JP6338442B2 - Solid-state imaging device, distance measuring device, and imaging device - Google Patents

Description

  The present invention relates to a solid-state image sensor, and more particularly to a solid-state image sensor used in a distance measuring device.

  AF distance detection technology is known for digital still cameras and digital video cameras. In relation to such AF distance detection technology, Patent Document 1 proposes a solid-state imaging device in which some pixels of the solid-state imaging device have a ranging function and are detected by a phase difference method. . This phase difference method acquires two images (hereinafter referred to as distance measurement images) that have passed through different areas on the pupil of the imaging optical system, and triangulation using stereo images based on the distance between the distance measurement images. This is a method of detecting a distance using. According to this, unlike the conventional contrast method, since it is not necessary to move the lens to measure the distance, high-speed and high-precision AF is possible.

Japanese Patent No. 4835136

  However, depending on the shooting conditions, the image quality of the distance measurement image deteriorates, and there is a problem that the distance measurement accuracy decreases. In general, since the exit pupil position of the imaging optical system changes depending on the zoom or focus state, the design pupil position of the imaging element and the exit pupil position of the imaging optical system do not necessarily match. When the design pupil position of the image sensor and the exit pupil position of the imaging optical system are different, the amount of eccentricity of the pupil region through which the light beam received by each distance measurement pixel passes depends on the position of each distance measurement pixel in the image sensor. Change. As the amount of eccentricity increases, a difference occurs between the pupil transmittance distributions of the light beams forming the two distance measurement images. Also, due to vignetting, a difference occurs between the pupil transmittance distributions of the light beams forming the two distance measurement images.

  If the exposure time is determined in accordance with a distance measurement image having a relatively low pupil transmittance of the light beam forming the distance measurement image, the distance measurement image having a relatively high pupil transmittance is likely to be saturated. Conversely, if the exposure time is determined in accordance with a distance measurement image having a relatively high pupil transmittance of the light beam forming the distance measurement image, the light amount of the distance measurement image having a relatively low pupil transmittance is likely to be insufficient. As described above, since the quality of the distance measurement image is degraded, the distance measurement accuracy is degraded. When a subject with a large contrast ratio is photographed, the degradation of the quality of the distance measurement image becomes a problem.

  In view of the above problems, an object of the present invention is to suppress degradation in the quality of a ranging image even when a subject having a large contrast ratio is photographed.

The first aspect of the present invention is:
A solid-state imaging device including a plurality of pixels that photoelectrically convert a subject image formed by an imaging optical system,
At least a part of the plurality of pixels is a ranging pixel in which a first photoelectric conversion unit and a second photoelectric conversion unit are provided side by side along a first direction,
When the region of the solid-state imaging device is divided into a first region and a second region by a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction,
More than 80% of the ranging pixels in the first region and within a region that is more than a predetermined distance away from a straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The capacity of the first photoelectric conversion unit is larger than the capacity of the second photoelectric conversion unit,
80% or more of the ranging pixels in the second region and within the region separated by the predetermined distance or more from the straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The capacity of the second photoelectric conversion unit is larger than the capacity of the first photoelectric conversion unit,
It is a solid-state image sensor.
The second aspect of the present invention is:
More than 80% of the ranging pixels in the first area have a capacitance of the first photoelectric conversion unit larger than that of the second photoelectric conversion unit,
80% or more of the ranging pixels in the second region have a capacity of the second photoelectric conversion unit larger than a capacity of the first photoelectric conversion unit.
It is a solid-state image sensor.
The second aspect of the present invention is:
A solid-state imaging device including a plurality of pixels that photoelectrically convert a subject image formed by an imaging optical system,
At least a part of the plurality of pixels is a ranging pixel in which a first photoelectric conversion unit and a second photoelectric conversion unit are provided side by side along a first direction,
When the region of the solid-state imaging device is divided into a first region and a second region by a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction,
More than 80% of the ranging pixels in the first region and within a region that is more than a predetermined distance away from a straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The volume of the first photoelectric conversion unit is larger than the volume of the second photoelectric conversion unit,
80% or more of the ranging pixels in the second region and within the region separated by the predetermined distance or more from the straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The volume of the second photoelectric conversion unit is larger than the volume of the first photoelectric conversion unit;
It is a solid-state image sensor.
The third aspect of the present invention is:
A solid-state imaging device including a plurality of pixels that photoelectrically convert a subject image formed by an imaging optical system,
At least a part of the plurality of pixels is a ranging pixel in which a first photoelectric conversion unit and a second photoelectric conversion unit are provided side by side along a first direction,
When the region of the solid-state imaging device is divided into a first region and a second region by a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction,
More than 80% of the ranging pixels in the first region and within a region that is more than a predetermined distance away from a straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The impurity concentration of the first photoelectric conversion unit is higher than the impurity concentration of the second photoelectric conversion unit,
80% or more of the ranging pixels in the second region and within the region separated by the predetermined distance or more from the straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The impurity concentration of the second photoelectric conversion unit is higher than the impurity concentration of the first photoelectric conversion unit;
It is a solid-state image sensor.
The fourth aspect of the present invention is:
A solid-state imaging device including a plurality of pixels that photoelectrically convert a subject image formed by an imaging optical system,
At least a part of the plurality of pixels is a ranging pixel in which a first photoelectric conversion unit and a second photoelectric conversion unit are provided side by side along a first direction,
When the region of the solid-state imaging device is divided into a first region and a second region by a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction,
More than half the distance measurement pixels in the first area and within the area that is more than a predetermined distance away from a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction, The area of the imaging surface of the first photoelectric conversion unit is larger than the area of the imaging surface of the second photoelectric conversion unit,
More than half the distance measurement pixels in the second area and within the area that is more than the predetermined distance from the straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The area of the imaging surface of the second photoelectric conversion unit is larger than the area of the imaging surface of the first photoelectric conversion unit;
The ranging pixel has a waveguide, and guides incident light to the first photoelectric conversion unit or the second photoelectric conversion unit according to an incident angle to the ranging pixel.
It is a solid-state image sensor.

  According to the present invention, it is possible to suppress degradation in the quality of a distance measurement image, and therefore, distance measurement accuracy is improved. This is particularly effective when a subject with a large contrast ratio is photographed.

1 is a configuration diagram of a digital camera according to a first embodiment. Arrangement and structure of solid-state imaging device according to embodiment 1 Sensitivity characteristics of ranging pixels according to the first embodiment Light flux received by the photoelectric conversion unit according to the first embodiment Modification of solid-state imaging device according to Embodiment 1 Arrangement and structure of solid-state imaging device according to embodiment 2 Sensitivity characteristics of ranging pixels according to the second embodiment Light flux received by the photoelectric conversion unit according to the second embodiment Arrangement and structure of solid-state imaging device according to embodiment 3 Modification of solid-state imaging device according to embodiment 3 Arrangement and structure of solid-state imaging device according to embodiment 4 Arrangement and structure of solid-state imaging device according to embodiment 6 Sensitivity characteristics of ranging pixels according to Embodiment 6 Arrangement and structure of solid-state imaging device shown in embodiment 7 Ranging Pixel Structure Using Waveguide

  Hereinafter, a distance measuring apparatus according to an embodiment of the present invention will be described with reference to the drawings. In that case, the same number is attached to the same function in all the drawings, and the repeated explanation is omitted.

[Embodiment 1]
<Camera>
A digital camera (imaging device) 100 according to this embodiment is shown in FIG. In FIG. 1, the digital camera 100 includes an imaging optical system 101, a solid-state image sensor 103, and an arithmetic processing unit 104. The solid-state imaging device 103 is disposed on the optical axis 102 of the imaging optical system 101, and the imaging optical system 101 forms a subject image on the solid-state imaging device 103.

The solid-state image sensor 103 includes a photodiode that detects light and generates a charge as a photoelectric conversion element (photoelectric conversion unit). The method of transferring the generated charge may be arbitrary. That is, the solid-state imaging device 103 may be a CCD (Charge Coupled Device) or a CMOS (Complementary Metal Oxide Semiconductor).

The solid-state image sensor 103 includes a plurality of pixels, all of which are ranging pixels. The ranging pixel includes two photoelectric conversion units, and is configured to selectively receive light from different pupil regions of the imaging optical system 101, respectively. Since the subject image signal having parallax is obtained from the two photoelectric conversion units of the ranging pixel, the subject distance can be calculated. Note that some of the plurality of pixels may be normal imaging pixels. The imaging pixel includes only one photoelectric conversion unit and receives light from the entire pupil region of the imaging optical system 101.

