Image sensor and imaging device

Abstract

The present invention provides an image sensor and an imaging device. An image sensor includes: a plurality of microlenses; and a pixel array that, with respect to each of the microlenses, has: a pair of first regions formed at a first depth from a surface onto which light is incident, a pair of second regions formed at a second depth deeper than the first depth, and a plurality of connection regions that respectively connect the pair of first regions and the pair of second regions. The direction in which the pair of second regions corresponding to each microlens is arranged is a first direction, and the direction in which the pair of first regions is arranged is either the first direction or a second direction orthogonal to the first direction.

Description

Technical Field

[0001] The present invention relates to an image sensor and an imaging device equipped with the image sensor, in which a plurality of pixel units each having a plurality of photoelectric conversion units are two-dimensionally arranged. Background Art

[0002] As one of the focus detection methods performed in an imaging device, a so-called imaging surface phase difference method is known: in this method, a pupil division signal pair formed in an image sensor is used to obtain a pupil division signal pair, and the pupil division signal pair is used to perform focus detection by the phase difference method.

[0003] In such an image sensor, International Publication No. 2016 / 53519 discloses suppressing color mixing of a phase difference detection signal into a captured image signal by forming photoelectric conversion portions at different depths to perform photoelectric conversion on visible light in different wavelength ranges. Further, it also discloses: by forming a pair of regions mainly for photoelectrically converting red light in a deep part and a pair of regions mainly for photoelectrically converting blue light in a shallow part, and arranging these pairs of regions in directions orthogonal to each other, to obtain focus detection signals having a phase difference in two different directions based on one pixel.

[0004] However, the image sensor disclosed in International Publication No. 2016 / 53519 uses transfer transistors for transmitting signals from respective photoelectric conversion portions formed at different depths, or vertical transfer transistors generally used for transmitting signals from a plurality of photoelectric conversion portions formed at different depths. In the case where transfer transistors are provided in respective photoelectric conversion portions, the number of transfer transistors is large and the area of the light receiving region becomes small. In this case, there are problems of high manufacturing cost and small saturated charge amount that the photoelectric conversion portion can hold. Further, in the case of using vertical transfer transistors, there is a problem that the manufacturing cost of the image sensor increases because the process for forming the transistors is complicated. Summary of the Invention

[0005] The present invention has been made in view of the above circumstances, and the present invention increases the saturated charge amount with a simple configuration while maintaining the number of pupil division directions of the phase difference signal for focus detection at 2.

[0006] According to the present invention, there is provided an image sensor including: a plurality of microlenses; and a pixel array that, with respect to each of the microlenses, has: a pair of first regions formed at a first depth from a surface on which light is incident, a pair of second regions formed at a second depth deeper than the first depth, and a plurality of connection regions that respectively connect the pair of first regions and the pair of second regions, wherein the direction in which the pair of second regions corresponding to each microlens is arranged is a first direction, and the direction in which the pair of first regions is arranged is the first direction or a second direction orthogonal to the first direction.

[0007] According to the present invention, there is provided an imaging device including: an image sensor and a focus detection component. The image sensor includes: a plurality of microlenses; and a pixel array that, with respect to each of the microlenses, has a pair of first regions formed at a first depth from a light incident surface, a pair of second regions formed at a second depth deeper than the first depth, and a plurality of connection regions that respectively connect the pair of first regions and the pair of second regions. The focus detection component is configured to obtain pupil division signals respectively corresponding to divided pupil regions based on signals output from the pixel array, and perform phase difference focus detection based on the pupil division signals, wherein the direction in which the pair of second regions corresponding to each microlens is arranged is a first direction, and the direction in which the pair of first regions is arranged is the first direction or a second direction orthogonal to the first direction.

[0008] Other features of the present invention will become apparent from the following description of exemplary embodiments (with reference to the drawings). BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The drawings incorporated in and constituting a part of this specification illustrate embodiments of the present invention and, together with the description, serve to explain the principles of the present invention.

[0010] Figure 1 is a diagram schematically showing the overall configuration of an image sensor according to an embodiment of the present invention.

[0011] Figure 2 is an equivalent circuit diagram of a pixel according to a first embodiment.

[0012] Figure 3 is a schematic diagram showing the basic layout of elements constituting a pixel according to a first embodiment.

[0013] Figures 4A to 4C is schematically showing Figure 3 a cross-sectional view of the cross-sectional structure of the pixel shown.

[0014] Figures 5A to 5D is Figures 4A to 4CA plan view of the photosensitive region of the pixel shown at various depths in the z direction.

[0015] Figure 6 Is schematically shown including Figure 3 A cross-sectional view of the cross-sectional structure of the pixel including the photosensitive region and the N-type connection region of the pixel shown.

[0016] Figures 7A to 7C Is a schematic diagram showing the first arrangement of the semiconductor region in the first embodiment.

[0017] Figures 8A to 8C Is a schematic diagram showing the second arrangement of the semiconductor region in the first embodiment.

[0018] Figures 9A to 9C Is a schematic diagram showing the third arrangement of the semiconductor region in the first embodiment.

[0019] Figures 10A to 10C Is a schematic diagram showing the fourth arrangement of the semiconductor region in the first embodiment.

[0020] Figure 11 Is a diagram schematically showing a pixel array in the range of 4 rows × 2 columns in the first embodiment.

[0021] Figure 12 Is a schematic diagram for explaining the pixel array in the modification example.

[0022] Figure 13 Is a diagram schematically showing a pixel array in the range of 2 rows × 2 columns in the modification example.

[0023] Figure 14 Is a diagram schematically showing a pixel array in the range of 2 rows × 2 columns in the modification example.

[0024] Figure 15 Is an equivalent circuit diagram of two pixels according to the second embodiment.

[0025] Figure 16A And 16B Is a schematic diagram showing the fifth arrangement of the semiconductor region in the second embodiment.

[0026] Figure 17A And 17B Is a schematic diagram showing the sixth arrangement of the semiconductor region in the second embodiment.

[0027] Figure 18A And 18B Is a schematic diagram showing the seventh arrangement of the semiconductor region in the second embodiment.

[0028] Figure 19 Is a diagram schematically showing a pixel array in the range of 4 rows × 4 columns in the second embodiment.

[0029] Figure 20 is a block diagram showing a schematic configuration of an imaging device according to a third embodiment.

[0030] Figure 21 is a diagram showing the relationship between pixels divided in the x direction and a partial pupil region according to a third embodiment.

[0031] Figure 22 is a conceptual diagram showing an example of a pupil intensity distribution in a partial pupil region according to a third embodiment.

[0032] Figure 23 is a diagram showing a sensor entrance pupil of an image sensor according to a third embodiment.

[0033] Figure 24 is a diagram showing a schematic relationship between an image offset amount and a defocus amount between parallax images according to a third embodiment.

[0034] Figures 25A to 25C is a diagram showing the relationship between a partial pupil region and an exit pupil of an imaging optical system with respect to pixels located at a peripheral image height in the x direction of an image sensor according to a third embodiment.

[0035] Figures 26A to 26C is a diagram showing according to a third embodiment Figures 25A to 25C a diagram showing the relationship of the x-direction dependence between the exit pupil and the pupil intensity distribution in each state shown.

[0036] Figure 27 is a diagram showing pixel positions according to a third embodiment.

[0037] Figure 28 is a diagram showing the relationship between pixels divided in the y direction and a partial pupil region according to a third embodiment.

[0038] Figures 29A to 29C is a diagram showing the relationship between a partial pupil region and an exit pupil of an imaging optical system with respect to pixels located at a peripheral image height in the y direction of an image sensor according to a third embodiment.

[0039] Figures 30A to 30C is a diagram showing according to a third embodiment Figures 29A to 29C a diagram showing the relationship of the y-direction dependence between the exit pupil and the pupil intensity distribution in each state shown.

[0040] Figures 31A to 31C is a diagram showing the relationship between a partial pupil region and an exit pupil of an imaging optical system with respect to pixels located at a peripheral image height in the x and y directions of an image sensor according to a third embodiment.

[0041] Figures 32A to 32C A diagram showing the x-direction dependence and y-direction dependence relationships between the exit pupil and the pupil intensity distribution in each of the states shown according to the third embodiment. Figures 31A to 31C A diagram showing the relationship between the exit pupil and the pupil intensity distribution in each of the states shown according to the third embodiment. Detailed Description of the Invention

[0042] Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. Note that the following embodiments are not intended to limit the scope of the claimed invention, and no limitation is made that the invention requires a combination of all the features described in the embodiments. Two or more of the multiple features described in the embodiments may be appropriately combined. In addition, the same or similar configurations are given the same reference numerals, and redundant descriptions thereof are omitted.

[0043] <First Embodiment>

[0044] Figure 1 A diagram schematically showing the overall configuration of an image sensor 100 according to a first embodiment of the present invention. The image sensor 100 includes a pixel array (pixel unit) 101, a vertical selection circuit 102, a column circuit 103, and a horizontal selection circuit 104.

