JP3218479B2 - Focus detection device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a focus detecting device such as a camera.

[0002]

2. Description of the Related Art Conventionally, a subject image formed by a pair of light beams coming from a pair of regions having different exit pupils of a photographing optical system is converted into a pair of sensors by a focus detection optical system having a pair of re-imaging lenses. By forming a pair of subject images on the light receiving section and performing arithmetic processing on a pair of subject image signals generated by the above-described sensor according to the light intensity of the pair of subject images, the relative displacement of the pair of subject images is obtained. 2. Description of the Related Art There is known a camera system including a focus detection device for calculating an amount and detecting a defocus amount between an image forming surface of a photographing lens and a film surface based on the shift amount.

FIG. 3 shows an example of the configuration of a focus detection optical system when a focus detection area is arranged at the center of a photographing screen. In this figure, the focus detection optical system 6 includes a field mask 60 having an opening 61, a condenser lens 62, an aperture mask 65 having a pair of aperture openings 63 and 64, and a pair of re-imaging lenses 66 and 67. The pair of primary images formed by the photographing optical system 3 are converted into a pair of light receiving portions 7 on the sensor 70 by the pair of re-imaging lenses 66 and 67.
The images are re-imaged on the images 1 and 72 as a pair of secondary images. Light receiving section 7
Reference numerals 1 and 72 each include a plurality of pixels.

A field mask 60 and a condenser lens 62
Is located near the expected focal plane of the imaging optical system 3 and divides the center of the opening 61, the center (optical axis) of the condenser lens 62, and the center of the pair of aperture openings 63 and 64 into two equal parts. The point that bisects the center of the pair of re-imaging lenses 66 and 67 and the point that bisects the center of the pair of light receiving units 71 and 72 are located on the optical axis of the photographing optical system 3. Are located in

[0005] An arrow X on the right side of the figure indicates a long side (horizontal) on the screen.
Y is the short (vertical) direction, and the opening 61
Sets a focus detection area extending in the horizontal direction on the screen. In the configuration of FIG. 3, a pair of aperture openings 63 and 64 are provided.
A pair of regions 33 symmetrical with respect to the optical axis of the surface 30 near the exit pupil of the imaging optical system 3 by the condenser lens 62;
The light beam projected onto the light-receiving area 34 and passing through this area is first formed as a pair of primary images near the field mask 60. The primary image further passes through a condenser lens 62 and a pair of aperture openings 63 and 64, and is passed by a pair of re-imaging lenses 66 and 67 onto a pair of light receiving sections 71 and 72 of the sensor 7 to form a pair of 2. It is formed as the next image.

FIG. 4 shows an example in which the focus detection area is arranged off the optical axis on the photographing screen. Elements having the same functions as those in FIG. 3 are denoted by the same reference numerals. The field mask 60 and the condenser lens 62 are arranged near the planned focal plane of the imaging optical system 3, but the center of the opening 61 is arranged at a position away from the optical axis of the imaging optical system 3. The center (optical axis) of the condenser lens 62, the point at which the center of the pair of aperture openings 63 and 64 is bisected, and the pair of re-imaging lenses 66 and 67
And the point bisecting the center of the pair of light receiving units 71 and 72 are also located at positions away from the optical axis of the imaging optical system 3.

In the configuration shown in FIG. 4, a pair of aperture openings 6 are provided.
The lenses 3 and 64 are projected by the condenser lens 62 onto a pair of regions 33 and 34 symmetrical with respect to the optical axis of the surface 30 near the exit pupil of the photographing optical system 3. A pair of primary images is formed near the field mask 60. The primary image is further converted to a condenser lens 6
2 and a pair of secondary images are formed on a pair of light receiving sections 71 and 72 of the sensor 7 by a pair of re-imaging lenses 66 and 67 through a pair of aperture openings 63 and 64.