  The arithmetic processing unit 104 includes a CPU, a DSP, and a memory that stores a program. The arithmetic processing unit 104 detects the distance of a subject and acquires a subject image by executing the program. Note that the arithmetic processing unit 104 may be implemented by an ASIC. A distance measuring device is configured by the distance measuring function (distance calculating unit) of the imaging optical system 101, the solid-state imaging device 103, and the arithmetic processing unit 104. Further, the imaging device is configured by the ranging device and the imaging function (subject image acquisition unit) of the arithmetic processing unit 104. Regarding the subject image acquisition function in the arithmetic processing unit 104, a well-known technique can be adopted, and thus detailed description thereof is omitted in this specification.

<Distance pixel>
FIG. 2A is a diagram illustrating the arrangement of the distance measurement pixels in the solid-state image sensor 103. The ranging pixel 110 is a ranging pixel arranged in the peripheral region 105 in the −X direction of the solid-state imaging device 103, and the ranging pixel 111 is arranged in the peripheral region 106 in the + X direction of the solid-state imaging device 103. Ranging pixels. The peripheral region 105 is a region separated from the straight line 107 passing through the center of the solid-state image sensor 103 and parallel to the direction (Y direction) perpendicular to the X direction by a predetermined distance or more in the −X direction. On the other hand, the peripheral area 106 is an area separated from the straight line 107 in the + X direction by a predetermined distance or more. In the example of FIG. 2A, the predetermined distance is 1/6 times the length of the solid-state image sensor 103 in the X direction. The X direction corresponds to the first direction in the present invention. However, the predetermined distance is preferably 0.40 times or more, and more preferably 0.25 times or more the X-direction length of the solid-state imaging device 103. The reason for this will be described later.

  FIG. 2B is a cross-sectional view illustrating the configuration of the ranging pixel 110 and the ranging pixel 111. The ranging pixel 110 and the ranging pixel 111 have a microlens 112 and a substrate 113 from the light incident side. The microlens 112 is made of SiO2 or the like that is transparent in the wavelength band to be detected, and the substrate 113 is made of a material such as Si that has absorption in the wavelength band to be detected.

  In the substrate 113, two photoelectric conversion units are formed that are arranged symmetrically in the X direction with the optical axis of the microlens 112 as the center. The photoelectric conversion unit in the −X direction of the ranging pixel 110 is referred to as a photoelectric conversion unit 121, the photoelectric conversion unit in the + X direction is referred to as a photoelectric conversion unit 122, and the photoelectric conversion unit in the −X direction of the ranging pixel 111 is referred to as a photoelectric conversion unit 123. The + X direction photoelectric conversion unit is referred to as a photoelectric conversion unit 124. The photoelectric conversion units 121 and 123 arranged on the negative side in the X direction correspond to the first photoelectric conversion unit, and the photoelectric conversion units 122 and 124 arranged on the positive side in the X direction correspond to the second photoelectric conversion unit. To do. That is, the first photoelectric conversion unit and the second photoelectric conversion unit are two photoelectric conversion units provided side by side along the X direction.

  The photoelectric conversion unit 121 and the photoelectric conversion unit 122, and the photoelectric conversion unit 123 and the photoelectric conversion unit 124 in the imaging plane (XY plane) of the substrate 113 have the same shape. Further, the photoelectric conversion unit 121 is longer than the photoelectric conversion unit 122 in the direction perpendicular to the imaging surface (Z direction). Further, the photoelectric conversion unit 124 is longer than the photoelectric conversion unit 123 in the direction perpendicular to the imaging surface (Z direction). Hereinafter, the length of the photoelectric conversion unit in the Z direction is referred to as the depth of the photoelectric conversion unit.

  The deeper the photoelectric conversion unit, the larger the capacity of the photoelectric conversion unit. That is, in the region (105) that is more than a predetermined distance in the −X direction from the Y-direction straight line 107 passing through the center of the solid-state imaging device 103, the capacitance of the photoelectric conversion unit 121 on the −X direction side is the photoelectric on the + X direction side. It is larger than the capacity of the converter 122. On the other hand, in a region (106) that is a predetermined distance or more away in the + X direction from the Y-direction straight line 107 passing through the center of the solid-state image sensor 103, the capacitance of the photoelectric conversion unit 124 on the + X direction side is photoelectric conversion on the −X direction side. It is larger than the capacity of the part 123.

  The photoelectric conversion portion is formed by implanting ions such as boron into the substrate 113. Further, a wiring (not shown) is provided on the substrate 113, and charges generated in the photoelectric conversion unit are transferred to the signal processing circuit by the wiring.

<Sensitivity characteristics of photoelectric converter>
By adopting such a configuration, the incident angle dependence in the XZ section of the sensitivity of the photoelectric conversion unit 121 and the photoelectric conversion unit 122 and the sensitivity of the photoelectric conversion unit 123 and the photoelectric conversion unit 124 can be made different from each other. . FIG. 3 shows the incident angle dependence of the sensitivity of each photoelectric conversion unit. The sensitivity characteristics (dotted lines) of the photoelectric conversion unit 122 and the photoelectric conversion unit 124 have high sensitivity to incident light from the minus direction (−X direction) and low sensitivity to incident light from the plus direction (+ X direction). On the other hand, the sensitivity characteristics (broken lines) of the photoelectric conversion unit 121 and the photoelectric conversion unit 123 have low sensitivity to incident light from the minus direction (−X direction) and high sensitivity to incident light from the plus direction (+ X direction). . That is, the photoelectric conversion unit 121 and the photoelectric conversion unit 123 selectively select the light flux that passes through the pupil regions 131 and 133 (first pupil region) in the + X direction, which is a part of the exit pupil 130 of the imaging optical system. It is designed to receive light. On the other hand, the photoelectric conversion unit 122 and the photoelectric conversion unit 124 selectively select a light beam that passes through the pupil regions 132 and 134 (second pupil region) in the −X direction, which is a part of the exit pupil 130 of the imaging optical system. It is designed to receive light.

<Micro lens>
In addition, since the two photoelectric conversion portions are arranged at symmetrical positions with the optical axis of the microlens 111 as the central axis and the opening shapes are the same, the sensitivity characteristic is a symmetrical shape with the light incident angle of 0 degree as an axis. It has become. Therefore, when the distance between the exit pupil 130 of the imaging optical system 101 and the solid-state imaging device 103 is infinite, the pupil region 131 and the pupil region 132, and the pupil region 133 and the pupil region 134 are centrosymmetric on the exit pupil 130. Become.

  That is, the position of the center of gravity of the pupil transmittance distribution of the pupil region 131 and the pupil transmittance distribution of the pupil region 132 are different. Similarly, the position of the center of gravity of the pupil transmittance distribution of the pupil region 133 and the pupil transmittance distribution of the pupil region 134 are also different. Here, the direction of the straight line connecting the center of gravity of the pupil transmittance distribution of the pupil region 131 (133) and the transmittance distribution of the pupil region 132 (134) is the X direction.

<The reason why pupil transmittance is different and its problem>
When the distance between the exit pupil 130 and the solid-state image sensor 103 is always kept infinite, the pupil transmittances of the pupil region 131 and the pupil region 132 are ideally equal to each other due to the sensitivity characteristics shown in FIG. Accordingly, the amount of light incident on the photoelectric conversion unit 121 and the photoelectric conversion unit 122 is equal. Similarly, the pupil transmittances of the pupil region 133 and the pupil region 134 are also equal to each other, and the amounts of light incident on the photoelectric conversion unit 123 and the photoelectric conversion unit 124 are equal.

  However, in an actual digital camera as described below, the pupil transmittances of the pupil region 131 and the pupil region 132 and the pupil transmittances of the pupil region 133 and the pupil region 134 are not equal to each other. The reason will be described.

  Due to the demand for miniaturization of the digital camera 100, the exit pupil position of the imaging optical system 101 changes depending on the zoom state. In general, on the telephoto side, the position of the exit pupil 130 is far from the solid-state image sensor, and on the wide-angle side, the position of the exit pupil 130 is closer to the solid-state image sensor. When the photographing lens has an inner focus or rear focus configuration, the position of the exit pupil 130 also varies depending on the focus state. Therefore, in the ranging pixel 110 located in the peripheral region away from the center of the solid-state image sensor 103 in the X direction, the pupil transmittance of the pupil region 131 and the pupil region 132 and the pupil transmission of the pupil region 133 and the pupil region 134 are transmitted. The rates are not equal to each other.

