JPH07107575B2 - Focus detection device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a focus detection device for detecting the focus adjustment state of an objective lens based on the brightness distribution of a subject, and particularly to the focus adjustment state based on the brightness distribution in a plurality of directions. To a device for detecting.

<Prior Art> Conventionally, a passive type automatic focus detection device detects a defocus amount of a photographing lens from a luminance distribution in only one specific direction of a subject, as shown in, for example, Japanese Patent Laid-Open No. 58-150918. Was configured. Therefore, the focus cannot be detected at all for the subject having no luminance distribution in the direction in which the focus detection device can detect. For example, the passive focus detection device of a camera generally has a detection ability only for the luminance distribution in the horizontal direction, so that a horizontal striped pattern of evenly spaced horizontal stripes or a striped pattern such as window blinds The subject had a drawback that it was difficult to focus.

<Problems to be Solved by the Invention> The present invention eliminates the above-mentioned drawbacks and is capable of detecting the focus adjustment state of a photographing lens regardless of the appearance or pattern of a subject and a signal after image detection. It is intended to provide a device that can be easily processed.

In order to achieve this object, an optical system composed of two pairs of image forming lens pairs having optical axes parallel to each other is formed in the image forming lens in order to form an image with a parallax from a light beam passing through the same objective lens. The optical system includes: a stop having corresponding apertures; and a single photoelectric conversion device having a plurality of pixel rows for receiving light forming the image, wherein the optical system includes a first imaging lens pair and a second imaging lens. The lens pair is arranged so that the lines connecting the optical axes of the imaging lenses intersect with each other, and the adjacent imaging lenses have a shape in which a part of the green is linearly cut and joined. A device for detecting a focus adjustment state of an objective lens based on a signal relating to an image interval, which is composed of a collar-shaped portion surrounding the image, and the image having the parallax has different directions.
It is assumed that a group is formed, and the amount of change in each image interval due to the defocus of the objective lens is made substantially equal.

<Embodiment> An embodiment of the present invention will be described below with reference to FIGS. Although FIG. 1 illustrates a part of the focus detection optical system, when the focus detection optical system is incorporated in a single-lens reflex camera, it is assumed that the taking lens is arranged in front of this optical system. To do.

In the figure, 101 is a visual field mask having linear openings in a cross shape, 102
Is a field lens, 103 is a porous diaphragm, 104 is a secondary imaging lens body, and 105 is a photoelectric conversion device having four pixel columns. Further, X is the optical axis of the photographing system. Visual field mask 101
Is located in the vicinity of the planned imaging surface of the taking lens (not shown) (including the case of the planned imaging surface), and the aerial image of the subject formed by the taking lens is transferred by the secondary imaging lens body 104 to the photoelectric conversion device 105. An image is re-formed on each pixel column 105a-105d.
Here, the perforated diaphragm 103 has openings 103a, 103c arranged vertically with the optical axis of the imaging system interposed therebetween, and openings 103b, 103d arranged horizontally (FIG. 2). Image lens body 104
The vertically-aligned positive lens units 104a and 104c and the horizontally-aligned positive lens units 104b and 104d have a configuration in which the optical axes of the lens units are parallel to each other (FIG. 3). Ten
On the image 5, the aerial image in the opening of the field mask 101 is separated into four images with parallax and re-imaged. In the secondary imaging lens body 104 of FIG. 3, the pairs of positive lens portions 104a and 104c, 104b and 104d aligned vertically and horizontally are arranged so that the lines connecting the optical axes of the respective lens portions intersect with each other. In addition, the distance between the optical axes of the positive lens parts is smaller than the diameter, and it is shaped like the lens is cut out after the green part has been cut, and it is useful for increasing the effective light amount. . In addition, there is a lower collar-shaped part around each positive lens part,
This portion is useful for gripping during assembly and attachment to the lens barrel.

The light flux transmitted through the field mask 101 and the field lens 102 and incident on the aperture opening 103a is then incident on the lens portion 104a of the secondary imaging lens body 104, and on the photoelectric conversion device 105, as shown in FIG. A field mask image 109a is formed on the. The pixel array 105a is located inside the field-of-view mask image 109a, where the illuminance distribution of the subject image is extracted as an electric signal. Similarly, the light flux transmitted through the aperture opening 103b and the lens portion 104b forms the visual field mask image 109b, the light flux transmitted through 103c and 104c forms the visual field mask image 109c, and the light flux transmitted through 103d and 104d the visual field mask image 109d. Since each is formed,
The subject image is photoelectrically converted by the pixel columns 105b, 105c, 105d.