In the above two configuration examples, the intensity distribution of the pair of secondary images is photoelectrically converted by the light receiving units 71 and 72 to become an electric object image signal, and is subjected to A / D conversion and digitization. The data is subjected to a filtering process for cutting high-frequency components and low-frequency components that cause an error in detection, and becomes a pair of subject image data. An example of an algorithm for calculating an image shift between a pair of subject image signals will be described below. Assuming that the subject image data is ai and bi (where i = 1 to m), first, a correlation amount C (L) is obtained by a differential correlation algorithm shown in the equation (1).

[0009]

## EQU1 ## where L is an integer and is a relative image shift amount in units of the pitch of the light receiving elements of the pair of light receiving element output data. In addition, the range taken by the parameter i in the integration operation の of Expression 1 is appropriately determined according to the shift amount L and the number m of data.

In the calculation result of Equation 1, the correlation amount C (L) is minimized at the shift amount L = kj where the data correlation is high.
Next, the image shift amount X that gives the minimum value C (X) of the correlation amount with respect to the continuous correlation amount is obtained by using the three-point interpolation method shown in Expression 2.

[0011]

Where the parameters D, SLO
P is represented by the equation of Equation 3 and the equation of Equation 4.

[0012]

(Equation 3)

[0013]

## EQU4 ## The defocus amount DEF of the object image plane with respect to the predetermined focal plane is calculated from the image shift amount X obtained by the equation (2).
Can be obtained by the equation of Expression 5.

[0014]

In the equation (5), P is a pitch in a direction in which light receiving elements constituting each light receiving section of the photoelectric conversion means 4 are arranged,
K is a coefficient determined by the configuration of the focus detection optical system.
In addition, in order to determine the reliability of the image displacement amount X obtained by the expression (2), it is determined that the image displacement amount X is reliable when the expression (6) is satisfied, and the image displacement amount X is validated. Indicates that the focus cannot be detected.

[0015]

In the equation (6), L is a predetermined value, and the minimum value C (X) / SLOP of the correlation amount is 0 (zero) when the pair of images completely match, and Becomes lower and becomes higher.

[0016]

The focus detection devices shown in FIGS. 3 and 4 have the following problems.
As shown in FIG. 3, in the configuration example of the focus detection optical system in which the focus detection area is arranged at the center of the screen, a pair of re-imaging lenses 66 and 6 are used.
7 is arranged eccentrically with respect to the optical axis of the photographing optical system 3, the primary image having an intensity distribution as shown by a thick line in FIG. The intensity distribution of the secondary image, that is, the point spread function (PSF: POINT SPREAD FUNCTION) in the direction in which the light receiving portions of the re-imaging lens are arranged becomes asymmetric with respect to one of the pair of re-imaging lenses 66 as shown in FIG. On the other hand, it also becomes asymmetric with respect to 67 as shown in FIG.

As shown in FIG. 4, in a configuration example of the focus detection optical system in which the focus detection area is arranged outside the center of the screen, the pair of re-imaging lenses 66 and 67 are asymmetric with respect to the optical axis of the photographing optical system 3. Due to the eccentric arrangement, the light receiving sections 71, 7
For the primary image having the intensity distribution as shown by the bold line in FIG. 14 along the arrangement direction 2, the intensity distribution of the secondary image, that is, the point spread function (PSF: PO
In the case of INT SPREAD FUNCTION), the spread is large as shown in FIG. 15 for one of the pair of re-imaging lenses, and the spread is small for the other as shown in FIG.

In any case, the disparity of the point spread functions occurs due to the unequal arrangement of the pair of re-imaging lenses. As described above, when the image shift detection calculation is performed using the subject image signal obtained by photoelectrically converting the intensity distribution of the pair of secondary images by the light receiving unit, the coincidence between the respective images is low. Even when it is determined that the focus cannot be detected because the image is not satisfied, or even when the focus can be detected, the degree of coincidence is low, so that a detection error occurs in the image shift amount X obtained by Expression (2).

