Image sensor

Abstract

The present invention provides an image sensor capable of detecting focus even during periods of significant defocus. [Solution] The image sensor comprises a plurality of first pixel rows having a plurality of first pixels, a plurality of second pixel rows having a plurality of second pixels, a plurality of third pixels that output signals used for imaging, and a circuit that selects either the first pixel row or the second pixel row based on information regarding the exit pupil position of the imaging optical system, wherein the first pixels receive light through an opening provided at a first position of the light-shielding portion, and the second pixels receive light through an opening provided at a second position of the light-shielding portion, and the image sensor comprises a first region where only one of the plurality of first pixel rows and the plurality of second pixel rows and the third pixels are arranged, and a second region which is the same size as the first region but has a higher image height in the first direction than the first region, and where both the plurality of first pixel rows and the plurality of second pixel rows and the third pixels are arranged.

Description

Technical Field

[0001] The present invention relates to an imaging device.

Background Art

[0002] Conventionally, an imaging device provided with focus detection pixels for performing pupil division type focus detection on an imaging surface has been known. Conventionally, there has been a problem that it is difficult to perform focus detection when the focus is greatly off (at large defocus).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

[0004] According to a first aspect, an imaging device includes: a plurality of first pixel columns having a plurality of first pixels arranged along a first direction, which receive light transmitted through an imaging optical system and output a signal used for focus detection; a plurality of second pixel columns having a plurality of second pixels arranged along the first direction, which receive light transmitted through the imaging optical system and output a signal used for focus detection; a plurality of third pixels which receive light transmitted through the imaging optical system and output a signal used for imaging; and a circuit which selects either one of the first pixel columns and the second pixel columns based on information regarding the position of the exit pupil of the imaging optical system. The first pixel receives light through an aperture provided at a first position of a light shielding portion, the second pixel receives light through an aperture provided at a second position different from the first position of the light shielding portion, and the imaging device includes: a first region in which only one of the plurality of first pixel columns and the plurality of second pixel columns and the third pixels are arranged; and a region having the same size as the first region, in which the image height in the first direction is higher than that in the first region, and both the plurality of first pixel columns and the plurality of second pixel columns and the third pixels are arranged.

Brief Description of the Drawings

[0005] [Figure 1] This is a schematic cross-sectional view showing the configuration of a camera system according to the first embodiment of the present invention. [Figure 2] This is a schematic plan view showing the imaging surface 20 of the image sensor 210. [Figure 3] This is an explanatory diagram of the imaging and focus detection pixel 301. [Figure 4] This is an explanatory diagram of the first focus detection pixels 302NR and NL. [Figure 5] This is an explanatory diagram of the second focus detection pixels 302SR and SL. [Figure 6] This is an explanatory diagram of the third focus detection pixels 302LR and LL. [Figure 7] This is a schematic plan view showing the imaging surface 20 of the image sensor 210. [Figure 8] This is an enlarged view of a portion of the first region 60a. [Figure 9] This is an enlarged view of a portion of area 60b in region 2. [Figure 10] This is an enlarged view of a portion of area 60c in region 3. [Figure 11] This is a flowchart of the control processing performed by the body CPU 220. [Figure 12] This is a flowchart of the AF process called from step S30 in Figure 11. [Figure 13] This is a flowchart of the AF process called from step S30 in Figure 11. [Figure 14] This is a schematic perspective view showing the structure of the image sensor 210a. [Figure 15] This is a schematic cross-sectional view showing the imaging / focus detection pixel 301 and the focus detection pixel 302. [Figure 16] This is a magnified plan view of a portion of the first region 60a of the image sensor 210a. [Figure 17] This is a magnified plan view of a portion of the second region 60b of the image sensor 210a. [Figure 18]This is a magnified plan view of a portion of the third region 60c of the image sensor 210a. [Figure 19] This is a schematic cross-sectional view showing the light beam from a light point P on the image plane S to be combined with the image sensor 210a. [Figure 20] This is an explanatory diagram of the first focus detection pixel related to a modified example. [Figure 21] This is an explanatory diagram of the imaging surface related to a modified example. [Figure 22] This is an explanatory diagram of the imaging and focus detection pixel related to a modified example. [Modes for carrying out the invention]

[0006] (First Embodiment) Figure 1 is a schematic cross-sectional view showing the configuration of a camera system according to the first embodiment of the present invention. In Figure 1, the parts particularly relevant to the present invention are shown, and other parts constituting the camera system 1 are omitted from the illustration. Hereinafter, the parts particularly relevant to the present invention will be described, and the description of other parts will be omitted.

[0007] Camera system 1 is a so-called single-lens reflex digital camera. Camera system 1 consists of an interchangeable lens 100 and a camera body 200. The user can select one of several types of interchangeable lenses 100 that are compatible with (can be attached to) the camera body 200, attach it to the camera body 200, and take pictures. Figure 1 shows one example of the several types of interchangeable lenses 100.

[0008] The camera body 200 and the interchangeable lens 100 have, for example, a bayonet-type lens mount mechanism. The interchangeable lens 100 is attached to the camera body 200 by fitting the mount portion of the interchangeable lens 100 into the mount portion of the camera body 200.

[0009] The interchangeable lens 100 includes an imaging optical system 110, a lens CPU 120, and an aperture 130. The imaging optical system 110 is composed of a lens 111 and a focusing lens 112. The imaging optical system 110 forms a subject image on the imaging surface of an image pickup device 210 described later. The position of the exit pupil of the imaging optical system 110 varies depending on the type of the interchangeable lens 100.

[0010] The camera body 200 includes an image pickup device 210, a body CPU 220, a focus adjustment unit 230, a ROM 240, and a display device 250. The image pickup device 210 is an image pickup device such as a CCD or a CMOS. The image pickup device 210 receives subject light that has passed through the imaging optical system 110 and outputs an imaging signal and a focus detection signal. The body CPU 220 is composed of a microprocessor (not shown) and its peripheral circuits. The body CPU 220 reads and executes a control program stored in advance in the ROM 240, which is a non-volatile storage medium, to control each part of the camera body 200 and perform data communication with the interchangeable lens 100. The focus adjustment unit 230 includes an actuator (not shown) and drives the focusing lens 112 in the optical axis direction based on the control of the body CPU 220 to adjust the focus of the imaging optical system 110. The display device 250 is a display device such as a liquid crystal display and displays a live view image, various setting screens, and the like.

[0011] The body CPU 220 includes a first focus detection unit 221 and a second focus detection unit 222 in software form. Each of these parts is realized software-wise by the body CPU 220 executing a predetermined control program stored in the ROM 240.

[0012] The first focus detection unit 221 detects the focus adjustment state of the imaging optical system 110 based on a focus detection signal output from an imaging and focus detection pixel described later. The second focus detection unit 222 detects the focus adjustment state of the imaging optical system 110 based on a focus detection signal output from a focus detection pixel described later.

[0013] (Description of the image pickup device 210) Figure 2(a) is a schematic plan view showing the imaging surface 20 of the image sensor 210. For the sake of explanation, in Figure 2(a), a coordinate system is set up where the X-axis is the direction parallel to the top and bottom edges of the imaging surface 20 (left-right direction on the paper), the Y-axis is the direction parallel to the left and right edges of the imaging surface 20 (top-bottom direction on the paper), and the Z-axis is the front-to-back direction of the imaging surface 20, i.e., the direction of the optical axis of the imaging optical system 110. The same coordinate system is set up in the following drawings as well.

[0014] Figure 2(a) shows a vertical line OX superimposed on the imaging surface 20, which bisects the imaging surface 20 in the left-right direction (X-axis direction), and a horizontal line OY, which bisects the imaging surface 20 in the up-down direction (Y-axis direction). The intersection of the vertical line OX and the horizontal line OY approximately coincides with the optical axis of the imaging optical system 110. Figure 2(b) shows a schematic diagram of an enlarged view of the region 20a near the center of the imaging surface 20 (near the intersection of the vertical line OX and the horizontal line OY) of the entire imaging surface 20 shown in Figure 2(a).

[0015] As shown in Figure 2(b), the imaging surface 20 has a large number of pixels 30 arranged in a two-dimensional square. The pixels 30 include two types of pixels: imaging and focus detection pixels 301 which are used for both imaging and focus detection, and focus detection pixels 302 which are used only for focus detection. However, in Figure 2(b), the two are not particularly distinguished. Below, we will first explain the imaging and focus detection pixels 301, paying particular attention to the imaging and focus detection pixel 301a located in the center of the imaging surface 20 (the intersection of the vertical line OX and the horizontal line OY).

[0016] Figure 3(a) is an enlarged plan view of the imaging and focus detection pixel 301a located at the center of the imaging plane 20 (the intersection of the vertical line OX and the horizontal line OY) among the pixels 30 shown in Figure 2(b), and Figure 3(b) is a cross-sectional view of the imaging and focus detection pixel 301a. The imaging and focus detection pixel 301a has a microlens 31, a color filter 32, and a pair of photoelectric conversion units 34L and 34R. The pair of photoelectric conversion units 34L and 34R have a shape that is approximately a semicircle obtained by dividing a circle along the Y-axis direction. That is, the pair of photoelectric conversion units 34L and 34R are arranged along the X-axis direction (focus detection direction). The width of the pair of photoelectric conversion units 34L and 34R in the X-axis direction is at most W1.

[0017] The microlens 31 focuses the incident light on the imaging and focus detection pixel 301a onto a pair of photoelectric conversion units 34L and 34R. This incident light enters the pair of photoelectric conversion units 34L and 34R via a color filter 32. The color filter 32 is formed to transmit either red, blue, or green light for each pixel, and not transmit other light. The color filter 32 of the imaging and focus detection pixel 301a is configured to form a so-called Bayer array. A wiring layer 33 is provided between the color filter 32 and the pair of photoelectric conversion units 34L and 34R. The wiring layer 33 is positioned between pixels so as not to obstruct the incident light to the pair of photoelectric conversion units 34L and 34R.

