JPH0197912A - Focus detector - Google Patents

Abstract

PURPOSE:To improve the focus detection precision of an objective lens by averaging the light quantity distribution of a background pattern included in the light quantity distribution of a pattern member. CONSTITUTION:A background pattern averaging control means 24 is provided which changes relative position relations of the light quantity distribution of a pattern member 26 projected on photoelectric transducer arrays 16a and 16b by a pattern member 26 included in an optical means and outputs stored electric signals of photoelectric transducer arrays 16a and 16b, on which the light quantity distribution of the pattern member 26 is projected, during this change to operating means 18 and 22 as a signal indicating the light quantity distribution of the pattern member 26 in case of calculation of an initial value indicating relative position relations of the light quantity distribution at the time of focusing an objective lens 14. Then, the light quantity distribution of the background pattern included in the light quantity distribution of the pattern member 26 is averaged. Thus, the focus detection precision of the objective lens is improved.

Description

Detailed Description of the Invention (Field of Application of the Invention) The present invention is suitable for cameras and the like, and forms a plurality of light intensity distributions by a light beam transmitted through a plurality of divided regions of the pupil of an objective lens. The present invention relates to an improvement of a focus detection device that detects the focus state of an objective lens from the relative positional relationship of light amount distribution.

(Background of the Invention) Various types of focus detection devices for cameras and the like have been proposed, one of which is shown in FIG.

In FIG. 7, l is the objective lens, 2 is the intended imaging plane of the objective lens l, 3 is a field mask, 4 is a field lens placed near the intended imaging plane 2, and 5 is the optical axis of the objective lens l. Two secondary imaging lenses 5a arranged symmetrically with respect to
, 5b, and 6 is a photoelectric conversion system consisting of two photoelectric conversion element rows 6a, 6b arranged behind the two secondary imaging lenses 5a, 5b. system, 7 is an aperture having two apertures 7a, 7b arranged corresponding to the two secondary imaging lenses 5a, 5b, 8 is the exit pupil of the objective lens l, divided into two areas 8a, 8b.

Note that the field lens 4 has openings 7a and 7b of the diaphragm 7.
is the area 8a of the exit pupil 8 of the objective lens 1.

It has the effect of forming an image near 8b, and the light flux transmitted through each region 8a, 8b is reflected in the photoelectric conversion element rows 6a, 6b.
A light intensity distribution (object image including objects that are out of focus) is formed on each of the above.

In the focus detection device shown in FIG. 7, when the imaging point of the objective lens l is in front of the intended imaging plane 2, the light amount distribution formed on the two photoelectric conversion element rows 6a and 6b, respectively, is When they are close to each other and the imaging point of the objective lens l is behind the intended imaging plane 2, the light intensity distributions formed on the two photoelectric conversion element arrays 6a and 6b are separated from each other. The state will be as follows. Moreover, since the amount of deviation in the light intensity distribution formed on the two photoelectric conversion element arrays 6a and 6b has a functional relationship with the amount of defocus of the objective lens l, if the amount of deviation is calculated by an appropriate calculation means, The direction and amount of defocus of lens l can be detected.

A method for calculating the amount of deviation between two light intensity distributions is disclosed, for example, in Japanese Patent Laid-Open No. 107313/1983 by the applicant of the present invention. The method will be briefly explained below.

The outputs photoelectrically converted by the two photoelectric conversion element arrays 6a and 6b are expressed as a (i) and b (i) −(
However, when i = 1 to N, N = number of elements in one photoelectric conversion element array), V(m), =
Σ min (a (i), b (−m to i+
) 1den-@sin (a(i+k), b(i-s)
) Calculate...(1) Calculate for different integers m (shift amounts).

Note that m1n (a, b) represents the smaller of two real numbers alb17), and the range of i for which the sum is calculated is for each subscript i, i
+ -m, i+, i-m enter the closed interval [1°N],
It is determined so that the fluctuation range of the first term and the second term of V (m) is equal to 1. Normally, V (m) for an integer m, which is the amount of deviation between the two light intensity distributions at
is not necessarily 0, but in such a case, a fractional value can be obtained using an appropriate interpolation method. The simplest interpolation method is linear interpolation, where V(no) and V(Ino
+1), the deviation amount M including fractions is calculated as follows (2).

The easiest way to calculate the amount of defocus d of the objective lens 1 from the amount of deviation M between the two light intensity distributions obtained in this way is to use the constant of proportionality K, assuming that both are approximately proportional. (M−δ.′) ... is obtained from (3). However, δ0 is the amount of deviation between the two light quantity distributions when the objective lens 1 is in the focused state. From now on, this δ
The initial deviation amount δ0, where 0 is defined as the initial deviation amount (initial value), is determined by design or by measurement during adjustment in the initial state.

