “S-shaped characteristic” as a result of calculation when the horizontal axis is k and the vertical axis is H _pf. Here, the curve 31 is k = 0.
Since H _pf = 0, it can be considered that the focus is in focus in this state. On the other hand, the curve 30 has a negative k and the curve 3
In the case of 2, k is positive and H _pf is zero, and the degree of blurring is calculated by the sign of k at this time, the front pin and the rear pin, and the absolute value of k. Depending on the way A and B are taken, if k is a negative pin, then the rear pin is in the result of the curve 30, and the front pin is in the case of the curve 32.

FIG. 5 shows only an electric circuit corresponding to a concrete embodiment of the present invention extracted from the AD converter 24 and the CPU circuit shown in FIG. 3. In FIG. 5, 101 is an analog electric signal from the light receiving element array 19. After being converted into a digital data group by the A / D converter 24, the contrast signal of the subject is calculated and the dead band width of the processing circuit of the similarly obtained focus signal is switched between wide and narrow. It is configured in.
It should be noted that, in recent years, it is generally common that these circuits are processed by software by a microcomputer or the like, and in the present embodiment also, this functional portion is configured by a microcomputer. However, in order to explain the function thereof, the functional block circuit portion is shown in an exploded manner in FIG. In FIG. 5, reference numeral 104 denotes an arithmetic processing unit, which is calculated from H _pf = Σ {| A _{n + k} -B _{n + 1} |-| A _{n + k + 1} -B _n |} as described above. , H _pf = 0 is obtained. And
The polarity of k0 obtained from this is the direction of defocus, and its absolute value is the amount of defocus.

On the other hand, there are several evaluation values indicating the contrast of the subject, but in this embodiment, they are calculated in the above-described process of calculating H _pf . It is the differential coefficient at the zero cross point of the S-shaped characteristic curve, that is, the slope. In the digital processing in the microcomputer, the method of obtaining this value is as follows. In addition to the description of the conventional example, k0 is variable from −a to a, and the point at which the value of H _pf changes from negative to positive is obtained by interpolation, but m ≦ k0 <m + 1. The value of the difference between H _{pf (m)} and H _{pf (m + 1)} is the contrast evaluation amount. In this way, the arithmetic unit of 104 outputs the contrast evaluation amount of 112.

A conversion unit 113 outputs a focus determination range (dead band width) signal 114 as a function of the contrast evaluation amount signal 112 from the calculation unit 104. Reference numeral 105 denotes a signal indicating the deviation amount k0, which includes information on the direction and amount of defocus.
Reference numeral 106 denotes a correction unit, which corrects an error caused by an optical cause occurring in the optical system in front of the light receiving element array, but since it is not directly related to the present invention, its details are omitted. Reference numeral 108 denotes a comparison / determination unit that compares the signal 107 and the output signal 114 of the conversion unit 113, and as a result, compares the magnitude relationship between the focus shift amount and the dead zone width, and when the focus shift amount> the dead zone width, the focus shift is performed. The drive signal 109 is generated according to the direction information. Regarding the conversion characteristic of the dead zone width with respect to the contrast evaluation amount in the conversion section 113, the conversion section 113 is set to the characteristic that the higher the contrast, the narrower the dead zone. Specifically, the conversion unit is configured such that the output with respect to the input takes the reciprocal or complements with the constant.

In consideration of the variable range of the dead zone width as the conversion unit 113, it is considered that the method using the table can set a value close to an ideal value, but it is not always necessary to convert continuously. Instead, for example, the converted value may be formed stepwise at least in two or more steps.

With the configuration described above, when the contrast of the subject is low, the conversion unit 113 outputs the signal 114 indicating the large dead band by the contrast evaluation amount signal 112 from the calculation unit 104. As long as it does not take a large value, that is, if the deviation amount is not large, the motor drive circuit 110
No output is applied to. As a result, the stability of the fifth illustrated processing system is increased, and for a subject having a small contrast, which is likely to include a false signal, a small amount of focus adjustment with respect to a picking is performed. On the other hand, as the contrast increases, the conversion unit 113 outputs the signal 114 corresponding to the narrow dead band width, and the comparison determination unit 108 increases the responsiveness. As a result, the distance information, that is, the shift amount signal 105, makes the motor drive circuit 110 highly sensitive and repeats the accurate focusing operation.

As described above, in the present invention, by evaluating the contrast amount, a dead zone corresponding to this is set, and both the stability and the sensitivity of the automatic focusing system are harmonized according to the contrast amount. Since I am trying to adjust it,
An optimum focus adjustment operation is possible each time, and a highly accurate automatic focus adjustment operation is possible for various subjects, which is extremely effective as an automatic focus adjustment device for various cameras.

[Brief description of drawings]

FIG. 1 is a block diagram showing the principle of an optical system of an automatic focusing device to which the present invention is applied. FIG. 2 is an operation explanatory diagram of the device shown in FIG. 1. FIG. 3 is a whole configuration of an automatic focusing device to which the present invention is applied. FIG. 4 is a diagram for explaining the operation of the apparatus shown in FIG. 3 FIG. 5 is an electric circuit diagram showing an embodiment of the present invention 104: a calculation unit for calculating a contrast evaluation amount and a deviation amount 113: a dead zone according to the contrast evaluation amount Conversion unit 108 that forms a signal corresponding to the width: Comparison determination unit 110 that compares the output of the conversion unit 113 and the signal corresponding to the amount of deviation 110: Photographic lens drive diagram Motor drive circuit

Claims (1)

[Claims] 1. A light receiving circuit for receiving a light flux from a subject,
A focus detection circuit that outputs a focus signal indicating a focus state based on the output of the light receiving circuit, a focus determination circuit that determines whether the focus signal is a value within a predetermined dead zone width, and a contrast state of a subject is detected. And a adjusting circuit for changing the dead zone width according to the contrast state detected by the contrast state detecting circuit, and widening the dead zone width when the contrast state is low. Focus detection device.