[0018] The dashed line 41 shown in Figure 3(b) represents the optical axis of the microlens 31 (hereinafter referred to as the optical axis 41). The dashed line 42 shown in Figure 3(b) represents the center of the division point between the photoelectric conversion unit 34L and the photoelectric conversion unit 34R (hereinafter referred to as the dividing line 42). The photoelectric conversion unit 34L and the photoelectric conversion unit 34R have shapes symmetrical with respect to the dividing line 42 and are arranged symmetrically with respect to the dividing line 42.

[0019] The pair of photoelectric conversion units 34L and 34R are, for example, photodiodes, and output a photoelectric conversion signal obtained by photoelectrically converting incident light. The imaging and focus detection pixel 301a outputs the photoelectric conversion signal output by the photoelectric conversion unit 34L and the photoelectric conversion signal output by the photoelectric conversion unit 34R separately. The signal obtained by adding these pair of photoelectric conversion signals is substantially the same as the imaging signal output by a single photoelectric conversion unit that substantially matches the external shape of the pair of photoelectric conversion units 34L and 34R. In other words, the imaging and focus detection pixel 301a can output an imaging signal.

[0020] On the other hand, the photoelectric conversion signal output by the photoelectric conversion unit 34L and the photoelectric conversion signal output by the photoelectric conversion unit 34R correspond to a pair of light beams that have passed through a pair of regions of the exit pupil of the imaging optical system 110. Therefore, by detecting the phase difference between the signal consisting of the photoelectric conversion signals output by the photoelectric conversion unit 34R of the numerous imaging and focus detection pixels 301a arranged in a line along the X-axis direction (a predetermined focus detection direction, also referred to as the first direction) and the signal consisting of the photoelectric conversion signals output by the photoelectric conversion unit 34L of those numerous imaging and focus detection pixels 301a, the focus evaluation value (defocus amount) of the imaging optical system 110 can be calculated. In other words, the imaging and focus detection pixels 301a can also output a focus detection signal that enables focus detection calculation using a well-known phase difference method.

[0021] As described above, the pair of photoelectric conversion units 34L and 34R of the imaging and focus detection pixel 301a output a photoelectric conversion signal that can be used as both an imaging signal and a pupil-splitting focus detection signal. The process of adding the pair of photoelectric conversion signals to form an imaging signal, and the process of converting the pair of photoelectric conversion signals into a focus detection signal, may be performed by the imaging and focus detection pixel 301a, by a dedicated circuit provided within the image sensor 210, by a dedicated circuit provided outside the image sensor 210, or by the body CPU 220.

[0022] Next, we will focus on another imaging and focus detection pixel 301b (see Figure 2(b)), which is located four pixels to the right (+X direction) of the image and focus detection pixel 301a. Figure 3(c) is a cross-sectional view of the imaging and focus detection pixel 301b. Like the imaging and focus detection pixel 301a, the imaging and focus detection pixel 301b also has a microlens 31, a color filter 32, and a pair of photoelectric conversion units 34L and 34R. However, as is clear from comparing Figure 3(b) and Figure 3(c), in the imaging and focus detection pixel 301a, the position of the optical axis 41 of the microlens 31 and the position of the dividing line 42, which represents the center of the division position of the pair of photoelectric conversion units 34L and 34R, are approximately coincident in the X direction, whereas in the imaging and focus detection pixel 301b, the optical axis 41 and the dividing line 42 are separated by a distance of 40 in the X direction.

[0023] Thus, in the imaging and focus detection pixel 301b, which is located away from the center of the imaging surface 20, the position of the optical axis 41 of the microlens 31 and the dividing line 42 of the pair of photoelectric conversion units 34L and 34R are separated, and this distance 40 increases as the distance from the center of the imaging surface 20 increases. The light beam that has passed through the imaging optical system 110 is incident on the imaging and focus detection pixel 301b from the upper left direction to the lower right direction of the paper in Figure 3(c). Therefore, in order to improve light collection, the microlens 31 is positioned with a displacement of 40 corresponding to the angle of the incident light. Although only the displacement in the X-axis direction is explained in Figures 3(b) and (c), similarly in the Y-axis direction, the positions of the microlens 31 and the pair of photoelectric conversion units 34L and 34R are also positioned differently.

[0024] Next, we will describe the focus detection pixels 302. In addition to the imaging and focus detection pixels mentioned above, the image sensor 210 has several types of focus detection pixels, including the first focus detection pixels 302NL and 302NR, the second focus detection pixels 302SL and 302SR, the third focus detection pixels 302LL and 302LR, the fourth focus detection pixels 302SNL and 302SNR, and the fifth focus detection pixels 302LNL and 302LNR. The locations of these focus detection pixels will be described in detail later, but first, we will describe the structure of these focus detection pixels below.

[0025] Figure 4(a) is a cross-sectional view of the first focus detection pixel 302NR. The first focus detection pixel 302NR includes a microlens 31, a photoelectric conversion unit 34, and a light-shielding member 35. The differences between the aforementioned imaging and focus detection pixel 301b and the first focus detection pixel 302NR are that it does not have a color filter 32, it has one undivided photoelectric conversion unit 34 instead of a pair of photoelectric conversion units 34L and 34R, and it has a light-shielding member 35.

[0026] The light-shielding member 35 is a thin film (light-shielding film) that blocks incident light. An opening 36NR is provided at a predetermined position (referred to as the first position) of the light-shielding member 35. Only the light beam that passes through the opening 36NR enters the photoelectric conversion unit 34, and other light beams are blocked by the light-shielding member 35 and do not enter the photoelectric conversion unit 34. The width of the opening 36NR in the X-axis direction (focus detection direction) is W2, which is smaller than the width W1 in the X-axis direction of the photoelectric conversion unit 34L and photoelectric conversion unit 34R shown in Figure 3.

[0027] Figure 4(b) is a cross-sectional view of the first focus detection pixel 302NL. The first focus detection pixel 302NL has almost the same configuration as the first focus detection pixel 302NR, and the difference from the first focus detection pixel 302NR is that the light-shielding member 35 has an opening 36NL instead of an opening 36NR. The opening 36NL has the same size and shape as the opening 36NR, but its position on the light-shielding member 35 is different from that of the opening 36NR.

[0028] The apertures 36NR and 36NL are located to the left and right of the dashed-dotted line 44N. In other words, the dashed-dotted line 44N is a straight line that defines the division center of apertures 36NR and 36NL. In the following explanation, the dashed-dotted line 44N will be referred to as the division line 44N. The first focus detection pixel 302NR outputs a received signal corresponding to the light image of the subject formed to the right of the division line 44N, and the first focus detection pixel 302NL outputs a received signal corresponding to the light image of the subject formed to the left of the division line 44N. In other words, the first focus detection pixels 302NR and 302NL each receive a pair of pupil-divided light beams and output a pair of focus detection signals.

[0029] The widths of the apertures 36NR and 36NL in the focus detection direction (X-axis direction) are smaller than the widths of the photoelectric conversion units 34R and 34L in the focus detection direction (X-axis direction) as explained in Figures 3(a) to (c). In other words, the photoelectric conversion units 34 of the first focus detection pixels 302NR and 302NL receive a more limited beam of light than the photoelectric conversion units 34R and 34L of the imaging and focus detection pixel 301. Therefore, especially when the amount of defocus is large, that is, when the subject to be focused is significantly out of focus, the focus detection signal output from the first focus detection pixels 302NR and 302NL has a stronger contrast than the focus detection signal output from the imaging and focus detection pixel 301. As a result, even when it is difficult to detect a phase difference with the focus detection signal output from the imaging and focus detection pixel 301, it is possible to detect the phase difference with the focus detection signal output from the first focus detection pixels 302NR and 302NL.

[0030] In the following explanation, it may be said that "the width of the photoelectric conversion unit 34 in the focal detection direction (X-axis direction) is smaller than the width of the photoelectric conversion unit 34R and 34L in the focal detection direction (X-axis direction)," meaning that the width of the apertures 36NR and 36NL in the focal detection direction is smaller than the width of the photoelectric conversion unit 34R and 34L in the focal detection direction (X-axis direction). In other words, "the width of the photoelectric conversion unit 34 in the focal detection direction (X-axis direction)" refers not to the actual width of the photoelectric conversion unit 34, but to the width of the incident range of the incident light on the photoelectric conversion unit 34. Therefore, "the width of the photoelectric conversion unit 34 in the focal detection direction (X-axis direction) is smaller than the width of the photoelectric conversion unit 34R and 34L in the focal detection direction (X-axis direction)" includes cases where the photoelectric conversion unit 34 is actually formed to have a small width, or where a light-shielding member 35 or the like is provided so that the incident range of light incident on the photoelectric conversion unit 34 has a small width.

[0031] The dividing line 44N approximately coincides with the dividing line 42 that bisects the photoelectric conversion unit 34 in the left-right direction (X-axis direction) of the paper. Figures 4(a) and (b) illustrate the principal ray 43N of the exit pupil of the imaging optical system 110 that passes through the vertex of the microlens 31. The principal ray 43N forms an angle θN with respect to the optical axis 41 of the microlens 31. The position of the dividing line 44N corresponds to the position of the principal ray 43N assumed by the first focus detection pixels 302NR and 302NL.