[0045] A plurality of pixels 105 are arranged in a matrix in the pixel array 101. The output of the vertical selection circuit 102 is input to the pixels 105 via a pixel drive wiring group 107, and the pixel signals of the pixels 105 in the row selected by the vertical selection circuit 102 are read out to the column circuit 103 via output signal lines 106. One output signal line 106 may be provided for each pixel column or a plurality of pixel columns, or a plurality of output signal lines 106 may be provided for each pixel column. The signals read out in parallel via the plurality of output signal lines 106 are input to the column circuit 103, and the column circuit 103 performs processes such as signal amplification, noise reduction, and A / D conversion, and holds the processed signals. The horizontal selection circuit 104 sequentially, randomly, or simultaneously selects the signals held in the column circuit 103, such that the selected signals are output to the outside of the image sensor 100 via a horizontal output line and an output unit (not shown).

[0046] By sequentially performing the operation of outputting the pixel signals of the row selected by the vertical selection circuit 102 to the outside of the image sensor 100 while changing the row to be selected in the vertical selection circuit 102, a two-dimensional imaging signal or a focus detection signal can be read out from the image sensor 100.

[0047] Figure 2 An equivalent circuit diagram of the pixel 105 in this embodiment.

[0048] Each pixel 105 has two photodiodes 201 (PDA) and 202 (PDB) as a photoelectric conversion unit. The signal charge that is photoelectrically converted and accumulated in the PDA 201 corresponding to the amount of incident light is transferred via a transfer switch (TXA) 203 to a floating diffusion unit (FD) 205 that constitutes a charge storage unit. In addition, the signal charge that is photoelectrically converted and accumulated in the PDB 202 is transferred via a transfer switch (TXB) 204 to the FD 205. The reset switch (RES) 206 is turned on to reset the FD 205 to the voltage of a constant voltage source VDD. In addition, by turning on the RES 206, TXA 203, and TXB 204 simultaneously, the PDA 201 and PDB 202 can be reset.

[0049] When a selection switch (SEL) 207 for selecting a pixel is turned on, an amplification transistor (SF) 208 converts the signal charge accumulated in the FD 205 into a voltage, and the converted signal voltage is output from the pixel to an output signal line 106. In addition, the gates of the TXA 203, TXB 204, RE 206, and SEL 207 are connected to a pixel drive wiring group 107 and are controlled by a vertical selection circuit 102.

[0050] In the following description, in the present embodiment, the signal charge accumulated in the photoelectric conversion unit is an electron, the photoelectric conversion unit is formed of an N-type semiconductor, and a P-type semiconductor is used to separate the N-type semiconductor. However, the signal charge may be a hole, the photoelectric conversion unit may be formed of a P-type semiconductor, and an N-type semiconductor may be used to separate the P-type semiconductor.

[0051] Next, in the pixel having the above configuration, an operation of reading signal charges from the PDA 201 and PDB 202 after a predetermined charge accumulation period has elapsed since the PDA 201 and PDB 202 were reset will be described. First, when the SEL 207 of each pixel 105 in the row selected by the vertical selection circuit 102 is turned on and the source of the SF 208 and the output signal line 106 are connected, a voltage corresponding to the voltage of the FD 205 can be read out to the output signal line 106. Then, the RES 206 is turned on and then off, thereby resetting the potential of the FD 205. After that, the process waits until the output signal line 106 affected by the voltage fluctuation of the FD 205 stabilizes, and the column circuit 103 processes the voltage of the stable output signal line 106 as a signal voltage N and holds the processed signal voltage N.

[0052] Thereafter, TXA203 is turned on and then off, thereby transferring the signal charge stored in PDA 201 to FD205. The voltage of FD 205 drops by an amount corresponding to the amount of signal charge stored in PDA201. Thereafter, the process waits until the output signal line 106 affected by the voltage fluctuation of FD 205 stabilizes, and the column circuit 103 extracts the voltage of the stable output signal line 106 as the signal voltage A, processes it, and holds the processed signal voltage A.

[0053] Thereafter, TXB 204 is turned on and then off, thereby transferring the signal charge stored in PDB 202 to FD205. The voltage of FD 205 drops by an amount corresponding to the amount of signal charge stored in PDB 202. Thereafter, the process waits until the output signal line 106 affected by the voltage fluctuation of FD 205 stabilizes, and the column circuit 103 extracts the voltage of the stable output signal line 106 as the signal voltage (A + B), processes it, and holds the processed signal voltage (A + B).

[0054] Based on the difference between the signal voltage N and the signal voltage A obtained in this way, the signal A corresponding to the amount of signal charge stored in PDA201 can be obtained. In addition, based on the difference between the signal voltage A and the signal voltage (A + B), the signal B corresponding to the amount of signal charge accumulated in PDB 202 can be obtained. This difference calculation can be performed by the column circuit 103 or after these signals are output from the image sensor 100. The phase difference signal can be obtained by independently using the signals A and B, and the captured image signal can be obtained by adding the signals A and B. Alternatively, in the case where the difference calculation is performed after the signal voltage is output from the image sensor 100, the captured image signal can be obtained by obtaining the difference between the signal voltage N and the signal voltage (A + B).

[0055] · Basic structure of pixel

[0056] Next, reference will be made to Figures 3 to 6 to describe in detail the basic structure of the pixel 105 of this embodiment.

[0057] Figure 3 is a schematic diagram showing the basic layout of the elements constituting the pixel 105 according to this embodiment. In Figure 3 the horizontal direction is the x direction, the vertical direction is the y direction, and the protruding direction is the z direction. In addition, in this embodiment, the "plan view" means observing from the z direction or the -z direction with respect to a plane (xy plane) substantially parallel to the surface of the semiconductor substrate on which the transistor gate is arranged. In addition, in this embodiment, the "horizontal" direction means the x direction, the "vertical" direction means the y direction, and the "depth" direction means the z direction.

[0058] In Figure 3 , the configurations identical to those in Figure 1 and 2 are given the same reference numerals, and their detailed description will be omitted. In Figure 3 , 301 represents a microlens (ML); 303 represents the gate electrode of TXA 203; 304 represents the gate electrode of TXB 204; 306 represents the gate electrode of RES 206; 307 represents the gate electrode of SEL 207; 308 represents the gate electrode of SF 208; and 310 represents a voltage supply line.

[0059] PDA 201 includes a cumulative region 311, a photosensitive region 313, and an N-type connection region 315, and PDB 202 includes a cumulative region 312, a photosensitive region 314, and an N-type connection region 316. These cumulative regions 311 and 312, photosensitive regions 313 and 314, and N-type connection regions 315 and 316 are made of N-type semiconductors. The areas of the photosensitive regions 313 and 314 are larger than those of the cumulative regions 311 and 312, respectively. In addition, as will be described in detail with reference to Figures 4A to 4C , the cumulative regions 311 and 312 are formed at a first depth, while the photosensitive regions 313 and 314 are formed at a second depth different from the first depth. In addition, for clarity, the region that mainly generates charges in response to incident light is referred to as the "photosensitive region", and the region where the generated charges mainly accumulate is referred to as the "cumulative region". However, there is no clear distinction between the charge generation region and the charge accumulation region. In response to the light that has arrived, charges are also generated in the cumulative regions 311 and 312, and a part of the generated charges also stays in the photosensitive regions 313 and 314.

[0060] Figures 4A to 4C is a diagram schematically showing the basic cross-sectional structure of the pixel 105. Figure 4A is Figure 3 a schematic diagram of the A-A' cross-section of Figure 4B is Figure 3 a schematic diagram of the B-B' cross-section of Figure 4C is Figure 3Schematic diagram of the C-C' cross-section. In this embodiment, pixel 105 is disposed on semiconductor substrate 401, and semiconductor substrate 401 has a first surface and a second surface opposite to the first surface. The first surface is the front surface of semiconductor substrate 401, and the second surface is the back surface of semiconductor substrate 401. The direction from the second surface to the first surface is the positive direction in the Z direction. Transistor gate electrodes, multilayer wiring structures, etc. are disposed on the first surface (front surface) side of semiconductor substrate 401. In addition, optical structures such as color filter 406 and ML 301 that commonly cover two photodiodes of each pixel are disposed on the second surface (back surface) side of semiconductor substrate 401, and light is incident on the second surface (back surface) side.

[0061] As Figure 4A shown, in semiconductor substrate 401, P-type semiconductor region 400, and accumulation regions 311 and 312 and photosensitive regions 313 and 314 surrounded by P-type semiconductor region 400 are disposed. Similar to accumulation regions 312 and photosensitive region 314, accumulation regions 311 and photosensitive region 313 also have different shapes in the plan view, and partial regions thereof overlap in the plan view. In addition, as described above, accumulation regions 311 and 312 and photosensitive regions 313 and 314 are formed at different depths, and accumulation regions 311 and 312 are formed at a depth (first depth) closer to the first surface side, and photosensitive regions 313 and 314 are formed at a depth (second depth) closer to the second surface side. In P-type semiconductor region 400, accumulation separation region 402 separates accumulation region 311 and accumulation region 312, and photosensitive separation region 403 separates photosensitive region 313 and photosensitive region 314.

[0062] As Figure 4B shown, accumulation region 311 and photosensitive region 313 are connected in the depth direction via N-type connection region 315. In addition, as Figure 4C shown, accumulation region 312 and photosensitive region 314 are connected in the depth direction via N-type connection region 316.