5 (a) and 5 (b) are sectional views showing this optical action. In the figure, 109 is a photographic lens, a field mask 103 is arranged in the vicinity of the primary imaging plane of the photographic lens, and a field lens 102 is arranged so as to project the porous diaphragm 103 onto the exit pupil of the photographic lens. Therefore, as shown in FIG. 6, four regions 108a, 108b, 108c, 10 separated from each other on the exit pupil 107 are provided.
The field mask images 109a, 109b, 109 are generated by the light flux that has passed through 8d.
c, 109d is formed. That is, the areas 108a, 108 on the exit pupil
A focus detection optical system of a phase difference detection method is configured for each of the light fluxes passing through c and 108b, 108d.

Therefore, in the front focus state in which an object in front of the subject is imaged on the planned image formation plane, as shown in FIG. 7A, the illuminance distribution of the image in the visual field mask image 109 shows an arrow from the in-focus state. In the rear focus state in which the object behind the subject is imaged on the planned image formation surface, the object moves in the direction of the arrow shown in FIG. 7B.

FIG. 8 shows photoelectric conversion outputs in the front-focused state and the rear-focused state. The pixel rows 105a and 105c and the pixel rows 105b and 10 when a point image is formed in the center of the field mask 101 by the photographing lens.
5d photoelectric conversion output. The photoelectric conversion outputs of the pixel rows 105a and 105b correspond to the A series, and the photoelectric conversion outputs of the pixel rows 105c and 105d correspond to the B series. FIG. 8 (a) shows the optical conversion output when the taking lens is at the in-focus position,
The figure (b) is a front pin state, and the figure (c) is a rear pin state.
The distance between the two images is narrow at the front pin and wide at the rear pin.
By detecting the relative distance between the two images, it becomes possible to calculate the defocus amount of the taking lens. The outline of the arithmetic method will be described later, but the detailed description is omitted because it is out of the scope of the present invention.

9 to 12 show another embodiment according to the present invention. In FIG. 9, 116 is a field mask having a square aperture whose diagonal lines are aligned in the vertical and horizontal directions, 102 is a field lens, 117 is a porous diaphragm, 118 is a secondary imaging lens body, and 119.
Is a photoelectric conversion device having vertical and horizontal pixel rows. It is assumed that the photographing lens is arranged in front of this row as well. The diaphragm 117 has four apertures 117a, 117b, 117c, 117d each having a point-symmetrical shape (Fig. 10), and the secondary imaging lens body 118 has four lens parts 1 at corresponding positions.
18a, 118b, 118c, 118d and aberration correcting prism parts 118e, 1
It has 18f, 118g and 118h. That is, the light flux that has entered the aperture opening 118a then enters the lens unit 118a and enters the prism unit 11a.
Emitted from 8e, the field mask 116 is placed on the photoelectric conversion device 119.
Forming an image of. Inside the field-of-view mask image 125a, pixel rows 11
9a, the pixel row 119b is located inside the field-of-view mask image 125b, the pixel row 119c is located inside the field-of-view mask image 125c, and the pixel row 119d is located inside the field-of-view mask image 125d. Is taken out as.

FIG. 10 illustrates details of the secondary imaging lens body 118. The pair of convex lens parts 118a and 118c and the pair of 118b and 118d are assembled so that the optical axis spacing is smaller than the diameter of each lens, and further, symmetrically in the horizontal and vertical directions at the exit end of the cast body. so,
A pair of correction prisms 118e and 118g tilted in the direction of the pixel row, 11
A pair of 8f and 118h is formed. The correction prism is intended to remove aberrations such as distortion generated in the separated image due to the difference between the position where the lens unit 118a looks into the subject and the position where the lens unit 118c looks into the subject. The objective is achieved by tilting the exit surface around the optical axis position of. Therefore, the same in-focus position is maintained regardless of where in the focus detection field the subject image enters.

FIG. 11 is a detailed view of the diaphragm. The aperture shape of the diaphragm is set so as to maximize the symmetry of the blurred image and the baseline length of the focus system. That is, the blurred image due to the diaphragm as shown in FIG. 2 is as shown in FIGS. 8B and 8C, and the two images are in a non-congruent blurred state. The reason why the blurred images are not congruent in this way is that the aperture openings 103a and 103c are shaped so as not to overlap with the other even if they are moved in parallel, and this is the same for 103b and 103d. . That is, in the case of the shape shown in FIG. 2, since the two sets of openings are in a mirror image relationship, the blurred image also becomes a mirror image relationship as shown in FIG. The result is not. On the other hand, the diaphragm shape shown in FIG. 11 is
When 117a is translated, it overlaps with 117c, and when 117b is translated, it overlaps with 117d. Therefore,
The blurred image has a congruent shape as shown in FIG. 13 and can be superimposed on the other by parallel movement of one, and the phase difference between the two images can be detected with high accuracy.