As a method for improving equation (3) and obtaining the amount of defocus d with higher accuracy, the applicant has also proposed a calculation method that takes into account the nonlinearity between the amount of defocus d and the amount of deviation M in the light intensity distribution. Although it is disclosed in JP-A-59-107311, the details are omitted here.

As is clear from equation (3), in such a focus detection device, the initial deviation amount δ. Since it is based on
In order to perform accurate focus detection, it is necessary that the initial deviation amount δ0 is always constant! be.

However, initially, the initial deviation amount δ. Even if the value is a certain value, each member constituting the system changes with aging and changes in the external environment, and the initial deviation amount δ0 may not always be kept constant. In the figure, when the distance between the two secondary imaging lenses 5a and 5b expands and contracts due to changes in temperature and humidity, the initial shift amount δ changes accordingly. will also change. Although the amount of expansion and contraction due to temperature and humidity is generally considered to be small, the amount of deviation in the light intensity distribution relative to the amount of defocus of the objective lens 1 is usually 1/10 to 1/10/10.
Since the amount of initial deviation δ0 is very small, even a slight change in the initial deviation amount δ0 becomes an obstacle in accurately detecting the focal state of the objective lens 1.

In particular, when plastic is used as an optical member, the amount of expansion and contraction due to changes in temperature and humidity is greater than that of glass, and when the change in refractive index is added, the effect on detection accuracy is extremely large.

One way to avoid these problems is to install a temperature/humidity sensor inside the system and make corrections according to the detected value, but a sensor with good accuracy and easy handling, especially a sensor for humidity, is recommended. The current situation is that we cannot find anything suitable.

Even if such a sensor were available, it would require space and increase costs. Furthermore, the influence of temperature and humidity, which changes from moment to moment, gradually permeates through the surface of each component, making it difficult to perform accurate corrections based solely on the instantaneous temperature and humidity at the time of correction. be.

In view of the above problems, the applicant of the present application has already proposed a countermeasure (Japanese Patent Application No. 61-247195).

FIG. 8 shows a part of the embodiment in the above-mentioned prior application, and shows the photoelectric conversion element arrays 6a, 6b in FIG. 7 and projected images 9a, 9b (focus In Figure 0, which shows the positional relationship with the light-shielding parts (including those that do not match), the shaded area is the light-shielding part. In terms of utilizing the entire area of the photoelectric conversion element arrays 6a, 6b, the projected images 9a, 9b of the field mask 3 are formed to include the photoelectric conversion element arrays 6a, 6b, as shown in FIG. However, in the embodiment of the prior application, as shown in FIG.
0a, lOb, lla, 11b are photoelectric conversion element rows 6a,
It is characteristic that it falls within 8b. In this state, when the objective lens 1 is directed toward the subject, if the subject is even slightly bright, light will appear on the photoelectric conversion element arrays 6a and 6b.
The edge portions of the field mask 3 are projected as edge images 10a-11b.

Assuming that the field mask 3 coincides with the planned image plane 2,
For example, edge images 10a, l on the same side of projected images 9a, 9b
If we calculate equations (1) and (2) only in the vicinity of ob, the resulting value, the amount of deviation M between the two light intensity distributions, will be
is considered to be the initial deviation amount at this point, and this value is designated as δ'. Therefore, this value δ′ is 6 in equation (3).
Used as 0, d=K(M-δ')...
(If set to 0, the defocus amount d can be calculated accurately even if there is a change in the initial deviation amount due to changes over time or environmental changes.Alternatively, in the sense of correcting the initial deviation amount δ0 Δ=δ′−δ. (5), and the calculation of d=K (M−(δ.+Δ)) (6) may be performed.

According to such a method, the initial deviation amount δ0 can be corrected without using a special device.

However, in this method, when calculating the initial shift amount δ0, tob is taken as information into the edge image lO of the visual field mask 3 against the background of the brightness of the visual field, so the objective lens l may be out of focus. When you are
Accurate correction may be difficult due to the inclusion of background patterns.

FIG. 10 shows this state, and the same parts as in FIG. 7 are given the same reference numerals. In FIG. 1θ, the object 12 and the field mask 3 are not in an imaging relationship with respect to the objective lens l, and the illustration focuses on the light beam that converges on the end point of the field mask 3. As is clear from this figure, when the objective lens l is out of focus,
The light amount distribution formed on the photoelectric conversion element arrays 6a, 6b as projected images 9a, 9b (FIG. 8) of the field mask 3 depends on the pattern of the regions 13a, 13b on the subject 12.