[0063] In addition, in Figure 4BIn [the figure], the region 404 in the accumulation region 311 has a shape that is recessed in the Z direction and filled with a P-type semiconductor, and the recessed region 404 suppresses the generation of charge accumulation in the region where the first surface side of the accumulation region 311 overlaps with the N-type connection region 315 in the plan view. As a result, when the signal charges accumulated in the accumulation region 311 in the PDA 201 are transferred to the FD 205, the residual of the signal charges in the PDA 201 is suppressed. Note that the present invention is not limited to forming the recessed region 404 in order to suppress the residual of the signal charges, and other methods such as reducing the impurity concentration of a part of the accumulation region 311 may also be used as long as the residual of the signal charges can be suppressed.

[0064] In addition, as Figure 4C shown, the lengths of the accumulation regions 311 and 312 in the Z direction are shorter than the lengths of the accumulation regions 311 and 312 in the cross section shown in Figure 4A and 4B The region 405 corresponding to the shortened lengths of the accumulation regions 311 and 312 in the Z direction is filled with a P-type semiconductor. The reason for changing the length in this way will be described in detail later.

[0065] Figures 5A to 5D are XY cross-sectional views of the PDA 201 and the PDB 202 at different depths in the Z direction. Figure 5A is a cross-sectional view taken along the line E-E’ of Figures 4A to 4C , Figure 5B is a cross-sectional view taken along the line F-F’ of Figures 4A to 4C , Figure 5C is a cross-sectional view taken along the line G-G’ of Figures 4A to 4C , Figure 5D is a cross-sectional view taken along the line H-H’ of Figures 4A to 4C . As Figure 5D shown, in the partial regions of the accumulation regions 311 and 312 that are located far from the gate electrodes 303 and 304, as described with reference to Figure 4C , the accumulation regions 311 and 312 are not formed, and the cutout regions 405 of the P-type semiconductor replace the accumulation regions 311 and 312.

[0066] Figure 6 is a cross-sectional view taken along the line D-D’ of Figure 3 . By taking along Figure 3The line D-D’ extends in the XY plane, and the cumulative regions 311 and 312, the photosensitive regions 313 and 314, and the N-type connection regions 315 and 316 are shown in the same figure. When light is incident on the second surface of the semiconductor substrate 401 via the ML 301 during the charge accumulation period, electrons (signal charges) are mainly generated by photoelectric conversion in the photosensitive regions. Most of the signal charges generated in the photosensitive region 313 move to the cumulative region 311 via the N-type connection region 315 and accumulate there. In addition, most of the signal charges generated in the photosensitive region 314 move to the cumulative region 312 via the N-type connection region 316 and accumulate there. In order to achieve the signal charge transfer from the photosensitive region to the cumulative region, it is desirable that the potential affecting the electrons monotonically decreases on the charge transfer path from the photosensitive region to the cumulative region.

[0067] · Arrangement examples of the cumulative region and the photosensitive region

[0068] Since the cumulative region and the photosensitive region are arranged at different depths, the shapes of the cumulative region and the photosensitive region can be designed to extend in different directions. Hereinafter, with reference to Figures 7A to 10C the arrangement examples of the cumulative region and the photosensitive region will be described. The configurations having the same functions as those described above Figure 3 and Figures 4A to 4C are denoted by the same reference numerals.

[0069] Figures 7A to 7C is a schematic diagram showing the first arrangement of the semiconductor region according to the present embodiment. In the following description, the pixel 105 having the first arrangement is referred to as the pixel 700.

[0070] Figure 7A is an exploded perspective view of the cumulative regions 311 and 312, the photosensitive regions 313 and 314, the N-type connection regions 315 and 316, the gate electrode 303 of the TXA 203, the gate electrode 304 of the TXB 204, and the FD 205 in the pixel 700. In the first arrangement, all the cumulative regions 311 and 312 and the photosensitive regions 313 and 314 extend in the y direction, that is, in the same direction.

[0071] Figure 7B is a schematic plan view showing the positional relationship of the cumulative regions 311 and 312, the photosensitive regions 313 and 314, the N-type connection regions 315 and 316, the gate electrode 303 of the TXA 203, the gate electrode 304 of the TXB 204, and the FD 205 in the pixel 700 in a plan view. In the first arrangement, since the photosensitive regions 313 and 314 where charges are generated by photoelectric conversion are arranged in the x direction, a phase difference signal with the pupil division direction in the x direction can be obtained. The reference numeral 701 indicates the division direction of the phase difference signal.

[0072] Figure 7C It is a schematic plan view showing the positional relationship between the cumulative separation region 402 and the photosensitive separation region 403 of the pixel 700 in a plan view. In the first arrangement, both the cumulative separation region 402 and the photosensitive separation region 403 extend in the y direction.

[0073] Figures 8A to 8C It is a schematic view showing a second arrangement of the semiconductor region according to the present embodiment. In the following description, the pixel 105 having the second arrangement is referred to as the pixel 800.

[0074] Figure 8A It is an exploded perspective view of the cumulative regions 311 and 312, the photosensitive regions 313 and 314, the N-type connection regions 315 and 316, the gate electrode 303 of TXA 203, the gate electrode 304 of TXB 204, and FD 205 in the pixel 800. In the second arrangement, the cumulative regions 311 and 312 extend in the y direction, and the photosensitive regions 313 and 314 extend in the x direction orthogonal to the y direction, that is, in different directions.

[0075] Figure 8B It is a schematic plan view showing the positional relationship between the cumulative regions 311 and 312, the photosensitive regions 313 and 314, the N-type connection regions 315 and 316, the gate electrode 303 of TXA 203, the gate electrode 304 of TXB 204, and FD 205 of the pixel 800 in a plan view. In the second arrangement, since the photosensitive regions 313 and 314 that generate charges by photoelectric conversion are arranged in the y direction, a phase difference signal with the pupil division direction in the y direction can be obtained. The reference numeral 801 indicates the division direction of the phase difference signal.

[0076] Figure 8C It is a schematic plan view showing the positional relationship between the cumulative separation region 402 and the photosensitive separation region 403 of the pixel 800 in a plan view. In the second arrangement, the cumulative separation region 402 extends in the y direction, and the photosensitive separation region 403 extends in the x direction.

[0077] As described above, in the case where the extending directions of the photosensitive regions 313 and 314 are orthogonal to each other with respect to the extending directions of the accumulation regions 311 and 312, the photosensitive separation region 403 overlaps with the accumulation regions 311 and 312 in the plan view. In the regions of the accumulation regions 311 and 312 that overlap with the photosensitive separation region 403 in the plan view, the N-type concentration becomes thinner due to the influence of the implantation of the P-type impurities used to form the photosensitive separation region 403. Therefore, by shortening the thicknesses of the accumulation regions 311 and 312 in the z direction at positions away from the gate electrode 303 of TXA 203 and the gate electrode 304 of TXB 204, the residual of the signal charges in the region where the N-type concentration becomes low is suppressed. However, as long as the residual of the signal charges can be suppressed, other methods such as reducing the impurity concentration in a part of the accumulation regions 311 and 312 can also be used.

[0078] Figures 9A to 9C is a schematic diagram showing a third arrangement of the semiconductor regions according to the present embodiment. In the following description, the pixel 105 having the third arrangement is referred to as the pixel 900. In the third arrangement, the positions of the recessed region 404 and the N-type connection regions 315 and 316 are different from the positions of Figures 7A to 7C the first arrangement shown.

[0079] Figure 9A is an exploded perspective view of the accumulation regions 311 and 312, the photosensitive regions 313 and 314, the N-type connection regions 315 and 316, the gate electrode 303 of TXA 203, the gate electrode 304 of TXB 204, and FD 205 in the pixel 900. In the third arrangement, all of the accumulation regions 311 and 312 and the photosensitive regions 313 and 314 extend in the same direction (y direction) as in the first arrangement.

[0080] Figure 9B is a schematic plan view showing the positional relationship of the accumulation regions 311 and 312, the photosensitive regions 313 and 314, the N-type connection regions 315 and 316, the gate electrode 303 of TXA 203, the gate electrode 304 of TXB 204, and FD 205 in the pixel 900 in the plan view. In the third arrangement, as in the first arrangement, since the photosensitive regions 313 and 314 are arranged in the x direction, a phase difference signal with the pupil division direction in the x direction can be obtained. The reference numeral 901 indicates the division direction of the phase difference signal.

[0081] Figure 9C is a schematic plan view showing the positional relationship between the accumulation separation region 402 and the photosensitive separation region 403 in the pixel 900 in the plan view. In the third arrangement, as in the first arrangement, both the accumulation separation region 402 and the photosensitive separation region 403 extend in the y direction.

[0082] Figures 10A to 10C is a schematic diagram showing a fourth layout of a semiconductor region according to the present embodiment. In the following description, the pixel 105 with the fourth layout is referred to as pixel 1000. In the fourth layout, the positions of the recessed region 404 and the N-type connection regions 315 and 316 are different from those of the Figures 8A to 8C second layout shown, which results in different connections between the accumulation regions 311 and 312 and the photosensitive regions 313 and 314.

[0083] Figure 10A is an exploded perspective view of the accumulation regions 311 and 312, photosensitive regions 313 and 314, N-type connection regions 315 and 316, gate electrode 303 of TXA 203, gate electrode 304 of TXB 204, and FD 205 in pixel 1000. In the fourth layout, the accumulation regions 311 and 312 extend in the y direction, and the photosensitive regions 313 and 314 extend in the x direction orthogonal to the y direction in the plan view.