[0032] Figure 5(a) is a cross-sectional view of the second focus detection pixel 302SR, and Figure 5(b) is a cross-sectional view of the second focus detection pixel 302SL. The second focus detection pixels 302SR and 302SL have almost the same configuration as the first focus detection pixels 302NR and 302NL shown in Figures 4(a) and (b), respectively. The difference from the first focus detection pixels 302NR and 302NL is that the light-shielding member 35 has openings 36SR and 36SL instead of openings 36NR and 36NL.

[0033] Openings 36SR and 36SL are approximately the same size and shape as openings 36NR and 36NL, respectively. Openings 36SR and 36SL are located to the left and right of the dashed-dot line 44S. In other words, the dashed-dot line 44S is a straight line that defines the dividing center of openings 36SR and 36SL. In the following explanation, the dashed-dot line 44S will be referred to as the dividing line 44S.

[0034] The dividing line 44S is located relatively to the right of the dividing line 44N shown in Figures 4(a) and (b). Therefore, the dividing line 44S is located at a predetermined distance to the right of the dividing line 42 that bisects the photoelectric conversion unit 34 in the left-right direction (X-axis direction) of the paper.

[0035] Figures 5(a) and (b) illustrate the principal ray 43S of the exit pupil of the imaging optical system 110, which passes over the top of the microlens 31. The principal ray 43S forms an angle θS with respect to the optical axis 41 of the microlens 31. The angle θS is larger than the angle θN shown in Figures 4(a) and (b). In other words, even if the exit pupil of the imaging optical system 110 is in a relatively close position and subject light does not enter the apertures 36NR and 36NL of the first focus detection pixels 302NR and 302NL, subject light enters the apertures 36SR and 36SL. The position of the dividing line 44S corresponds to the position of the principal ray 43S assumed by the second focus detection pixels 302SR and 302SL. Thus, while aperture 36NR is located in the first position, aperture 36SR is located in a second position different from the first position.

[0036] Figure 6(a) is a cross-sectional view of the third focus detection pixel 302LR, and Figure 6(b) is a cross-sectional view of the third focus detection pixel 302LL. The third focus detection pixels 302LR and 302LL have almost the same configuration as the first focus detection pixels 302NR and 302NL shown in Figures 4(a) and (b), respectively. The difference from the first focus detection pixels 302NR and 302NL is that the light-shielding member 35 has openings 36LR and 36LL instead of openings 36NR and 36NL.

[0037] Openings 36LR and 36LL are approximately the same size and shape as openings 36NR and 36NL, respectively. Openings 36LR and 36LL are located to the left and right of the dashed-dot line 44L. In other words, the dashed-dot line 44L is a straight line that defines the dividing center of openings 36LR and 36LL. In the following explanation, the dashed-dot line 44L will be referred to as the dividing line 44L.

[0038] The dividing line 44L is located relatively to the left of the dividing line 44N shown in Figures 4(a) and (b). Therefore, the dividing line 44L is located at a predetermined distance to the left of the dividing line 42 that bisects the photoelectric conversion unit 34 in the left-right direction (X-axis direction) of the paper.

[0039] Figures 6(a) and (b) illustrate the principal rays 43S of the exit pupil of the imaging optical system 110, passing through the vertex of the microlens 31. The principal ray 43L forms an angle θL with respect to the optical axis 41 of the microlens 31. The angle θL is smaller than the angle θN shown in Figures 4(a) and (b). In other words, even when the exit pupil of the imaging optical system 110 is at a relatively distant position and subject light does not enter the apertures 36NR and 36NL of the first focus detection pixels 302NR and 302NL, subject light still enters the apertures 36LR and 36LL. The position of the dividing line 44L corresponds to the position of the principal ray 43L assumed by the third focus detection pixels 302LR and 302LL.

[0040] The focus detection pixels further include fourth focus detection pixels 302SNL and 302SNR (both not shown) and fifth focus detection pixels 302LNL and 302LNR (both not shown). The fourth focus detection pixels 302SNL and 302SNR correspond to the first focus detection pixels 302NL and 302NR, respectively, and the division center of the opening provided in the light-shielding member 35 lies between the division line 44N for the first focus detection pixels 302NL and 302NR and the division line 44L for the third focus detection pixels 302LL and 302LR. In other words, the focus detection signal can be suitably output when the exit pupil of the imaging optical system 110 is farther away than the position assumed by the first focus detection pixels 302NL and 302NR, and closer than the position assumed by the third focus detection pixels 302LL and 302LR. The fifth focus detection pixels 302LNL and 302LNR correspond to the first focus detection pixels 302NL and 302NR, respectively, and the division center of the opening provided in the light-shielding member 35 lies between the division line 44N in the first focus detection pixels 302NL and 302NR and the division line 44S in the second focus detection pixels 302SL and 302SR. In other words, the focus detection signal can be suitably output when the exit pupil of the imaging optical system 110 is closer than the position assumed by the first focus detection pixels 302NL and 302NR, and further than the position assumed by the second focus detection pixels 302SL and 302SR.

[0041] As described above, the image sensor 210 has five types of focus detection pixels (first focus detection pixels 302NL, 302NR; second focus detection pixels 302SL, 302SR; third focus detection pixels 302LL, 302LR; fourth focus detection pixels 302SNL, 302SNR; fifth focus detection pixels 302LNL, 302LNR), and each of these five types of focus detection pixels is assumed to have a different exit pupil position. In other words, each corresponds to a different imaging optical system 110. The position of the exit pupil changes, for example, when the interchangeable lens 100 is changed to a different type, or when the zoom position is changed in a so-called zoom lens with a variable focal length. However, with the camera system 1 of this embodiment, a focus detection signal can be reliably obtained from any of the focus detection pixels regardless of the position of the exit pupil.

[0042] Next, we will explain the arrangement of the various types of focus detection pixels described above. Figure 7 is a schematic plan view of the imaging surface 20 (i.e., the image capture screen) of the image sensor 210. Now, let's consider five regions obtained by dividing the imaging surface 20 into five equal parts in the left-right direction. Each region has the outline of a rectangle with its long side parallel to the vertical line OX. Of these five regions, the region containing the center of the imaging surface 20 is defined as the first region 60a. The two regions adjacent to the first region 60a are defined as the second region 60b. Furthermore, the remaining two regions adjacent to the left and right ends of the imaging surface 20 are defined as the third region 60c. In this embodiment, multiple focus detection pixel rows 61, in which focus detection pixels are arranged in a single row in the left-right direction (X-axis direction), are provided within the first region 60a, the second region 60b, and the third region 60c.

[0043] The second region 60b is located at a longer distance from the optical axis of the imaging optical system 110 in the left-to-right direction (X-axis direction) than the first region 60a. In other words, the second region 60b is a region with a higher image height than the first region 60a. The third region 60c is located at a longer distance from the optical axis of the imaging optical system 110 in the left-to-right direction (X-axis direction) than both the first region 60a and the second region 60b. In other words, the third region 60c is a region with a higher image height than both the first region 60a and the second region 60b.

[0044] Figure 8 shows an enlarged view of a portion of the first region 60a. In Figure 8, the letters "R," "G," and "B" represent imaging and focus detection pixels 301 having red, green, and blue color filters 32, respectively, and the letters "NL" and "NR" represent the first focus detection pixels 302NL and 302NR, respectively.

[0045] As shown in Figure 8, in the focus detection pixel array 61N within the first region 60a, first focus detection pixels 302NL and 302NR are arranged alternately at regular intervals d along the left-right direction (X-axis direction). For example, from left to right, the pixels 30 are arranged as follows: first focus detection pixel 302NL, multiple imaging and focus detection pixels 301a, first focus detection pixel 302NR, multiple imaging and focus detection pixels 301a, first focus detection pixel 302NL, ... In other words, within the first region 60a, some of the imaging and focus detection pixels 301 are replaced with first focus detection pixels 302NL and 302NR.

[0046] As shown in Figure 7, the second region 60b has more focus detection pixel arrays 61 than the first region 60a. Figure 9 shows an enlarged view of a part of the second region 60b. In Figure 9, the second focus detection pixels 302SL, 302SR, and the third focus detection pixels 302LL, 302LR are represented by the letters "SL," "SR," "LL," and "LR," respectively. Within the second region 60b, there are multiple instances of three types of focus detection pixel arrays 61: focus detection pixel arrays 61N, focus detection pixel arrays 61S, and focus detection pixel arrays 61L.

[0047] In the focus detection pixel array 61N, the first focus detection pixels 302NL and 302NR are arranged alternately at regular intervals d along the left-right direction (X-axis direction). In the focus detection pixel array 61S, the second focus detection pixels 302SL and 302SR are arranged alternately at regular intervals d along the left-right direction (X-axis direction). In the focus detection pixel array 61L, the third focus detection pixels 302LL and 302LR are arranged alternately at regular intervals d along the left-right direction (X-axis direction). In other words, within the second region 60b, a portion of the imaging and focus detection pixels 301 are replaced by the first focus detection pixels 302NL and 302NR, the second focus detection pixels 302SL and 302SR, and the third focus detection pixels 302LL and 302LR.

[0048] As shown in Figure 7, the third region 60c has even more focus detection pixel arrays 61 than the second region 60b. Figure 10 shows an enlarged view of a part of the third region 60c. In Figure 10, the fourth focus detection pixels 302SNL, 302SNR, and the fifth focus detection pixels 302LNL, 302LNR are represented by the letters "SNL", "SNR", "LNL", and "LNR", respectively. Within the third region 60c, there are multiple instances of five types of focus detection pixel arrays 61: focus detection pixel arrays 61N, focus detection pixel arrays 61S, focus detection pixel arrays 61L, focus detection pixel arrays 61SN, and focus detection pixel arrays 61LN.

