Symmetric polarization filter for autofocus pixel structures

Abstract

This invention provides an image sensor and apparatus for a phase-detection autofocus process. A symmetrical polarization filter includes a first polarizer defining one of a first plurality of apertures and a second polarizer adjacent to the first polarizer defining a second plurality of apertures. The first plurality of apertures are mirror-symmetrical with respect to a horizontal axis of the symmetrical polarization filter between the first and second polarizers. The horizontal axis can be defined as a symmetry axis of the symmetrical polarization filter that is coplanar with both the first and second polarizers.

Description

Technical Field

[0001] The present invention generally relates to image sensors, and particularly (but not exclusively), to CMOS image sensors and their applications. Prior Art

[0002] Image sensors have become ubiquitous and are now widely used in digital cameras, cellular phones, security cameras, as well as in medical, automotive, and other applications. As image sensors are integrated into a wider range of electronic devices, there is a desire to enhance their functionality, performance metrics, and the like in as many ways as possible (e.g., resolution, power consumption, dynamic range, etc.) through both device architecture design and image acquisition processing.

[0003] A typical image sensor operates in response to image light reflected from an external scene being incident on the image sensor. The image sensor comprises an array of pixels having photosensitive elements (e.g., photodiodes) that absorb a portion of the incident image light and generate an image charge upon absorption. The image charge of each pixel can be measured as an output voltage of each photosensitive element, which varies in response to the incident image light. In other words, the amount of image charge generated is proportional to the intensity of the image light, and this image charge is used to generate a digital image (e.g., image data) representing the external scene. Simple diagram description

[0004] Non-limiting and non-exhaustive embodiments of the present invention are described with reference to the following figures, wherein like reference numerals refer to similar parts throughout the various views unless otherwise specified. Where appropriate, not all instances of a component are necessarily labeled to avoid cluttering the drawings. The figures are not necessarily to scale, emphasis instead being placed on illustrating the principles being described.

[0005] 1A illustrates a top view of a minimal repeating unit of an image sensor having a shared microlens and a polarization pixel according to an embodiment of the present invention.

[0006] 1B-1D illustrate exemplary sub-pixels and grouped polarized pixels of an imaging system according to an embodiment of the present invention.

[0007] FIG. 2A is a schematic diagram showing an asymmetric polarization filter used in this technology.

[0008] FIG. 2B is a schematic diagram illustrating a symmetrical polarization filter according to an embodiment of the present invention.

[0009] 3 is a diagram illustrating an example process for forming polarizer elements for defining the example symmetric polarizing filter of FIG. 2B , according to an embodiment of the present invention.

[0010] FIG4 is a functional block diagram of an imaging system according to an embodiment of the present invention, the imaging system including an image sensor having polarization pixels that implement a symmetrical polarization filter as described in the exemplary embodiment of FIG2B through FIG3 .

[0011] Corresponding reference characters indicate corresponding components throughout the several views of the drawings. Those skilled in the art will appreciate that elements in the figures are illustrated for simplicity and clarity and are not necessarily drawn to scale. For example, the dimensions of some elements in the figures may be exaggerated relative to other elements to help improve understanding of the various embodiments of the present invention. Additionally, common but well-understood elements that are useful or necessary in a commercially feasible embodiment are often not depicted to facilitate a more unobstructed view of these various embodiments of the present invention. Implementation Method

[0012] Embodiments of devices, systems, and methods are described herein, each including or otherwise relating to an image sensor having a symmetrical polarization filter. In the following description, numerous specific details are set forth to provide a thorough understanding of the examples. However, those skilled in the art will recognize that the techniques described herein can be practiced without employing one or more of these specific details, or with other methods, components, materials, and so forth. In other instances, well-known structures, materials, or operations are not shown or described in detail to avoid obscuring certain aspects.

[0013] Advances in semiconductor processing technology have enabled the fabrication of complementary metal oxide semiconductor devices (e.g., image sensors, processors, displays, and the like) with increasingly smaller feature sizes. This has enabled the miniaturization of many devices and the integration of multiple sensor features into a single sensor array. For example, a hybrid image sensor may include both an image pixel structure and a polarization pixel structure to facilitate both image capture and autofocus operations on a single sensor. However, as described in more detail with reference to FIG. 2A , polarization filters based on a conventional design face several process and performance limitations. For example, conventional polarizer configurations include four different polarizer filters formed by different mask patterns, which increases manufacturing complexity. As another example, conventional polarizer configurations typically block incident light in the central region of the component photodiodes that comprise the polarization pixel, thereby reducing the quantum efficiency of the polarization pixel structure and impairing PDAF performance. Therefore, there is a need for improved polarizer configurations that address the shortcomings of conventional polarization pixel structures.

[0014] Embodiments described herein utilize an image sensor having an architecture that may include an improved polarization filter configuration to provide improved phase information of an external scene relative to conventional polarizer configurations (e.g., for phase detection autofocus, depth mapping, and the like). Thus, an image sensor architecture may include a plurality of polarization pixel structures configured to provide surface normal angle information of an external scene. Advantageously, the phase information provided by a plurality of sub-pixels can be combined with the surface normal angle information to provide a depth map compatible with a variety of scenes. For example, the phase information from a plurality of sub-pixels can provide phase information of edges of an external scene compared to one another or relative depth information (e.g., when switching to different objects in the external scene).

[0015] Embodiments of the present invention include symmetrical polarizing filter configurations based at least in part on uniform linear polarizer patterns for multi-directional phase detection autofocus (PDAF). This phase information can then be used to generate multi-directional guidance regarding how an objective lens of an imaging system should be adjusted when a point of interest (POI) is out of focus. Compared to conventional multi-polarizer configurations that include multiple different polarizer patterns with varying extinction ratios and performance characteristics, the symmetrical polarizing filters of the present invention offer improved quantum efficiency, improved extinction ratio consistency, and improved manufacturing efficiency.

[0016] Figures 1A through 1D illustrate representative views of an image sensor 100. It should be understood that the views presented in Figures 1A through 1D may omit certain elements of image sensor 100 to avoid obscuring the details of the present invention. Therefore, some elements of image sensor 100 may be unlabeled or otherwise omitted in Figures 1A through 1D. It should further be understood that, in some embodiments, image sensor 100 may omit elements shown in Figures 1A through 1D.