Furthermore, in order to maximize the baseline length of the focus detection system after satisfying such conditions, the maximum diameters of the four openings are made circular according to the exit pupil shape of the photographing lens.

Further, in order to effectively use the light quantity, a square aperture as shown in FIG. 14 may be used.

In the diaphragm shape shown here, the center of gravity intervals of the pair of diaphragm apertures are set to be the same in order to equalize the base lengths of the respective focus detection optical systems. That is, 103a- in FIG.
The intervals of center of gravity of the openings 103c and 103b-103d, 117a-117c and 117b-117d in FIG. 11 and 136a-136c and 136b-136d in FIG. 14 are equal. By doing so, for example, the relative movement amount of the illuminance distribution in the field mask image shown in FIG. 7 with respect to the defocus of the objective lens is 109a-109c and 109b-109d.
Is equal to

12 is a view showing a state of a field mask image on a photoelectric conversion element in the focus detection optical system shown in FIG. 12 and FIG. 9. The field mask is a secondary imaging lens body 118 having four lens parts.
Are separated into four images and projected. The points a, b, c, d in the field mask opening shown in FIG. 9 correspond to a ', b', c ', d'in FIG. 12, respectively.

In FIG. 9, an image of a subject is formed in the vicinity of the field mask 116 by a photographing lens (not shown), and the light flux passing through the aperture further passes through the perforated diaphragm 117 and the secondary image forming lens body 118.
As a result, two pairs of images having parallax are formed on the photoelectric conversion device 119. At that time, as in the case of FIG.
After receiving the image forming action at a and the aberration correcting action at the prism portion 118e, an image is formed on the pixel row 119a. Further, the light flux that has passed through the opening 117c forms an image on the positive lens portion 118c, and the prism portion
After receiving the aberration correction action at 118g, an image is formed on the pixel column 119c. The separated images on the pixel rows 119a and 119c are symmetrically moved in the vertical direction in accordance with the focus adjustment of the photographing lens (not shown), as in FIG. Also, multi-hole diaphragm
The luminous fluxes that have passed through 117b and 117d are pixel rows 119b and 119d, respectively.
An image is formed on the upper surface and moves symmetrically in the horizontal direction according to the focus adjustment of a photographing lens (not shown).

The defocus amount of the photographing lens can be calculated as follows from the phase difference of the image projected on the pixel array of the photoelectric conversion device.

d is the defocus amount of the taking lens, Z is the relative shift amount of the two images, M is the imaging magnification of the secondary imaging system, and Fo is the ray passing through the center of the pupil-divided partial area of the taking lens. Calculation of d = (Fo / M) Z / {1 + FoZ / (Mg)} (1) where g is the distance between the film surface and the exit pupil surface of the photographic lens The formula holds. Thereby, the differential focus amount of the photographing lens can be calculated from the relative displacement amount Z of the two images. The relative displacement amount Z of the two images is, for example, Japanese Patent Laid-Open No. 58-14.
It can be determined by the method disclosed in No. 2306.

Alternatively, the deviation amount Z may be obtained after adding the outputs of two sets of vertical and horizontal pixel columns. FIG. 15 is an explanatory view of this image shift amount detecting method, in which 130 to 135 represent outputs of pixel columns, (a) -C is a pixel column 105a, and (a) -D is 105.
c, (b) -E corresponds to 105b, and (b) -F corresponds to 105d. Further, FIG. 14 (c) is a combination of the outputs shown in (a) and (b), and (c) -G is the sensor output.
130 and 132 (c) -H add sensor outputs 131 and 133, respectively. If the relative image shift amount is obtained from the thus obtained combined sensor outputs 134 and 135 by the above method, the defocus amount of the taking lens can be calculated. If such a process is performed, the contrast of the image on the pixel row is low as shown in FIG. 15 (b), and even if the image shift amount cannot be detected with high accuracy by only the pixel row output, the image on the other pixel row is detected. If a high-contrast image is formed on the pixel array, the combined output of the pixel rows will have a high contrast as shown in FIG. 15 (c), and the amount of image deviation can be detected with high accuracy. It only needs to be done once. This effect is particularly effective when an image having no contrast is projected on one set of pixel rows, for example, a vertical stripe object.

FIG. 16 is a diagram showing the position of the distance measuring field within the finder field of the camera. In the figure, 126 is a finder field, 127 and 128 are distance measuring fields, and 129 is a distance measuring unit instruction mark. The distance measuring unit instruction mark is written on a focusing plate (not shown) of the camera. Further, the subject image incident on the distance measuring field 127 is re-imaged on the photoelectric conversion pixel columns 105b, 105d or 119b, 119d, and the subject image incident on the distance measuring field 128 is photoelectric conversion pixel columns 105a, 105c or 119a. , 119c is re-imaged.