[0084] Figure 10B is a schematic plan view showing the positional relationship of the accumulation regions 311 and 312, photosensitive regions 313 and 314, N-type connection regions 315 and 316, gate electrode 303 of TXA 203, gate electrode 304 of TXB 204, and FD 205 in pixel 1000 in the plan view. In the fourth layout, since the photosensitive regions 313 and 314 that generate charges through photoelectric conversion are arranged in the y direction, a phase difference signal with a pupil division direction of the y direction can be obtained. Reference numeral 1001 indicates the division direction of the phase difference signal.

[0085] Figure 10C is a schematic plan view showing the positional relationship between the accumulation separation region 402 and the photosensitive separation region 403 in pixel 1000 in the plan view. In the fourth layout, the accumulation separation region 402 extends in the y direction, and the photosensitive separation region 403 extends in the x direction.

[0086] Similarly, in the fourth layout, in the regions of the accumulation regions 311 and 312 that overlap with the photosensitive separation region 403 in the plan view, due to the influence of the implantation of P-type impurities for forming the photosensitive separation region 403, the N-type concentration becomes thinner. Therefore, by shortening the thickness of the accumulation regions 311 and 312 in the z direction at positions away from the gate electrode 303 of TXA 203 and the gate electrode 304 of TXB 204, the residual of signal charges in the region where the N-type concentration becomes low is suppressed. However, as long as the residual of signal charges can be suppressed, other methods such as reducing the impurity concentration in a part of the accumulation regions 311 and 312 can also be used.

[0087] AsFigures 7A to 10C As shown, in the first to fourth arrangements, the cumulative regions 311 and 312, the gate electrodes 303 of TXA 203, the gate electrodes 304 of TXB 204, and the FD 205 are arranged in the same manner. Therefore, the layouts of RES 206, SF 208, SEL 207, and the metal wiring can be made the same in the first to fourth arrangements. As a result, the capacitance of FD 205 can be made the same for all pixels, and the gain for converting the signal charge transferred to FD 205 into voltage and the gain for reading the voltage of SF 208 can be set to be equal for all pixels.

[0088] · Pixel layout

[0089] Next, an example of the pixel layout of pixels 700, 800, 900, and 1000 having the above configuration will be described.

[0090] Figure 11 is a schematic diagram showing the pixel layout of the pixel array 101 in the range of 4 rows × 2 columns. By arranging a large number of Figure 11 the pixel sets of 4 rows × 2 columns shown, a captured image signal and a phase difference signal can be obtained. In Figure 11 it, ML 301, the photosensitive regions 313 and 314, the N-type connection regions 315 and 316, the gate electrodes 303 of TXA 203, the gate electrodes 304 of TXB 204, FD 205, the division direction of the phase difference signal, and the color filter 406 are shown.

[0091] The color filter (CFR) 1101 having R (red) spectral transmittance characteristics, the color filter (CFG) 1102 having G (green) spectral transmittance characteristics, and the color filter 1103 having B (blue) spectral transmittance characteristics are used as the color filter 406. These color filters 406 are arranged in a Bayer array. In addition, in Figure 11 the example shown, the pixels are arranged in the order of pixel 700, pixel 700, pixel 900, and pixel 900 from the top to the bottom of the right column. In addition, the pixels are arranged in the order of pixel 700, pixel 800, pixel 900, and pixel 1000 from the top to the bottom of the left column.

[0092] By using the phase difference signal obtained based on pixel 700 and pixel 900, a phase difference signal with the pupil division direction in the x-direction can be obtained. In addition, by using the phase difference signal obtained based on pixel 800 and pixel 1000, a phase difference signal with the pupil division direction in the y-direction can be obtained. In the case where the contrast in the x-direction is high (such as when the subject has a pattern close to a vertical stripe pattern, etc.), the phase difference signal with the pupil division direction in the x-direction is more effective. On the other hand, in the case where the contrast in the y-direction is high (such as when the subject has a pattern close to a horizontal stripe pattern, etc.), the phase difference signal with the pupil division direction in the y-direction is more effective.

[0093] The cumulative regions of the pixels arranged as described above are divided into type 1 and type 2. In type 1, the cumulative regions (the cumulative region 311 of pixel 700, the cumulative region 311 of pixel 800, the cumulative region 312 of pixel 900, and the cumulative region 312 of pixel 1000) are adjacent to both the recessed region 404 and the notch region 405 of the P-type semiconductor. In type 2, the cumulative regions (the cumulative region 312 of pixel 700, the cumulative region 312 of pixel 800, the cumulative region 311 of pixel 900, the cumulative region 311 of pixel 1000) are adjacent only to the notch region 405 of the P-type semiconductor. Since the volume of the N-type region of the cumulative region in type 1 is smaller than the volume of the N-type region of the cumulative region in type 2, the amount of signal charge that can be stored in the cumulative region in type 1 can be smaller.

[0094] Since the imaging signal is a signal obtained by adding signal A and signal B, the saturation charge amount of the imaging signal is uniform among all pixels. On the other hand, since the phase difference signal independently uses signal A and signal B, the saturation charge amount of the phase difference signal is different between the first arrangement and the second arrangement and between the third arrangement and the fourth arrangement. Considering the above characteristics, as Figure 11 shown, a large number of pixel sets arranged as such are arranged on a plane to form a pixel array 101, and by performing signal processing such as averaging or using only the signals of specific pixels according to the pixel arrangement, the saturation charge amount of the phase difference signal obtained in the overall region of the pixel array 101 can be made uniform.

[0095] For example, in a case where the intensity distributions of the phase difference signals obtained only from pixel 700 and the intensity distributions of the phase difference signals obtained only from pixel 900 have significantly different shapes from each other, there is a high possibility that the phase difference signal of either pixel 700 or 900 will saturate. In such a case, among the phase difference signals based only on the phase difference signal of pixel 700 and the phase difference signal based only on pixel 900, it can be assumed that the phase difference signal having a smaller signal intensity difference between signal A and signal B will include a saturated phase difference signal. Therefore, the phase difference signal having a larger signal intensity difference between signal A and signal B based on pixel 700 or 900 is used for focus detection. By doing so, the phase difference signal based on the pixel having a larger saturation charge amount can be used, and thus the saturation charge amount of the phase difference signal to be used can be made uniform over the entire area of the pixel array.

[0096] In addition, by arranging the pixels as in the present embodiment, even when the mutual relationship between the accumulation regions 311 and 312, the N-type connection regions 315 and 316, and the P-type regions 404 and 405 changes due to alignment variations during manufacturing and the charge amount that can be accumulated in the accumulation regions 311 and 312 deviates from the designed structure, the saturation charge amount of the phase difference signal can be made uniform by performing signal processing such as averaging or using only specific signals according to the arrangement of the semiconductor regions in each pixel.

[0097] As described above, according to the first embodiment, when the pupil division direction is two, the capacitance and saturation charge amount of the FD can be made uniform, and thus image data that does not require interpolation processing can be acquired.

[0098] <Modification Example>

[0099] Next, a modification example of the first embodiment will be described. Components having the same functions as those in the first embodiment are denoted by the same reference numerals, and their descriptions will be appropriately omitted or simplified. In this modification example, another example of the pixel layout will be described.

[0100] Figure 12 is a schematic diagram showing the pixel array 101 in the modification example of the first embodiment, and in Figure 12 the partial pixel array 1201 on the left side and the partial pixel array 1202 on the right side with respect to the center in the x direction have different pixel layouts.

[0101] Figure 13 is a schematic diagram showing the pixel layout of the partial pixel array 1201 in the range of 2 rows × 2 columns, and by arranging a large number of Figure 13The partial pixel array 1201 is configured with the pixel set of 2 rows × 2 columns shown. The color filters are arranged as CFG 1102 and CFB 1103 from top to bottom in the right column, and CFR 1101 and CFG 1102 from top to bottom in the left column. The arrangement of the semiconductor regions is shown by pixels 900 and pixels 900 from top to bottom in the right column and pixels 900 and pixels 1000 from top to bottom in the left column.

[0102] Figure 14 is a schematic diagram showing the pixel layout of the partial pixel array 1202 in the range of 2 rows × 2 columns, and by arranging a large number of Figure 14 The partial pixel array 1202 is configured with the pixel set of 2 rows × 2 columns shown. The color filters are arranged as CFG 1102 and CFB 1103 from top to bottom in the right column, and CFR 1101 and CFG 1102 from top to bottom in the left column. The arrangement of the semiconductor regions is shown by pixels 700 and pixels 700 from top to bottom in the right column and pixels 700 and pixels 800 from top to bottom in the left column.

[0103] In the partial pixel array 1201, when detecting the phase difference in the x direction, the phase difference signal obtained according to pixel 900 is used. In pixel 900, the photosensitive region connected to the type 2 accumulation region having a larger saturation charge amount than the type 1 accumulation region is arranged on the side with a relatively small x coordinate in the pixel. On the other hand, in the partial pixel array 1202, when performing phase difference detection in the x direction, the phase difference signal obtained according to pixel 700 is used. In pixel 700, the photosensitive region connected to the type 2 accumulation region is arranged on the side with a relatively large x coordinate in the pixel. By arranging the pixels in this way, when a relatively large amount of light is incident on the side far from the center of the pixel array, the saturation charge amount of the phase difference signal can be increased in each pixel.