[0049] In the focus detection pixel sequence 61SN, the fourth focus detection pixels 302SNL and 302SNR are arranged alternately at regular intervals d along the left-right direction (X-axis direction). In the focus detection pixel sequence 61LN, the fifth focus detection pixels 302LNL and 302LNR are arranged alternately at regular intervals d along the left-right direction (X-axis direction). In other words, within the third region 60c, a portion of the imaging and focus detection pixels 301 are replaced by the first focus detection pixels 302NL and 302NR, the second focus detection pixels 302SL and 302SR, the third focus detection pixels 302LL and 302LR, the fourth focus detection pixels 302SNL and 302SNR, and the fifth focus detection pixels 302LNL and 302LNR.

[0050] Next, the focus detection process by the first focus detection unit 221 and the second focus detection unit 222 will be described. At the start of the focus detection process, one focus detection area is pre-set on the shooting screen. The focus detection area may be set manually by the user using an operating element such as a button (not shown), or it may be set automatically by the body CPU 220 through a well-known main subject recognition process such as face recognition.

[0051] In the focus detection process, the second focus detection unit 222 first performs a focus detection calculation in the focus detection area based on the focus detection signal output from the focus detection pixel 302. This focus detection calculation is more suitable for detecting large defocus, i.e., a state where the image is significantly out of focus, compared to the focus detection calculation performed by the first focus detection unit 221 (details will be described later). Therefore, in the following description, the focus detection calculation performed by the second focus detection unit 222 will be referred to as the focus detection calculation for large defocus.

[0052] When the second focus detection unit 222 detects a defocus amount greater than a predetermined amount, i.e., when a large defocus is detected, the focus adjustment unit 230 drives the focusing lens 112 based on the detection result. On the other hand, when a defocus amount less than a predetermined amount is detected, i.e., when the imaging optical system 110 is in focus to some extent, the first focus detection unit 221 performs a focus detection calculation in the focus detection area based on the focus detection signal output from the imaging and focus detection pixel 301. Then, the focus adjustment unit 230 drives the focusing lens 112 based on the detection result. Here, the focus detection calculation performed by the first focus detection unit 221 is more suitable for focus detection when there is no large defocus compared to the focus detection calculation performed by the second focus detection unit 222 (details will be described later). Therefore, in the following description, the focus detection calculation performed by the first focus detection unit 221 will be referred to as the focus detection calculation for small defocus.

[0053] The following describes the focus detection calculation for large defocus performed by the second focus detection unit 222. First, the second focus detection unit 222 selects a focus detection pixel sequence 61 from the vicinity of the focus detection area according to the position of the exit pupil of the imaging optical system 110. For example, if the focus detection area is near the center of the shooting screen and only a focus detection pixel sequence 61N is present in that vicinity, then that focus detection pixel sequence 61N is necessarily selected.

[0054] On the other hand, if the focus detection area is located away from the center of the shooting screen, and there are three focus detection pixel sequences 61L, 61N, and 61S in its vicinity, one of the focus detection pixel sequences 61 is selected from among them depending on the position of the exit pupil of the imaging optical system 110. Specifically, if the exit pupil of the imaging optical system 110 is at a relatively distant position (further than a predetermined distance), focus detection pixel sequence 61L is selected. Conversely, if the exit pupil of the imaging optical system 110 is at a relatively close position (close within a predetermined distance), focus detection pixel sequence 61S is selected. If it is neither (at an intermediate position), focus detection pixel sequence 61N is selected.

[0055] Next, the second focus detection unit 222 acquires a focus detection signal from the focus detection pixels in the selected focus detection pixel sequence 61. For example, if the focus detection pixel sequence 61N is selected, then the first focus detection pixels 302NL and 302NR are arranged in a row within that sequence. The second focus detection unit 222 takes a pair of signals: an output signal obtained by arranging the outputs of a large number of first focus detection pixels 302NL (the light-receiving outputs of the photoelectric conversion unit 34), and an output signal obtained by arranging the outputs of a large number of first focus detection pixels 302NR (the light-receiving outputs of the photoelectric conversion unit 34), and uses these as a pair of focus detection signals. Then, it performs a correlation calculation to calculate the phase difference between these two pairs of focus detection signals and calculates the amount of defocus. Since such calculations are well known, an explanation is omitted.

[0056] The reason why the focus detection calculation performed by the second focus detection unit 222 is suitable for focus detection during large defocus is that the photoelectric conversion unit 34 of the focus detection pixel 302 is partially shielded from light by the light-shielding member 35. When the focus detection signal obtained from the imaging and focus detection pixel 301 is significantly out of focus (i.e., when the amount of defocus is very large), it becomes a smooth signal with unclear contrast. It is difficult to accurately detect the phase difference of such a signal by correlation calculation.

[0057] In contrast, the photoelectric conversion unit 34 of the focus detection pixel 302 has its incident light restricted by the light-shielding member 35, and the width of the area into which the light beam that has passed through the imaging optical system 110 is incident in the focus detection direction (X-axis direction) is smaller than the width of the photoelectric conversion units 34L and 34R of the imaging and focus detection pixel 301 in the same direction.

[0058] Generally, when the aperture 130 of the imaging optical system 110 is reduced, the depth of field increases, and a subject image with less blur is obtained. The focus detection pixel 302 can achieve a similar effect. In other words, even when the imaging optical system 110 is significantly out of focus (i.e., the amount of defocus is very large), the focus detection signal obtained from the focus detection pixel 302 is a signal with higher contrast (a steeper, more accurate representation of the original subject image) compared to the focus detection signal obtained from the imaging and focus detection pixel 301. Such a signal makes it easier to perform phase difference detection by correlation calculation compared to the focus detection signal obtained from the imaging and focus detection pixel 301. Therefore, even when the focus is significantly out of focus, more accurate focus detection calculations can be performed compared to the imaging and focus detection pixel 301.

[0059] As described above, the second focus detection unit 222 selects an appropriate focus detection pixel array 61 according to the position of the exit pupil of the imaging optical system 110, so that the light beam from the exit pupil is not kicked out and does not interfere with focus detection. In the first region 60a, even when the exit pupil of the imaging optical system 110 is far away, the angle of incidence of the light beam from the exit pupil in the X-axis direction is not as large as in the second region 60b and the third region 60c, so even with the focus detection pixel array 61N, the light beam from the exit pupil is not kicked out, and the focus detection pixel array 61N alone is sufficient. For this reason, only the focus detection pixel array 61N is arranged in the first region 60a. The reason why there are fewer types of focus detection pixels arranged in the second region 60b than in the third region 60c is the same.

[0060] Next, the focus detection calculation for small defocus performed by the first focus detection unit 221 will be explained. The first focus detection unit 221 first acquires focus detection signals from imaging and focus detection pixels 301 near the focus detection area. For example, it selects a number of imaging and focus detection pixels 301 arranged in a horizontal row near the focus detection area. Then, it combines the output signals obtained by arranging the light-receiving outputs of the photoelectric conversion units 34L of the selected imaging and focus detection pixels 301, and the output signals obtained by arranging the light-receiving outputs of the photoelectric conversion units 34R of those imaging and focus detection pixels 301, and uses these as a pair of focus detection signals. The first focus detection unit 221 performs a correlation calculation to calculate the phase difference of this pair of focus detection signals and calculates the amount of defocus. Since such calculations are well known, an explanation will be omitted.

[0061] The photoelectric conversion units 34L and 34R of the imaging and focus detection pixel 301 are not limited by the amount of incident light, unlike the photoelectric conversion unit 34 of the focus detection pixel 302. Therefore, the amount of light converted photoelectrically is larger than that of the focus detection pixel 302. In other words, the signal amount of the focus detection signal is larger than that of the focus detection pixel 302. Therefore, when the amount of defocus is relatively small, the amount of defocus obtained by the second focus detection unit 222 through focus detection calculation is more accurate than the amount of defocus obtained by the first focus detection unit 221 through focus detection calculation. Therefore, in this embodiment, when there is small defocus, focus adjustment is performed based on the result of focus detection by the first focus detection unit 221.

[0062] Figure 11 is a flowchart of the control process executed by the body CPU 220. When the camera body 200 is powered on, the body CPU 220 reads the control program, including this control process, from the ROM 240 and starts executing it.

[0063] First, in step S10, the body CPU 220 starts the periodic operation of the image sensor 210, that is, reading out the imaging signal at predetermined intervals (e.g., 1 / 60th of a second) to create a live view image. In step S20, the body CPU 220 determines whether a predetermined autofocus operation (e.g., half-pressing the shutter release switch) has been performed. If no autofocus operation has been performed, the body CPU 220 proceeds to step S30.

[0064] In step S30, the body CPU 220 reads an imaging signal from the imaging and focus detection pixels of the image sensor 210 to create a live view image. The imaging signal read here is a signal for creating a live view image, and it is not necessary to read a signal from all imaging and focus detection pixels. For example, decimal readout is performed, reading every three pixels. In step S40, the body CPU 220 updates the live view image displayed on the display device 250. That is, it creates a live view image using the imaging signal read in step S30 and displays it on the display device 250. After that, the body CPU 220 proceeds to step S20.

[0065] On the other hand, if the autofocus operation is not performed in step S20, the body CPU 220 proceeds to step S50. In step S50, the body CPU 220 performs the autofocus (AF) operation described later. In step S60, the body CPU 220 determines whether a predetermined release operation (for example, fully pressing the release switch) has been performed. If the release operation has not been performed, the body CPU 220 proceeds to step S70.

[0066] In step S70, the body CPU 220 reads an imaging signal from the imaging and focus detection pixels of the image sensor 210 to create a live view image. In step S80, the body CPU 220 updates the live view image displayed on the display device 250. In other words, it creates a live view image using the imaging signal read in step S70 and displays it on the display device 250. After that, the body CPU 220 proceeds to step S20.

