JPH05232380A - Focus detecting device for photographing device - Google Patents

Abstract

PURPOSE:To provide a focus detecting device for a photographing device in which errors in frequency detection caused by the influence of the end of a photoelectric conversion element train are reduced and whose constitution is simple. CONSTITUTION:A film surface 15 is constituted between the CCD parts 17a, 17b of a CCD sensor 17 so that it may equivalently exist. The CCD part 17a consists of 100 photoelectric conversion elements, such as the photoelectric conversion elements 17a1-17a100, and the photoelectric conversion elements are arranged in a shape obtained by dividing a doughnut shape into 100 equal parts in a circumferential direction. Then, the photoelectric conversion elements are symmetric with respect to the circumferential direction, and there is no real discrimination of an origin but 100 elements are assumed to be arranged toward the element 17a2 by considering the element 17a1 as the origin for the sake of convenience. The photoelectric conversion element 17b consists of 100 photoelectric conversion elements, such as the photoelectric conversion elements 17b1-17b100 in the same way.

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 for a photographing device such as a camera, which detects a focus on a photographing surface using a contrast method using a contrast of a subject image and measures a distance to the subject. Regarding

[0002]

2. Description of the Related Art Conventionally, in a focus detecting device for a photographing apparatus using a contrast method, a photoelectric conversion element array arranged in a line or an area for detecting the contrast is arranged horizontally or vertically with respect to the screen. The output is extracted linearly and continuously, and the focus state is detected by detecting the spatial frequency component of the subject image that is peculiar to the image quality in the focused state. Therefore, a method of extracting a component of a predetermined frequency using a filter, a method of obtaining a differential value, a difference, etc.,
A frequency component (MTF) method using Fourier transform is known. Further, JP-A-59-146008 discloses a technique of arranging light receiving elements for distance measurement in an annular shape.

In this technique, a split mask or a fly-eye lens is used to split the subject light incident on the same image formation planned position according to the incident angle, and the divided subject light is measured by two sets of light receiving element rows. To do. Then, the subject images received by the two sets of light receiving element rows are laterally displaced depending on the front focus and the rear focus with respect to the planned imaging position. The present invention relates to a phase difference type focus detection method capable of measuring the state of focus deviation and the magnitude of deviation by detecting the magnitude and direction of this deviation.

As described above, according to the present technology, by arranging the light receiving elements on the circumference centered on the optical axis, the distances of the respective elements from the optical axis are made constant, and the subject due to the difference in the distance from the optical axis. A feature is that a measurement error caused by a difference in incident angle of light or a light amount is eliminated.

[0005]

However, in the above-mentioned technique, pixels composed of photoelectric conversion elements are arranged linearly,
Since this is a method of evaluating the degree of blurring of the subject image in that direction as the contrast, in the filter processing and the difference processing, a large amount of error is included in the detection value at the end portion. In particular, when the frequency is extracted by filtering or Fourier transform, the error in the contrast information at the low frequency side becomes large due to the end thereof.

Further, since the horizontal and vertical contrasts of the subject are detected independently, in order to deal with all the subjects having various frequencies in the vertical and horizontal directions, the contrast in a plurality of directions of the screen is processed, which is a complicated structure. Need.

Further, in the case of the focus detection by the phase difference method disclosed in the above-mentioned Japanese Patent Laid-Open No. 59-146008, it is necessary to perform a process for obtaining the amount of deviation between the two obtained subject images. It can be said that it is irrelevant to the process of extracting the contrast information such as the component amount of the predetermined frequency and the spatial frequency component included in the image.

The present invention has been made in view of the above problems, and an object of the present invention is to provide a focus detection apparatus which eliminates a frequency detection error due to the influence of an end portion of a photoelectric conversion element array and has a simple structure. To do.

[0009]

In order to achieve the above object, in a focus detecting apparatus of the present invention, an image forming optical system for forming an image of a subject light and a near image forming surface of the image forming optical system are provided. A photoelectric conversion element array arranged in a ring, a reading means for reading out the output of the photoelectric conversion element array as a continuous brightness distribution signal, and a continuous brightness distribution signal read by the reading means for arithmetic processing The present invention is characterized by comprising a calculating means for calculating the frequency component of the luminance distribution and a detecting means for detecting the focus state of the imaging optical system based on the frequency component calculated by the calculating means.