[0104] In this modification example, the microlens is optically designed to be eccentric toward the intersection point according to the distance from the intersection point of the pixel array and the optical axis, so that for all pixels, the center of the pupil division intersects the optical axis at a fixed distance from the pixel array. Hereinafter, this distance will be referred to as the sensor pupil distance. When the exit pupil distance of the imaging lens is less than the sensor pupil distance, more light is incident on the side of each pixel far from the intersection point of the optical axis than on the side of each pixel close to the intersection point. In this modification example, since the photosensitive region connected to the accumulation region having a larger saturation charge amount is arranged on the side far from the optical axis in each pixel, the saturation charge amount of the phase difference signal becomes large.

[0105] Therefore, by adopting a pixel array layout as in the modification of the first embodiment, the saturation charge amount of the phase difference signal can be increased according to the relationship between the sensor pupil distance and the exit pupil distance of the imaging lens.

[0106] <Second Embodiment>

[0107] Next, a second embodiment of the present invention will be described.

[0108] In the second embodiment, a configuration in which the pixel signal reading circuit composed of FD 205 and its subsequent elements is shared by two pixels will be described. Configurations having the same functions as those in the first embodiment are denoted by the same reference numerals, and their detailed descriptions will be omitted or simplified.

[0109] Figure 15 is an equivalent circuit diagram of two pixels in the second embodiment. To distinguish the photodiodes of these two pixels, the pixel on the upper side of the figure is referred to as pixel 105U, and the two photodiodes of pixel 105U are hereinafter referred to as PD1A 1501 and PD1B 1502. In addition, the pixel on the lower side of the figure is referred to as pixel 105D, and the two photodiodes of pixel 105D are referred to as PD2A 1503 and PD2B 1504.

[0110] PD1A 1501 and PD1B 1502 share a microlens 301, and PD2A 1503 and PD2B 1504 share a microlens. These two pixels (i.e., pixel 105U and pixel 105D) share the pixel signal reading circuit composed of FD 205, RES 206, SEL 207, and SF208.

[0111] The signal charge obtained by photoelectric conversion according to the incident light amount on PD1A 1501 and accumulated in PD1A 1501 is transmitted to FD 205 via the transmission switch (TX1A) 1505. In addition, the signal charge obtained by photoelectric conversion and accumulated in PD1B1502 is transmitted to FD 205 via the transmission switch (TX1B) 1506. Similarly, the signal charge obtained by photoelectric conversion and accumulated in PD2A 1503 is transmitted to FD 205 via the transmission switch (TX2A) 1507, and the signal charge obtained by photoelectric conversion and accumulated in PD2B 1504 is transmitted to FD 205 via the transmission switch (TX2B) 1508.

[0112] · Arrangement Examples of Accumulation Region and Photosensitive Region

[0113] Figure 16A and 16BIt is a schematic diagram showing a fifth layout of semiconductor regions of two pixels according to the second embodiment. In the following description, two pixels 105U and 105D having the fifth layout are referred to as a pixel set 1600.

[0114] Figure 16A It is an exploded perspective view of the accumulation regions 1601, 1602, 1603 and 1604, photosensitive regions 1611, 1612, 1613 and 1614, N-type connection regions 1621, 1622, 1623 and 1624, gate electrodes 1631 of TX1A 1505, gate electrodes 1632 of TX1B 1506, gate electrodes 1633 of TX2A 1507, gate electrodes 1634 of TX2B 1508, and FD 205 in the pixel set 1600.

[0115] Figure 15 The illustrated PD1A 1501 is composed of the accumulation region 1601, the photosensitive region 1611, and the N-type connection region 1621, and PD1B 1502 is composed of the accumulation region 1602, the photosensitive region 1612, and the N-type connection region 1622. PD2A 1503 is composed of the accumulation region 1603, the photosensitive region 1613, and the N-type connection region 1623, and PD2B 1504 is composed of the accumulation region 1604, the photosensitive region 1614, and the N-type connection region 1624. PD1A 1501 and PD1B 1502 constituting the pixel 105U are arranged on the side having a relatively large y coordinate with respect to FD205, and PD2A 1503 and PD2B 1504 constituting the pixel 105D are arranged on the side having a relatively small y coordinate with respect to FD 205.

[0116] Figure 16B It is a schematic plan view showing the positional relationship in plan view of the accumulation regions 1601, 1602, 1603 and 1604, photosensitive regions 1611, 1612, 1613 and 1614, N-type connection regions 1621, 1622, 1623 and 1624, gate electrodes 1631 of TX1A 1505, gate electrodes 1632 of TX1B 1506, gate electrodes 1633 of TX2A 1507, gate electrodes 1634 of TX2B 1508, and FD 205 in the pixel set 1600. Since the photosensitive regions 1611 and 1612 of the pixel 105U and the photosensitive regions 1613 and 1614 of the pixel 105D all extend in the y direction, a phase difference signal with a pupil division direction in the x direction can be obtained based on both the pixels 105U and 105D. The arrow 1650 indicates the division direction of the phase difference signal.

[0117] Figure 17A and 17BIt is a schematic diagram showing a sixth layout of semiconductor regions of two pixels according to the second embodiment. In the following description, two pixels 105U and 105D having the sixth layout are referred to as pixel set 1700.

[0118] Figure 17A It is an exploded perspective view of the accumulation regions 1601, 1602, 1603 and 1604, photosensitive regions 1611, 1612, 1613 and 1614, N-type connection regions 1621, 1622, 1623 and 1624, gate electrodes 1631 of TX1A 1505, gate electrodes 1632 of TX1B 1506, gate electrodes 1633 of TX2A 1507, gate electrodes 1634 of TX2B 1508, and FD 205 in pixel set 1700. In addition, Figure 17B It is a schematic plan view showing the positional relationship of the components of pixel set 1700 in a plan view.

[0119] Pixel set 1700 has the same configuration as pixel set 1600, except that the extending directions of the photosensitive regions 1613 and 1614 of pixel 105D are different. Similarly to pixel set 1600, pixel 105U has photosensitive regions 1611 and 1612 extending in the y direction and can obtain a phase difference signal with a pupil division direction of the x direction. On the other hand, in pixel 105D, since the photosensitive regions 1613 and 1614 extend in the x direction, a phase difference signal with a pupil division direction of the y direction can be obtained. Arrows 1750 and 1751 indicate the division directions of the phase difference signals.

[0120] Figure 18A and 18B It is a schematic diagram showing a seventh layout of semiconductor regions of two pixels according to the second embodiment. In the following description, two pixels 105U and 105D having the seventh layout are referred to as pixel set 1800.

[0121] Figure 18A It is an exploded perspective view of the accumulation regions 1601, 1602, 1603 and 1604, photosensitive regions 1611, 1612, 1613 and 1614, N-type connection regions 1621, 1622, 1623 and 1624, gate electrodes 1631 of TX1A 1505, gate electrodes 1632 of TX1B 1506, gate electrodes 1633 of TX2A 1507, gate electrodes 1634 of TX2B 1508, and FD 205 in pixel set 1800. In addition, Figure 18B It is a schematic plan view showing the positional relationship of the components of pixel set 1800 in a plan view.

[0122] Except that the extending directions of the light-sensitive regions 1613 and 1614 of the pixel 105U are different, the pixel set 1800 has the same configuration as the pixel set 1600. Similarly to the pixel set 1600, the pixel 105D has light-sensitive regions 1613 and 1614 extending in the y direction and can obtain a phase difference signal with the pupil division direction in the x direction. On the other hand, in the pixel 105U, since the light-sensitive regions 1611 and 1612 extend in the x direction, a phase difference signal with the pupil division direction in the y direction can be obtained. Arrows 1850 and 1851 indicate the division directions of the phase difference signals.

[0123] · Pixel layout

[0124] Next, an example of the pixel arrangement of the pixel sets 1600, 1700, and 1800 having the above configuration will be described.

[0125] Figure 19 is a schematic diagram showing the pixel layout of the pixel array 101 in the range of 4 rows × 4 columns in the second embodiment. By arranging a large number of Figure 19 the pixel sets in the range of 4 rows × 4 columns shown on a plane, a captured image signal and a phase difference signal can be obtained. In Figure 19 it shows ML 301, light-sensitive regions 1611, 1612, 1613, and 1614, N-type connection regions 1621, 1622, 1623, and 1624, gate electrodes 1631, 1632, 1633, and 1634, FD 205, the division directions of the phase difference signals, and the color filter 406. It shows CFR 1101, CFG 1102, and CFB 1103 as the color filter 406 arranged in a Bayer array.

[0126] In addition, in Figure 19 the example shown, the pixel set 1700 is arranged in the first column from the left, the pixel set 1600 is arranged in the second column, the pixel set 1800 is arranged in the third column, and the pixel set 1600 is arranged in the fourth column. In addition, in the third column and the fourth column, the pixel sets are arranged to be offset by one pixel in the y direction with respect to the pixel sets in the first column and the second column.