  Even if the distance between the exit pupil 130 and the solid-state image sensor 103 is always kept infinite, the amount of light passing through the periphery of the exit pupil 130 may decrease due to vignetting in the imaging optical system 101. . In that case, in the ranging pixels located in the peripheral region of the solid-state imaging device, the pupil transmittances of the pupil region 131 and the pupil region 132 and the pupil transmittances of the pupil region 133 and the pupil region 134 are not equal to each other.

  FIGS. 4A and 4B show a case where the exit pupil 130 is at a finite distance from the solid-state image sensor 103 (e.g., when the exit pupil 130 is photographed on the wide-angle side of the zoom lens).

  FIG. 4A shows the state of the light beam received by the ranging pixel 110 in the peripheral region 105 in the −X direction (the negative direction of the first direction) from the center of the solid-state image sensor 103. The photoelectric conversion unit 121 receives the light beam 141 from the pupil region 131, and the photoelectric conversion unit 122 receives the light beam 142 from the pupil region 132. As can be seen from FIG. 4A, the spread angle of the light beam 141 is larger than the spread angle of the light beam 142. Accordingly, the pupil transmittance of the pupil region 131 is higher than that of the pupil region 132, and the amount of light received by the photoelectric conversion unit 121 is greater than the amount of light received by the photoelectric conversion unit 122.

  FIG. 4B shows a state of the light beam received by the ranging pixel 111 in the peripheral region 106 in the + X direction (positive direction of the first direction) from the center of the solid-state image sensor 103. The photoelectric conversion unit 123 receives the light beam 143 from the pupil region 133, and the photoelectric conversion unit 124 receives the light beam 144 from the pupil region 134. As can be seen from FIG. 4B, the divergence angle of the light beam 144 is larger than the divergence angle of the light beam 143. Accordingly, since the pupil transmittance of the pupil region 134 is higher than that of the pupil region 133, the amount of light received by the photoelectric conversion unit 124 is larger than the amount of light received by the photoelectric conversion unit 123.

  In the conventional solid-state imaging device having the same capacity of the two photoelectric conversion units as disclosed in Patent Document 1, there is a tradeoff between insufficient light quantity and saturation. That is, when the exposure time is determined in accordance with the photoelectric conversion units 122 and 123 that receive a relatively small amount of light, the photoelectric conversion units 121 and 124 that receive a large amount of light are likely to be saturated. On the other hand, when the exposure time is determined in accordance with the photoelectric conversion units 121 and 124 that receive a relatively large amount of light, a shortage of light occurs in the photoelectric conversion units 122 and 123 that receive a small amount of light. If saturation or light quantity shortage occurs, the quality of the distance measurement image decreases, and the distance measurement accuracy decreases. In particular, when a subject having a large contrast ratio is photographed, the degradation of the quality of the distance measurement image becomes a problem.

<Effect due to change in capacitance of photoelectric conversion unit>
In the solid-state imaging device 103 according to the present embodiment, the depth of the photoelectric conversion units 121 and 124 that receive a relatively large amount of light is made deeper than the depth of the photoelectric conversion units 122 and 123 that receive a relatively small amount of light. Yes. That is, the capacitances of the photoelectric conversion units 121 and 124 that receive light in the pupil regions 131 and 134 having higher pupil transmittance than the photoelectric conversion units 122 and 123 that receive light from the pupil regions 132 and 133 having low pupil transmittance. Has increased. Accordingly, it is possible to simultaneously solve the shortage of light amount in the photoelectric conversion units 122 and 123 and the saturation in the photoelectric conversion units 121 and 124 that receive the light that has passed through the pupil regions 131 and 134.

  In order to change the depth of the photoelectric conversion portion, the depth of ion implantation may be changed. Further, the impurity concentration in the photoelectric conversion unit may be changed. That is, the impurity concentration in the second photoelectric conversion unit may be higher than the impurity concentration in the first photoelectric conversion unit. The higher the impurity concentration, the stronger the potential gradient and the deeper the effective depth of the photoelectric conversion unit. In order to increase the impurity concentration, the concentration of ions to be implanted may be increased.

<Distance detection processing>
A subject distance calculation process performed by the arithmetic processing unit 104 will be described. The arithmetic processing unit 104 acquires a first ranging image from signals obtained from the photoelectric conversion unit 121 of the ranging pixel 110 and the photoelectric conversion unit 123 of the ranging pixel 111. Similarly, the arithmetic processing unit 104 acquires a second ranging image from signals obtained from the photoelectric conversion unit 122 of the ranging pixel 110 and the photoelectric conversion unit 124 of the ranging pixel 111. The arithmetic processing unit 104 obtains an image shift amount between these two ranging images. The image shift amount may be calculated by a known method using a correlation value or the like. When the image shift amount between the two distance measurement images is obtained, the distance of the subject can be calculated based on the principle of triangulation.

<Summary>
Thus, when the design pupil position of the solid-state image sensor 103 and the exit pupil position of the imaging optical system 101 are different, the pupil transmittances of the pupil regions corresponding to the two photoelectric conversion units in the ranging pixel are different. As can be seen from FIG. 2 and FIG. 4, there is a magnitude relationship between the transmittances of the two pupil regions, with a straight line 107 passing through the center of the solid-state imaging device 103 and perpendicular to the X direction (first direction) as a boundary line. Reverse. Therefore, the positional relationship along the X direction of the photoelectric conversion unit having a large amount of received light is reversed between the region 105 and the region 106 with the straight line 107 as a boundary. That is, in the first embodiment, in the area 105, the photoelectric conversion unit 121 positioned in the −X direction is larger than the received light amount of the photoelectric conversion unit 122 positioned in the + X direction in the ranging pixel 110. On the other hand, in the area 106, the photoelectric conversion unit 124 positioned in the + X direction in the distance measurement pixel 111 is larger than the received light amount of the photoelectric conversion unit 123 positioned in the + X direction.

  As shown in this embodiment, the capacity of the photoelectric conversion unit that receives light from the pupil region with high pupil transmittance is set larger than the capacity of the photoelectric conversion unit that receives light from the pupil region with low pupil transmittance. As a result, it is possible to avoid the problem of light quantity reduction and saturation. That is, even if the exposure time is set so that a sufficient amount of light can be received from a pupil region having a low pupil transmittance, saturation can be avoided because the capacity of the other photoelectric conversion unit is larger than that. Therefore, even when a subject having a high contrast ratio is photographed, neither light amount deficiency nor saturation occurs, and the degradation of the quality of the distance measurement image can be suppressed and accurate distance measurement can be performed.

  As described above, by adopting the present embodiment, it is possible to improve the ranging accuracy and the quality of the captured image over the entire surface of the solid-state imaging device regardless of the zoom state and the focus state.

  In addition, in the conventional solid-state imaging device, a method of increasing both the capacities of the first photoelectric conversion unit and the second photoelectric conversion unit is also conceivable. However, increasing the capacity of the photoelectric conversion unit more than necessary is not preferable because it leads to an increase in power consumption and a decrease in reading speed. In this embodiment, the capacity of the photoelectric conversion unit that receives a relatively small amount of light is relatively small, and the capacity of the photoelectric conversion unit that receives a relatively large amount of light is relatively large. An increase and a decrease in reading speed can be suppressed.

  More preferably, the magnitude of the lateral drift electric field in the second photoelectric conversion part having a relatively large capacity is made larger than the magnitude of the drift electric field in the first photoelectric conversion part having a relatively small capacity. This is because the charge transfer rate is slower as the capacity of the photoelectric conversion unit is larger. The magnitude of the drift electric field of the second photoelectric converter having a relatively large capacity of the photoelectric converter is made larger than the magnitude of the drift electric field of the first photoelectric converter having a relatively small capacity of the photoelectric converter. Thus, the difference in charge transfer rate can be suppressed. Specifically, the impurity distribution may be inclined in the horizontal direction by performing multiple implantations while shifting in the horizontal direction.