[0017] FIG1A illustrates a top view 100-A of an image sensor 100 having a shared microlens 115 and a polarized pixel 104, according to an embodiment of the present invention. Image sensor 100 includes an image pixel 102, which may represent any image pixel included in a plurality of image pixels that collectively form image sensor 100. Image pixel 102 may be a repeating unit of image sensor 100 and include a plurality of subpixels 103 (e.g., a blue subpixel 103-B, a green subpixel 103-G, and a red subpixel 103-R) and a plurality of polarized pixels 104 (e.g., P1, P2, P3, and P4). In this context, the term "subpixel" refers to a group of pixels that, as part of a multi-color image pixel 102, together generate image data corresponding to the indicated color filter. Image sensor 100 includes, among other things, a plurality of photodiodes 105 (e.g., B1, B2, B3, B4, G1, G2, G3, G4, R1, R2, R3, R4, P1, P2, P3, P4, etc.) arranged as a photodiode array in rows (e.g., R1, R2, R3, ..., RY) and columns (e.g., C1, C2, C3, ..., CX), a plurality of sub-pixel color filters 110 (e.g., 110-G, 110-B, 110-R), a plurality of polarization filters 107, a plurality of shared microlenses 115, and a plurality of unshared microlenses 117. Elements of image sensor 100 are optically aligned to form a plurality of sub-pixels 103 and a plurality of polarized pixels 104 of image sensor 100 for capturing images (eg, color images, depth map images, video, and the like) while also providing phase detection autofocus.

[0018] In some embodiments, the plurality of photodiodes 105 are arranged in a regular, repeating manner such that the plurality of photodiodes 105 are positioned or otherwise disposed at regular intervals within a semiconductor material to form a square or rectangular photodiode array. Individual photodiodes included in the plurality of photodiodes 105 may correspond to doped regions within respective portions of the semiconductor material that respond to incident light (e.g., the doped regions may form a PN junction that generates an image charge proportional to a magnitude or intensity of the incident light). It should be understood that multiple photodiodes 105 are disposed within the semiconductor material (see, for example, Figures 1B to 1C) and can be optically aligned with respective sub-pixel color filters (e.g., 110-B, 110-G, or 110-R), polarization filters 107, shared microlenses 115, non-shared microlenses 117, or the like to form, for each imaging pixel of image sensor 100, an individual sub-pixel 103 included in the plurality of sub-pixels 103 or an individual polarization pixel 104 included in the plurality of polarization pixels 104.

[0019] View 100-A depicted in FIG1A generally refers to a given sub-pixel included in the plurality of sub-pixels 103 or a given polarized pixel included in the plurality of polarized pixels 104 by referring to a labeled photodiode, without necessarily showing an exemplary size, shape, or position of the given photodiode. Furthermore, view 100-A may omit the labeling of some elements included in image sensor 100. However, it should be noted that exemplary cross-sectional views of image sensor 100 are shown in FIG1B-1C , and an exemplary relative size comparison of certain elements is shown in FIG1D .

[0020] Referring back to FIG1A , image pixel 102 represents an individual image pixel 102 included in a plurality of image pixels 102 that collectively form image sensor 100. Image pixel 102 includes a plurality of polarized pixels 104 (e.g., P1, P2, P3, P4) and a plurality of sub-pixels 103 (e.g., a green sub-pixel 103-G, a blue sub-pixel 103-B, and a red sub-pixel 103-R). Each of the plurality of polarized pixels 104 includes a non-shared microlens 117 and a polarizing filter 107 optically aligned over a respective photodiode (e.g., a corresponding photodiode associated with P1, P2, P3, or P4) included in a photodiode array (e.g., a plurality of photodiodes 105). At least a subset of the plurality of sub-pixels 103 includes a shared microlens 115 and a sub-pixel color filter 110 optically aligned above a group of photodiodes (e.g., B1, B2, B3, and B4). In the illustrated embodiment, the group of photodiodes associated with a given one of the plurality of sub-pixels 103 includes four photodiodes arranged together in a 2×2 pattern. Thus, a shared microlens 110 can extend over two columns and two rows included in the photodiode array, overlying the four photodiodes.

[0021] In some embodiments, polarized pixels 104 are grouped together to cover an area similar to that of individual sub-pixels 103 of sub-pixels 103. In the illustrated embodiment, four polarized pixels included in polarized pixels 104 are arranged adjacent to each other in a 2×2 pattern to form a respective group of polarized pixels (e.g., P1, P2, P3, and P4) at least partially surrounded by sub-pixels 103. More specifically, an imaging pixel 102 may include three sub-pixels (e.g., a green sub-pixel 103-G, a blue sub-pixel 103-B, and a red sub-pixel 103-R) that complement polarized pixels 104 (e.g., a group of polarized pixels including P1, P2, P3, and P4) to form a repeating unit (e.g., imaging pixel 102) of image sensor 100. In one embodiment, a group of polarized pixels 104 of an image pixel 102 includes four polarized pixels 104 (e.g., the group of polarized pixels located in rows 7-8 and columns 5-6 of FIG. 1A ) (as indicated by the symbols "···" in the figure), which are collectively surrounded by a first group of sub-pixels 103 (e.g., the sub-pixels located in rows 5-10 and columns 3-8). The first group of sub-pixels includes four green sub-pixels 103-G, two blue sub-pixels 103-B, and two red sub-pixels 103-R. As shown in FIG. 1A , the four green sub-pixels 103-G are arranged diagonally adjacent to the four polarized pixels, while the two blue sub-pixels 103-B and the two red sub-pixels 103-R are arranged laterally adjacent to the four polarized pixels (e.g., vertically or horizontally relative to the four polarized pixels when viewed from a top-down perspective as shown in FIG. 1A ).

[0022] As shown in FIG1A , image pixel 102 is a panchromatic image pixel (e.g., capable of generating an image signal in response to incident light that collectively represents at least the visible portion of the electromagnetic spectrum) and includes a plurality of subpixels (e.g., a red subpixel 103-R, a green subpixel 103-G, and a blue subpixel 103-B). Each subpixel 103 may include a corresponding shared microlens 115 to direct light passing through a subpixel color filter 110 toward a respective photodiode group included in the photodiode array. In contrast, a non-shared microlens 117 is non-shared with respect to the plurality of photodiodes 105 and thus directs light passing through a polarization filter 107 toward a respective photodiode 105, forming a corresponding polarized pixel 104. In some embodiments, polarized pixels 104 are grouped together (e.g., in 2×2 groups). As shown in FIG1A , according to embodiments of the present invention, there may be a spacing and / or a separation distance between individual photodiodes, microlenses, and sub-pixel color filters. Furthermore, in some embodiments, other components (e.g., vias, wiring, circuits, isolation trenches, and the like) may be disposed within the spacing.

[0023] Each microlens (e.g., shared microlens 115 and / or non-shared microlens 117) may be formed of a polymer (e.g., polymethyl methacrylate, polydimethylsiloxane, etc.) or other materials, and may be shaped to have an optical power for converging, diverging, or otherwise directing light incident on the microlens through a corresponding optically aligned one of the plurality of sub-pixel color filters 110 and / or polarization filters 107 to one or more photodiodes included in the plurality of photodiodes 105.