[0019] Even in the case of a general subject, the secondary image formed on the light receiving section is a convolution of the subject image intensity distribution and the PSF, so that the same problem as described above occurs. Therefore, the present invention solves the above problem, and even if the degree of coincidence of the intensity distribution of the pair of secondary images formed by the pair of re-imaging lenses decreases, focus detection cannot be disabled,
An object is to enable highly accurate focus detection.

[0020]

In order to solve the above-mentioned problems, the present invention provides a photographing optical system for forming a subject image on a predetermined focal plane according to FIG.
A sensor that includes a pair of light receiving units and generates a pair of output signals corresponding to the light intensity distribution on the pair of light receiving units, and a subject image formed in a focus detection area of a part of the planned focal plane, a focus detecting optical system comprising a pair of re-imaging lens for re-imaged on the pair of light receiving portions, the pair
Image of the point light source on the pair of light receiving portions by the re-imaging lens of
Are different from each other when re-imaging
And the difference between the pair of point spread functions.
For the pair of output signals, the degree of coincidence of the intensity distribution is reduced, a shaping process for restoring the degree of coincidence of the intensity distribution is performed based on the pair of point spread functions, and the pair of processes is performed by the shaping process. A shaping means for generating a signal; and a focus detection calculating means for performing an image shift detection calculation on the pair of processing signals to calculate a focus state of the photographing optical system.

[0021]

With respect to a pair of secondary images re-formed on the light receiving section by the pair of re-imaging lenses, the sensor outputs a pair of output signals (subjects) corresponding to respective light intensity distributions. Generating a pair of re-imaging lenses
Are different from each other, the degree of coincidence of the intensity distribution of each output signal (subject image signal) is low .
In this case, the above-mentioned shaping means performs the paired point image distribution.
A different shaping process is performed on the pair of output signals (subject image signal) based on the function, and a shaping process is performed so as to restore the degree of coincidence of the intensity distribution. By generating a processing signal, the above-described focus detection calculation means performs an image shift detection calculation on the pair of processing signals to calculate a focus state of the imaging optical system,
Due to the restoration of the degree of coincidence by the shaping means, focus detection is not disabled, and highly accurate focus detection can be performed.

[0022]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described with reference to FIGS. FIG. 2 is a block diagram of an embodiment of the present invention. FIG. 2 shows a state where the lens 2 is replaceable with respect to the camera body 1 and the lens 2 is mounted on the body 1.

A light beam from a subject passing through a photographing optical system 3 built in the lens 2 is split in a direction of a finder 9 and a sub-mirror 5 by a main mirror 4 on the body 1 side constituted by a half mirror. The light beam deflected toward the body bottom surface by the sub-mirror 5 is guided to a focus detection optical system 6 arranged near the conjugate plane of the film. This focus detection optical system 6 is the focus detection optical system shown in FIG. It has the same configuration as the system 6.

The re-imaging lenses 66 and 6 of the focus detection optical system 6
The intensity distribution of the pair of secondary images formed by the photoelectric conversion unit 7 is photoelectrically converted by the light receiving units 71 and 72 of the sensor 7 to generate an electrical pair of subject image signals. The CPU 8 in FIG. 2 includes an AD conversion unit 80 and a RAM 81 (random access memory)
, A shaping unit 82, and a focus detection calculating unit 83. A pair of electrical subject image signals are AD-converted by the AD conversion unit 80 to be digital data. Where AD
The pair of converted subject image data are respectively denoted by ai and bi.
(Where i = 1 to m).

The pair of subject image data is temporarily stored in the RAM 8
The shaping means 82 performs shaping processing on a pair of subject image data stored in the RAM 81 and stores the shaped data a ′ i and b ′ i in the RAM 81 again. The focus detection calculating means 83 calculates the pair of subject image data a ′ i and b ′ i stored in the RAM 81 by using the above-described algorithm of the image shift calculation calculation using the formulas (1) to (6). The defocus amount of the photographing optical system 3 is calculated.