<Identification of distance to surrounding area and modification of first embodiment>
The difference in pupil transmittance between the two pupil regions corresponding to the two photoelectric conversion units of the ranging pixel 110 increases as the distance between the ranging pixel 110 and the straight line 107 passing through the center of the solid-state image sensor 103 increases. In particular, a region separated from the straight line 107 by at least 0.25 times the X-direction length of the solid-state image sensor 103 (
The difference in pupil transmittance is large in a region within 1/4 of the entire X direction end of the solid-state image sensor 103. It should be noted that in a region separated from the straight line 107 by 0.40 times or more of the length of the solid-state imaging device in the X direction (a region within 1/10 of the entire X-direction end of the solid-state imaging device 103), The difference becomes noticeable. Therefore, the distance between the straight line 107 and the peripheral regions 105 and 106 is preferably 0.40 times or more, and more preferably 0.25 or more times the length of the solid-state image sensor 103 in the X direction.

  In addition, it is more preferable that the capacitance difference between the two photoelectric conversion units is increased as the distance measurement pixels are arranged farther from the center (straight line 107) of the solid-state imaging device according to the difference in pupil transmittance. FIG. 5A is an example in which the capacitance difference between the two photoelectric conversion units is increased according to the position of the ranging pixel. In FIG. 5A, a region 105 is a region on the −X direction side with respect to the straight line 107, and is a region whose distance from the straight line 107 is 1/6 or more of the X direction length of the solid-state image sensor 103. Similarly, the region 106 is a region on the + X direction side of the straight line 107, and is a region whose distance from the straight line 107 is 1/6 or more of the X direction length of the solid-state imaging device 103. On the lower side of FIG. 5A, the depth of the photoelectric conversion unit is shown for four distance measuring pixels surrounded by a chain line in the peripheral regions 105 and 106. In this example, the capacity of the photoelectric conversion units 121 and 124 corresponding to the pupil region having a high pupil transmittance is increased toward the periphery, and the capacitance of the photoelectric conversion units 122 and 123 corresponding to the pupil region having a low pupil transmittance is increased. Is made smaller. However, other methods can be adopted as a method of increasing the difference in capacity as it goes to the periphery. For example, it is conceivable that the capacities of the photoelectric conversion units 121 and 124 are fixed and the capacities of the photoelectric conversion units 122 and 123 are made smaller toward the peripheral part. On the contrary, it is also conceivable that the capacities of the photoelectric conversion units 122 and 123 are fixed, and the capacities of the photoelectric conversion units 121 and 124 are increased toward the periphery.

  Alternatively, all the capacities of the photoelectric conversion units 121 to 124 may be changed. It is also conceivable that the photoelectric conversion units 122 and 123 have a larger reduction rate than the photoelectric conversion units 121 and 124. In general, the pixels located in the peripheral area of the solid-state imaging device have a smaller incident light amount. Therefore, the photoelectric conversion units 122 and 123 are smaller in the photoelectric conversion units 121 and 124 while the capacity of each of the photoelectric conversion units 121 to 124 is smaller. It is preferable to increase the reduction rate compared to 124. In other words, the capacity of the photoelectric conversion unit is reduced in the peripheral part by decreasing the capacity of the photoelectric conversion part as the pixel is far from the center and making the reduction rate according to the distance different among the photoelectric conversion parts in the pixel. It is preferable to increase the difference. By doing in this way, the problem of a light quantity fall and saturation can be avoided also between different ranging pixels.

(Modification 2. Distance measuring pixel near the center)
Contrary to the above, in the vicinity of the center of the solid-state image sensor 103, that is, in an area less than a predetermined distance from the straight line 107 passing through the center of the solid-state image sensor 103 and perpendicular to the X direction, two pupils corresponding to the two photoelectric conversion units. The difference in pupil transmittance between regions is small. Specifically, if the distance from the straight line 107 is less than 0.25 times the length of the solid-state image sensor 103 in the X direction, the difference in pupil transmittance is small. Therefore, in the central region 108 of the solid-state imaging device 103, there is no need to provide a difference in the capacity of the two photoelectric conversion units in the ranging pixel, or photoelectric conversion corresponding to a pupil region having a relatively high pupil transmittance. The capacity of the part may be relatively small. In the vicinity of the center of the solid-state image sensor 103, even if the capacity of the photoelectric conversion unit that receives light passing through the pupil region having a small pupil transmittance is greater than the capacity of the other photoelectric conversion unit, The quality degradation is small and is not a problem.

FIG. 5B is a diagram illustrating an example of this modification. In FIG. 5B, a region 108 surrounded by a chain line is a region whose distance from the straight line 107 is within 1/6 of the X-direction length of the solid-state image sensor 103. In the example of FIG. 5B, in the area 108, the ranging pixel 111 is also provided on the −X direction side of the straight line 107, and the ranging pixel 110 is also provided on the + X direction side of the straight line 107. Yes. As described above, the distance measurement pixels in the area 108 may have the same capacity of the two photoelectric conversion units. Alternatively, in the area 108, the −X direction side of the straight line 107 may be all the ranging pixels 111, and the + direction side of the straight line 107 may be all the ranging pixels 110.

  Alternatively, the two regions 105 and 106 may be adjacent to each other without providing the central region. That is, all regions on the −X direction side from the straight line 107 may be the region 105, and all regions on the + X direction side from the straight line 107 may be the region 106. This case corresponds to the case where the predetermined distance is zero.

(Modification 3. Tolerance of ranging pixels that do not satisfy the condition)
In the above description, in the region 105 and the region 106, the capacitance of the photoelectric conversion unit corresponding to the pupil region having a high pupil transmittance is set larger than that of the other photoelectric conversion unit for all the ranging pixels. However, some distance measurement pixels in the region 105 and the region 106 may have a capacitance of a photoelectric conversion unit corresponding to a pupil region having a low pupil transmittance that is equal to or less than the capacitance of the other photoelectric conversion unit. Absent.

  At least in the region on the negative side in the X direction from the center of the solid-state imaging device 103 (in the region 105), the capacitance of the photoelectric conversion unit (121) on the negative direction side in the X direction is in the positive direction in the X direction. It suffices if there is a ranging pixel larger than the capacity of the photoelectric conversion unit (122) on the side. Similarly, in the region on the positive side in the X direction from the center of the solid-state image sensor 103 (in the region 106), the capacitance of the photoelectric conversion unit (124) on the positive direction side in the X direction is It suffices if there is a ranging pixel larger than the capacity of the photoelectric conversion unit (123) on the negative direction side. About the ranging pixel which has such a structure, the quality fall of a ranging image can be suppressed.

  However, in order to effectively suppress the degradation in the quality of the distance measurement image, it is preferable that the capacitance satisfies the above-described conditions for at least a predetermined ratio of the distance measurement pixels in each region. That is, it is preferable to relatively increase the capacity of the photoelectric conversion unit corresponding to a pupil region having a high pupil transmittance in at least a predetermined ratio (for example, more than half) of distance measurement pixels. Here, it is preferable that the above-mentioned conditions are satisfied for at least half of the ranging pixels, and it is more preferable that the ratio increases. For example, it is more preferable that the predetermined ratio is 80% or more.

(Deformation 4. Presence of normal imaging pixels)
All the pixels of the solid-state image sensor 103 may be distance measurement pixels, or only a part may be distance measurement pixels. When all the pixels are distance measurement pixels, a captured image can be acquired by calculating the sum of the distance measurement images acquired by the two photoelectric conversion units. When some of the pixels are distance measuring pixels, the other pixels are normal imaging pixels including one photoelectric conversion unit that receives light from the entire pupil region of the imaging optical system 101. In this case, a captured image at the distance measurement pixel may be acquired by the same method as described above, or may be obtained by complementing with a captured image acquired by a normal image pickup pixel provided around the distance measurement pixel.

[Embodiment 2]
<Distance pixel>
The solid-state image sensor 203 in the second embodiment is optimized when the exit pupil 130 of the imaging optical system 101 is in a position close to the solid-state image sensor 203.

  6A is a diagram showing the arrangement of the ranging pixels 210 and the ranging pixels 211 included in the solid-state imaging device 203, and FIG. 6B shows the configuration of the ranging pixels 210 and 211. It is sectional drawing.

The distance measurement pixel 210 and the distance measurement pixel 211 according to the present embodiment are different from the distance measurement pixel 110 and the distance measurement pixel 111 according to the first embodiment in the shape of the microlens and the depth of the photoelectric conversion unit. The microlens 212 is an exit pupil 130 at a position close to the solid-state image sensor 203.
The principal ray passing through the center of the light is eccentrically arranged so as to be incident between the photoelectric conversion unit 221 and the photoelectric conversion unit 222. That is, in the ranging pixel 210 in the peripheral region 205 in the minus direction (−X direction) with respect to the center of the solid-state image sensor 203, the microlens 212 is arranged eccentrically in the plus direction (+ X direction). On the other hand, in the ranging pixels 211 in the peripheral region 206 in the plus direction (+ X direction) with respect to the center of the solid-state image sensor 203, the microlens 212 is arranged eccentrically in the minus direction (−X direction).