[0024] As part of forming a full-color image pixel, image pixel 102 may include a plurality of sub-pixel color filters 110 each having a respective spectral photoresponse. The term "spectral photoresponse" describes the portion of the electromagnetic spectrum that a color filter (e.g., sub-pixel color filters 110-G, 110-B, 110-R, and the like) transmits. For example, a spectral photoresponse corresponding to the color green (G) indicates that the color filter will transmit a portion of the electromagnetic spectrum corresponding to green light, while substantially absorbing or reflecting other portions of the electromagnetic spectrum that do not correspond to green light. Similarly, a spectral photoresponse corresponding to full color or broadband indicates that the color filter substantially transmits a portion of the electromagnetic spectrum corresponding to the visible spectrum of light, while substantially absorbing or reflecting regions of the electromagnetic spectrum outside the visible range (e.g., UV, IR, etc., if the photodiode has a spectral response outside the visible range). In some embodiments, the spectral photoresponses of the blue, green, red, and broadband sub-pixel color filters correspond to frequency ranges within the electromagnetic spectrum of approximately 450 nm to 490 nm, 520 nm to 560 nm, 635 nm to 700 nm, and 400 nm to 700 nm, respectively. In some embodiments, the plurality of sub-pixel color filters 110 included in the plurality of sub-pixels 103 may individually or in combination have a spectral photoresponse corresponding to any one of red, green, blue, panchromatic (e.g., clear or white), yellow, cyan, magenta, or other colors.

[0025] It should be understood that image sensor 100 can be fabricated using semiconductor device processing and CMOS-compatible microfabrication techniques known to those skilled in the art. In one embodiment, fabrication of image sensor 100 may include: providing a semiconductor material (e.g., a silicon wafer having a front side and a back side); forming a mask or template (e.g., made of cured photoresist) on the front side of the semiconductor material by photolithography to provide a plurality of exposed regions on the front side of the semiconductor material; doping (e.g., by ion implantation, chemical vapor deposition, physical vapor deposition, and the like) the exposed portions of the semiconductor material to form a plurality of photodiodes 105 extending from the front side of the semiconductor material into the semiconductor material; removing the mask or template (e.g., by dissolving the cured photoresist with a solvent); and planarizing (e.g., by chemical mechanical planarization or polishing) the front side of the semiconductor material.

[0026] In the same or another embodiment, photolithography can similarly be used to form a plurality of sub-pixel color filters 110 (e.g., via a cured colored polymer having a desired spectral photoresponse), a plurality of shared microlenses 115, a plurality of unshared microlenses 117 (e.g., polymer-based microlenses having a target shape and size formed from a master mold or template), and a plurality of polarizing filters (e.g., a metal grid or wire structured or otherwise configured to linearly polarize light to a predetermined degree). It should be understood that the described techniques are merely exemplary and not exhaustive, and that other techniques may be utilized to fabricate one or more components of image sensor 100.

[0027] As described in more detail with reference to FIG3 , embodiments of the present invention relate to an improved polarization pixel 104 configuration that reduces manufacturing process complexity. In the context of the semiconductor manufacturing techniques described above, a conventional polarization pixel 104 configuration may include four different polarization filters formed by photolithographic patterning of a metal film or other medium to define four different polarizers, as described in more detail with reference to FIG2A . Therefore, conventional polarization configurations involve four separate photolithographic operations to define the four different polarization filter patterns. In contrast, a polarization filter may include four substantially identical polarizers 205 (see FIG2B ) that are rotated relative to one another. These can be patterned in a single photolithographic operation, for example, involving rotating a semiconductor substrate support platen relative to an optical photolithographic source (e.g., a patterning laser).

[0028] FIG1B illustrates a cross-sectional view 100-AA′ along line AA′ of image sensor 100 shown in FIG1A , according to an embodiment of the present invention. Specifically, cross-sectional view 100-AA′ is along a row of image sensor 100 , which includes a plurality of sub-pixels (e.g., green sub-pixel 103-G and blue sub-pixel 103-B) but does not include any polarized pixels 104 as shown in FIG1A .

[0029] Referring back to FIG. 1B , image sensor 100 includes a plurality of shared microlenses 115, a plurality of sub-pixel color filters 110, and a plurality of photodiodes 105 formed in respective portions of semiconductor material 101 (e.g., silicon). Sub-pixel color filters 110 (e.g., B represents a blue spectrum light response, G represents a green spectrum light response, and the like) are optically aligned with a group of photodiodes (e.g., B2 and B4) and a shared microlens 115 to form a corresponding sub-pixel included in a plurality of sub-pixels (e.g., 103-G, 103-B, and the like). Thus, a sub-pixel color filter 110 is positioned overlying a corresponding group of photodiodes. Similarly, a shared microlens 115 is optically aligned with a group of photodiodes and a sub-pixel color filter 110 or otherwise positioned overlying the photodiodes and sub-pixel color filter 110.

[0030] Portions of semiconductor material 101 (e.g., 101-2, 101-4, and the like) can be defined to correspond to individual photodiodes 105. It should be understood that individual photodiodes can extend laterally across a portion of the entire cross-sectional area of ​​their respective portions of semiconductor material 101. In some embodiments, a gap is defined between adjacent photodiodes 105 (e.g., the region of semiconductor material 101 between photodiodes B2 and B4) so ​​that additional structures can be formed in the gap to improve image sensor performance (e.g., isolation trenches, floating diffusion regions, and the like).

[0031] FIG1C illustrates a cross-sectional view 100-BB′ along line BB′ of image sensor 100 shown in FIG1A , according to an embodiment of the present invention. Specifically, cross-sectional view 100-BB′ is along a row of image sensor 100 , which includes a plurality of sub-pixels 103 (e.g., red sub-pixels 103-R and the like) and a plurality of polarized pixels 104 (e.g., 104-P1, 104-P3, and the like), as shown in FIG1A .

[0032] Referring back to FIG. 1C , image sensor 100 includes a plurality of shared microlenses 115 , a plurality of unshared microlenses 117 , a plurality of sub-pixel color filters 110 (e.g., R indicates a red spectrum light response, and the like), a plurality of polarization filters 107 , a plurality of photodiodes 105 disposed in respective portions of semiconductor material 101 (e.g., 101 - 1 , 101 - 3 , and the like), and an optional deep trench isolation structure 109 .

[0033] In the illustrated embodiment, a given polarization pixel (e.g., first polarization pixel 104-P1) of an imaging pixel of image sensor 100 includes a first photodiode (e.g., P1) disposed in a respective portion (e.g., 101-1) of semiconductor material 101, a non-shared microlens (e.g., 117-1) optically aligned with and / or disposed overlying the first photodiode, and a polarization filter (e.g., 107-1) disposed between the first photodiode and the non-shared microlens. In some embodiments, polarization filter 107 is configured as a linear polarizer having a patterned linear polarizer to achieve a polarization of light incident on the polarization filter. Polarization filter 107 can be formed, for example, by patterning a metal film disposed overlying semiconductor material 101 between the formation of the photodiode and the placement of color filter 110 and / or microlenses 115 and / or 117. As described in more detail with reference to FIG2B , the separation distance 210 and orientation angle θ of a given polarizing filter 107 can be configured to impart a linear polarization to incident light. In conventional polarizers, the polarizer angle θ can be 0°, 45°, 90°, or 135°, as described in more detail with reference to FIG2A . In contrast, in some embodiments of the present invention, the polarizer angle θ can be defined as approximately 22.5° or approximately -22.5° relative to a common axis defined about line BB', with multiples of approximately 45° being added for subsequent polarized pixels 104, as described in more detail with reference to FIG2B .