[0010]

That is, in the focus detecting apparatus of the present invention, when the light to be imaged is formed by the image forming optical system, the photoelectric conversion element array arranged substantially annularly near the image forming surface of the image forming optical system. The output is read by the reading means as a continuous luminance distribution signal. Then, the continuous luminance distribution signal read by the reading means is processed by the calculating means,
The frequency component of the brightness distribution is calculated. Then, the focus state of the imaging optical system is detected by the detection means based on the frequency component calculated by the calculation means.

[0011]

1 is a diagram showing the configuration of a first embodiment of the present invention.

As shown in FIG. 1A, a photoelectric conversion element array 2 is annularly two-dimensionally arranged near the image forming surface of an image forming optical system 1 for forming an image of light from a subject. The photoelectric conversion element array 2 is connected to a subject luminance distribution detection unit 3 that detects the subject luminance distribution in a ring shape.

A subject image contrast detector 4 for detecting contrast information of the subject image is connected to the subject luminance distribution detector 3, and the subject image contrast detector 4 has a focus state of the subject image. The focus detection unit 5 for detecting the is connected.

In such a configuration, the output of the ring-shaped photoelectric conversion element array 2 arranged two-dimensionally in a ring is read by the subject brightness distribution detection unit 3 according to the array so that the starting point and the ending point coincide.

The signal from the subject luminance distribution detecting section 3 is treated as a repetitive signal in order to be processed by the subject image contrast detecting section 4 in a practically endless state. All of 2 will receive the same treatment and be effectively used. Further, since the repeated processing is performed as an actual continuous signal, the contrast information of the spatial frequency of the low-frequency subject image can be handled. Furthermore, since the pixel rows are arranged in a ring shape, the spatial frequency can be detected by all methods on the screen.

Further, as shown in FIG. 1 (b), by using the annular photoelectric conversion element arrays 2a and 2b each having a predetermined optical path length, the annular photoelectric conversion from the imaging optical system 1 with different optical path lengths is performed. The subject brightness distribution detecting unit 3 detects each subject brightness distribution from a plurality of output signals from the element arrays 2a and 2b, the subject image contrast unit 4 obtains each contrast information, and the focus detecting unit 5 orthogonally crosses the optical axis. It is also possible to detect the focus shift amount and the shift direction with respect to the focus state on the reference plane.

Further, as shown in FIG. 1C, an optical system driving section 6 for changing the optical path lengths of the imaging optical system 1 and the photoelectric conversion element array 2 is provided, and the photoelectric conversion element array 2 with different optical path lengths is provided. It is also possible to use the output signal from the optical disc and the information of the position and the optical path length of the imaging optical system 1 when the output signal is detected.

Further, as shown in FIG. 1D, an image pickup section 7 for taking out a subject image on a reference plane orthogonal to the optical axis.
Is provided, and the imaging optical system 1 is also used as a photographing optical system,
By detecting the focus state on the imaging surface, TTL (T
It can also be configured to be applied to a focus detecting device of the hrough the lens type.

Further, the focus detection section 5 uses a plurality of sets of output signals from the photoelectric conversion element array 2 with different optical path lengths on the optical axis of the imaging optical system 1, so that the reference plane orthogonal to the optical axis. The distance from the subject to the subject can be detected, and the present invention can be applied to a focus detection device that is an external light distance measuring device. Hereinafter, the present invention will be described with reference to a TTL type frequency component (MTF) of a camera.
A second embodiment applied to the type focus detection device will be described. FIG. 2 is a diagram showing a schematic cross section of the camera.

As shown in the figure, in the body 18 of the camera, a half mirror is provided on the optical path of the light passing through the lens 10 for photographing so as to divide the subject image, and a quick return function is provided. A main mirror 13 is provided.

On the optical path of the light reflected by the main mirror 13, a pentaprism 11 and an eyepiece lens 1 for transferring a subject image through a screen 19.
Two are provided.

On the other hand, a sub-mirror 14 is provided on the optical path of the light transmitted through the main mirror 13, and a split prism 16 is provided on the optical path of the light reflected by the sub-mirror 14.