  Further, the photoelectric conversion unit 222 is deeper than the photoelectric conversion unit 221, and the photoelectric conversion unit 223 is deeper than the photoelectric conversion unit 224. That is, in the region (205) that is more than a predetermined distance in the −X direction from the Y-direction straight line 107 passing through the center of the solid-state imaging device 203, the capacitance of the photoelectric conversion unit 222 on the + X direction side is equal to the photoelectric on the −X direction side. It is larger than the capacity of the conversion unit 221. On the other hand, in a region (206) that is more than a predetermined distance away in the + X direction from the Y-direction straight line 107 passing through the center of the solid-state imaging device 203, the capacitance of the −X direction side photoelectric conversion unit 223 is + X direction side photoelectric conversion. It is larger than the capacity of the part 224.

<Sensitivity characteristics of photoelectric converter>
FIG. 7 shows the incident angle dependence of the sensitivity of each photoelectric conversion unit. The sensitivity characteristics (dotted lines) of the photoelectric conversion unit 222 and the photoelectric conversion unit 224 have high sensitivity to incident light from the minus direction (−X direction) and low sensitivity to incident light from the plus direction (+ X direction). On the other hand, the sensitivity characteristics (broken line) of the photoelectric conversion unit 221 and the photoelectric conversion unit 223 have low sensitivity to incident light from the minus direction (−X direction) and high sensitivity to incident light from the plus direction (+ X direction). . However, the sensitivity characteristics of the photoelectric conversion units 221 and 222 are shifted in the negative direction with respect to the sensitivity characteristics of the photoelectric conversion units 121 and 122. Similarly, the sensitivity characteristic of the photoelectric conversion unit 223 and the sensitivity characteristic of the photoelectric conversion unit 224 are shifted in the positive direction with respect to the sensitivity characteristics of the photoelectric conversion units 123 and 124.

<The reason why pupil transmittance is different and its problem>
When such a solid-state image sensor 203 is used, the pupil region 131 and the pupil region 132 are centrosymmetric on the exit pupil 130 when the distance between the exit pupil 130 of the imaging optical system 101 and the solid-state image sensor 203 is short. (FIG. 8 (a)). Similarly, the pupil region 133 and the pupil region 134 are also centrosymmetric on the exit pupil 130 (FIG. 8B). Therefore, if the distance between the exit pupil 130 and the solid-state image sensor 203 is always kept at this distance, the pupil transmittances of the pupil region 131 and the pupil region 132 are ideally equal to each other, and the photoelectric conversion unit 221 and the photoelectric conversion unit 222 The amount of incident light is equal. Similarly, the pupil transmittances of the pupil region 133 and the pupil region 134 are also equal to each other, and the amounts of light incident on the photoelectric conversion unit 223 and the photoelectric conversion unit 224 are equal.

  However, as described above, the pupil position of the imaging optical system varies depending on the zoom state, the focus state, and the like. Further, even if the distance between the exit pupil 130 and the solid-state image sensor 203 is always kept close, there may be a case where the amount of light is reduced due to vignetting of the imaging optical system. For the above reasons, in the ranging pixels located in the peripheral region of the solid-state imaging device, the pupil transmittances of the pupil region 131 and the pupil region 132 and the pupil transmittances of the pupil region 133 and the pupil region 134 are not equal to each other.

  FIGS. 8C and 8D show the case where the exit pupil 130 is at a distance far from the solid-state imaging device 203 (e.g., when the image is taken on the telephoto side of the zoom lens).

FIG. 8C shows the state of the light beam received by the ranging pixel 210 in the peripheral region 205 in the −X direction (negative side in the first direction) of the solid-state image sensor 203. The photoelectric conversion unit 221 receives the light beam 141 from the pupil region 131, and the photoelectric conversion unit 222 receives the light beam 142 from the pupil region 132. As can be seen from FIG. 8C, the divergence angle of the light beam 142 is larger than the divergence angle of the light beam 141. Accordingly, since the pupil transmittance of the pupil region 132 is higher than that of the pupil region 131, the amount of light received by the photoelectric conversion unit 222 is larger than the amount of light received by the photoelectric conversion unit 221.

  FIG. 8D shows the state of the light beam received by the ranging pixel 211 in the peripheral region 206 in the + X direction (positive side in the first direction) of the solid-state image sensor 203. The photoelectric conversion unit 223 receives the light beam 143 from the pupil region 133, and the photoelectric conversion unit 224 receives the light beam 144 from the pupil region 134. As can be seen from FIG. 8D, the divergence angle of the light beam 143 is larger than the divergence angle of the light beam 144. Accordingly, since the pupil transmittance of the pupil region 133 is higher than that of the pupil region 134, the amount of light received by the photoelectric conversion unit 223 is larger than the amount of light received by the photoelectric conversion unit 224.

  The position of the photoelectric conversion unit with a large amount of received light is opposite to that of the first embodiment, but if there is a difference in the amount of light received between the two photoelectric conversion units, the trade-off between insufficient light quantity and saturation is the same as in the first embodiment. Problems arise.

<Effect due to change in capacitance of photoelectric conversion unit>
In the solid-state imaging device 203 according to the present embodiment, the depth of the photoelectric conversion units 222 and 223 that receive a relatively large amount of light is made deeper than the depth of the photoelectric conversion units 221 and 224 that receive a relatively small amount of light. Yes. That is, the capacitances of the photoelectric conversion units 222 and 223 that receive light in the pupil regions 132 and 133 having higher pupil transmittance than the photoelectric conversion units 221 and 224 that receive light from the pupil regions 131 and 134 having low pupil transmittance. Has increased. This solves simultaneously the shortage of light in the photoelectric conversion units 221 and 224 that receive the light that has passed through the pupil regions 131 and 134 and saturation in the photoelectric conversion units 222 and 223 that receive the light that has passed through the pupil regions 132 and 133. It becomes possible to do.

  Since the method of changing the depth of each photoelectric conversion unit has already been described in the first embodiment, description thereof is omitted here.

<Effect of this embodiment>
As described above, by adopting the present embodiment, as in the case of the first embodiment, it is possible to improve the ranging accuracy over the entire surface of the solid-state imaging device irrespective of the zoom state and the focus state.

  In the present embodiment, various modifications shown in the first embodiment can be adopted.

[Embodiment 3]
In contrast to the solid-state imaging device according to the first and second embodiments, the capacitance is changed depending on the depth of the photoelectric conversion unit, whereas the distance measuring pixel in the solid-state imaging device 303 according to the third embodiment is the size within the plane of the photoelectric conversion unit ( The capacity is changed by changing the area. In order to change the in-plane size of the photoelectric conversion portion, the size of the region where ions are implanted into the substrate 113 may be changed.

  Hereinafter, as in the first embodiment, a case where the exit pupil 130 of the imaging optical system 101 is optimized when it is located at an infinite distance from the solid-state image sensor 303 will be described. When the exit pupil 130 is optimized when it is close to the solid-state imaging device, the positional relationship of the photoelectric conversion units is set along the first direction as in the relationship between the first and second embodiments. And invert it.

FIG. 9A is a diagram showing the arrangement of distance measuring pixels in the solid-state image sensor 303. FIGS. 9B and 9C are cross-sectional views showing the configuration of the ranging pixels 310 and 311. Although only the photoelectric conversion unit is shown in this figure, a microlens 112 is provided on the upper part of the photoelectric conversion unit as in the first embodiment. The ranging pixel 310 is a ranging pixel located in the peripheral region 305 in the minus direction (−X direction) of the solid-state imaging device 303. In the ranging pixel 310, the length of the photoelectric conversion unit 321 in the second direction (Y direction) perpendicular to the first direction is longer than that of the photoelectric conversion unit 322 (FIG. 9B). On the other hand, the ranging pixel 311 is a ranging pixel located in the peripheral region 306 in the plus direction (+ X direction) of the solid-state imaging device 303. In the ranging pixel 311, the length of the photoelectric conversion unit 324 in the second direction (Y direction) perpendicular to the first direction is longer than that of the photoelectric conversion unit 323 (FIG. 9C). That is, the photoelectric conversion units 321 and 324 that receive light from the pupil region with relatively high pupil transmittance than the photoelectric conversion units 322 and 323 that receive light from the pupil region with relatively low pupil transmittance. The capacity is increasing. This simultaneously solves the shortage of light in the photoelectric conversion unit that receives light that has passed through the pupil region with low pupil transmittance and saturation in the photoelectric conversion unit that receives light that has passed through the pupil region with high pupil transmittance. Is possible.