[0034] In some embodiments, for a given image pixel, polarized pixels 104 are grouped together (e.g., 104-P1 and 104-P3) in a 2×2 pattern (e.g., as shown in FIG. 1A ). A group of polarized pixels 104 can provide varying degrees of linear polarization to incident light. For example, a group of four polarized pixels arranged in a 2×2 pattern (e.g., P1, P2, P3, and P4 shown in FIG. 1A and partially shown in FIG. 1C ) can include polarizer angles that impart linear polarizations of approximately 22.5°, approximately 67.5°, approximately 112.5°, and approximately 157.5°, respectively.

[0035] As shown in FIG1C , polarizing filter 107 is positioned between sub-pixel color filter 110 and photodiode 105. In some embodiments, sub-pixel color filter 110 (e.g., 110-1) optically aligned with polarizing pixel 107 may exhibit a spectral response corresponding to green or broadband light. In some embodiments, polarizing pixel 104 may omit a sub-pixel color filter 110. In some embodiments, polarizing filters 107 have a negligible thickness, such that the thickness of sub-pixel color filters 110 positioned above polarizing pixel 104 and sub-pixels 103 is substantially equal. Sub-pixel color filter 110 may be planarized (e.g., by chemical mechanical polishing or other planarization techniques) such that the microlenses (e.g., shared microlenses 115 and unshared microlenses 117) are positioned on a substantially flat surface within the tolerances of image sensor applications.

[0036] In the illustrated embodiment, image sensor 100 includes a deep trench isolation structure 109 (e.g., formed from an oxide (e.g., silicon oxide) within semiconductor material 101) disposed between adjacent photodiodes (e.g., P1 and P3) within a plurality of photodiodes 105 to reduce electrical crosstalk between the photodiodes. As illustrated, deep trench isolation structure 109 extends to a depth within semiconductor material 101 exceeding the depth of the plurality of photodiodes 105. Deep trench isolation structure 109 may be omitted in favor of a shallow trench isolation structure extending within semiconductor material 101 to a portion of the depth of the plurality of photodiodes 105. In some embodiments, isolation structure 109 is omitted for at least a portion of photodiodes 105.

[0037] FIG1D illustrates a relative size comparison between first subpixel color filter 110, polarizing filter 107, shared microlens 115, and unshared microlens 117 included in an imaging pixel of image sensor 100 shown in FIG1A through FIG1C , according to an embodiment of the present invention. Subpixel 103 includes a group of photodiodes 105-1 arranged in a 2×2 pattern optically aligned with subpixel color filter 110. Polarizing pixel 104 includes photodiode 105-2 optically aligned with polarizing filter 107. As shown, shared microlens 115 has a first area, while unshared microlens 117 has an area smaller than the first area of ​​shared microlens 115. Subpixel color filter 110 has a third area larger than the first area of ​​shared microlens 115. Polarizing filter 107 has a fourth area larger than the second area of ​​unshared microlens 117. The third area of ​​sub-pixel color filter 110 is larger than the fourth area of ​​unshared microlens 117. In some embodiments, the relative size comparison between sub-pixel color filter 110, polarization filter 107, shared microlens 115, and unshared microlens 117 of a given sub-pixel 103 and polarization pixel 104 may be representative of each and every sub-pixel and / or polarization pixel included in image sensor 100 depicted in FIG. 1A through FIG. 1C .

[0038] FIG2A is a schematic diagram illustrating an asymmetric polarization filter used in the prior art. The illustrated asymmetric polarization filter is labeled "Prior Art" to indicate that the configuration shown is typical of conventional configurations for polarized pixels 104. As described with reference to FIG1A through FIG1D , the asymmetric polarization filter may be or include a metal layer that has been patterned, for example, by selective deposition and etching of a metal film, to form four distinct polarizers. The polarizers in FIG2A are defined by the quadrants bounded by axes BB' and CC', such that polarizer BC is seen in the upper left corner, polarizer BC' is seen in the upper right corner, polarizer B'C' is seen in the lower right corner, and polarizer B'C is seen in the lower left corner. The polarizer angles Θ shown in FIG2A are: for polarizer BC, Θ=0°; for polarizer BC', Θ=45°; for polarizer B'C', Θ=90°; for polarizer B'C, Θ=135°, where Θ is defined as an angle relative to BB'.

[0039] The axes D-D', E-E', F-F', and G-G', which are parallel to the centerline of BB' or CC' in each polarizer, are superimposed on an asymmetric polarizing filter. For example, E-E' is parallel to BB' and positioned to serve as the horizontal centerline of polarizers BC and B'C. Similarly, G-G' is parallel to BB' and positioned to serve as the horizontal centerline of polarizers BC' and B'C'. In contrast, D-D' is parallel to CC' and positioned to serve as the vertical centerline of polarizers BC and BC'. Similarly, F-F' is parallel to CC' and positioned to serve as the vertical centerline of polarizers B'C' and B'C.

[0040] In Figures 2A and 2B, the polarizers are shown as shaded areas defining a negative blank space aligned with the apertures. In contrast to positive structures such as polarization gratings, metal film polarizers transmit light through apertures defined by selectively depositing and / or removing metal material according to a pattern. As shown in Figure 2A, conventional asymmetric polarization filters are patterned with an opaque metal film material overlying the central region of each polarizer. For example, the central region of polarizer BC is indicated by the intersection of axes D-D' and EE' of the overlying metal film material, rather than by an aperture. The same is true for polarizers BC', B'C', and B'C. While not limited to a single physical structure for photodiode operation, the quantum efficiency of photocurrent generation (e.g., the conversion fraction of incident photons to electrons in a doped semiconductor material) is typically highest in the central region. Thus, blocking incident light in and around the central region of a photodiode reduces quantum efficiency and impairs the performance of the polarized pixel.

[0041] Furthermore, the number of apertures is defined by the spacing between them, so the number of apertures varies depending on the polarizer angle θ and the size and shape of the polarizer. Therefore, it can be seen that the number of apertures varies between polarizers in an asymmetric polarization filter. For example, the asymmetric polarization filter defines six apertures for polarizers BC and B'C', and eight apertures for polarizers BC' and B'C. Similarly, the width of the apertures is at least partially constrained by the polarization efficiency. Therefore, the configuration of the asymmetric polarization filter in FIG2A results in polarizers having different numbers of apertures and different transmission areas (referring to the area of ​​the aggregated apertures). Thus, even for completely unpolarized incident light, the transmission of the incident light through the polarizers of the asymmetric polarization filter differs between polarizers BC / B'C' and polarizers B'C / BC'. In this context, transmission refers to the fraction of incident light that is transmitted through the polarizer relative to the total light incident on the polarizer.