[0067] On the other hand, if the shutter release operation was performed in step S60, the body CPU 220 proceeds to step S90. In step S90, the body CPU 220 reads out the imaging signal from the imaging and focus detection pixels 301 of the image sensor 210. The imaging signal read out here is the signal of the recorded image data (main image data). Since it is desirable that the recorded image data be of the highest possible quality, signals are read out from all imaging and focus detection pixels 301. In step S100, the body CPU 220 performs interpolation of the focus detection pixels 302. That is, since an imaging signal cannot be obtained from the position where the focus detection pixels 302 exist, the body CPU 220 pseudo-generates the imaging signal that should have been obtained from that position (if there were imaging and focus detection pixels 301 at that position) based on the imaging signals obtained from the surrounding imaging and focus detection pixels 301.

[0068] In step S110, the body CPU 220 creates recorded image data and stores it in a storage medium (not shown) (e.g., a memory card). In step S120, the body CPU 220 determines whether a predetermined power-off operation (e.g., pressing the power switch) has been performed. If the power-off operation has not been performed, the body CPU 220 proceeds to step S20. On the other hand, if the power-off operation has been performed, the body CPU 220 terminates the process shown in Figure 11.

[0069] Figure 12 is a flowchart of the AF process called from step S30 in Figure 11. First, in step S200, the second focus detection unit 222 selects one focus detection pixel sequence from the focus detection pixel sequences near the focus detection area, based on the current information of the imaging optical system 110 (e.g., the position of the exit pupil). For example, if the focus detection area is located in the third region 60c and the exit pupil of the imaging optical system 110 is somewhat far away, the focus detection pixel sequence 61L is selected. The current information of the imaging optical system 110 can be received by the body CPU 220 from the lens CPU 120 through data communication between the lens CPU 120 and the body CPU 220.

[0070] In step S210, the second focus detection unit 222 reads a focus detection signal from a focus detection pixel 302 included in the focus detection pixel sequence selected in step S200. In step S220, the second focus detection unit 222 performs a focus detection calculation based on the focus detection signal read in step S210, i.e., a focus detection calculation for large defocus.

[0071] In step S230, the second focus detection unit 222 determines whether a defocus amount greater than a predetermined amount, i.e., a major defocus, has been detected based on the focus detection calculation for major defocus performed in step S220. If a defocus amount greater than a predetermined amount is detected, the second focus detection unit 222 proceeds to step S240. In step S240, the focus adjustment unit 230 drives the focusing lens 112 based on the defocus amount obtained from the focus detection calculation for major defocus. After that, the second focus detection unit 222 proceeds to step S200.

[0072] On the other hand, if no amount of defocus exceeding a predetermined amount is detected in step S230 (for example, if the detected amount of defocus is less than the predetermined amount, or if no amount of defocus was detected), the second focus detection unit 222 proceeds to step S250. In step S250, the first focus detection unit 221 reads out a light-receiving signal (light-receiving output) from the imaging and focus detection pixel 301. As mentioned above, the signal read out here can be treated as both an imaging signal and a focus detection signal.

[0073] In step S260, the body CPU 220 updates the live view image displayed on the display device 250. That is, it treats the signal read in step S250 as an imaging signal, creates a live view image, and displays it on the display device 250. In step S270, the first focus detection unit 221 selects one pixel sequence consisting of imaging and focus detection pixels 301 arranged in the focus detection direction (X-axis direction) from the vicinity of the focus detection area. In step S280, the first focus detection unit 221 uses the signal corresponding to the pixel sequence selected in step S270 from the signals read in step S250 as the focus detection signal and performs a focus detection calculation based on that focus detection signal, i.e., a focus detection calculation for small defocus.

[0074] In step S290, the body CPU 220 determines whether the focus detection calculation for minor defocus performed in step S280 was successful. If, for example, phase difference detection cannot be performed due to low contrast of the subject image, and focus detection fails, the body CPU 220 proceeds to step S300. In step S300, the body CPU 220 determines whether the most recent focus detection calculation for major defocus performed in step S220 was successful. If the amount of defocus was successfully calculated by the most recent focus detection calculation for major defocus, the body CPU 220 proceeds to step S310.

[0075] In step S310, the focus adjustment unit 230 drives the focusing lens 112 based on the amount of defocus obtained by the focus detection calculation for large defocus. After that, the body CPU 220 proceeds to step S200.

[0076] In step S300, if the focus detection calculation for the most recent large defocus failed to calculate the amount of defocus, the body CPU 220 proceeds to step S320. In step S320, the focus adjustment unit 230 performs a so-called search drive (scan drive), which moves the focusing lens over a certain range. After that, the body CPU 220 proceeds to step S200.

[0077] In step S290, if the focus detection calculation for minor defocus performed in step S280 was successful, the body CPU 220 proceeds to step S330. In step S330, the body CPU 220 determines whether the imaging optical system 110 is in focus, that is, whether the amount of defocus detected in step S280 is sufficiently small (smaller than a predetermined threshold). If it is not in focus, that is, if the amount of defocus is greater than or equal to the predetermined threshold, the body CPU 220 proceeds to step S340. In step S340, the focus adjustment unit 230 drives the focusing lens 112 based on the amount of defocus obtained by the focus detection calculation for minor defocus. After that, the body CPU 220 proceeds to step S250. The reason for proceeding to step S250 instead of step S200 here is that it is already clear that the image is in focus to some extent (not in a state of major defocus). If the focus was achieved in step S330, that is, if the amount of defocus was less than a predetermined threshold, the body CPU 220 terminates the AF process. As described above, in the AF process of this embodiment, a focus detection calculation for large defocus is performed first, and if a large defocus state is detected, the lens is driven based on the result of the focus detection calculation for large defocus without performing the focus detection calculation for small defocus, which has a high probability of failure.

[0078] The camera system according to the first embodiment described above provides the following effects and advantages. (1) The image sensor 210 is arranged with a plurality of imaging and focus detection pixels 301 into which incident light is incident within a predetermined range, and a plurality of types of focus detection pixels 302 into which incident light is incident within a range smaller than that of the imaging and focus detection pixels 301. The image sensor 210 has a first region 60a into which one type of focus detection pixel 302 is located, and a second region 60b into which the image height is higher than that of the first region 60a and more types of focus detection pixels 302 are located. In this way, focus detection can be performed appropriately whether the image is in a large or small defocus state.

[0079] (2) The multiple types of focus detection pixels 302 have a width in a predetermined direction of the aperture that is smaller than that of the imaging and focus detection pixel 301. In this way, focus detection can be performed appropriately whether the state is large or small in focus.

[0080] (3) The first region 60a is equipped with first focus detection pixels 302NR and 302NL corresponding to the first exit pupil position, and the second region 60b is equipped with the first focus detection pixels 302NR and 302NL, and second focus detection pixels 302SR and 302SL corresponding to a second exit pupil position different from the first exit pupil position. In this manner, even if the position of the exit pupil changes due to reasons such as replacing the interchangeable lens 100 with a different type, focus detection can be performed appropriately.

[0081] (4) The distance in the X-axis direction from the optical axis of the imaging optical system 110 (center of the imaging surface 20) to the first region 60a is shorter than the distance in the same direction from the optical axis of the imaging optical system 110 (center of the imaging surface 20) to the second region 60b. In this way, only the minimum number of focus detection pixels 302 can be placed in positions where there is little possibility of subject light being knocked off, and the image quality of the captured image data is improved.

[0082] (5) The image sensor 210 includes an imaging and focus detection pixel 301 into which incident light is incident within a predetermined range, first focus detection pixels 302NR, NL into which incident light is incident within a range smaller than that of the imaging and focus detection pixel 301 and which correspond to a first exit pupil position, and second focus detection pixels 302SR, SL into which incident light is incident within a range smaller than that of the imaging and focus detection pixel 301 and which correspond to a second exit pupil position different from the first exit pupil position. The image sensor 210 has a first region 60a into which the first focus detection pixels 302NR, NL are located, and a second region 60b into which the image height is higher than that of the first region 60a and which includes the first focus detection pixels 302NR, NL and the second focus detection pixels 302SR, SL. This makes it easier to manufacture the image sensor 210 compared to the case where two photoelectric conversion units 34 are provided for a single focus detection pixel 302.

[0083] (6) The image sensor 210 has a first focus detection pixel 302NR and a second focus detection pixel 302SR that receive light from one of the pair of exit pupils, and a first focus detection pixel 302NL and a second focus detection pixel 302SL that receive light from the other of the pair of exit pupils. This eliminates the need to individually modify the photoelectric conversion unit 34, thereby reducing the manufacturing cost of the image sensor 210.

[0084] (7) The image sensor 210 further includes third focus detection pixels LR and LL, into which incident light is incident in an area smaller than that of the imaging and focus detection pixel 301, and a third region 60c, which has a higher image height than the second region 60b and in which the first focus detection pixels 302NR and NL, the second focus detection pixels 302SR and SL, and the third focus detection pixels 302LR and LL are arranged. In this way, more precise focus detection can be performed according to the exit pupil position.

[0085] (8) Multiple types of focus detection pixels 302 are arranged in multiples along the X-axis. This allows for focus detection using a phase difference detection method.

[0086] (9) The first focus detection unit 221 performs focus detection based on the focus detection signal output from the imaging and focus detection pixel 301. The second focus detection unit 222 performs focus detection based on the focus detection signal (photoelectric conversion signal) output from the first focus detection pixels 302NR and 302NL when the exit pupil of the imaging optical system 110 is in the first position, and performs focus detection based on the focus detection signal output from the second focus detection pixels 302SR and 302SL when the exit pupil of the imaging optical system 110 is in the second position. In this way, appropriate focus detection can always be performed regardless of whether the image is in a small or large defocus state, or whether the exit pupil of the imaging optical system 110 is in the first or second position.