  In addition, in the conventional solid-state imaging device, a method of increasing both the capacities of the two photoelectric conversion units is also conceivable. However, as described in the first and second embodiments, there is a problem that power consumption increases unnecessarily. Further, when the size of the photoelectric conversion unit is changed, the pixel area of the distance measuring pixel is limited, so that it is difficult to increase the capacity of the two photoelectric conversion units. In this embodiment, the capacity of the photoelectric conversion unit that receives a relatively small amount of light is relatively small, and the capacity of the photoelectric conversion unit that receives a relatively large amount of light is relatively large. The pixel area can prevent the degradation of the quality of the distance measurement image.

<Relationship between light receiving area and photoelectric conversion unit>
As shown in FIGS. 9D and 9E, the photoelectric conversion units 321 and 324 having relatively large capacitance (large area) are not completely included in the light receiving regions 325 of the ranging pixels 310 and 311. Is preferred. That is, it is preferable that a part of the photoelectric conversion units 321 and 324 is located outside the light receiving region 325. In other words, it can be expressed that the photoelectric conversion units 321 and 324 having a relatively large capacity preferably include a non-light-receiving region where incident light does not reach. The reason why such a configuration is preferable will be described below. Note that the light receiving area is an area where incident light reaches on the substrate surface of the distance measuring pixel, and is determined by a condensing characteristic by the microlens 111, vignetting by wiring, and the like.

  When the photoelectric conversion unit is completely included in the light receiving region, the amount of light incident on the photoelectric conversion unit increases as the size of the photoelectric conversion unit in the in-plane direction increases. Therefore, when the photoelectric conversion units 321 and 324 having a larger capacity are completely included in the light receiving region 325, the amount of light incident on the photoelectric conversion units 321 and 324 also increases. If the photoelectric conversion units 321 and 324 are not included in the light receiving region, light is not received in a portion not included in the light receiving region (the hatched region in FIG. 9C), and thus it is not necessary to increase the amount of incident light. In order not to include the photoelectric conversion units 321 and 324 in the light receiving region, the photoelectric conversion units 321 and 324 are extended to the lower part of the wiring, or a part of the photoelectric conversion units 321 and 324 is covered with a light shielding film. ,realizable.

  The above configuration is preferable because it is possible to more easily solve the shortage of light amount in the photoelectric conversion unit with a small amount of received light and the saturation in the photoelectric conversion unit with a large amount of received light.

<FD sharing>
Advantage of changing the capacitance of the photoelectric conversion unit by changing the length in the direction (Y direction) perpendicular to the first direction (X direction), thereby making it easier to provide a transistor (Tr), a floating diffusion (FD), and the like. There is also. Specifically, as shown in FIGS. 10A and 10B, the upper end position of the photoelectric conversion units 322 and 323 in the Y direction matches the upper end position of the photoelectric conversion units 321 and 324 in the Y direction. The lengths in the Y direction of the conversion units 322 and 323 are shortened. That is, the lower end position of the photoelectric conversion units 322 and 323 in the Y direction is set to the + Y direction side of the lower end position of the photoelectric conversion units 321 and 324 in the Y direction. With such a configuration, a space below the lower end in the Y direction of the photoelectric conversion units 322 and 323 is vacant, so that circuit elements such as Tr and FD can be provided here. Note that the Y-direction length of the photoelectric conversion units 322 and 323 may be shortened with the lower end position in the Y direction being the same as that of the photoelectric conversion units 321 and 324. In this case, a space is vacant above the upper ends of the photoelectric conversion units 322 and 323 in the Y direction. In addition to these methods, if the Y-direction length of the photoelectric conversion units 322 and 323 is shortened so that the Y position at the center of the photoelectric conversion units 322 and 323 is shifted from the Y position at the center of the ranging pixels 310 and 311, the same applies. The effect is obtained.

  Furthermore, as shown in FIGS. 10C and 10D, in the two distance measuring pixels adjacent in the Y direction, the center of the photoelectric conversion units 322 and 323 in each distance measuring pixel is set to the two distance measuring pixels. It is preferable to shift in a direction away from the adjacent portion. Specifically, in the ranging pixel on the upper side in the Y direction, the upper end in the Y direction of the photoelectric conversion unit 322 is matched with the upper end in the Y direction on the photoelectric conversion unit 321, and the ranging pixel on the lower side in the Y direction is used. The lower end in the Y direction is made to coincide with the lower end in the Y direction of the photoelectric conversion unit 321. By shortening the Y direction length of the photoelectric conversion unit 321 under such conditions, a space can be provided between the photoelectric conversion units 322 of the two distance measuring pixels. It is preferable to arrange an FD (charge detection unit) shared by the two distance measuring pixels in this space. By sharing the FD among a plurality of pixels, the degree of freedom of layout such as wiring can be increased within a limited pixel area.

[Embodiment 4]
The solid-state imaging device 403 in the present embodiment changes the capacitance of the photoelectric conversion unit by changing the length in the first direction (X direction).

  In the following, as in the first embodiment, a case where the exit pupil 130 of the imaging optical system 101 is optimized when it is at a position at infinity from the solid-state imaging device 403 is shown. When the exit pupil 130 is optimized when it is close to the solid-state imaging device, the positional relationship of the photoelectric conversion units is set along the first direction as in the relationship between the first and second embodiments. And invert it.

  FIG. 11A is a diagram illustrating the arrangement of the distance measurement pixels 410 and 411 included in the solid-state image sensor 403. FIGS. 11B and 11C are cross-sectional views showing the configurations of the distance measuring pixels 410 and 411. FIG. The ranging pixel 410 is a ranging pixel located in the peripheral area 405 in the minus direction (−X direction) of the solid-state imaging device 403. In the distance measuring pixel 410, the length of the photoelectric conversion unit 421 in the X direction (first direction) is longer than that of the photoelectric conversion unit 422 (FIG. 11B). On the other hand, the ranging pixel 411 is a ranging pixel located in the peripheral region 406 in the plus direction (+ X direction) of the solid-state imaging device 403. In the distance measurement pixel 411, the length of the photoelectric conversion unit 424 in the X direction (first direction) is longer than that of the photoelectric conversion unit 423 (FIG. 11C). This simultaneously solves the shortage of light in the photoelectric conversion unit that receives light that has passed through the pupil region with low pupil transmittance and saturation in the photoelectric conversion unit that receives light that has passed through the pupil region with high pupil transmittance. Is possible.

  As in the third embodiment, it is preferable that the photoelectric conversion units 421 and 424 having large capacities (large areas) are not completely included in the light receiving regions 425 of the ranging pixels 410 and 411. That is, it is preferable that the photoelectric conversion units 421 and 424 having a long length in the X direction include a non-light receiving region that does not receive incident light.

<Change in baseline length>
In addition, when the centers of the two photoelectric conversion units in one ranging pixel are shifted from the center of the light receiving region 425 in the X direction (first direction), the pupil division characteristics can be controlled. Therefore, it is preferable. In particular, as shown in FIGS. 11D and 11E, if the centers of the two photoelectric conversion units are shifted to the photoelectric conversion units 422 and 423 having a small capacity, the pupil division characteristic is not changed without largely changing the wiring layout. Can be controlled.

[Embodiment 5]
In the first to fourth embodiments, the capacitance is changed by changing any one of the effective depth, the length in the Y direction, and the length in the X direction of the photoelectric conversion unit. In the present embodiment, the capacitance of the photoelectric conversion unit is changed by combining these plural methods. The capacity of the photoelectric conversion unit is determined by three products of the effective depth of the photoelectric conversion unit, the length in the direction perpendicular to the X direction, and the length in the X direction. By combining these, light that passes through a pupil region that has a relatively high pupil transmittance than the capacity of a photoelectric conversion unit that selectively receives light that passes through a pupil region that has a relatively low pupil transmittance is received. What is necessary is just to enlarge the capacity | capacitance of a photoelectric conversion part.