[0042] It is understood that non-uniform transmission between the polarizers that make up a polarization filter can introduce errors in PDAF or other signals generated using polarized pixels. For example, the extinction ratio (ER) is an important figure of merit used to characterize linear polarizers. ER is defined as the ratio of the transmittance of incident polarized light that is substantially aligned with the polarization angle of a given polarizer to the transmittance of incident polarized light that is substantially orthogonal to the polarization angle of the given polarizer. The mathematical formula for the extinction ratio is: ,in is the extinction ratio, T1 is the maximum transmittance of the polarizer, which occurs when the polarizer axis is parallel to the polarization plane of the incident polarized light beam. T2 is the minimum transmittance of the polarizer, which occurs when the polarizer axis is perpendicular to the polarization plane of the incident polarized light beam.

[0043] The extinction ratios described above are defined for a single polarizer, whose transmittance T is measured using two differently polarized light sources. In some cases, the extinction ratio can also be measured for a single light source and two polarizers. In these cases, the extinction ratio depends on the polarization angle difference between the two polarizers. This polarization angle difference can be used, for example, to determine the selectivity of a given polarizer for incident light. In the context of the asymmetric polarization filter of Figure 2A, the ER of the different polarizers comprising the asymmetric polarization filter also depends at least in part on the area fraction of the apertures, such that different polarizers will exhibit different ER values ​​due, at least in part, to the different number, relative size, and spacing of the apertures. This difference introduces errors in PDAF measurements and other processing of the signals generated by polarized pixels 104. This error is typically addressed through sensor calibration or software correction, which increases the computational resource requirements of the autofocus process and can introduce delays that impair autofocus performance.

[0044] FIG2B illustrates a schematic diagram of an example symmetric polarizing filter 200 according to an embodiment of the present invention. Example symmetric polarizing filter 200 includes a plurality of polarizers 205, which can be positioned overlying photodiodes 105 of polarized pixels 104, as described in greater detail with reference to FIG1C . Polarizers 205 of example symmetric polarizing filter 200 include a plurality of apertures 215 defined in a patterned material 220 separated by a spacing 210. The combination of a photodiode 105 and a polarizer 205, and in some cases, a color filter layer 110 and / or a shared microlens 115, together define a polarized pixel structure 104.

[0045] In some embodiments, the pitch 210 can be determined at least in part based on the wavelength of the incident light, with a wider pitch being used to polarize longer wavelengths and a shorter pitch being used to polarize shorter wavelengths. Thus, the pitch 210 can be approximately 0.1 μm, approximately 0.2 μm, approximately 0.3 μm, approximately 0.4 μm, approximately 0.5 μm, approximately 0.6 μm, approximately 0.7 μm, approximately 0.8 μm, approximately 0.9 μm, approximately 1 μm, approximately 1.1 μm, approximately 1.2 μm, approximately 1.3 μm, approximately 1.4 μm, approximately 1.5 μm, approximately 1.6 μm, approximately 1.7 μm, approximately 1.8 μm, approximately 1.9 μm, approximately 2.0 μm, or greater, including fractions and interpolations thereof. The number of apertures 215 can be determined by the size of the polarization pixels 104, as previously described. For larger polarized pixels 104, more apertures 215 can be defined for a single polarizer 205. Advantageously, a larger number of apertures 215 improves the signal-to-noise ratio characteristics, but also increases the size of the sensor. In some embodiments, the polarized pixel 104 may be about 0.1 μm 2, about 0.2 μm 2, about 0.3 μm 2, about 0.4 μm 2, about 0.5 μm 2, about 0.6 μm 2, about 0.7 μm 2, about 0.8 μm 2, about 0.9 μm 2, about 1.0 μm 2, about 2 μm 2, about 3 μm 2, about 4 μm 2, about 5 μm 2, about 10 μm 2, about 15 μm 2, about 20 μm 2, about 25 μm 2, about 30 μm 2, about 35 μm 2, about 40 μm 2, about 45 μm 2, about 50 μm 2, about 55 μm 2, about 60 μm 2, about 65 μm 2, about 70 μm 2, about 75 μm 2, or about 80 μm 2. 2, about 85 μm 2, about 90 μm 2, about 95 μm 2, about 100 μm 2, about 105 μm 2, about 110 μm 2, about 115 μm 2, about 120 μm 2, about 125 μm 2, about 130 μm 2, about 135 μm 2, about 140 μm 2, about 145 μm 2, and about 150 μm 2, including fractions and interpolations thereof.Thus, the number of apertures 215 can be based at least in part on the size of the polarization pixel 104 and the wavelength of the incident light, wherein the polarizer 205 includes about 5 apertures 215, about 10 apertures 215, about 15 apertures 215, about 20 apertures 215, about 25 apertures 215, about 30 apertures 215, about 35 apertures 215, about 40 apertures 215, about 45 apertures 215, about 50 apertures 215, about 55 apertures 215, about 60 apertures 215, about 65 apertures 215, about 70 apertures 215, or about 80 apertures 215. 5. Approximately 75 pores 215, approximately 80 pores 215, approximately 85 pores 215, approximately 90 pores 215, approximately 95 pores 215, approximately 100 pores 215, approximately 105 pores 215, approximately 110 pores 215, approximately 115 pores 215, approximately 120 pores 215, approximately 125 pores 215, approximately 130 pores 215, approximately 135 pores 215, approximately 140 pores 215, approximately 145 pores 215, and approximately 150 pores 215, including fractions and interpolated values ​​thereof.

[0046] Although illustrated as including four polarizers 205, the example symmetric polarizing filter 200 may include a single polarizer 205, two polarizers 205, three polarizers 205, four polarizers 205, five polarizers 205, six polarizers 205, seven polarizers 205, eight polarizers 205, or more. A polarizer 205 of the example symmetric polarizing filter 200 is characterized by a polarization angle θ1 defined relative to a vertical axis of the polarized pixel 104. In the example symmetric polarizing filter 200, the vertical axis corresponds to axis BB', but it should be understood that the term "vertical" is used to indicate a relative alignment rather than an absolute direction.

[0047] Compared to the asymmetric polarizing filter of FIG2A , the example symmetric polarizing filter 200 exhibits pattern symmetry across the CC' axis, such that a first polarizer 205-1 and a third polarizer 205-3, which are laterally adjacent to each other, are substantially symmetrical with a second polarizer 205-2 and a fourth polarizer 205-4, respectively. The second polarizer 205-2 and the fourth polarizer 205-4 are vertically adjacent to the first polarizer 205-1 and the third polarizer 205-3, respectively. In this context, symmetry describes a mirror reflection of the pattern of apertures 215 about the CC' axis, such that the first polarizer 205-1 and the second polarizer 205-2 define the same number of apertures 215, which are aligned at substantially the same absolute angle relative to the CC' axis. Therefore, axis CC' represents a transverse axis of example-symmetric polarizing filter 200, where the transverse axis is defined as an axis of symmetry of example-symmetric polarizing filter 200 that is coplanar with the first polarizer and the second polarizer. In this context, coplanarity refers to a property of axis CC' such that a line defined by two points on CC' lies in a plane parallel to a surface of example-symmetric polarizing filter 200. Example-symmetric polarizing filter 200 can be a three-dimensional layer, and axis CC' can therefore define a plane passing through the layer. Therefore, the plane of axis CC' can be substantially orthogonal to the surface of example-symmetric polarizing filter 200.