A charge coupled device (CCD) sensor 17 is provided on the optical path of the subject light having different optical path lengths divided by the division prism 16. Further, the CCD sensor 17 includes a CCD section 17a including a photoelectric conversion element array arranged in a ring and a CCD.
It is configured by the portion 17b. In addition, a central processing unit (CPU) 18 for performing various kinds of signal processing and detecting a focus state is provided in the body 18.

In such a structure, the light made into a parallel light flux by the lens 10 is reflected by the main mirror 13, and the reflected light is incident on the pentaprism 11 via the screen 19 and reflected by the reflecting surface thereof. The light is guided to the eyepiece lens 12.

On the other hand, the light transmitted through the main mirror 13 is reflected by the sub mirror 14, and the reflected light is split by the split prism 16 into lights having different optical path lengths, which are received by the CCD sections 17a and 17b of the CCD sensor 17, respectively. It Then, the focus state is detected by the CPU 18 according to the light receiving results of the CCD units 17a and 17b. Next, FIGS. 3A and 3B show the CCD sensor 17
It is a figure which shows the relationship between the CCD parts 17a and 17b of FIG. As shown in the figure, the film surface 15 is equivalently arranged between two photoelectric conversion element arrays on the optical axis, that is, between the CCD units 17a and 17b.
The optical path length difference between the film surface 15 and the CCD unit 17a is da, and the optical path length between the film surface and the CCD unit 17b is db.
(Here, da = db). Figure 3 (c) is a CCD
FIG. 3 is a diagram showing the arrangement of photoelectric conversion elements on the sensor 17. As shown in the figure, the CCD section 17a and the CCD section 17b have the same structure and are formed on the CDD section 17, respectively.

The CCD section 17a includes a photoelectric conversion element 17a1.
To 17a100 consisting of 100 photoelectric conversion elements. These photoelectric conversion elements are formed in a donut shape divided into 100 equal parts in the circumferential direction. Each photoelectric conversion element is symmetrical in the circumferential direction, and although there is no actual starting point distinction, 17a2 is set as a starting point at 17a1 for the sake of convenience.
It is assumed that 100 pieces are arranged in the direction of. CCD section 17
Similarly for b, photoelectric conversion elements 17b1 to 17b1
It consists of 100 photoelectric conversion elements up to b100. In addition, on the CCD sensor 17, an interface unit 17c is arranged for exchanging data with the outside and controlling the integration timing to each pixel. FIG. 4 shows the CC
It is a figure which shows the connection relationship of the D sensor 17 and CPU18.

The signals of the CCD units 17a and 17b, which are two annularly arranged photoelectric conversion element arrays on the CCD unit 17, are:
Controlled by the interface section 17c, the data of each pixel is converted into photoelectric conversion elements 17a1 to 17a100, 17b1 to 17.
The data is output to the CPU 18 in the order of b100.

This signal is converted into a digital signal by the analog / digital converter 20 in the CPU 18,
The signal from the CCD unit 17a is supplied to the subject brightness distribution detection unit 2
The signal from the CCD section 17b is separated by the subject brightness distribution detecting section 22.

Then, the separated signals are processed by the subject image MTF detectors 23 and 24 to extract the components of a predetermined frequency. The frequency component value by the CCD unit 17a is MTFa, and the frequency component value by the CCD unit 17b is MTFb. The MTF ratio focus detection unit 25 calculates (MTFa
The value of / MTFb) and the data specific to the taking lens are converted into a signal of a focus shift amount and a shift direction.

Here, when compared with FIG.
The D section 17a and the CCD section 17b correspond to the ring-shaped photoelectric conversion element array 2, the interface 17c, the analog / digital (A / D) converter 20, and the subject brightness distribution detection section 21.
Reference characters a and 22a correspond to the subject brightness distribution detection unit 3. The subject image MTF detectors 23a and 24b correspond to the subject image contrast detector 4, and the MTF ratio focus detector 25 corresponds to the focus detector 5. Hereinafter, detection of frequency components will be described with reference to FIG.

As shown in the figure, the signals from the photoelectric conversion elements 17a1 to 17a100 of the CCD 17a are arranged in a ring shape and the photoelectric conversion elements 17a1 and 17a100 are adjacent to each other. However, it can be treated as a continuous waveform. Therefore, the error in extracting the component of the predetermined frequency by performing the Fourier transform by the known method, especially the error in the low frequency detection is reduced. Further, in a linear arrangement, the values at both ends are discontinuous and cannot be treated as a repetitive waveform, so that the error when Fourier transform is performed becomes large. The operation of the CPU 18 will be described below with reference to the flowchart shown in FIG.