That is, if the patterns of the region 13a and the region 13b are significantly different, the respective patterns will be superimposed on the edge images 10a and 10b of the field mask 3, and the two light intensity distributions will be different as shown in FIG. It becomes an obstacle to the search.

(Objective of the Invention) An object of the present invention is to solve the above-mentioned problems, and to reduce the trouble caused by the mixing of background patterns when correcting the initial value of the relative positional relationship of the light intensity distribution. Furthermore, it is an object of the present invention to provide a focus detection device that can improve focus detection accuracy of an objective lens.

(Characteristics of the Invention) In order to achieve the above object, the present invention uses a pattern member included in an optical means to calculate each photoelectron when calculating an initial value representing the relative positional relationship of the light amount distribution when an objective lens is focused. While changing the relative positional relationship of the pattern member light quantity distribution projected onto the conversion element array over a predetermined range,
providing background pattern averaging control means for outputting the accumulated electrical signals of each photoelectric conversion element array onto which the pattern member light amount distribution is projected during the change to the calculation means as a signal representing the pattern member light amount distribution; The present invention is characterized in that the light amount distribution of the background pattern included in the pattern member light amount distribution is averaged.

(Example of the invention) FIG. 1 shows an embodiment of the invention.

14 is an objective lens, 15 is a photoelectric conversion system, 16a, 16b
is a photoelectric conversion element array having two series of charge storage functions, 17
18 is a calculation circuit, 19 is a drive control circuit, 20 is a motor for moving the objective lens 14, 21 is an extraction circuit, 22 is a correction value calculation circuit, 23 is a storage circuit, and 24 is a monitor sensor for monitoring the amount of light. This is a background pattern averaging control circuit. Note that the AF optical system below the field mask 3 shown in FIG. 5 is omitted.

In this embodiment, a correction value Δ
The basic idea of seeking
Since the invention is the same as the invention disclosed in No. 247195, when calculating the correction value Δ, the photoelectric conversion element arrays 16a and 16b near the position where the edge portion of the field mask 3 corresponding to the pattern member of the present invention is projected is It is only necessary to consider the output value; even if we take into account changes in the position of the edge images 10a' and 10b of the visual field mask 3 due to environmental changes or secular changes, it is usually sufficient to have output information from around 5 to 6 elements. many.

First, the operation when performing normal automatic focus adjustment (hereinafter referred to as AF mode) will be explained.

For example, when AF mode is selected.

A by pressing the first stage of the release button (not shown), etc.
F operation is started. The light transmitted through different regions of the pupil of the objective lens 14 forms a light amount distribution on the two photoelectric conversion element arrays 16a and 16b, respectively, by an AF optical system (not shown). The photoelectric conversion element arrays 16a and 16b accumulate the output of the monitor sensor 17- until it reaches a certain value, and when the accumulation is completed, the output of each element is sent to the arithmetic circuit 18 as one time-series signal. 18 uses the obtained signal and the correction value Δ of the initial shift amount δ0 at that time from the memory circuit 23 to calculate the equations (L), (2), and (6), and calculates the focus deviation. Calculate the quantity d. The drive control circuit 19 drives the motor 20 by a necessary amount according to the obtained defocus amount d, and brings the objective lens 14 into focus.

Next, the operation when correcting the initial deviation amount δ0 (hereinafter referred to as correction mode) will be described.

When the correction mode is selected by operating the operating member, etc.
For example, the correction operation is started by pressing the first stage of a release button (not shown). The light from the objective lens 14 illuminates the field mask 3, and an AF optical system (not shown) forms edge images 10a and 10b of the field mask 3, that is, field mask edge light intensity distributions, on two photoelectric conversion element rows 16a and 16b, respectively. Ru. This visual field mask edge light amount distribution is obtained by superimposing the light amount distribution of the background pattern. The photoelectric conversion element array lea, 16b is the monitor sensor 1
The arithmetic circuit 18 accumulates the output of the objective lens 14 until it reaches a certain value.
The amount of defocus d is calculated. The background pattern averaging control circuit 24 converts the obtained defocus amount d as follows: d → + d-e-a, where ε is the planned imaging plane 2 in FIG.
a is a constant consisting of the distance of the code material from the field mask 3 to the field mask 3 measured from .