[0127] By arranging the pixel sets in this way, a phase difference signal with the pupil division direction in the x direction can be obtained based on the pixels (R pixels) covered by CFR 1101, the pixels (B pixels) covered by CFB 1103, and the pixels (Gr pixels) covered by CFG 1102 arranged in the same row as the pixels covered by CFR 1101. In addition, a phase difference signal with the pupil division direction in the y direction can be obtained based on the pixels (Gb pixels) covered by CFG 1102 arranged in the same row as the pixels (B pixels) covered by CFB 1103.

[0128] Here, the characteristics of the Gb pixels arranged in the first column and the Gb pixels arranged in the third column will be described.

[0129] The Gb pixels arranged in the first column are the pixels 105D of the pixel set 1700. The charge converted from the light incident on the photosensitive region 1613 with a larger y coordinate is accumulated in the accumulation region 1603, and the charge converted from the light incident on the photosensitive region 1614 with a smaller y coordinate is accumulated in the accumulation region 1604. Here, the accumulation region 1603 in the pixel set 1700 is of type 1 and is adjacent to both the recessed region 404 and the notch region 405 of the P-type semiconductor, and the accumulation region 1604 is of type 2 and is only adjacent to the notch region 405 of the P-type semiconductor.

[0130] The Gb pixels arranged in the third column are the pixels 105U of the pixel set 1800. The charge converted from the light incident on the photosensitive region 1611 with a larger y coordinate is accumulated in the accumulation region 1601, and the charge converted from the light incident on the photosensitive region 1612 with a smaller y coordinate is accumulated in the accumulation region 1602. Here, the accumulation region 1601 in the pixel set 1800 is of type 2, and the accumulation region 1603 is of type 1.

[0131] Since the volume of the N-type region of the accumulation region of type 1 is smaller than the volume of the N-type region of the accumulation region of type 2, the amount of signal charge that can be stored in the accumulation region of type 1 can be smaller. That is, in the Gb pixels arranged in the first column, the saturation charge amount of the photosensitive region with a smaller y coordinate is greater than the saturation charge amount of the photosensitive region with a larger y coordinate. On the other hand, in the Gb pixels arranged in the third column, the saturation charge amount of the photosensitive region with a larger y coordinate is greater than the saturation charge amount of the photosensitive region with a smaller y coordinate.

[0132] Therefore, as Figure 19 shown, by alternately arranging the Gb pixels (first column) in which the saturation charge amount of the photosensitive region with a smaller y coordinate is greater than the saturation charge amount of the photosensitive region with a larger y coordinate and the Gb pixels (third column) in which the saturation charge amount of the photosensitive region with a larger y coordinate is greater than the saturation charge amount of the photosensitive region with a smaller y coordinate for every two columns, in the case where a larger amount of light is incident on the photosensitive region whose y coordinate in the Gb pixel is greater than the other photosensitive regions in the Gb pixel, and in the case where a larger amount of light is incident on the photosensitive region whose y coordinate in the Gb pixel is less than the other photosensitive regions in the Gb pixel, non-saturated signals can be obtained. That is, the saturation charge amounts of the phase difference signals from the pupil division pixels divided in the vertical direction can be made uniform.

[0133] As described above, according to the second embodiment, even with a configuration in which the readout circuit is shared by two pixels, the same effect as that of the first embodiment can be obtained.

[0134] <Third Embodiment>

[0135] Next, a third embodiment of the present invention will be described.

[0136] Figure 20 FIG. is a block diagram showing a schematic configuration of an imaging device according to a third embodiment of the present invention. The imaging device of this embodiment includes an image sensor 100 having the configuration described above, an overall control / operation unit 2, an instruction unit 3, a timing generation unit 4, an imaging lens unit 5, a lens drive unit 6, a signal processing unit 7, a display unit 8, and a recording unit 9.

[0137] The imaging lens unit 5 forms an optical image of a subject on the image sensor 100. Although represented by one lens in the figure, the imaging lens unit 5 may include a plurality of lenses including a focusing lens, a zoom lens, and an aperture, and the imaging lens unit 5 may be detachable from the main body of the imaging device or may be configured integrally with the main body.

[0138] The image sensor 100 has the configuration described in the above embodiments, and converts the light incident through the imaging lens unit 5 into an electrical signal and outputs it. Signals are read out from each pixel of the image sensor 100, so that a pupil division signal available for phase difference focus detection and an imaging signal as a signal of each pixel can be obtained.

[0139] The signal processing unit 7 performs predetermined signal processing such as correction processing on the signal output from the image sensor 100, and outputs a pupil division signal for focus detection and an imaging signal for recording.

[0140] The overall control / operation unit 2 drives and controls the entire imaging device comprehensively. In addition, the overall control / operation unit 2 also performs calculations for focus detection using the pupil division signal processed by the signal processing unit 7, and performs arithmetic processing for exposure control on the image signal and predetermined signal processing such as imaging and compression for generating an image for recording / playback.

[0141] The lens drive unit 6 drives the imaging lens unit 5, and performs focus control, zoom control, aperture control, etc. on the imaging lens unit 5 according to a control signal from the overall control / operation unit 2.

[0142] The instruction unit 3 accepts inputs such as a shooting execution instruction, a drive mode setting for the imaging device, and various other settings and selections input from the outside through, for example, a user's operation, and sends them to the overall control / operation unit 2.

[0143] The timing generation unit 4 generates timing signals for driving the image sensor 100 and the signal processing unit 7 according to the control signals from the overall control / operation unit 2.

[0144] The display unit 8 displays preview images, playback images, and information such as the drive mode settings of the imaging device.

[0145] The recording unit 9 is provided with a recording medium (not shown) and records the imaging signals for recording. Examples of the recording medium include semiconductor memories such as flash memories. The recording medium may be detachable from the recording unit 9 or may be built-in.

[0146] Next, an arithmetic method for calculating the defocus amount according to the pupil division signal in the overall control / operation unit 2 will be described.

[0147] First, reference will be made to Figures 21 to 27 Describe the phase difference focus detection in the x direction in this embodiment.

[0148] Figure 21 is a cross-sectional view of the pixel 700 or pixel 900 whose pupil division direction taken along the line A-A' shown is the x direction, and the pupil plane at a position separated from the imaging plane 2000 of the image sensor 100 by a distance Ds in the negative direction of the z axis. In Figure 3 x, y, and z indicate the coordinate axes on the imaging plane 2000, x Figure 21 x, y p y p y, and z p z indicate the coordinate axes on the pupil plane.

[0149] The pupil plane and the light receiving surface (second surface) of the image sensor 100 are substantially in a conjugate relationship via ML 301. Therefore, the light beam that has passed through the partial pupil region 2001 is received in the photosensitive region 313 or the accumulation region 311. In addition, the light beam that has passed through the partial pupil region 2002 is received in the photosensitive region 314 or the accumulation region 312. The signal charges photoelectrically converted near the boundary between the photosensitive region 313 and the photosensitive region 314 are randomly transferred to the accumulation region 311 or the accumulation region 312. Therefore, at the boundary between the partial pupil region 2001 and the partial pupil region 2002, the signal gradually switches as the x coordinate increases, and the x-direction dependence of the pupil intensity distribution has a shape as shown in Figure 22 shown. Hereinafter, the pupil intensity distribution corresponding to the photosensitive region 313 and the accumulation region 311 is referred to as the first pupil intensity distribution 2101, and the pupil intensity distribution corresponding to the photosensitive region 314 and the accumulation region 312 is referred to as the second pupil intensity distribution 2102.

[0150] Next, with reference to Figure 23 , the sensor entrance pupil of the image sensor 100 will be described. In the image sensor 100 of the present embodiment, according to the image height coordinates of the pixels on the two-dimensional plane, the ML S 301 of each pixel is continuously offset toward the center of the image sensor 100. That is, each ML 301 is arranged to be more eccentric toward the center as the image height of the pixel becomes higher. The center of the image sensor 100 and the optical axis of the imaging optical system are changed according to the following mechanism, but are substantially the same, where the mechanism reduces the influence of blur caused by camera shake or the like by driving the imaging optical system or the image sensor 100. As a result, in the pupil plane located at a distance Ds from the image sensor 100, the first pupil intensity distribution 2101 and the second pupil intensity distribution 2102 of each pixel 700 or pixel 900 arranged at each image height coordinate of the image sensor 100 substantially match. That is, in the pupil plane located at a distance Ds from the image sensor 100, the first pupil intensity distribution 2101 and the second pupil intensity distribution 2102 of all the pixels of the image sensor 100 substantially match.

[0151] Hereinafter, the first pupil intensity distribution 2101 and the second pupil intensity distribution 2102 are referred to as the "sensor entrance pupil" of the image sensor 100, and the distance Ds is referred to as the "sensor pupil distance" of the image sensor 100. It is not necessary to configure all pixels to have the same sensor pupil distance. For example, the sensor pupil distances of pixels up to 80% of the image height may be substantially the same, or the pixels may be configured to have different sensor pupil distances for each row or for each detection region.

[0152] Figure 24 A schematic relationship diagram between the image offset amount and the defocus amount between the parallax images is shown. The image sensor 100 (not shown) of the present embodiment is aligned on the imaging plane 2000, and as in Figure 21 , the exit pupil of the imaging optical system is divided into a partial pupil region 2001 and a partial pupil region 2002.