<Effects of the Invention> As described above, by enabling focus detection based on subject luminance distributions in a plurality of directions, for example, a horizontal stripe pattern sweater that cannot be detected by the conventional bushy focus detection device. , It became possible to focus on the window blinds, and it became possible to eliminate subjects that were not good at it. Further, there is an effect that the focus detection accuracy becomes constant regardless of the direction of the object coverage distribution.

In addition, since Fo can be handled as a constant in the above-described formula for calculating the defocus amount of the taking lens, there is an advantage that the calculation process can be facilitated.

On the other hand, after performing output addition between the pair of sensors,
Add signals 130 and 132 and signals 131 and 133 in the figure to obtain signals 134 and 135.
After the image is created, the image shift amount calculation is performed to detect the defocus amount of the objective lens. According to this, the image shift amount can be calculated only once.

In this case, even if a flat object image is incident in a certain pixel column direction, since it has contrast in the other pixel column directions, focus detection cannot occur.

[Brief description of drawings]

FIG. 1 is a perspective view of a focus detection apparatus according to a first embodiment of the present invention, FIG. 2 is a plan view of a diaphragm, FIG. 3 is a detailed view of a secondary imaging lens body, and FIG. 4 is a photoelectric conversion device. Explanatory drawing which shows the upper visual field mask image. FIG. 5 is a sectional view of the focus detection device, FIG. 6 is an explanatory view showing a pupil division state of the objective lens, FIG. 7 is an explanatory view showing an illuminance distribution moving direction in the field mask image, and FIG. 8 is an objective lens. FIG. 9 is an explanatory view showing the sensor output according to the in-focus state, FIG. 9 is a perspective view showing another embodiment of the focus detection device,
FIG. 10 is a detailed view of the secondary imaging lens body, FIG. 11 is a plan view of the diaphragm, FIG. 12 is an explanatory view showing a field mask image on the photoelectric conversion device, and FIG. 13 is an in-focus state of the objective lens. FIG. 14 is a plan view showing a modified example of the aperture, FIG. 15 is an explanatory view showing an example of addition of pixel row outputs, FIG.
The figure is an explanatory view showing a finder field and a distance measuring unit when the present focus detection device is applied to a camera. In the figure, 101,116 …… field mask, 102 …… field lens, 10
3,117 ... Porous diaphragm, 103a-103d, 117a-117d ... Aperture,
104,118 ... Secondary imaging lens body, 104a-104d, 118a-118d
...... Positive lens unit, 105,119 …… Photoelectric conversion device, 105a
-105d, 119a-119d ... Pixel columns (photoelectric conversion means).

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Akira Akashi, Akira Akashi, 770 Shimonoge, Takatsu-ku, Kawasaki-shi, Kanagawa Canon Inc., Tamagawa Plant (72) Akira Ishizaki, 770 Shimonoge, Takatsu-ku, Kawasaki-shi, Kanagawa Canon Inc. Inside the Tamagawa Plant (72) Keishi Otaka Keishi Otaka, 770 Shimonoge, Takatsu-ku, Kawasaki City, Kanagawa Prefecture Canon Inc. Inside the Tamagawa Plant (72) Takeshi Koyama 770, Shimonoge, Takatsu-ku, Kawasaki City, Kanagawa Canon Inc., Tamagawa Plant (56) References JP-A-58-27110 (JP, A)

Claims (2)

[Claims] 1. A pair of imaging lenses having optical axes parallel to each other for forming an image with parallax from a light beam passing through the same objective lens.
An optical system comprising a set, a diaphragm having openings respectively corresponding to the image forming lenses, and a single photoelectric conversion device having a plurality of pixel rows for receiving the light forming the image. The first image forming lens pair and the second image forming lens pair are arranged so that the lines connecting the optical axes of the image forming lenses intersect with each other, and the adjacent image forming lens pairs form a part of the green line. A device for detecting a focus adjustment state of an objective lens based on a signal related to an image distance, which has a parallax and is composed of a portion having a shape that is excised and coupled and is surrounded by a collar-shaped portion. A focus detection device characterized in that two sets of images are formed in different directions, and the amount of change in each image interval due to defocusing of the objective lens is substantially equal. 2. A focus adjustment state after synthesizing an output from a pixel row receiving light from one of the imaging lens pairs and an output from a pixel row receiving light from another imaging lens pair. The focus detection device according to claim 1, wherein signal processing is performed to detect the.