[0048] In this context, "substantially" refers to properties of polarizer 205, spacing 210, and / or aperture 215 that are within typical manufacturing tolerances of semiconductor fabrication processes. Thus, apertures 205 that are substantially aligned along a polarization angle refer to apertures aligned within typical manufacturing tolerances of semiconductor fabrication processes. It should be understood that different semiconductor fabrication processes may exhibit different tolerances. Therefore, "substantially" may include a tolerable degree of non-uniformity. Similarly, "substantially symmetric" may include a tolerable degree of deviation from perfect symmetry, as various mechanical, optical, and chemical differences may arise between two different patterning and / or deposition / removal sequences involved in converting a polarizer design into a fabricated polarizer 205 on a semiconductor substrate. Furthermore, in this context, "absolute angle" refers to an unsigned magnitude of an angle (e.g., |θ 1|). Advantageously, each polarizer 205 can be at least partially patterned by rotating a semiconductor wafer relative to an optical photolithography source using the same pattern 310 (see FIG. 3 ). Thus, the aperture 215 in a given polarizer 205 can be described by two angles: a polarization angle θ1 and a rotated polarization angle θ2 relative to the polarization angle θ1.

[0049] In some embodiments, defining polarization angle θ1 allows for forming a set of polarizers 205 without repeating angular orientations as a means of reducing crosstalk between polarizers 205. For example, while angular orientations can technically be defined between 0 and 360 degrees, angles separated by 180 degrees behave similarly, if not identically. Thus, the range of angular orientations relative to the vertical axis BB' can be limited to a non-zero polarization angle θ1 to 180 degrees. Thus, polarization angle θ1 can be defined at least in part based on the number of polarizers 205 included in a polarization filter. For example, in a polarizing filter comprising four component polarizers, as illustrated in FIG2B , the polarization angle θ1 can have a magnitude of approximately 45 degrees or less, approximately 40 degrees or less, approximately 35 degrees or less, approximately 30 degrees or less, approximately 25 degrees or less, approximately 20 degrees or less, approximately 15 degrees or less, approximately 10 degrees or less, approximately 5 degrees or less, approximately 1 degree or less, greater than 0, and fractions and interpolations thereof. In an illustrative example, the polarization angle θ1 can be approximately 22.5° or approximately −22.5° relative to the BB′ axis. In the illustrated example, θ1 can be defined such that a positive angle is defined as a clockwise rotation relative to the BB′ axis. Thus, a negative angle can be defined as a counterclockwise rotation relative to the BB′ axis.

[0050] A rotated polarization angle θ2 can then be defined such that the third polarizer 205-3 is selective with respect to the first polarizer 205-1 and such that the two polarizers 205 forming a polarization filter are not substantially aligned. Thus, the rotated polarization angle θ2 can be based at least in part on the number of polarizers 205 included in the polarization filter. In an example polarizing filter 200 including four polarizers 205 (e.g., in a 2×2 grouping of two columns and two rows), the rotated polarization angle θ2 can be about 90 degrees or less, about 85 degrees or less, about 80 degrees or less, about 75 degrees or less, about 70 degrees or less, about 65 degrees or less, about 60 degrees or less, about 55 degrees or less, about 50 degrees or less, about 45 degrees or less, about 40 degrees or less, about 35 degrees or less, about 30 degrees or less, about 25 degrees or less, about 20 degrees or less, about 15 degrees or less, about 10 degrees or less, about 5 degrees or less, about 1 degree or less, or less, including fractions and interpolations thereof. In an illustrative example, the rotated polarization angle θ2 can be about 45° or about −45° relative to a positive clockwise rotation from the BB′ axis. As shown, the polarization angle θ2 can be rotated relative to the polarization angle θ1 so that the total angular orientation of the aperture 215 of a given polarizer 205 is the sum of θ1 and θ2. For example, in a symmetrical polarizing filter 200, the first polarizer 205 is oriented with a θ1 value of approximately 22.5 degrees and a θ2 value of approximately 45 degrees, so that the angular orientation of the aperture 215 of the third polarizer 205-3 laterally adjacent to the first polarizer 205-1 is approximately 67.5 degrees.

[0051] In some embodiments, symmetry across a transverse axis (e.g., CC') reproduces a consistent rotated polarization angle θ2 for each subsequent polarizer 205 included in a given polarization filter. In the example symmetric polarization filter 200, the fourth polarizer 205-4 is rotated approximately 90 degrees relative to polarization angle θ1, based at least in part on a reflection of the angular rotation of the third polarizer 205-3 across transverse axis CC'. Similarly, the second polarizer 205-2, which is vertically adjacent to the first polarizer 205-1 and laterally adjacent to the fourth polarizer 205-4, is rotated approximately 135 degrees relative to θ1 and defines an angular orientation of approximately 157.5 degrees relative to the BB' axis.

[0052] For values ​​of Θ1 and Θ2, this symmetry condition is found where 2Θ1 = Θ2. Thus, in the case of a symmetrical polarization filter comprising more than six polarizers 205 (e.g., in a 2×3 grouping of two columns and three rows), the value of Θ1 can be approximately 15 degrees and the value of Θ2 can be approximately 30 degrees, such that an upper row of three polarizers 205 can be described by angular orientations of 15 degrees, 45 degrees, and 75 degrees, and a lower row reflected about axis C-C' can be described by angular rotations of 105 degrees, 135 degrees, and 165 degrees. In the exemplary six-polarizer configuration, no two polarizers are described by the same angular orientation and are separated by at least the value of Θ2. As with the example symmetric polarization filter 200, the hexapoleonic polarizer 205 configuration described above can be formed with a consistent number of apertures 215 and a consistent spacing 210, wherein each pattern 310 is defined using the same optical photolithographic pattern. Advantageously, the processing complexity of a symmetric polarization filter can be lower than that of an asymmetric polarization filter, which includes at least two different patterns and four orientations.

[0053] As previously described, the extinction ratio of a polarizer 205 describes the selectivity of the polarizer 205 in transmitting angular components of incident light aligned with the angular orientation of the polarizer 205 and attenuating angular components of incident light that are not aligned with the angular orientation of the polarizer 205. Compared to the conventional asymmetric polarizing filter of FIG. 2A , the example symmetric polarizing filter 200 of FIG. 2B includes four substantially identical polarizers 205 that have been rotated and / or reflected. As such, the extinction ratios of each component polarizer 205 of the example symmetric polarizing filter 200 can be substantially equal. Advantageously, providing an example symmetric polarizing filter 200 with substantially equal ER values ​​for each polarizer 205 improves performance and reduces errors in PDAF and other measurements compared to conventionally used asymmetric polarizing filters.