As shown in the figure, first, the CCD sensor 17
Are driven to obtain signals of the luminance distribution of the subject image with different optical path lengths by the CCD units 17a and 17b (step S1).
01). Next, with respect to the luminance distribution information, the data of each pixel is digitally converted using the A / D converter 20 (step S102).

Then, from the digitally converted values, data corresponding to the respective pixel columns, that is, the CCD units 17a and 17b are collected (steps S103 and S104). These are sequential data according to the pixel arrangement. Next, a predetermined frequency component is extracted from the data corresponding to the CCD unit 17a by using Fourier transform (step S105).

Since the extraction frequency in this case corresponds to a large focus shift, per 1 mm on the film surface,
A frequency of about 0.2 to 1 Hz is set to about 1 to 10 Hz for detecting a medium focus shift, and a frequency of about 10 to 20 Hz for enhancing focusing accuracy in the vicinity of focusing. In this embodiment, this predetermined frequency is 0.5, 5, 15 Hz. Then, the intensities for the three types of frequencies detected as a result of the Fourier transform are set to MTFa0.5, MTFa5, and MTFa15, respectively. Similarly, MTFb0.5 corresponding to the CCD unit 17b
, MTFb5, MTFb15 are calculated (step S106). Next, for each of these three types of frequencies, a ratio is obtained from the frequency components (step S107). As is well known, the value of this ratio is (MTFb
/ MTFa) or (MTFa-MTFb) / (MT
Fa + MTFb) and the like.

Next, these calculated frequency component ratio values are converted into focus shift amounts and directions in the form of table reference of frequency component ratio data based on known lens characteristics (step S108). In this case, the two sensor units 1
The optical path differences da and db from the film surface of 7a and 17b are used. In this way, the amount and direction of focus shift on the film surface for two types of frequencies are detected. Hereinafter, with reference to the flowchart of FIG. 7, a process when the focus shift amounts of the three different frequencies do not match will be described. Each calculated frequency 0.5,
The frequency component ratios at ₅ , 15 Hz are R _0.5 ,
Let them be R ₅ and R ₁₅ . First, the frequency component ratio R _0.5 is converted into focus shift information d _0.5 using the data of the lens characteristics and the optical path length (step S200).

Similarly, the frequency component ratios R ₅ and R ₁₅ are converted into focus shift information d ₅ and d ₁₅ using the lens characteristics and the optical path length data, respectively (steps S201 and S20).
2).

The focus shift information d _0.5 , d ₅ , d
Reference numeral ₁₅ is a signal including the focus shift amount and direction. Then, the lower the frequency is, the higher the accuracy at the time of a large focus shift is. Therefore, the focus shift information at each frequency is checked with priority from the low frequency side. Next, it is checked whether the difference between the focus shift information d _0.5 and d ₅ is less than or equal to a predetermined value Th (step S203).

When this difference is large, it is judged that the accuracy of the focus shift information at a high frequency of 5 Hz or higher is low, and the focus shift information d is set to d _0.5 , and this routine is exited (step S207). When the difference is smaller than the predetermined value Th, d = (d _0.5 + d ₅ ) / 2 and the average value of d _0.5 and d ₅ is substituted into the focus shift information d (step S20).
4). Next, the difference between the focus shift information d and d ₁₅ is a predetermined value T.
It is checked whether it is smaller than h (step S205).

When this difference is small, it is judged that the reliability of the focus shift information d ₁₅ is also high, and the average value of d _0.5 , d ₅ , and d ₁₅ is substituted for the focus shift information d, and the routine is executed. Ends (step S206).

When the above difference is large, it is judged that the accuracy of the focus shift information d ₁₅ is poor, and d _0.5 and d ₅ obtained earlier are determined.
The routine is ended with the focus deviation information d depending on the average value of.

By the way, in the above-mentioned embodiment, two sets of photoelectric conversion element arrays are used, but it is also possible to detect the focus shift amount and the shift direction in the same way even with one set. In this case, it is only necessary to compare the contrast information of the subject images with different optical path lengths, and thus the distance between the lens and the photoelectric conversion element array may be changed using the optical system driving unit 6. Next, FIG. 8 is a diagram showing a configuration of the CCD sensor 17. As shown in the figure, the CCD sensor 17 has CCD parts 17a and 17b.
And an interface unit 17c.