Next, the calculation of shaping when the focus detecting optical system 6 is configured as shown in FIG. 3 will be described.
In the arrangement of the re-imaging lenses 66 and 67 in FIG. 3, as described above, the point image distribution of the re-imaging lenses 66 and 67 with respect to the thick line in FIG. FIG.
Since the arrangement of the re-imaging lens is symmetrical with respect to the optical axis with respect to the other 67, as shown in FIG.

This point image is sampled by a plurality of pixels of the light receiving units 71 and 72, and AD-converted data is as shown in FIGS. 8 and 9, respectively. In this state, the photographing optical system 3 is in a focused state, and the pair of images must match without any displacement. In FIG. 8, a (n-2), a (n-1),
Data other than a (n) and a (n + 1) have almost zero values, and in FIG. 9, b (n-1), b (n), b (n +
Data other than 1) and b (n + 2) are almost 0 (zero). The value of each data is a (n−2) = k1, a
(N-1) = k2, a (n) = k3, a (n + 1) = k
4, b (n-1) = k4, b (n) = k3, b (n +
1) = k2, b (n + 2) = k1, and the two image data do not match in this state.

Then, the shaping means 82 performs shaping processing on the image data sequence ai in FIG. 8 with a weighting coefficient sequence Ai (i = 1 to 4) as shown in FIG. (Convolution operation) is performed, and the data is converted into a symmetric point spread function data string a ′ i by the processing. Also, for the image data sequence bi of FIG. 9, the weighting coefficient sequence Bi (i =
By performing the shaping process (convolution operation) in 1) to 4), the data is converted into a symmetric point spread function data sequence b′i.

However, the value of the weight coefficient is A1 = α × k4, A
2 = α × k3, A3 = α × k2, A4 = α × k1, B1
= Α × k1, B2 = α × k2, B3 = α × k3, B4 =
α × k4, the value of which is obtained by calculating the point spread function of the re-imaging lens of the focus detection optical system by an optical design program in advance, and is fixedly stored in the shaping means 82.

The above-mentioned convolution operation is performed on the image data sequence ai in FIG. 8 by the formula of Expression 7 and on the image data sequence bi in FIG. 'i (i = 2 to m-2), b' i (i =
3 to m-1) are calculated.

[0031]

(Equation 7)

[0032]

## EQU8 ## The shaping process data a'i thus obtained,
As shown in FIG. 12 and FIG. 13, the individual values of b ′ i completely match without any deviation. Therefore, the focus detection calculating means 83 uses the shaping processing data a ′ i and b ′ i. , The focus can be detected without error.

In FIGS. 12 and 13, the values of the shaping process data a ′ i and b ′ i have values other than 0 (zero) only for i = n−3 to n + 3, and are expressed by the following equation (9).

[0034]

As described above, the shaping process is performed on the image data sequence ai with the weight coefficient sequence Ai (i = 1 to 4), and the weight coefficient sequence Bi (i = 1 to 4) is performed on the image data sequence bi. 12), the values of the shaping data a ′ i and b ′ i can completely match without any deviation as shown in FIGS. 12 and 13.

The above description has been made on the case where the object is a point image, but the present invention is also applicable to a general object image other than the point image. For example, assuming that the point spread functions of the re-imaging lenses 66 and 67 are p (x) and p '(x) (symmetrical to p (x)), the intensity of the subject image on one light receiving portion of the sensor The distribution function g (x) is the intensity distribution function f (x) of the subject image.
And the point spread function p (x) convolution f (x) ＠
given by p (x), the point spread function p (x)
The function g ′ (x) = g (x) by further performing convolution shaping processing with the function p ′ (x)
{P ′ (x) = f (x)} p (x) {p ′ (x)} is obtained. The intensity distribution function G (x) of the object image on the other light receiving portion of the sensor is a convolution f (x) of the intensity distribution function f (x) of the object image and the point spread function p ′ (x). ) ＠P ′ (x), which is a function p ′ (x) symmetrical to the point spread function p ′ (x) and obtained by further performing convolution shaping processing, G ′ (x ) = G (x) ＠p (x) = f (x) ＠ ｛p ′
(X) {p (x)}, and the function g ′ (x) and the function G ′
(X) becomes equal.