[0054] FIG3 illustrates an example process 300 for defining the constituent polarizer 200 elements of the example symmetric polarizing filter 200 of FIG2B , according to an embodiment of the present invention. Example process 300 is illustrated using exemplary angular rotations θ1 and θ2 of a pattern 310, where pattern element 311 represents the front face of an aperture 215 to be defined in an opaque filter material 220 (e.g., a metal line). The constituent operations 301 through 307 of example process 300 are described in functional terms, but it should be understood that each operation may include one or more sub-operations performed as part of the design and fabrication of a polarizer 205, as described in greater detail with reference to FIG1A through FIG1C . As one of ordinary skill in the art will appreciate, the sequence of operations 301 through 307 may be partially sequential and partially parallel, as some CMOS processes are localized to precise regions of a substrate surface, while other processes are applied uniformly across a wide region of the substrate surface.

[0055] In operation 301, aperture pattern 310 is rotated by polarization angle θ1 to define rotated pattern 315. Compared to the asymmetric polarization filter of FIG. 2A , the rotation of pattern 310 can be continuously applied to define each polarizer 205 of the symmetric polarization filter 200. In this manner, a plurality of polarization pixels 104 can be defined using pattern 310 transferred to a semiconductor substrate via optical photolithography, rather than using multiple different patterns as is known in the art. As previously described, the use of multiple different patterns introduces non-uniformity in the transmission characteristics and extinction coefficients of different polarization pixels 104, which can serve as a source of error in PDAF calculations.

[0056] In operation 303, an aperture frame 320 is applied to the rotated pattern 310 to define apertures 215 of the first polarizer 205-1. As described in more detail with reference to FIG2B, the first polarizer 205-1 includes apertures 215 separated (offset) by a spacing 210 defined in the filter material 220. Based at least in part on the fabrication process for transferring a micropattern from a design onto the semiconductor substrate of the image sensor 100, the aperture frame 320 can be a different size than the fabricated first polarizer 205-1.

[0057] In operation 305, rotated pattern 315 is further rotated by θ2 to redefine rotated pattern 315. As shown, pattern 310, rotated pattern 315, and redefined rotated pattern 325 are identical except for the angular orientation of pattern elements 311 relative to a vertical axis (e.g., aligned with BB' in FIG. 2B ). For example, pattern elements 311 of pattern 310 are aligned substantially parallel to the vertical direction, while elements 311 of rotated pattern 315 are aligned substantially parallel to θ1, and elements 311 of redefined rotated pattern 325 are substantially aligned with the angle defined by the sum of θ1 and θ2. In some embodiments, operation 307 may optionally be continued by rotating aperture frame 320 rather than rotating pattern 315. For example, aperture frame 320 may be rotated by θ2 to define third polarizer 205-3 rather than further rotating rotated pattern 315.

[0058] In some embodiments, the second polarizer 205-2 and the fourth polarizer 205-4 are defined by specular reflections of the third polarizer 205-3 and the first polarizer 205-1, respectively, across a transverse axis (e.g., axis C-C'). In some embodiments, the second polarizer 205-2 and the fourth polarizer 205-4 are defined by repeated instances of operation 305 (e.g., rotating pattern 315 by successive increments of θ). Thus, example process 300 allows the use of a single pattern 310 to define an arbitrary number of polarizers 205. In the context of conventional quadruple polarizer 205 designs, example process 300 provides improved performance characteristics by providing four polarizers 205 having substantially identical extinction coefficients and transmission values.

[0059] At operation 307, the pattern defined by aperture frame 320 is transferred to the image sensor as part of fabricating polarizer 205. As one of ordinary skill in the art will appreciate, operation 307 may include multiple sub-operations for depositing a film of opaque filter material 220, depositing a photoresist layer overlying opaque filter material 220, patterning the photoresist layer according to pattern 310, and selectively removing opaque filter material 220 to define aperture 215. In an illustrative example, opaque filter material 220 may be or include a metal that may be deposited in a uniform layer overlying photodiode 105, for example, by physical vapor deposition or other thin film deposition techniques. The opaque filter material may be patterned according to patterns 315 and 325 to define polarizer 205 by photolithographic patterning followed by a deposition / etching process.

[0060] FIG4 is a functional block diagram of an imaging system 402 according to an embodiment of the present invention. The imaging system 402 includes an image sensor 400 having polarized pixels 104 that implement a symmetrical polarizing filter as described in the exemplary embodiment of FIG2B through FIG3 . Image sensor 400 may have a structure corresponding to image sensor 100 illustrated in FIG1A through FIG1D , including the symmetrical polarizing filter 200 described in more detail with reference to FIG2B . Imaging system 402 includes: image sensor 400, which generates an electrical signal or image signal in response to incident light 470; objective lens 465, which has adjustable optical power for focusing on one or more points of interest within an external scene 403; and controller 450, which controls, among other things, the operation of image sensor 400 and objective lens 465. Image sensor 400 is one possible implementation of image sensor 100 illustrated in FIG1A through FIG1B . Image sensor 400 is a simplified schematic diagram showing a semiconductor material 401 with a plurality of photodiodes 405 disposed within respective portions of semiconductor material 401, a plurality of color filters 410, and a plurality of microlenses 415. Controller 450 includes one or more processors 452, memory 454, control circuitry 456, readout circuitry 458, and function logic 460.

[0061] Controller 450 includes logic and / or circuitry for controlling the operation of various components of imaging system 402 (e.g., during the pre-stage, post-stage, and in-situ stages of image and / or video acquisition). Controller 450 can be implemented as hardware logic (e.g., a dedicated integrated circuit, a field programmable gate array, a system-on-chip, etc.), software / firmware logic executed on a general-purpose microcontroller or microprocessor, or a combination of both hardware and software / firmware logic. In one embodiment, controller 450 includes a processor 452 coupled to a memory 454 that stores instructions for execution by controller 450 and / or one or more other components of imaging system 402. When executed, the instructions may cause imaging system 402 to perform operations associated with various functional modules, logic blocks, or circuits of imaging system 402, including any one or a combination of control circuitry 456, readout circuitry 458, function logic 460, image sensor 400, objective lens 465, and any other components of imaging system 402 (illustrated or otherwise). A memory system is a non-transitory computer-readable medium that may include, but is not limited to, a volatile (e.g., RAM) or non-volatile (e.g., ROM) storage system that can be read by controller 450. It should further be understood that controller 450 may be a monolithic integrated circuit, one or more discrete interconnected electrical components, or a combination thereof. Additionally, in some embodiments, one or more electrical components may be coupled together to collectively function as controller 450 to coordinate the operations of imaging system 502.

[0062] Control circuitry 456 can control the operating characteristics of photodiode array 405 (e.g., exposure duration, when to capture digital images or video, and the like). Readout circuitry 458 reads or otherwise samples analog signals from individual photodiodes (e.g., reading out electrical signals generated by each of the plurality of photodiodes 405 in response to incident light to generate image signals used to capture an image frame, and the like) and can include amplifier circuitry, analog-to-digital (ADC) circuitry, image buffers, and the like. In the illustrated embodiment, readout circuitry 458 is included in controller 450, but in other embodiments, readout circuitry 458 may be separate from controller 450. Function logic 460 is coupled to readout circuitry 458 to receive image data, demosaic the image data, and generate one or more image frames. In some embodiments, the electronic signals and / or image data may be manipulated or otherwise processed by function logic 460 (e.g., applying post-image effects such as cropping, rotating, removing red eye, adjusting brightness, adjusting contrast, etc.).