The optical system drive unit 6 includes a motor, a reduction gear train, a lens drive unit 6 including a power control circuit for the motor, a CCD sensor 17 and a lens for detecting the optical path length of the lens 10. The position detector 27 is included.

With such a configuration, the CCD sensor 17
Outputs the subject brightness distribution information, and the optical system driving unit 6 outputs the optical path length of the optical system when the CCD sensor 17 measures the subject image brightness distribution information.

As a result, similarly to the first embodiment described above, the focus state on the film surface can be detected by detecting the contrast of the object by measuring the object image a plurality of times.

If the distance da between the CCD unit 17a and the film surface is set to "0", the detected contrast becomes equivalent to the contrast on the film surface, and the degree of focus is obtained by one contrast detection. It is possible to detect. In this case, it is difficult to detect the direction of the focus shift, but the degree of focus shift can be detected.

Further, when the characteristics of the position and the focus shift of the lens 10 and the focus shift of the lens 10 can be understood, the phase difference focus detection unit disclosed by the applicant of the present application in Japanese Patent Application Laid-Open No. 2-282333 is used. Similar to the conventional device for outputting the object distance, the object distance can be detected in the first embodiment described above. As described above, the present invention can be applied to a distance measuring device that detects a subject distance, and can be similarly applied to a passive distance measuring device. The passive range finder will be described below.

Since the conventional external light type passive distance measuring device for a camera is a triangular distance measuring method using a phase difference as shown schematically in FIG. 9 (b), the longer a certain base line length is, the more distance detection is performed. The accuracy of is increased.

On the other hand, the passive distance measuring apparatus to which the present invention is applied has a lens 28 for image formation and a half mirror 29 for splitting the subject light, as shown schematically in FIG. 9 (a). And a mirror 30 for guiding a part of the split light onto the CCD sensor 17, a CCD sensor 17 and a CCD section 17a and a CCD section 17b formed thereon.

With such a structure, the base line length is not required, so that a small distance measuring device can be manufactured. still,
The signal processing may be performed on the basis of the in-focus point of the subject at infinity instead of the film surface as in the first embodiment.

An application example of the present invention will be described below with reference to FIG. 10 in the case of using a CCD sensor for image pickup used in, for example, a video camera, in which pixels are two-dimensionally evenly spread.

In the figure, the square mark is the pixel 100, and the circled pixel is the annular photoelectric conversion element 10 used in the present invention.
It is 1. The subject brightness detection unit 3 sequentially extracts adjacent pixel data of circled pixels from all the pixels to detect a ring-shaped subject brightness distribution. In this way, it is possible to use a general area sensor. Further, as shown in FIG. 11A, it is possible to use a part of the photoelectric conversion element array a plurality of times within an array of one round.

Further, according to the arrangement order shown in FIG. 11B, it is possible to handle the subject brightness distribution in an 8-shaped annular arrangement by switching the distance measuring areas to a wide area.

By handling the areas in a circular shape in accordance with the arrangement order shown in FIG. 11C, it is possible to switch the distance measuring areas such that the areas are divided into a plurality of narrow areas.

As described above in detail, according to the present invention, the photoelectric conversion elements for detecting the subject image are arranged two-dimensionally in a ring shape, and the output of the photoelectric conversion element array is arranged so that the start point and the end point coincide with each other. Since it is read out according to the array, it is possible to perform processing as a repetitive signal in a practically endless state when detecting the subject image contrast from the subject luminance distribution information. Due to such an effect, all the photoelectric conversion elements are treated equally and can be effectively used.

Further, since the repeated processing is performed as a realistic continuous signal, the contrast information of the low frequency component of the spatial frequency of the subject image can be handled and the focus state can be accurately detected even when the focus is largely defocused. Like Further, since the pixel rows are arranged in a ring shape, the spatial frequency in all directions on the screen can be detected.

Although the embodiments of the present invention have been described above, it is needless to say that the present invention is not limited to these and can be improved and changed. For example, in the present invention, the photoelectric conversion elements are arranged in an annular shape, but the photoelectric conversion elements are not limited to being arranged on the circumference centered on the optical axis, and may be arranged so that their start points and end points coincide.