Note that ＠ indicates a convolution operation. For example, f (x) ＠p (x) is ∫f (z) × p (z−x) d
z. Next, a description will be given of a shaping operation when the focus detection optical system 6 is configured as shown in FIG. FIG.
In the arrangement of the re-imaging lenses 66 and 67, as described above, the point image distribution of the re-imaging lenses 66 and 67 with respect to the thick line in FIG. The distance between the re-imaging lens 67 and the optical axis is closer to the optical axis than the re-imaging lens 66 and the amount of eccentricity is reduced.

This point image is sampled by a plurality of pixels of the light receiving units 71 and 72, and AD-converted data is as shown in FIGS. 17 and 18, respectively. In this state, the photographing optical system 3 is in a focused state, and the pair of images must match without any displacement. In FIG. 17, a (n-2), a
Data other than (n-1), a (n), a (n + 1), and a (n + 2) are almost 0, and in FIG. 9, b (n
Data other than -1), b (n) and b (n + 1) are almost 0
Is the value of The value of each data is a (n−2) = k1,
a (n-1) = k2, a (n) = k3, a (n + 1) =
k2, a (n + 2) = k1, b (n-1) = k4, b
(N) = k5, b (n + 1) = k4, and the two image data do not match as they are.

Therefore, the shaping means 82 has the same waveform function as in FIG. 1 for the image data sequence ai in FIG.
9, a shaping process (convolution operation) is performed on the weight coefficient sequence Ai (i = 1 to 3), and the calculation is converted into a point spread function data sequence a ′ i. Also, a shaping process (convolution operation) is performed on the image data sequence bi in FIG. 18 using the weight coefficient sequence Bi (i = 1 to 5) having the same waveform function as in FIG. It is converted into a distribution function data sequence b'i.

However, the value of the weight coefficient is A1 = α × k4, A
2 = α × k5, A3 = α × k4, B1 = α × k1, B2
= Α × k2, B3 = α × k3, B4 = α × k2, B1 =
α × k1, the value of which is obtained by calculating the point spread function of the re-imaging lens of the focus detection optical system by an optical design program in advance, and is fixedly stored in the shaping means stage.

The above-mentioned convolution operation is further performed on the image data sequence ai in FIG.
Each of the eight image data strings bi is performed by the equation 11 to form shaping data a ′ i (i = 2 to m−1),
b ′ i (i = 3 to m−2) is detected.

[0041]

(Equation 10)

[0042]

## EQU11 ## The shaping process data a 'thus obtained
As shown in FIGS. 21 and 22, the individual values of i, b ′ i completely match without any deviation, and therefore, the focus detection calculation means using the shaping data a ′ i, b ′ i 83
Accordingly, if the calculation of the focus detection is performed, the focus can be detected without error.

In FIGS. 21 and 22, the values of the shaping data a ′ i and b ′ i have values other than 0 only for i = n−3 to n + 3.

[0044]

As described above, the shaping process is performed on the image data sequence ai with the weight coefficient sequence Ai (i = 1 to 3), and the weight coefficient sequence Bi (i = 1 to 5) is performed on the image data sequence bi. 21), the values of the shaping process data a ′ i and b ′ i can completely match without any deviation as shown in FIGS. 21 and 22.