[Embodiment 6]
In the said Embodiments 1-5, the capacity | capacitance is changed by the same method about all the photoelectric conversion parts. In the present embodiment, the method of setting the capacity of the photoelectric conversion unit is changed by the distance measurement pixel. FIG. 12A is a diagram showing the arrangement of the ranging pixels in the solid-state imaging device 403 according to the present embodiment, and FIG. 12B is a cross-sectional view showing the configuration of the ranging pixels. As shown in the figure, in this embodiment, the distance measuring pixel 412 whose capacity is changed according to the length in the X direction and the distance measuring pixel 413 whose capacity is changed according to the depth of the photoelectric conversion unit are arranged close to each other. Yes.

  Further, the centers of the two photoelectric conversion units of the ranging pixel 412 are shifted from the center of the light receiving region 425 in the X direction. In addition, the centers of the two photoelectric conversion units of the ranging pixel 413 coincide with the center of the light receiving region 425. In other words, the deviation of the center of the two photoelectric conversion units in the ranging pixel 412 from the center of the light receiving region is the magnitude of the deviation of the center of the two photoelectric conversion units in the ranging pixel 413 from the center of the light receiving region. Is different.

  FIGS. 12A and 12B show only the ranging pixels in the peripheral region 405 in the −X direction, but the positional relationship of the photoelectric conversion units is reversed in the first direction also in the peripheral region in the + X direction. Are arranged. By adopting such a configuration, it is possible to perform distance measurement with higher accuracy regardless of the zoom state or the subject. The principle will be described below.

  FIG. 13 shows the sensitivity characteristics of the photoelectric conversion unit in each ranging pixel. The pupil region received by the photoelectric conversion unit 426 is closer to the center than the pupil region received by the photoelectric conversion unit 428, and the pupil region received by the photoelectric conversion unit 427 is more than the pupil region received by the photoelectric conversion unit 429. Close to the surroundings. Further, the amount of light received by the photoelectric conversion unit 426 is larger than the amount of light received by the photoelectric conversion unit 428, and the amount of light received by the photoelectric conversion unit 427 is smaller than the amount of light received by the photoelectric conversion unit 429.

  When distance measurement is performed using the distance measurement image acquired by the photoelectric conversion unit 428 and the distance measurement image acquired by the photoelectric conversion unit 427, the distance between the divided pupil regions is long, so that the distance measurement accuracy is improved. . On the other hand, when the ranging image acquired by the photoelectric conversion unit 426 and the ranging image acquired by the photoelectric conversion unit 429 are used, the amount of received light is large and the quality of the ranging image is improved.

  Therefore, by switching the photoelectric conversion unit to be used according to the zoom state or the subject, it is possible to select which of the distances between the pupil regions and the quality of the distance measurement image is to be prioritized.

<Proximity placement of ranging pixels>
It is desirable that the ranging pixel 412 and the ranging pixel 413 are arranged in the vicinity so as to receive a part of the light beam from the same subject. Desirably, it is arranged within a distance of 4 pixels, more desirably within a distance of 2 pixels. FIG. 12 shows an example in which the ranging pixels 412 and the ranging pixels 413 are arranged adjacent to each other.

[Embodiment 7]
In the present embodiment, a case where the present invention is applied to a solid-state imaging device 503 that performs distance measurement by pupil division in the Y direction is shown.

  In the following, as in the first embodiment, a case where the exit pupil 130 of the imaging optical system 101 is optimized when it is at a position at infinity from the solid-state imaging device 403 is shown. In the case where the exit pupil 130 is optimized when it is close to the solid-state imaging device, the magnitude relationship of the capacitance of the photoelectric conversion unit is changed along the Y direction in the same way as the relationship between the first and second embodiments. What is necessary is just to invert.

  FIG. 14A is a diagram illustrating an arrangement of the ranging pixels 510 and 511 included in the solid-state image sensor 503. FIG. FIGS. 14B and 14C are cross-sectional views showing the configuration of the ranging pixels 510 and 511. Photoelectric conversion units 521 and 523 and photoelectric conversion units 522 and 524 are formed in the substrate of the ranging pixels 510 and 511 and are arranged at positions symmetrical with respect to the Y direction around the optical axis of the microlens 112. . The photoelectric conversion units 521 and 523 and the photoelectric conversion units 522 and 524 have the same shape in the substrate in-plane direction. With such a configuration, the incident angle dependency of the sensitivity of the photoelectric conversion units 521 and 523 and the photoelectric conversion units 522 and 524 in the YZ section can be made different from each other.

  Further, the capacities of the photoelectric conversion units 521 and 524 that receive a relatively large amount of light are larger than the capacities of the photoelectric conversion units 522 and 523 that receive a relatively small amount of light. Specifically, with respect to the ranging pixel 510 arranged in the peripheral region 505 in the −Y direction of the solid-state imaging device 503, the photoelectric conversion unit 521 on the −Y direction side is closer to the photoelectric conversion unit 522 on the −Y direction side. The capacity is increasing. Further, with respect to the ranging pixels 511 arranged in the peripheral region 506 in the + Y direction of the solid-state imaging device 503, the capacitance of the photoelectric conversion unit 524 on the + Y direction side is larger than that on the −Y direction side of the photoelectric conversion unit 523. ing. In addition, the method of changing a capacity | capacitance can employ | adopt the arbitrary methods demonstrated above.

  As a result, the light quantity shortage in the photoelectric conversion units 522 and 523 that receive light that has passed through the pupil region with low pupil transmittance and the saturation in the photoelectric conversion units 521 and 524 that receive light that has passed through the pupil region with high pupil transmittance are achieved. Can be solved at the same time. As described above, with respect to a subject composed of line segments extending in the X direction, it is possible to improve the ranging accuracy and the quality of the captured image over the entire surface of the solid-state imaging device regardless of the zoom state and the focus state.

  Ranging pixels that perform pupil division in the X direction and ranging pixels that perform pupil division in the Y direction may be arranged at the same time. Each of these ranging pixels is configured to have a large capacity of a photoelectric conversion unit corresponding to a pair of two photoelectric conversion units and a pupil region having a relatively high pupil transmittance. By adopting such a configuration, it is possible to perform distance measurement for both a subject constituted by a line segment extending in the X direction and a subject constituted by a line segment extending in the Y direction. In this case as well, the capacity of the photoelectric conversion unit may be determined according to the magnitude relationship of pupil transmittance.

[Others]
In the above embodiment, a method using a microlens is used as a pupil division method, but the method is not limited to this. As shown in FIG. 15A, the waveguide 114 may be used and division may be performed according to the waveguide mode. The dominant waveguide mode that couples light incident on the waveguide through the first pupil region is different from the dominant waveguide mode that couples light incident on the waveguide through the second pupil region. . Therefore, the light beam that has passed through the first pupil region can be selectively guided to the first photoelectric conversion unit, and the light beam that has passed through the second pupil region can be selectively guided to the second photoelectric conversion unit. As shown in FIG. 15B, a pupil division waveguide 115 and a waveguide 116 for guiding light to the photoelectric conversion unit may be used. As shown in FIG. These microlenses and a waveguide for guiding light to the photoelectric conversion unit may be used at the same time. By using the waveguide, it is possible to efficiently guide the light incident on the pixel to the photoelectric conversion unit, obtain a higher-quality ranging image, and perform more accurate ranging. Also in the modification shown in FIG. 15, the eccentricity of the microlens shown in the second embodiment may be combined.

  Although various embodiments and modifications thereof have been described above, it is possible to configure the present invention by combining these contents as much as possible.