[0087] (10) If the focus detection result by the second focus detection unit 222 indicates a defocus of a predetermined amount or more, the focus adjustment unit 230 adjusts the focus of the imaging optical system 110 based on the focus detection result. If the focus detection result by the second focus detection unit 222 does not indicate a defocus of a predetermined amount or more, the focus adjustment unit 230 adjusts the focus of the imaging optical system 110 based on the focus detection result by the first focus detection unit 221. In this way, when there is a large defocus, it is not necessary to operate the first focus detection unit 221, which reduces the processing load on the body CPU 220 and reduces power consumption.

[0088] (Second Embodiment) The camera system according to the second embodiment has the same configuration as the camera system according to the first embodiment, but the content of the AF processing performed by the body CPU 220 differs from that of the first embodiment. The AF processing in the second embodiment will be described below.

[0089] Figure 13 is a flowchart of the AF process called from step S50 in Figure 11 in this embodiment. First, in step S400, the first focus detection unit 221 reads out the received signal (received output) from the imaging and focus detection pixel 301. As mentioned above, the signal read out here can be treated as both an imaging signal and a focus detection signal.

[0090] In step S410, the body CPU 220 updates the live view image displayed on the display device 250. In other words, it treats the signal read in step S400 as an imaging signal, creates a live view image, and displays it on the display device 250.

[0091] In step S415, the first focus detection unit 221 selects one pixel sequence from the vicinity of the focus detection area, consisting of imaging and focus detection pixels 301 arranged in the focus detection direction (X-axis direction). In step S420, the first focus detection unit 221 uses the signal read out in step S400 that corresponds to the pixel sequence selected in step S415 as the focus detection signal, and performs a focus detection calculation based on that focus detection signal, i.e., a focus detection calculation for small defocus.

[0092] In step S430, the body CPU 220 determines whether the focus detection calculation for minor defocus performed in step S420 was successful. If, for example, phase difference detection cannot be performed due to the imaging optical system 110 being significantly out of focus, i.e., in a state of major defocus, and the focus detection fails, the body CPU 220 proceeds to step S440.

[0093] In step S440, the second focus detection unit 222 selects one focus detection pixel sequence from the focus detection pixel sequences near the focus detection area, based on information from the current imaging optical system 110 (e.g., the exit pupil position). In step S450, the second focus detection unit 222 reads a focus detection signal from the focus detection pixel 302 included in the focus detection pixel sequence selected in step S440. In step S460, the second focus detection unit 222 performs a focus detection calculation based on the focus detection signal read in step S450, i.e., a focus detection calculation for large defocus.

[0094] In step S470, the body CPU 220 determines whether the focus detection calculation for large defocus performed in step S460 was successful. If focus detection fails, the body CPU 220 proceeds to step S490. In step S490, the focus adjustment unit 230 performs a so-called search drive (scan drive), which moves the focusing lens over a certain range. After that, the body CPU 220 proceeds to step S400.

[0095] On the other hand, if the calculation of the amount of defocus by the focus detection calculation for large defocus is successful in step S470, the body CPU 220 proceeds to step S480. In step S480, the focus adjustment unit 230 drives the focusing lens 112 based on the amount of defocus obtained by the focus detection calculation for large defocus. After that, the body CPU 220 proceeds to step S400.

[0096] In step S430, if the calculation of the defocus amount by the focus detection calculation for small defocus is successful, the body CPU 220 proceeds to step S500. In step S500, the body CPU 220 determines whether the imaging optical system 110 is in focus, that is, whether the amount of defocus detected in step S420 is sufficiently small (smaller than a predetermined threshold). If it is not in focus, that is, if the amount of defocus is greater than or equal to the predetermined threshold, the body CPU 220 proceeds to step S510. In step S510, the focus adjustment unit 230 drives the focusing lens 112 based on the amount of defocus obtained by the focus detection calculation for small defocus. After that, the body CPU 220 proceeds to step S400. On the other hand, if it was in focus in step S500, that is, if the amount of defocus was smaller than the predetermined threshold, the body CPU 220 terminates the process shown in Figure 13. As described above, in the AF processing of this embodiment, a focus detection calculation for minor defocus is performed first, and only if this fails is a focus detection calculation for major defocus performed.

[0097] The camera system according to the second embodiment described above provides the following effects and advantages. (1) If the focus detection by the first focus detection unit 221 is successful, the focus adjustment unit 230 adjusts the focus of the imaging optical system 110 based on the result of the focus detection. If the focus detection by the first focus detection unit 221 fails, the focus adjustment unit 230 adjusts the focus of the imaging optical system 110 based on the result of the focus detection by the second focus detection unit 222. In this way, focus detection by the second focus detection unit 222 is performed only when there is a large defocus, which reduces the processing load on the body CPU 220 and reduces power consumption.

[0098] (Third embodiment) The camera system according to the third embodiment has the same configuration as the camera system according to the first embodiment, but differs from the first embodiment in that it has an image sensor 210a with a different structure instead of the image sensor 210. The image sensor 210a in the third embodiment will be described below.

[0099] Figure 14 is a schematic perspective view showing the structure of an image sensor 210a according to a third embodiment. The image sensor 210a has a microlens array 211 and a photoelectric conversion array 212. The microlens array 211 has a plurality of microlenses 31 arranged in a square in two dimensions. The photoelectric conversion array 212 has a plurality of photoelectric conversion units 34 arranged in a square in two dimensions. Note that the arrangement of the plurality of microlenses 31 does not have to be a square.

[0100] The microlens array 211 is positioned at a distance f from the light-receiving surface of the photoelectric conversion array 212 by the focal length f of the microlens 31. In other words, the microlens array 211 and the photoelectric conversion array 212 are positioned such that the focal position of the microlens 31 coincides with the light-receiving surface of the photoelectric conversion array 212.

[0101] In this embodiment, the diameter of one microlens 31 is greater than the width of one photoelectric conversion unit 34. In other words, one microlens 31 covers multiple photoelectric conversion units 34. Subject light passing through one microlens 31 is incident on the multiple photoelectric conversion units 34 corresponding to that microlens 31. As illustrated in Figure 14, the range 215 into which light passing through one microlens 31 is incident includes multiple photoelectric conversion units 34.

[0102] The image sensor 210a has multiple pixels 30. The pixels 30 include two types: an imaging and focus detection pixel 301 used for both imaging and focus detection, and a focus detection pixel 302 used only for focus detection. These two types of pixels will be described below.

[0103] Figure 15 is a schematic cross-sectional view showing the imaging / focus detection pixel 301 and the focus detection pixel 302. In Figure 15, the cross-section of the central portion of the imaging surface of the image sensor 210a is schematically shown. In the first embodiment, one pixel 30 had one microlens 31. In contrast, in this embodiment, multiple pixels 30 share one microlens 31. In other words, in this embodiment, the microlens 31 of one pixel 30 and the microlens 31 of another pixel 30 may be the same microlens 31.

[0104] The imaging and focus detection pixel 301 includes a microlens 31, a color filter 32, and a photoelectric conversion unit 34. Incident light entering the microlens 31 enters the photoelectric conversion unit 34 via the color filter 32. The color filter 32 is formed to transmit either red, blue, or green light for each pixel, and not transmit other light. The color filter 32 of the imaging and focus detection pixel 301 is configured to form a so-called Bayer array. The imaging and focus detection pixel 301 outputs a photoelectric conversion signal output by the photoelectric conversion unit 34.

[0105] In this embodiment, the imaging and focus detection pixel 301, similar to the imaging and focus detection pixel 301 in the first embodiment shown in Figure 3(c), is such that the part of the pixel that is farther from the center of the imaging plane has a distance between the optical axis 41 of the microlens 31 and the center of the photoelectric conversion unit 34, and this distance increases as the distance from the center of the imaging plane increases. This point is the same as in the first embodiment, so it is not illustrated or explained here.

[0106] The focus detection pixel 302 is the same as in the first embodiment, except that the microlens 31 is shared among multiple pixels 30. That is, the focus detection pixel 302 has a microlens 31, a color filter 32, a photoelectric conversion unit 34, and a light-shielding member 35. The incident light that enters the microlens 31 passes through the color filter 32, and a portion that is not blocked by the light-shielding member 35 enters the photoelectric conversion unit 34.

[0107] In this embodiment, as in the first embodiment, the focus detection pixel 302 includes multiple types of focus detection pixels, such as the first focus detection pixels 302NL and 302NR, the second focus detection pixels 302SL and 302SR, the third focus detection pixels 302LL and 302LR, the fourth focus detection pixels 302SNL and 302SNR, and the fifth focus detection pixels 302LNL and 302LNR. In the first embodiment, these focus detection pixels differed in the width of their apertures, but in this embodiment, the width of the apertures is the same for all of them.

[0108] Figure 16 is a magnified plan view of a portion of the first region 60a (Figure 7) of the image sensor 210a. In Figure 16, the letters "R," "G," and "B" represent imaging pixels 303 having red, green, and blue color filters 32, respectively, and the letters "NL" and "NR" represent first focus detection pixels 302NL and 302NR, respectively.

[0109] In the first embodiment, the focus detection pixel array 61N in the first region 60a had first focus detection pixels 302NL and 302NR arranged alternately at regular intervals d along the left-right direction (X-axis direction). In this embodiment, instead of the focus detection pixel array 61N, we consider a microlens array 610N in which microlenses 31 containing focus detection pixels 302 are arranged in a single line along the left-right direction (X-axis direction). In the microlenses 31 included in the microlens array 610N in the first region 60a, first focus detection pixels 302NL and 302NR are arranged alternately at regular intervals (every other pixel in Figure 16).