The above description has been made on the case where the subject is a point image, but the present invention is also applicable to a general subject image other than the point image. For example, assuming that the point spread functions of the re-imaging lenses 66 and 67 are p (x) and q (x), respectively, the intensity distribution function g (x) of the subject image on one light receiving portion of the sensor is Given by the convolution f (x) ＠p (x) of the intensity distribution function f (x) and the point spread function p (x),
By further performing the convolution shaping process using the function q (x), the function g ′ (x) = g (x) ＠q
(X) = f (x) {p (x) {q (x)} is obtained. The intensity distribution function G (x) of the subject image on the other light receiving portion of the sensor is represented by the intensity distribution function f of the subject image.
Convolution f of (x) and point spread function q (x)
(X) ＠q (x), and a function G ′ (x) = G (x) ＠p (x) = f () obtained by further performing convolution shaping processing with a function p (x) x) ＠ ｛q
(X) {p (x)}, so that the function g ′ (x) and the function G ′ (x) are equal.

The above-mentioned ＠ indicates a convolution operation in the same manner as described above. For example, f (x) ＠p (x) is ∫f
(Z) × p (z−x) dz is shown. As described above, according to the present embodiment, of each of the pair of subject image data formed by the pair of re-imaging lenses, the point spread function by the other re-imaging lens is convoluted and shaped. Thereby, the degree of coincidence between the pair of subject image data can be restored, and as a result, the focus detection accuracy can be improved.

Although the shaping process is performed by digital arithmetic processing after AD conversion of a pair of subject image signals, the shaping process is performed analogously by an analog filter in the state of the analog signal before the AD conversion. Is also good. Further, when the height of the light receiving portion of the sensor 7 in the direction perpendicular to the arrangement direction of the pixels cannot be ignored, the concept of the point image distribution in the above description may be replaced with a line image distribution.

In each of the above embodiments, the photographing optical system 3
Is a TTL type focus detection device that re-images a light beam having passed through as a subject image on a sensor by a focus detection optical system having a pair of re-imaging lenses.
As described above, even in an external light type focus detection device that forms a pair of images with a pair of lens systems and detects the image shift,
When the characteristics of the pair of lens systems are different and the point spread functions are different, the present invention can be applied.

[0049]

As described above, according to the present invention, even if the point spread functions of the pair of re-imaging lenses of the focus detecting optical system are different, highly accurate focus detection can be performed by the aforementioned shaping means. Since it becomes possible, it is not necessary to make the re-imaging lens aspherical in order to make the point spread functions coincide or to form the re-imaging lens with a plurality of optical elements, which is advantageous in cost.

In addition, since there is no restriction that the point spread functions coincide when the focus detection area is arranged off-axis, the focus detection area can be arranged farther from the optical axis.

[Brief description of the drawings]

FIG. 1 is a diagram corresponding to claims of the present invention.

FIG. 2 is a block diagram of a configuration showing an embodiment of the present invention.

FIG. 3 is a perspective view illustrating a configuration of a focus detection optical system when a focus detection area is arranged at the center of a shooting screen. ,

FIG. 4 is a perspective view for explaining a configuration of a focus detection optical system when a focus detection area is arranged outside an optical axis on a photographing screen.

FIG. 5 is a waveform diagram when the light intensity distribution of a pair of primary images obtained by the imaging optical system 3 is expressed along the parallel direction of the light receiving units 71 and 72 in FIG.

6 is a waveform diagram showing a light intensity distribution of a secondary image obtained by the re-imaging lens 66 in FIG.

FIG. 7 is a waveform diagram showing a light intensity distribution of a secondary image obtained by the re-imaging lens 67 in FIG.

FIG. 8 is a waveform diagram showing data when the light intensity distribution of the secondary image obtained by the re-imaging lens 66 in FIG. 3 is sampled by a plurality of pixels of the light receiving unit 71 and subjected to AD conversion. is there.

9 is a waveform diagram showing data when the light intensity distribution of the secondary image obtained by the re-imaging lens 67 in FIG. 3 is sampled by a plurality of pixels of the light receiving unit 72 and subjected to AD conversion. is there.