The background pattern averaging control circuit 24 outputs the converted defocus amount d to the drive control circuit 19, and the objective lens 1
4 to the averaging start position 14-1. With this operation, the objective lens 14 is set to a position where the defocus amount measured from the field mask 3 (hereinafter referred to as the corrected defocus amount) is approximately -aSS. When the lens setting of 0 or more is completed, the background pattern is averaged. The control circuit 24 sends (-ε+a) to the drive control circuit 19 as the defocus amount d. The drive control circuit 19 starts driving the objective lens 14 at a constant speed according to the defocus amount (-(+a). At the same time, the background pattern averaging control circuit 24 sends electrical signals to the two photoelectric conversion element arrays 16a and 16b. The movement of the objective lens 14 is completed and the averaging end position 14 is started.
-2, the photoelectric conversion element arrays 16a and 16b stop accumulating electrical signals. The extraction circuit 21 extracts, from among the time-series signals indicating the output values of each element accumulated during this period,
Element outputs necessary for the calculation, for example, output values of five elements near the positions where the edge images 10a and lOb of the visual field mask 3 are projected are extracted. The correction value calculation circuit 22 calculates the equations (1) and (2) based on the extracted output value, and calculates the deviation amount of the two edge images lOa and 10b (Fig. 8), that is, the field mask edge light intensity distribution. The value of M in equation (2) is calculated. Since this value is the initial deviation amount δ′ at that point,
If the difference from the original initial deviation amount δ0 is Δ=δ'-60, this becomes the correction value for the initial deviation amount δ0. The correction value Δ is stored in the storage circuit 23, and is used in subsequent calculations in the calculation circuit 18 in the AF mode.

Here, the effects of this embodiment will be explained with reference to FIGS. 2 to 4.

FIG. 2 shows the light amount distribution on the two photoelectric conversion element rows 16a and 18b of the field mask 3 illuminated with uniform light.

FIG. 3 shows the light intensity distribution of the object (background pattern) to be projected onto the photoelectric conversion element arrays 16a and 16b when the field mask 3 is not present, and the broken line indicates correction, and the defocus amount is +.
When a, the - dotted chain line indicates when the corrected defocus amount is -a. In this embodiment, the objective lens 14 is adjusted over the range where the corrected defocus amount is from -a to +a.
Since the photoelectric conversion element rows 16a and 16b accumulate the photoelectrically converted electrical signals without moving the photoelectric conversion element rows 16a and 16b, the light intensity distribution substantially projected onto the photoelectric conversion element rows 16a and 16b is as shown in FIG. The distribution is like the solid line in the same figure, which is a continuous overlapping of the dashed-dotted line and the broken line. This and field mask 3
Considering the light shielding caused by the photoelectric conversion element array 16
The field mask edge light intensity distribution detected by a and 16b is the fourth
The shape will be as shown in the figure. In this light quantity distribution, fine irregularities are filled in compared to the light quantity distribution shown by dotted lines or broken lines in FIG. 3, and the background pattern is averaged.

Furthermore, since the light intensity distributions are superimposed between corrected defocus amounts whose positive and negative values are approximately equal around the position of the field mask 3, the finally obtained field mask edge light intensity distribution in FIG. 4 has almost the same shape. As a result of the above, the above-mentioned problem caused by the background pattern contamination can be almost completely removed.

In the embodiment described above, first, the objective lens 14 was driven so that the corrected defocus amount was -a, but the position of the objective lens 14 when starting the correction mode operation was the corrected defocus amount + a. If it is close to , it is of course possible to drive the objective lens 14 so that the amount of corrected defocus changes from +a to -a.

Furthermore, the amount of defocus d and the photoelectric conversion element rows 16a, 16
Since the deviation amount M of the light intensity distribution on b is not necessarily proportional and has a certain nonlinearity, the edge image 10 of the field mask 3 on the photoelectric conversion element arrays 16a and 16b is taken into consideration.
It is also effective to set the front and rear corrected focus deviation amounts to be unequal so that the moving amounts of the projected image of the background pattern with respect to a and the fob are equal in positive and negative directions.

In the above embodiment, the correction value Δ of the initial deviation amount δ0 is determined from the positional change of the light intensity distribution at the edge portion of the field mask 3.
However, the present invention is not limited to this, but is also applicable to pattern members shown in FIGS. 5 and 6, as described in the aforementioned Japanese Patent Application No. 61-247195 It is. The one shown in FIG. 5 is one in which a plurality of openings 25 are provided as pattern members at the ends of the visual field mask 3, and this reduces calculation errors that tend to occur in pattern members with a small amount of information, such as edge portions. Can be done. In the device shown in FIG. 6, a light shielding member 26 is inserted in the center of the field of view by a drive mechanism (not shown) during the correction mode in order to eliminate errors at the periphery and center of the field of view caused by the field of view mask 3. .