[0110] For example, in Figure 16, the 6x6 array of 36 pixels 30 with the leftmost microlens 31a contains one first focus detection pixel 302NL. The 6x6 array of 36 pixels 30 with the microlens 31b to its right does not contain a focus detection pixel 302. The 6x6 array of 36 pixels 30 with the microlens 31c to the right of that contains one first focus detection pixel 302NR.

[0111] The pair of first focus detection pixels 302NL and 302NR are arranged point-symmetrically within each microlens 31 with respect to the principal ray from the assumed exit pupil position. For example, in the microlens row 610N shown in Figure 16, the position of the principal ray from the assumed exit pupil position is indicated by the circle LP. When focusing on each microlens 31, the first focus detection pixels 302NL and 302NR are arranged point-symmetrically with respect to this circle LP. Specifically, the first focus detection pixel 302NL is located immediately to the lower right of the circle LP, and the first focus detection pixel 302NR is located immediately to the upper left of the circle LP.

[0112] Figure 17 is a magnified plan view of a portion of the second region 60b (Figure 7) of the image sensor 210a. In Figure 17, the second focus detection pixels 302SL, 302SR, and the third focus detection pixels 302LL, 302LR are represented by the letters "SL," "SR," "LL," and "LR," respectively. Within the second region 60b, there are multiple microlens arrays 61 of three types: microlens array 610L, microlens array 610N, and microlens array 610S.

[0113] In the microlens 31 included in the microlens array 610N, first focus detection pixels 302NL and 302NR are arranged alternately at regular intervals (every other pixel in Figure 17). In the microlens 31 included in the microlens array 610S, second focus detection pixels 302SL and 302SR are arranged alternately at regular intervals (every other pixel in Figure 17). In the microlens 31 included in the microlens array 610L, third focus detection pixels 302LL and 302LR are arranged alternately at regular intervals (every other pixel in Figure 17).

[0114] The pair of second focus detection pixels 302SL and 302SR and the pair of third focus detection pixels 302LL and 302LR are arranged point-symmetrically within each microlens 31 with respect to the principal ray from the assumed exit pupil position. For example, in the microlens array 610S shown in Figure 17, the position of the principal ray from the assumed exit pupil position is indicated by the circle LP. When focusing on each microlens 31, the second focus detection pixels 302SL and 302SR are arranged point-symmetrically with respect to this circle LP. Specifically, the second focus detection pixel 302SL is located immediately to the lower right of the circle LP, and the second focus detection pixel 302SR is located immediately to the upper left of the circle LP. The same applies to the third focus detection pixels 302LL and 302LR.

[0115] The first focus detection pixels 302NL and 302NR are positioned near the optical axis of the microlens 31. In contrast, the second focus detection pixels 302SL and 302SR are positioned at a certain distance to the right of the optical axis of the microlens 31. Similarly, the third focus detection pixels 302LL and 302LR are positioned at a certain distance to the left of the optical axis of the microlens 31 (in the opposite direction to the second focus detection pixels 302SL and 302SR).

[0116] As described above, in this embodiment, the position of the focus detection pixels 302 differs depending on the assumed pupil position. That is, the focus detection pixels 302 are positioned according to the assumed pupil position. For example, the light from the assumed exit pupil position for the first focus detection pixels 302NL and 302NR forms a spot near the center of the microlens 31, so the first focus detection pixels 302NL and 302NR are positioned near the center of the microlens 31. On the other hand, for the second focus detection pixels 302SL and 302SR, which assume a short pupil position, the spot is formed on the side closer to the optical axis of the image sensor 210a with respect to the optical axis of the microlens 31, i.e., on the left side of Figure 16, so the second focus detection pixels 302SL and 302SR are positioned to the right of the optical axis of the microlens 31. The third focus detection pixels 302LL and 302LR are the opposite, and are positioned to the left of the optical axis of the microlens 31.

[0117] Figure 18 is a magnified plan view of a portion of the third region 60c (Figure 7) of the image sensor 210a. In Figure 18, the fourth focus detection pixels 302SNL, 302SNR, and the fifth focus detection pixels 302LNL, 302LNR are represented by the letters "SNL," "SNR," "LNL," and "LNR," respectively. Within the third region 60c, there are multiple microlens arrays 610 of five types: microlens array 610S, microlens array 610SN, microlens array 610N, microlens array 610LN, and microlens array 610L.

[0118] In the microlens 31 included in the microlens array 610N, first focus detection pixels 302NL and 302NR are arranged alternately at regular intervals (every other pixel in Figure 18). In the microlens 31 included in the microlens array 610S, second focus detection pixels 302SL and 302SR are arranged alternately at regular intervals (every other pixel in Figure 18). In the microlens 31 included in the microlens array 610L, third focus detection pixels 302LL and 302LR are arranged alternately at regular intervals (every other pixel in Figure 18). In the microlens 31 included in the microlens array 610SN, fourth focus detection pixels 302SNL and 302SNR are arranged alternately at regular intervals (every other pixel in Figure 18). In the microlens 31 included in the microlens array 610LN, fifth focus detection pixels 302LNL and 302LNR are arranged alternately at regular intervals (every other pixel in Figure 18).

[0119] The pair of fourth-focus detection pixels 302SNL, 302SNR and the pair of fifth-focus detection pixels 302LNL, 302LNR are arranged point-symmetrically within each microlens 31 with respect to the principal ray from the assumed exit pupil position. For example, in the microlens row 610SN shown in Figure 18, the position of the principal ray from the assumed exit pupil position is indicated by the circle LP. When focusing on each microlens 31, the fourth-focus detection pixels 302SNL and 302SNR are arranged point-symmetrically with respect to this circle LP. Specifically, the fourth-focus detection pixel 302SNL is located immediately to the lower right of the circle LP, and the fourth-focus detection pixel 302SNR is located immediately to the upper left of the circle LP. The same applies to the fifth-focus detection pixels 302LNL, 302LNR.

[0120] The fourth focus detection pixels 302SNL and 302SNR are positioned at a certain distance to the right of the optical axis of the microlens 31. The distance from the optical axis of the microlens 31 is shorter than that of the second focus detection pixels 302SL and 302SR. Similarly, the fifth focus detection pixels 302LNL and 302LNR are positioned at a certain distance to the left of the optical axis of the microlens 31 (the opposite direction from the fourth focus detection pixels 302SNL and 302SNR). The distance from the optical axis of the microlens 31 is shorter than that of the third focus detection pixels 302LL and 302LR.

[0121] Using the image sensor 210a configured as described above, a so-called refocus function can be realized, which synthesizes an image of an arbitrary image plane within a predetermined range in the optical axis direction (a so-called refocus image) from the imaging signal output by the imaging and focus detection pixel 301. The refocus function will be described below.

[0122] Figure 19 is a schematic cross-sectional view showing the light beam from a light point P on the image plane S to be combined with the image sensor 210a. In Figure 19, consider a light point P provided on the image plane S to be combined. The angle of light spread θ from this light point P toward the image sensor 12 is determined by the pupil size of the imaging optical system 110 (i.e., the aperture value of the imaging optical system 110). The aperture value of the microlens 31 is configured to be the same as or smaller than the aperture value of the imaging optical system 110. Therefore, a light beam emitted from this light point P and incident on a certain microlens 31 does not spread outside the region covered by that microlens 31.

[0123] Here, as shown in Figure 19, if the light beam from light point P is incident on five microlenses 31(1) to 31(5), then by integrating the amount of incident light (photoelectric conversion output of photoelectric conversion units 34(1) to 34(5)) on the light-receiving surface of the light beam 300(1) to 300(5) incident on these microlenses 31(1) to 31(5), the total amount of incident light limited to the pupil from light point P can be obtained. In other words, the amount of light from light point P (the pixel to be combined) on the image plane S to be combined can be obtained.

[0124] The body CPU 220 sets multiple light points P on a designated image plane S and identifies a microlens 31 into which the light beam from each light point P is incident. For each identified microlens 31, the body CPU 220 identifies which photoelectric conversion unit 34 the light beam from the light point P is incident on. The body CPU 220 calculates the pixel value of the light point P by integrating the photoelectric conversion outputs of the identified photoelectric conversion units 34. If the identified photoelectric conversion unit 34 is the photoelectric conversion unit 34 of the focus detection pixel 302, it is desirable to use the photoelectric conversion output obtained by interpolation calculation based on the photoelectric conversion outputs of the surrounding photoelectric conversion units 34 instead of the photoelectric conversion output from that photoelectric conversion unit 34.

[0125] Through the above process, the image of the image plane S designated as the target for synthesis is synthesized. The body CPU 220 can synthesize images of multiple different image planes from the imaging signal output from the image sensor 210a in a single imaging cycle. In other words, images of multiple image planes can be obtained from a single imaging result.

[0126] However, there are certain constraints on the position of the image plane that can be suitably combined while maintaining a certain resolution through the above processing. For example, it is known that combining the image plane near the vertex of the microlens 31 results in a decrease in resolution compared to combining image planes at other positions. Thus, there are constraints on the position of the image plane that can be combined. Therefore, the body CPU 220 recognizes the main subjects included in the imaging range using well-known subject recognition technology, etc., and drives the focusing lens 112 so that these main subjects are included in a range that can be suitably combined (hereinafter simply referred to as the combined range). In this embodiment, focus adjustment refers to adjusting the position of the focusing lens 112 so that the main subjects are included in the combined range.