[0153] For the defocus amount d, the distance from the imaging position of the subject to the imaging plane is given by |d|. The front focus state where the imaging position of the subject is closer to the subject than the imaging plane is represented by a negative value (d < 0), and the rear focus state where the imaging position of the subject is on the opposite side of the subject with respect to the imaging plane is represented by a positive value (d > 0). The in-focus state of the imaging position of the subject on the imaging plane is represented as d = 0. Figure 24An example is shown where the subject on the object plane 2301 is in the in-focus state (d = 0) and the subject on the object plane 2302 is in the front-focus state (d < 0). Both the front-focus state (d < 0) and the rear-focus state (d > 0) are referred to as defocus states (|d| > 0).

[0154] In the front-focus state (d < 0), in the light beam from the subject on the object plane 2302, the light beam that has passed through the partial pupil region 2001 (2002) converges once and then diverges to have a radius r1 (r2) with respect to the position G1 (G2) that is the center of gravity of the light beam, and a blurred image is formed on the imaging plane 2000. The blurred image is received by the photosensitive region 313 and the accumulation region 311, and the photosensitive region 314 and the accumulation region 312, and a parallax image is generated. Therefore, the generated parallax image is a blurred image of the subject, where the image of the subject on the object plane 2302 expands to have a radius r1 (r2) with respect to the position G1 (G2) that is the center of gravity.

[0155] The radius r1 (r2) of the blur of the subject image generally increases proportionally as the magnitude |d| of the defocus amount d increases. Similarly, the magnitude |p| of the image offset amount p (= G2 - G1) between the subject images of the parallax images also increases approximately proportionally as the magnitude |d| of the defocus amount d increases. Although the image offset direction of the subject images between the parallax images is opposite to the image offset direction in the front-focus state, the same relationship holds in the rear-focus state (d > 0). In the in-focus state (d = 0), the positions of the centers of gravity of the subject images in the parallax images are the same (p = 0), and no image offset occurs.

[0156] Therefore, for the two phase difference signals obtained by using the signals from the photosensitive region 313 and the accumulation region 311 and the signals from the photosensitive region 314 and the accumulation region 312, as the magnitude of the defocus amount of the parallax images increases, the magnitude of the image offset amount in the x direction between the two phase difference signals increases. Based on this relationship, phase difference focus detection is performed by converting the image offset amount calculated by correlating the image offset amounts in the x direction between the parallax images into a defocus amount. The conversion coefficient used at this time is referred to as the conversion coefficient Kx.

[0157] Next, the relationship between the conversion coefficient Kx and the offset of the pupil will be described. Figures 25A to 25C The relationship between the partial pupil regions 2001 and 2002 and the exit pupil 2400 of the imaging optical system in the pixels 700 or 900 located at the peripheral image height (R, 0) of the image sensor 100 is shown. In addition, Figures 26A to 26CShows the relationship of the x-direction dependence between the exit pupil and the pupil intensity distribution. In the x-direction dependence of the pupil intensity distribution, the area inside the exit pupil is shown by a solid line, and the area outside the exit pupil is shown by a dotted line. The position (R, 0) of pixel 700 or pixel 900 in the pixel array is a position offset from the center of the optical axis in the x+ direction, as Figure 27 shown.

[0158] Figure 25A Shows the case where the exit pupil distance D1 of the imaging optical system and the sensor pupil distance Ds of the image sensor 100 are the same. In this case, the pupil is divided into a first partial pupil region 2001 and a second partial pupil region 2002 near the center of the exit pupil 2400 of the imaging optical system. In this case, the range of the light beam to be received extends equally in the positive and negative directions from the x coordinate where the signal intensities of the first pupil intensity distribution 2101 and the second pupil intensity distribution 2102 intersect. As a result, the relationship between the pupil intensity distribution and the exit pupil is as Figure 26A shown.

[0159] On the other hand, as Figure 25B shown, when the exit pupil distance D1 of the imaging optical system is shorter than the sensor pupil distance Ds of the image sensor 100, a pupil deviation occurs between the exit pupil of the imaging optical system and the entrance pupil of the sensor 100 at the peripheral image height of the image sensor 100, and as a result, the exit pupil 2400 of the imaging optical system is unevenly divided, and the relationship between the pupil intensity distribution and the exit pupil is as Figure 26B shown.

[0160] Similarly, as Figure 25C shown, when the exit pupil distance D1 of the imaging optical system is longer than the sensor pupil distance Ds of the image sensor 100, a pupil deviation occurs between the exit pupil of the imaging optical system and the entrance pupil of the sensor 100 at the peripheral image height of the image sensor 100, and as a result, the exit pupil 2400 of the imaging optical system is unevenly divided, and the relationship between the pupil intensity distribution and the exit pupil is as Figure 26C shown.

[0161] As Figures 25A to 25C shown, the pupil intensity distribution region included in the exit pupil varies according to the exit pupil distance D1. Therefore, the conversion coefficient Kx representing the relationship between the amount of defocus and the magnitude of the image shift amount in the x direction between the phase difference signals has different values according to the exit pupil distance and the image height.

[0162] Next, referring to Figures 28 to 32C, the phase difference focus detection in the case where the pupil division direction is the y-direction will be described. The description of the parts that are the same as the phase difference focus detection in the x-direction will be omitted, and the differences will be described. In the phase difference focus detection in the y-direction, the image shift amount in the y-direction is calculated based on the phase difference signal of the pixel 800 or the pixel 1000 whose pupil division direction is the y-direction, and the image shift amount is converted into a defocus amount using the conversion coefficient Ky.

[0163] Figure 28 A cross-sectional view of the pixel 800 or 1000 whose pupil division direction in the photosensitive area is the y-direction, and the pupil plane at a position separated from the image sensor 100 by the sensor pupil distance Ds in the negative z-axis direction at the imaging plane 2000. In Figure 28 x, y, and z indicate the coordinate axes on the imaging plane 2000, and x p , y p and z p indicate the coordinate axes on the pupil plane.

[0164] The pupil plane and the light-receiving surface (second surface) of the image sensor 100 have a substantially conjugate relationship via ML 301. Therefore, the light beam that has passed through the partial pupil region 2701 is received in the photosensitive region 313, and the light beam that has passed through the partial pupil region 2702 is received in the photosensitive region 314. In addition, the light beam that has passed through the partial pupil region 2703 is received in the accumulation region 311, and the light beam that has passed through the partial pupil region 2704 is received in the accumulation region 312. The ratios of the light beams received in the photosensitive regions 313 and 314 and the ratios of the light beams received in the accumulation regions 311 and 312 are mainly determined by the wavelength of the received light.

[0165] Next, the relationship between the pupil shift and the conversion coefficient Ky will be described, where the conversion coefficient Ky represents the relationship between the image shift amount between the parallax images and the defocus amount when detecting the phase difference in the y-direction. Figures 29A to 29C Shows the relationship between the partial pupil regions 2701 to 2704 in the pixel 800 or 1000 located at the peripheral image height (0, U) of the image sensor 100 and the exit pupil 2400 of the imaging optical system. In addition, Figures 30A to 30C shows the relationship of the y-direction dependence between the exit pupil and the pupil intensity distribution. The position (0, U) of the pixel 800 or the pixel 1000 in the pixel array is a position shifted from the optical axis center in the y+ direction, as Figure 27 shown.

[0166] Figure 29AShows a case where the exit pupil distance D1 of the imaging optical system is the same as the sensor pupil distance Ds of the image sensor 100. In this case, the light beams to be received in the photosensitive regions 313 and 314 are split in the y direction near the center of the exit pupil 2400 of the imaging optical system. Therefore, in the pixels where the pupil splitting direction is the y direction, the y-direction dependence between the third pupil intensity distribution 2901 corresponding to the photosensitive region 313 and the fourth pupil intensity distribution 2902 corresponding to the photosensitive region 314 is as Figure 30A shown.

[0167] In addition, the light received in the accumulation regions 311 and 312 is not split in the y direction, which is the direction for calculating the image offset amount, and the amount of light is substantially the same. Therefore, in the pixels where the pupil splitting direction is the y direction, the y-direction dependence between the fifth pupil intensity distribution 2903 corresponding to the accumulation region 311 and the sixth pupil intensity distribution 2904 corresponding to the accumulation region 312 is as Figure 30A shown.

[0168] In the pixel 800, the photosensitive region 313 and the accumulation region 311, and the photosensitive region 314 and the accumulation region 312 are respectively connected via N-type connection regions, and in the pixel 1000, the photosensitive region 313 and the accumulation region 312, and the photosensitive region 314 and the accumulation region 311 are respectively connected via N-type connection regions. Therefore, in the pixel 800, signals corresponding to the combination of the third pupil intensity distribution 2901 and the fifth pupil intensity distribution 2903 and signals corresponding to the combination of the fourth pupil intensity distribution 2902 and the sixth pupil intensity distribution 2904 are output as phase difference signals.