FIG. 10 is a waveform diagram when the waveform diagram of FIG. 6 is inverted left and right.

FIG. 11 is a waveform diagram when the waveform diagram of FIG. 7 is inverted left and right.

FIG. 12 is a waveform diagram showing shaping processing data when the output signal of one light receiving unit 71 is convolved by the shaping means 82.

13 is a waveform chart showing shaping processing data when the output signal of the other light receiving section 72 is convolved by the shaping means 82. FIG.

FIG. 14 is a waveform diagram when the light intensity distribution of a pair of primary images is expressed along the parallel direction of the light receiving units 71 and 72 in FIG. 4, as in FIG.

FIG. 15 is a waveform diagram showing a light intensity distribution of a secondary image obtained by the re-imaging lens 66 in FIG.

16 is a waveform diagram showing a light intensity distribution of a secondary image obtained by the re-imaging lens 67 in FIG.

FIG. 17 is a waveform diagram showing data when the light intensity distribution of the secondary image obtained by the re-imaging lens 66 in FIG. 4 is sampled by a plurality of pixels of the light receiving unit 71 and subjected to AD conversion. is there.

18 is a waveform diagram showing data when the light intensity distribution of the secondary image obtained by the re-imaging lens 67 in FIG. 4 is sampled by a plurality of pixels of the light receiving unit 72 and subjected to AD conversion. is there.

FIG. 19 is the same waveform diagram as FIG.

FIG. 20 is the same waveform diagram as FIG.

21 is a waveform chart showing shaping processing data when the output signal of the light receiving unit 71 in FIG. 4 is convolved by the shaping means 82. FIG.

22 is a waveform diagram showing shaping processing data when the output signal of the light receiving section 72 in FIG. 4 is convolved by the shaping means 82. FIG.

[Explanation of symbols]

Reference Signs List 1 camera body, 2 lens, 3 shooting optical system 6 focus detection optical system, 66, 67 re-imaging lens,
71, 72 light receiving section 7 sensor, 82 shaping means, 83 focus detection calculating means

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G02B ^7/ 28-7/34 G03B 13/36

Claims (4)

(57) [Claims] An imaging optical system for forming an image of a subject on a predetermined focal plane includes a pair of light receiving units, and generates a pair of output signals corresponding to a light intensity distribution on the pair of light receiving units. a sensor, the object image formed on a part of the focus detection area of the predetermined focal plane, a pair of focus detecting optical system comprising a pair of re-imaging lens for re-imaged on the light receiving portion And said
A point light source is provided on the pair of light receiving portions by a pair of re-imaging lenses.
Are different from each other when re-imaging the image of
Focus detection optical system, and the degree of coincidence of the intensity distribution due to the difference between the pair of point spread functions
The intensity distribution for the pair of output signals whose
Shaping processing for restoring the degree of coincidence of the pair of point image distribution
Performing based on the number, shaping means for generating a pair of processing signals by the shaping processing, focus detection calculation means for performing image shift detection calculation on the pair of processing signals, and calculating the focus state of the imaging optical system, A focus detection device comprising: 2. The method according to claim 1, wherein the shaping unit calculates a pair of point spread functions corresponding to the pair of re-imaging lenses.
Each of the pair of output signals generated by the sensor corresponding to the pair of re-imaging lenses is stored in advance and the other of the pair of output signals not corresponding to the output signal.
2. The focus detection device according to claim 1 , wherein the point spread function of the imaging lens is convolved. 3. A method according to claim 1, wherein the part of the focus detection area is disposed on an optical axis of the photographing optical system, and the part of the pair of point image
The distributions of the cloth functions are asymmetrical, and
3. The method according to claim 1, wherein the light-emitting element is line-symmetric.
The focus detection device according to any one of the preceding claims. 4. The focus detection device according to claim 1, wherein the part of the focus detection area is disposed off the optical axis on the optical axis of the photographing optical system.