[0057]

According to the present invention, it is possible to provide a focus detection apparatus for a photographing apparatus which eliminates a frequency detection error due to the influence of the end portion of the photoelectric conversion element array and has a simple structure and high accuracy.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a first embodiment of the present invention and a modification thereof.

FIG. 2 is a diagram showing a configuration of a second embodiment in which the present invention is applied to a TTL type MTF type focus detection device for a camera.

3 (a) and 3 (b) are views showing the relationship between CCD parts 17a, 17b of the CCD sensor 17 and the optical axis of the film surface, and FIG. 3 (c) is a layout of photoelectric conversion elements on the CCD sensor 17. FIG. FIG.

FIG. 4 is a diagram showing a detailed configuration of a CPU 18.

FIG. 5 is a diagram for explaining detection of frequency components.

FIG. 6 is a flowchart for explaining the operation of CPU 18.

FIG. 7 is a flowchart for explaining processing when the focus shift amounts for three different types of frequencies do not match.

FIG. 8 is a diagram showing a configuration of a CCD sensor 17.

FIG. 9 is a diagram for explaining a passive distance measuring device.

FIG. 10 is a diagram for explaining an application example of the present invention in which a CCD sensor for imaging, which is used in, for example, a video camera, in which pixels are uniformly spread in two dimensions is used.

FIG. 11 is a diagram for explaining the arrangement and arrangement order of photoelectric conversion elements.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Imaging optical system, 2 ... Annular photoelectric conversion element array, 3 ... Subject brightness distribution detection part, 4 ... Subject image contrast detection part, 5
... focus detection unit, 6 ... optical system drive unit, 7 ... imaging unit.

[Procedure amendment]

[Submission date] April 30, 1992

[Procedure Amendment 1]

[Document name to be amended] Statement

[Correction target item name] 0009

[Correction method] Change

[Correction content]

[0009]

In order to achieve the above object, in a focus detection apparatus of the present invention, an image forming optical system for forming an image of a subject light and an annular shape near the image forming surface of the image forming optical system. The photoelectric conversion element array arranged in the above, the reading means for reading out the output of the photoelectric conversion element array as a continuous brightness distribution signal, and the continuous brightness distribution signal read by the reading means is processed to calculate the brightness. The present invention is characterized by comprising a calculating means for calculating the frequency component of the distribution and a detecting means for detecting the focus state of the imaging optical system based on the frequency component calculated by the calculating means.

[Procedure Amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0010

[Correction method] Change

[Correction content]

[0010]

That is, in the focus detection apparatus of the present invention, when the subject light is imaged by the imaging optical system, the output of the photoelectric conversion element array arranged in an annular shape near the imaging surface of the imaging optical system. But,
The reading means reads the continuous luminance distribution signal. Then, the continuous luminance distribution signal read by the reading means is arithmetically processed by the arithmetic means to calculate the frequency component of the luminance distribution. And
The focus state of the imaging optical system is detected by the detection means based on the frequency component calculated by the calculation means.

[Procedure 3]

[Document name to be amended] Statement

[Correction target item name] 0041

[Correction method] Change

[Correction content]

By the way, in the above-mentioned embodiment, two sets of photoelectric conversion element arrays are used, but it is also possible to detect the focus shift amount and the shift direction in the same way even with one set. In this case, it is only necessary to compare the contrast information of the subject images with different optical path lengths, and thus the distance between the lens and the photoelectric conversion element array may be changed using the optical system driving unit 6. Next, FIG. 8 is a diagram showing a configuration of the CCD sensor 17. As shown in the figure, the CCD sensor 17 comprises a CCD unit 17a and an interface unit 17c.

[Procedure amendment 4]

[Document name to be corrected] Drawing

[Correction target item name] Figure 8

[Correction method] Change

[Correction content]

[Figure 8]

Claims (1)

[Claims] 1. An image forming optical system for forming an image of light to be imaged, a photoelectric conversion element array arranged substantially annularly near an image forming surface of the image forming optical system, and an output of the photoelectric conversion element array being continuous. Read out as a brightness distribution signal, a calculation means for calculating a frequency component of the brightness distribution by processing the continuous brightness distribution signal read out by the read means, and a frequency calculated by the calculation means. A focus detection device comprising: a detection unit that detects a focus state of the imaging optical system based on a component.