[0169] On the other hand, in pixel 1000, signals corresponding to the combination of the third pupil intensity distribution 2901 and the sixth pupil intensity distribution 2904 and signals corresponding to the combination of the fourth pupil intensity distribution 2902 and the fifth pupil intensity distribution 2903 are output as phase difference signals. Since the fifth pupil intensity distribution 2903 and the sixth pupil intensity distribution 2904 are substantially constant with respect to the y direction and have substantially the same signal amount, the phase difference signals obtained from pixel 800 and pixel 1000 are substantially the same. Therefore, the positions of the centers of gravity of pixel 800 and pixel 1000 are equal in the y direction, and the conversion coefficient Ky representing the relationship between the amount of defocus of pixel 800 and pixel 1000 and the image shift amount in the y direction has substantially the same value. That is, signals obtained from the cumulative regions 311 and 312 that do not divide the pupil in the y direction are added as offset components to the signals obtained from the photosensitive regions 313 and 314 that divide the exit pupil in the y direction. However, since the offsets are equal, even if the combination of signals to be added from the photosensitive regions and the cumulative regions is different between pixel 800 and 1000, the conversion coefficient Ky does not change.

[0170] Figure 29B and 29C as well as Figure 30B and 30C show a case where the exit pupil distance D1 of the imaging optical system is shorter than or longer than the sensor pupil distance Ds in pixel 800 or pixel 1000 at the position (0, U). In this case, the relationship between the third pupil intensity distribution 2901 and the fourth pupil intensity distribution 2902 and the exit pupil is the same as in the case where the pupil division direction is the x direction. In addition, the fifth pupil intensity distribution 2903 and the sixth pupil intensity distribution 2904 are the same as the fifth pupil intensity distribution 2903 and the sixth pupil intensity distribution 2904 in the case of Figure 29A and 30A , and the signal amount is substantially constant and substantially equal with respect to the y direction. Therefore, the conversion coefficient Ky takes the same value for pixel 800 and pixel 1000.

[0171] Next, a description will be given with reference to Figures 31A to 32C In the phase difference detection in the case where the pupil division direction is the y direction, the pixel positions in the pixel array are at positions (R, U) shifted from the optical axis in the x+ direction and the y+ direction. As Figure 27 shown, the pixel position (R, U) in the pixel array is a position shifted by R in the x+ direction and by U in the y+ direction from the optical axis.

[0172] Figure 31AThe case where the exit pupil distance D1 of the imaging optical system is the same as the sensor pupil distance Ds of the image sensor is shown. In this case, similar to the case of the pixel 800 or the pixel 1000 at the position (0, U), since the third pupil intensity distribution 2901 and the fourth pupil intensity distribution 2902 are divided near the center of the exit pupil 2400 of the imaging optical system, and the fifth pupil intensity distribution 2903 and the sixth pupil intensity distribution 2904 are not divided in the y direction and have substantially the same amount of light, the distribution is as Figure 32A shown.

[0173] Figure 31B The relationship between the partial pupil regions 2701 to 2704 and the exit pupil distance D1 of the imaging optical system is shown in the case where the exit pupil distance D1 of the imaging optical system is shorter than the sensor pupil distance Ds of the image sensor 100. Since the exit pupil 2400 is divided into the partial pupil region 2701 corresponding to the photosensitive region 313 and the partial pupil region 2702 corresponding to the photosensitive region 314 in the same manner as the pixel 800 or the pixel 1000 at the position (0, U), the y-direction dependence of the third pupil intensity distribution 2901 and the fourth pupil intensity distribution 2902 is almost the same as the y-direction dependence of the pixel 800 or the pixel 1000 at the position (0, U), and has the characteristics as Figure 32B shown.

[0174] On the other hand, since the partial pupil region 2703 and the partial pupil region 2704 are not divided in the y direction, the fifth pupil intensity distribution 2903 and the sixth pupil intensity distribution 2904 do not have y-direction dependence. However, since the amount of light passing through the pupil region 2703 is less than the amount of light passing through the pupil region 2704, the fifth pupil intensity distribution 2903 is smaller than the sixth pupil intensity distribution 2904, and has Figure 32B the characteristics shown.

[0175] Similarly, in the case where the exit pupil distance D1 of the imaging optical system is longer than the sensor pupil distance Ds of the image sensor 100, the relationship between the partial pupil regions 2701 to 2704 and the exit pupil distance D1 of the imaging optical system is as Figure 31C shown. In this case, the y-direction dependence of the third pupil intensity distribution 2901 and the fourth pupil intensity distribution 2902 is almost the same as the y-direction dependence of the pixel 800 or the pixel 1000 at the position (0, U).

[0176] On the other hand, since the amount of light passing through the partial pupil region 2704 is less than the amount of light passing through the partial pupil region 2703, the sixth pupil intensity distribution 2904 is smaller than the fifth pupil intensity distribution 2903, and has Figure 32CThe characteristics shown. In this case, the offset components added to the photosensitive regions 313 and 314 that divide the exit pupil in the y direction are different between the pixel 800 and the pixel 1000. Therefore, the conversion coefficient Ky takes different values between the pixel 800 and the pixel 1000.

[0177] Therefore, in the phase difference focus detection in the y direction, by calculating the image offset amount based on the phase difference signals obtained from the pixel 800 and the pixel 1000, and by calculating the defocus amount using different conversion coefficients Ky, a more accurate defocus amount can be calculated.

[0178] Although the present invention has been described with reference to exemplary embodiments, it should be understood that the present invention is not limited to the disclosed exemplary embodiments. The scope of the claims will be given the broadest interpretation to cover all such modifications and equivalent structures and functions.

Claims

1. An image sensor, comprising: a plurality of microlenses; and a pixel array that, with respect to each of the plurality of microlenses, has: a pair of first regions formed at a first depth from a light-incident surface, a pair of second regions formed at a second depth deeper than the first depth, and a plurality of connection regions each connecting the pair of first regions and the pair of second regions, wherein the direction in which the pair of second regions corresponding to each microlens is arranged is a first direction, and the direction in which the pair of first regions is arranged is the first direction or a second direction orthogonal to the first direction, wherein the first regions generate charges by photoelectrically converting incident light, and the second regions accumulate the charges generated in the first regions via the connection regions, wherein the pixel array further includes: a conversion component configured to convert charges into a voltage, and a transmission component connected to each of the second regions in the second regions and configured to transmit the charges accumulated in the second regions to the conversion component, wherein the length of the second regions in the depth direction is shorter in a first partial region of the second regions than in a second partial region of the second regions, and the first partial region is located farther from the transmission component than the second partial region; or the impurity concentration is lower in a first partial region of the second regions than in a second partial region of the second regions, and the first partial region is located farther from the transmission component than the second partial region.

2. The image sensor according to claim 1, wherein the length of the second partial region in the depth direction is shorter in a region corresponding to the connection region than in a region not corresponding to the connection region.

3. The image sensor according to claim 1, wherein the impurity concentration in the second partial region is lower in a region corresponding to the connection region than in a region not corresponding to the connection region.

4. The image sensor according to claim 1, wherein the plurality of connection regions connect the pair of first regions and the pair of second regions in different combinations.

5. The image sensor according to claim 4, wherein the pixel array includes: a first arrangement in which the direction in which the pair of first regions is arranged is the first direction, and the pair of first regions is connected to the pair of second regions in a first combination via the plurality of connection regions; a second arrangement in which the direction in which the pair of first regions is arranged is the second direction, and the pair of first regions is connected to the pair of second regions in the first combination via the plurality of connection regions; a third arrangement in which the direction in which the pair of first regions is arranged is the first direction, and the pair of first regions is connected to the pair of second regions in a second combination different from the first combination via the plurality of connection regions; and Fourth arrangement, in which the direction in which the pair of first regions is arranged is the second direction, and the pair of first regions is connected to the pair of second regions in the second combination via the plurality of connection regions.

6. The image sensor according to claim 5, wherein, the first arrangement and the third arrangement are oriented such that one of each pair of second regions including a second partial region having a region corresponding to the connection region is arranged closer to the center side of the pixel array than the other of each pair of second regions.

7. The image sensor according to claim 1, wherein, each conversion member is connected to two pairs of second regions via the transmission member.

8. The image sensor according to claim 7, wherein, the combination of the two pairs of second regions includes a first combination and a second combination. In the first combination, the directions of both pairs of first regions corresponding to the two pairs of second regions are the first direction. In the second combination, the direction of one of the two pairs of first regions corresponding to the two pairs of second regions is the first direction, and the direction of the other pair of first regions is the second direction.

9. The image sensor according to claim 1, further comprising a Bayer color filter disposed between the microlens and the pixel array, wherein, the pixel array is configured such that a green color filter is disposed between the microlens and a pair of first regions arranged in the first direction.

10. An imaging device, comprising: the image sensor according to claim 1; and a focus detection member configured to obtain pupil division signals respectively corresponding to divided pupil regions based on signals output from the pixel array, and perform phase difference focus detection based on the pupil division signals.

11. The imaging device according to claim 10, wherein, phase difference focus detection is performed by using signals output from a pixel group in which each pixel has a second arrangement and signals output from a pixel group in which each pixel has a fourth arrangement. In the second arrangement, the direction in which the pair of first regions is arranged is the second direction, and the pair of first regions is connected to the pair of second regions in a first combination via the plurality of connection regions. In the fourth arrangement, the direction in which the pair of first regions is arranged is the second direction, and the pair of first regions is connected to the pair of second regions in a second combination via the plurality of connection regions.

12. The imaging device according to claim 11, wherein, for the pixel group in which each pixel has the second arrangement and the pixel group in which each pixel has the fourth arrangement, different conversion coefficients for calculating a defocus amount based on an image offset amount of the phase difference signal are used.

Images