(Correspondence between the invention and the embodiments) The field mask 3, the field lens 4° aperture 7, and the secondary imaging optical system 5 in FIG. 25 or the light shielding member 26 corresponds to the pattern member, the initial deviation amount δ0 or δ' corresponds to the initial value, the arithmetic circuit 18 and the correction value arithmetic circuit 22 in FIG. 24 correspond to background pattern averaging control means.

(Modified example) In the embodiment shown in FIG. 1, the objective lens 14 is moved over a range where the corrected defocus amount is equal to positive and negative, but as a simple method, it is possible to Instead of driving the objective lens by a certain amount or to the limit of movement in the direction in which the absolute value of the corrected focal point L1 shift amount increases, the photoelectric conversion element array may be made to accumulate electrical signals. As a result, the blur of the projected image of the background pattern on the photoelectric conversion element array increases due to an increase in the amount of defocus of the objective lens, and the background pattern is averaged.

Roughly simplifying the cylinder, the amount of defocus at the start of the correction mode is not calculated, the objective lens is unconditionally driven in a predetermined direction and by a predetermined amount, and then electrical signals of the photoelectric conversion element array are accumulated. However, a considerable effect can be obtained in terms of averaging the background pattern.

(Effects of the Invention) As described above, according to the present invention, when calculating the initial value representing the relative positional relationship of the light amount distribution when the objective lens is focused, each photoelectric While changing the relative positional relationship of the pattern member light quantity distribution projected onto the conversion element array over a predetermined range,
providing a background pattern averaging control means for outputting the accumulated electric signal of each photoelectric conversion element array onto which the pattern member light amount distribution is projected during the change to the calculation means as a signal representing the pattern member light amount distribution; As a result, the light intensity distribution of the background pattern included in the pattern member light intensity distribution is averaged, so that when correcting the initial value of the relative positional relationship of the light intensity distribution, problems caused by the background pattern being mixed in can be reduced. This allows the focus detection accuracy of the objective lens to be further improved.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention, and FIG.
5 is a front view showing another example of the pattern member in the present invention, and FIG. 6 is a diagram showing another example of the pattern member in the present invention. 7 is a schematic diagram showing a conventional focus detection device; FIG. 8 is a front view showing the positional relationship between the photoelectric conversion element array and the projected image of the field mask in the embodiment of the prior application; FIG. 9 is a front view showing the positional relationship between the conventional photoelectric conversion element array and the projected image of the field mask, FIG. 10 is a schematic diagram showing the state in which the light amount distribution of the background pattern is mixed into the light amount distribution of the pattern member, and FIG. The figure is a diagram showing the level of an electric signal representing the pattern member light amount distribution. !・・・・・・Objective lens, 2・Old・・・Planned focal plane, 3
...Field mask, 4...Field lens, 5...Secondary imaging optical system, 6a, 6b...
...Photoelectric conversion element array, 7...Aperture, 8...
...exit pupil, 8a, 8b...area, lOa
, lOb, lla, llb・-=-z-7 di-image, 12・
...Subject, 13a, 13b...area,
14...Objective lens, 16a, 16b...
...Photoelectric conversion element array, 18... Arithmetic circuit, 21
...Extraction circuit, 22...Correction value calculation circuit, 23...Storage circuit, 24...Background pattern averaging control circuit, 25... ...Aperture of field mask, 26... Light shielding member.

Claims (1)

[Claims] (1) an optical means that forms a plurality of light intensity distributions by at least two light beams transmitted through different regions of the pupil of an objective lens; a plurality of photoelectric conversion element arrays that convert the plurality of light intensity distributions into electrical signals; The light amount distribution when the objective lens is focused is determined from the relative positional relationship between a pattern member included in the means and which projects a pattern member light amount distribution onto each of the photoelectric conversion element arrays and the pattern member light amount distribution. and a calculation means for calculating an initial value representing a relative positional relationship and determining an amount of defocus of the objective lens based on the relative positional relationship of the light amount distribution and the initial value, the focus detection device comprising: When calculating the initial value, the relative positional relationship of the pattern member light amount distribution is changed over a predetermined range, and during the change, the accumulated electric signal of each photoelectric conversion element row on which the pattern member light amount distribution is projected is calculated. A focus detection device comprising background pattern averaging control means for outputting the background pattern averaging control means to the calculation means as a signal representing the light amount distribution of the pattern member.