[0127] Next, the focus detection calculation for small defocus performed by the first focus detection unit 221 will be described. The first focus detection unit 221 first acquires focus detection signals from imaging and focus detection pixels 301 near the focus detection area. For example, it selects a number of microlenses 31 arranged in a horizontal row near the focus detection area. Then, for each of the selected microlenses 31, it generates a signal by adding the light-receiving outputs of the photoelectric conversion unit 34 of the imaging and focus detection pixel 301 located on the left half of the microlens 31, and a signal by adding the light-receiving outputs of the photoelectric conversion unit 34 of the imaging and focus detection pixel 301 located on the right half of the microlens 31. These pairs of signals can be treated as a pair of focus detection signals, similar to the light-receiving outputs of the photoelectric conversion units 34L and 34R in the first embodiment. The first focus detection unit 221 performs a correlation calculation to calculate the phase difference of this pair of focus detection signals and calculates the amount of defocus. Since such calculations are well known, a detailed explanation will be omitted.

[0128] In the above explanation, the received light outputs of the photoelectric conversion units 34 of the imaging and focus detection pixels 301, which are arranged on the left and right halves of the microlens 31, were added together. However, by reducing the number of photoelectric conversion units 34 used for addition, a focus detection signal can be obtained that appears as if the amount of incident light has been limited. The fewer photoelectric conversion units 34 used for addition the more accurate the system becomes in handling large defocus, but on the other hand, the signal amount of the focus detection signal becomes smaller and the accuracy deteriorates. Therefore, for example, in the case of small defocus, it is desirable to add the received light outputs from many photoelectric conversion units 34. On the other hand, by generating a pair of focus detection signals using only the received light outputs from two (a pair) of photoelectric conversion units 34, it is possible to handle large defocus.

[0129] Furthermore, in cases of significant defocus, even a pair of focus detection signals using only the light-receiving outputs from the two (pair) photoelectric conversion units 34 may not be able to detect focus. In such cases, in this embodiment, focus adjustment is performed using the result of a focus detection calculation for significant defocus performed by the second focus detection unit 222.

[0130] Next, the focus detection calculation for large defocus performed by the second focus detection unit 222 will be explained. The second focus detection unit 222 first selects a microlens array 610 from the vicinity of the focus detection area according to the position of the exit pupil of the imaging optical system 110. For example, if the focus detection area is near the center of the shooting screen and only a microlens array 610N is present in that vicinity, then that microlens array 610N is inevitably selected.

[0131] On the other hand, if the focus detection area is located away from the center of the shooting screen, and there are three microlens rows 610S, 610N, and 610L nearby, one of the microlens rows 610 is selected from among them depending on the position of the exit pupil of the imaging optical system 110. Specifically, if the exit pupil of the imaging optical system 110 is in a relatively distant position (further than a predetermined distance), microlens row 610L is selected. Conversely, if the exit pupil of the imaging optical system 110 is in a relatively close position (close within a predetermined distance), microlens row 610S is selected. If it is neither (in an intermediate position), microlens row 610N is selected.

[0132] Next, the second focus detection unit 222 acquires a focus detection signal from the focus detection pixels 302 in the selected microlens array 610. For example, if microlens array 610N is selected, then first focus detection pixels 302NL and 302NR are arranged in a row within that microlens array 610N. The second focus detection unit 222 takes a pair of signals as a pair of focus detection signals: an output signal obtained by arranging the outputs of a large number of first focus detection pixels 302NL (the light-receiving outputs of the photoelectric conversion unit 34) and an output signal obtained by arranging the outputs of a large number of first focus detection pixels 302NR (the light-receiving outputs of the photoelectric conversion unit 34). Then, it performs a correlation calculation to calculate the phase difference of this pair of focus detection signals and calculates the amount of defocus. Since such calculations are well known, an explanation will be omitted.

[0133] The pair of focus detection signals obtained here are more capable of handling large defocus issues than the pair of focus detection signals using only the light-receiving outputs from the two (pair) photoelectric conversion units 34 described above, because the aperture is restricted by the light-shielding member 35.

[0134] The camera system according to the third embodiment described above provides the following effects and advantages. (1) A focus detection pixel 302 and multiple imaging and focus detection pixels 301 are provided, to which light passing through one microlens 31 is incident, and a focus detection pixel 302 and multiple imaging and focus detection pixels 301 are incident to which light passing through another microlens 31 is incident. In this way, even in a so-called refocus camera, appropriate focus detection is possible whether the camera is in a state of great defocus or a state of slight defocus.

[0135] The following modifications are also within the scope of the present invention, and it is possible to combine one or more of these modifications with the embodiments described above.

[0136] (Variation 1) In the embodiment described above, a light-shielding member 35 having an opening of a predetermined width was provided in the focus detection pixel 302. However, instead of providing the light-shielding member 35, the width of the photoelectric conversion unit 34 may actually be reduced.

[0137] Figure 20(a) illustrates a first focus detection pixel 302NR' having a photoelectric conversion section 34NR formed to match the position and size of the opening 36NR, instead of the light-shielding member 35 having an opening 36NR.

[0138] Furthermore, the first focus detection pixel 302NR and the first focus detection pixel 302NL may be a single pixel having a pair of photoelectric conversion units. That is, for one pixel, a photoelectric conversion unit 34 of a size corresponding to the aperture 36NR may be provided at a position corresponding to the aperture 36NR, and another photoelectric conversion unit 34 of a size corresponding to the aperture 36NL may be provided at a position corresponding to the aperture 36NL, thereby replacing the first focus detection pixels 302NR and NL in the above-described embodiment. The same applies to the second to fifth focus detection pixels.

[0139] Alternatively, for each pixel, a pair of photoelectric conversion units 34L and 34R similar to those of the imaging and focus detection pixel 301 can be provided, and both openings 36NL and NR can be provided in the light-shielding member 35. The same applies to the second to fifth focus detection pixels.

[0140] (Variation 3) If a light-shielding member 35 is provided on the focus detection pixel 302, there is a possibility that light beams that did not enter the photoelectric conversion unit 34 (light beams that entered the light-shielding member 35) may leak into the adjacent imaging and focus detection pixel 301. To prevent such stray light, an anti-reflective coating may be provided on the surface of the light-shielding member 35. Figure 20(b) illustrates a first focus detection pixel 302NR'' in which an anti-reflective coating 37 is provided on the surface of the light-shielding member 35.

[0141] (Modification 4) The number of types, arrangement, and arrangement of focus detection pixels 302 may differ from those of the embodiments described above. Similarly, the number of types, arrangement, shape, and arrangement pattern of the regions where focus detection pixels 302 are arranged, as exemplified by the first region 60a, second region 60b, and third region 60c, may differ from those of the embodiments described above.

[0142] Figure 21(a) shows an example in which the first region 60a, the second region 60b, and the third region 60c are arranged in an arc shape. Figure 21(b) shows an example in which the first region 60a, the second region 60b, and the third region 60c are arranged in concentric circles. Thus, various shapes and arrangement patterns are possible for the regions in which the focus detection pixels 302 are placed.

[0143] (Variation 5) The shape of the pair of photoelectric conversion units 34R and 34L of the imaging and focus detection pixel 301 may differ from that shown in Figure 3(a). For example, they may be rectangular, as shown in Figure 22(a). Also, the focus detection direction is not limited to the X-axis direction, but may be the Y-axis direction, for example. In this case, a pair of photoelectric conversion units 34T and 34B divided in the Y-axis direction may be provided, for example, as shown in Figure 22(b).

[0144] Furthermore, both the X-axis and Y-axis directions can be used as focus detection directions. For example, as shown in Figure 22(c), four photoelectric conversion units 34a, 34b, 34c, and 34d are provided. In this case, adding the received signals of the two photoelectric conversion units 34a and 34c located on the left side of the page yields a signal corresponding to the received signal of the photoelectric conversion unit 34L shown in Figure 22(a). Also, adding the received signals of the two photoelectric conversion units 34a and 34b located on the upper side of the page yields a signal corresponding to the received signal of the photoelectric conversion unit 34T shown in Figure 22(b). Therefore, by configuring the imaging and focus detection pixel 301 as shown in Figure 22(c), and providing a focus detection pixel 302 having photoelectric conversion units with reduced widths in the X-axis and Y-axis directions compared to the four photoelectric conversion units 34a, 34b, 34c, and 34d, both the X-axis and Y-axis directions can be used as focus detection directions.

[0145] The present invention is not limited to the embodiments described above, as long as the features of the present invention are not impaired, and other forms that can be conceivable within the scope of the technical idea of ​​the present invention are also included within the scope of the present invention. [Explanation of symbols]

[0146] 1…Camera system, 100…Interchangeable lens, 110…Imaging optical system, 112…Focusing lens, 120…Lens CPU, 130…Aperture, 200…Camera body, 210…Image sensor, 220…Body CPU, 221…First focus detection unit, 222…Second focus detection unit, 230…Focus adjustment unit, 240…ROM, 250…Display device, 301, 301a, 301b…Imaging and focus detection pixels, 302…Focus detection pixels

Claims

[Claim 1] It receives light transmitted through the imaging optical system and outputs a signal used for focus detection, and has multiple rows of first pixels arranged along a first direction, The imaging optical system receives light transmitted through it and outputs a signal used for focus detection, and has a plurality of second pixel rows having a plurality of second pixels arranged along the first direction, Multiple third pixels that receive light transmitted through the imaging optical system and output a signal used for imaging, A circuit that selects either the first pixel row or the second pixel row based on information regarding the exit pupil position of the imaging optical system, An image sensor comprising, The first pixel receives light through an opening provided at a first position in the light-shielding portion, The second pixel receives light through an opening provided at a second position different from the first position of the light-shielding portion, The image sensor comprises a first region where only one of the plurality of first pixel rows and the plurality of second pixel rows and the third pixel are arranged, and a second region which is the same size as the first region but has a higher image height in the first direction than the first region, and where both the plurality of first pixel rows and the plurality of second pixel rows and the third pixel are arranged. Image sensor.

Images