101: imaging optical system 103: solid-state imaging device 110, 111: ranging pixels 121, 122, 123, 124: photoelectric conversion unit

Claims (23)

A solid-state imaging device including a plurality of pixels that photoelectrically convert a subject image formed by an imaging optical system,
At least a part of the plurality of pixels is a ranging pixel in which a first photoelectric conversion unit and a second photoelectric conversion unit are provided side by side along a first direction,
When the region of the solid-state imaging device is divided into a first region and a second region by a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction,
More than 80% of the ranging pixels in the first region and within a region that is more than a predetermined distance away from a straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The capacity of the first photoelectric conversion unit is larger than the capacity of the second photoelectric conversion unit,
80% or more of the ranging pixels in the second region and within the region separated by the predetermined distance or more from the straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The capacity of the second photoelectric conversion unit is larger than the capacity of the first photoelectric conversion unit,
Solid-state image sensor. More than 80% of the ranging pixels in the first area have a capacitance of the first photoelectric conversion unit larger than that of the second photoelectric conversion unit,
80% or more of the ranging pixels in the second region have a capacity of the second photoelectric conversion unit larger than a capacity of the first photoelectric conversion unit.
The solid-state imaging device according to claim 1 . More than half the distance measurement pixels in the first area and within the area less than the predetermined distance from a straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The capacity of the second photoelectric conversion unit is equal to or greater than the capacity of the first photoelectric conversion unit,
More than half of the distance measurement pixels in the second area and within the area less than the predetermined distance from a straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The capacity of the first photoelectric conversion unit is greater than or equal to the capacity of the second photoelectric conversion unit.
The solid-state imaging device according to claim 2 . The predetermined distance is at least 0.40 times the length of the solid-state image sensor in the first direction.
The solid-state imaging device according to claim 2 or 3 . A distance measuring pixel having a larger distance from a straight line passing through the center of the solid-state imaging device and perpendicular to the first direction has a larger capacitance difference between the first photoelectric conversion unit and the second photoelectric conversion unit.
The solid-state image sensor of any one of Claim 1 to 4 . A distance measuring pixel that has a larger distance from a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction, than the distance measuring pixel that has a smaller distance, has the first photoelectric conversion unit and the second photoelectric sensor. A distance measuring pixel having a larger distance due to a smaller capacity of the converter and a decrease rate corresponding to the distance from the straight line is different between the first photoelectric converter and the second photoelectric converter. A large difference in capacitance between the first photoelectric conversion unit and the second photoelectric conversion unit;
The solid-state imaging device according to claim 5 . Since the length in the direction perpendicular to the imaging surface is different between the first photoelectric conversion unit and the second photoelectric conversion unit, there is a difference in capacitance between the first and second photoelectric conversion units.
The solid-state image sensor of any one of Claim 1 to 6 . Since the impurity concentration is different between the first photoelectric conversion unit and the second photoelectric conversion unit, there is a difference in capacitance between the first and second photoelectric conversion units.
The solid-state image sensor of any one of Claim 1 to 7 . Since the area on the imaging surface is different between the first photoelectric conversion unit and the second photoelectric conversion unit, there is a difference in capacitance between the first and second photoelectric conversion units.
The solid-state image sensor of any one of Claim 1 to 8 . The length in the first direction in the imaging surface is different between the first photoelectric conversion unit and the second photoelectric conversion unit.
The solid-state imaging device according to claim 9 . The centers of the first photoelectric conversion unit and the second photoelectric conversion unit are shifted from the center of the ranging pixel toward the photoelectric conversion unit having a small capacity along the first direction.
The solid-state image sensor according to claim 9 or 10 . Of the first and second photoelectric conversion units, the photoelectric conversion unit having a larger area on the imaging surface includes a non-light-receiving region where light does not enter.
The solid-state image sensor of any one of Claim 9 to 11 . A solid-state imaging device including a plurality of pixels that photoelectrically convert a subject image formed by an imaging optical system,
At least a part of the plurality of pixels is a ranging pixel in which a first photoelectric conversion unit and a second photoelectric conversion unit are provided side by side along a first direction,
When the region of the solid-state imaging device is divided into a first region and a second region by a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction,
More than 80% of the ranging pixels in the first region and within a region that is more than a predetermined distance away from a straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The length in the direction perpendicular to the imaging surface of the first photoelectric conversion unit is longer than the length in the direction perpendicular to the imaging surface of the second photoelectric conversion unit,
A straight line within the second region and passing through the center of the solid-state imaging device and perpendicular to the first direction
More than 80% of the distance measurement pixels in the region separated by the predetermined distance or more from the first photoelectric conversion unit have a length in a direction perpendicular to the imaging surface of the second photoelectric conversion unit. Longer than the length in the direction perpendicular to the imaging surface,
Solid-state image sensor. A solid-state imaging device including a plurality of pixels that photoelectrically convert a subject image formed by an imaging optical system,
At least a part of the plurality of pixels is a ranging pixel in which a first photoelectric conversion unit and a second photoelectric conversion unit are provided side by side along a first direction,
When the region of the solid-state imaging device is divided into a first region and a second region by a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction,
More than 80% of the ranging pixels in the first region and within a region that is more than a predetermined distance away from a straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The impurity concentration of the first photoelectric conversion unit is higher than the impurity concentration of the second photoelectric conversion unit,
80% or more of the ranging pixels in the second region and within the region separated by the predetermined distance or more from the straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The impurity concentration of the second photoelectric conversion unit is higher than the impurity concentration of the first photoelectric conversion unit;
Solid-state image sensor. The first region is located in a negative direction of the first direction from a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction,
The second region is located in a positive direction of the first direction from a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction.
The solid-state image sensor of any one of Claim 1 to 14 . The first region is located in a positive direction of the first direction from a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction,
The second region is located in a negative direction of the first direction from a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction.
The solid-state image sensor of any one of Claim 1 to 15 . The distance measuring pixel has a microlens, and the first photoelectric conversion unit and the second photoelectric conversion unit are arranged symmetrically with respect to the optical axis of the microlens.
The solid-state image sensor of any one of Claim 1 to 16 . The ranging pixel has a waveguide, and guides incident light to the first photoelectric conversion unit or the second photoelectric conversion unit according to an incident angle to the ranging pixel.
The solid-state image sensor of any one of Claim 1 to 17 . A solid-state imaging device including a plurality of pixels that photoelectrically convert a subject image formed by an imaging optical system,
At least a part of the plurality of pixels is a ranging pixel in which a first photoelectric conversion unit and a second photoelectric conversion unit are provided side by side along a first direction,
When the region of the solid-state imaging device is divided into a first region and a second region by a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction,
More than half the distance measurement pixels in the first area and within the area that is more than a predetermined distance away from a straight line that passes through the center of the solid-state imaging device and is perpendicular to the first direction, The area of the imaging surface of the first photoelectric conversion unit is larger than the area of the imaging surface of the second photoelectric conversion unit,
More than half the distance measurement pixels in the second area and within the area that is more than the predetermined distance from the straight line passing through the center of the solid-state imaging device and perpendicular to the first direction, The area of the imaging surface of the second photoelectric conversion unit is larger than the area of the imaging surface of the first photoelectric conversion unit;
The ranging pixel has a waveguide, and guides incident light to the first photoelectric conversion unit or the second photoelectric conversion unit according to an incident angle to the ranging pixel.
Solid-state image sensor. At least a part of the plurality of pixels includes a second distance measuring unit in which a third photoelectric conversion unit and a fourth photoelectric conversion unit are provided side by side along a second direction perpendicular to the first direction. Pixel,
When the region of the solid-state imaging device is divided into a third region and a fourth region by a straight line passing through the center of the solid-state imaging device and parallel to the first direction,
More than at least half of the second ranging pixels in the third region and in the region that is more than a predetermined distance away from a straight line that passes through the center of the solid-state imaging device and is perpendicular to the second direction. The ranging pixel has a capacity of the third photoelectric conversion unit larger than a capacity of the fourth photoelectric conversion unit,
More than at least half of the second ranging pixels in the fourth region and in the region that is more than a predetermined distance away from a straight line that passes through the center of the solid-state imaging device and is perpendicular to the second direction. The distance measurement pixel has a capacity of the fourth photoelectric conversion unit larger than a capacity of the third photoelectric conversion unit.
The solid-state image sensor of any one of Claim 1 to 19 . An imaging optical system;
The solid-state imaging device according to any one of claims 1 to 20 , wherein a subject image formed by the imaging optical system is photoelectrically converted.
A distance calculation unit that calculates a distance of a subject based on signals obtained from the first and second photoelectric conversion units of the solid-state imaging device;
Ranging device comprising. The first photoelectric conversion unit is configured to selectively receive light from a first pupil region that is a part of an exit pupil of the imaging optical system,
The second photoelectric conversion unit is configured to selectively receive light from a second pupil region that is a part of an exit pupil of the imaging optical system and is different from the first pupil region. And
Of the first and second pupil regions, the capacity of the photoelectric conversion unit corresponding to the pupil region having a high pupil transmittance is larger than the capacity of the photoelectric conversion unit corresponding to the pupil region having a low pupil transmittance,
The distance measuring device according to claim 21 . A distance measuring device according to claim 21 or 22 ,
A subject image acquisition unit for acquiring a subject image based on a signal obtained from the solid-state image sensor;
An imaging apparatus comprising:

Images