[0063] In the context of phase detection autofocus (PDAF) operation, PDAF describes a technique for generating an autofocus control signal based at least in part on a phase mismatch between two image signals generated by an image sensor. The magnitude and polarity of the phase mismatch can be used to generate a control signal to adjust the focus of a lens relative to the image sensor. Phase mismatch refers to a difference signal between two or more virtual images, such as those formed by underfocus or overfocus, which varies depending on the focal length of an objective lens relative to a sensor surface. The format and magnitude of the control signal are based at least in part on the hardware settings of imaging system 402 and the type of servo system and / or motor included to drive objective lens 465 or other lenses. In an illustrative example, if a phase mismatch signal indicates that the image is out of focus, a PDAF signal can be generated to bring the focus on image sensor 400 and / or generate visual data to indicate the out-of-focus condition.

[0064] A phase mismatch signal can be determined from the difference in photocurrent signals generated by pairs of relatively aligned polarized pixels. For example, a LR difference signal, a UD difference signal, or a LR / UD difference-sum signal combining both LR and UD can be used to determine phase mismatch and enhance signal strength, as a method for improving accuracy and reducing oversensitivity in the PDAF autofocus process. In some embodiments, the LR or UD difference signal is generated by summing the signals from photodiodes on each side of a directional boundary. For example, an LR difference signal can be generated by summing the signals from photodiodes comprising the first polarizer 205-1 and the fourth polarizer 205-4, and summing the signals from photodiodes comprising the second polarizer 205-2 and the third polarizer 205-3, and then calculating the difference between the two summed signals.

[0065] Using either the LR or UD difference signals allows for faster generation of the phase mismatch signal at the expense of accuracy and sensitivity. In some embodiments, a first group of polarized pixels 104 is configured to generate the LR difference signal, and a second group of polarized pixels 104 is configured to generate the UD difference signal. This allows both the LR and UD difference signals to be generated in the same sensor cycle rather than multiple sensor cycles, enabling the generation of an LR / UD difference-sum signal (e.g., using adjacent pixel structures).

[0066] Reference throughout this specification to "one example" or "one embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, the appearances of the phrases "in one example" or "one embodiment" in various places throughout this specification are not necessarily referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0067] For ease of description, spatially relative terms such as "below," "beneath," "above," "beneath," "above," "up," "top," "bottom," "left," "right," "center," "middle," and the like may be used herein to facilitate describing the relationship of one element or feature to another element or feature, as depicted in the figures. It should be understood that the spatially relative terms are intended to encompass different orientations of the device in use or operation in addition to the orientation depicted in the figures. For example, if the device in the figures is rotated or turned over, an element described as "below," "beneath," or "beneath" another element or feature would then be oriented "above" the other element or feature. Thus, the exemplary terms "below" or "beneath" would encompass both above and below orientations. The device can be otherwise oriented (rotated 90 degrees or at other orientations), and the spatially relative descriptors used herein should be interpreted accordingly. It should also be understood that when an element is referred to as being "between" two other elements, it can be the only element between the two other elements, or one or more intervening elements may also be present.

[0068] Throughout this specification, several technical terms are used. These terms have their ordinary meanings in the fields from which they originate, unless explicitly defined herein or the context of their use clearly indicates otherwise. It should be noted that component names and symbols may be used interchangeably throughout this document (e.g., Si versus silicon); however, both have the same meaning.

[0069] The procedures explained above can be implemented using software and / or hardware. The described techniques may comprise machine-executable instructions embodied in a tangible or non-transitory machine-readable (e.g., computer) storage medium. When executed by a machine (e.g., controller 450 in FIG. 4 ), the machine-executable instructions cause the machine to perform the described operations. Alternatively, the procedures may be embodied in hardware, such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or the like.

[0070] A tangible machine-readable storage medium includes any mechanism that provides (i.e., stores) information in a non-transitory form that can be accessed by a machine (e.g., a computer, a network device, a personal digital assistant, a manufacturing tool, or any device with one or more processors). For example, a machine-readable storage medium includes recordable and non-recordable media (e.g., read-only memory (ROM), random-access memory (RAM), magnetic disk storage media, optical storage media, flash memory devices, etc.).

[0071] The above description of the illustrated examples of the present invention, including what is described in the Abstract, is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Although specific examples of the present invention are described herein for illustrative purposes, those skilled in the art will recognize that various modifications are possible within the scope of the invention.

[0072] These modifications may be made to the embodiments of the present invention in light of the foregoing detailed description. The terms used in the accompanying claims should not be construed to limit the invention to the specific embodiments disclosed in the specification. Rather, the scope of the invention is determined entirely by the accompanying claims, which should be construed in accordance with applicable rules of claim interpretation.

[0073] 100: Image sensor 100-A: Top view 100-AA': Cross-section 100-BB': Cross-section 101: Semiconductor Materials 101-1: Part 101-2: Part 101-3: Part 101-4: Part 102: Image pixels 103-R: Red sub-pixel 103-B: Blue sub-pixel 103-G: Green sub-pixel 104: Polarized Pixel 104-P1: first polarization pixel 104-P3: Polarized pixel 105: Photodiode 105-1: Photodiode 105-2: Photodiode 107: Polarization filter 107-1: Polarization filter 109: Deep trench isolation structure 110: Sub-pixel color filter 110-1: Sub-pixel color filter 110-R: Sub-pixel color filter 110-B: Sub-pixel color filter 110-G: Sub-pixel color filter 115: Shared microlens 117: Non-shared microlens 117-1: Non-shared microlens 200: Symmetrical polarization filter 205-1: First polarizer 205-3: The third polarizer 205-2: Second polarizer 205-4: Fourth Polarizer 210: Separation distance / spacing 215: Porosity 220: Patterned material / opaque filter material 300:Procedure 301: Operation 303: Operation 305: Operation 307: Operation 310: Pattern 311: Pattern elements 315: Rotating Pattern 320: Pore frame 325: Rotating pattern 400: Image sensor 401: Semiconductor Materials 402: Imaging System 403: External scene 405: Photodiode 410: Color filter 415: Microlens 450: Controller 452:Processor 454:Memory 456: Control circuit 458: Readout circuit 460: Functional Logic 465:Objective lens C1,C2,C3,…,CX: rows R1, R2, R3, …, RY: columns

Claims

1. A symmetrical polarized filter, comprising: A first polarizer, which defines a first plurality of pores; And a second polarizer adjacent to the first polarizer, the second polarizer defining a second plurality of apertures, wherein the first plurality of apertures and the second plurality of apertures are mirror-symmetrical about a transverse axis of a symmetrical polarization filter between the first polarizer and the second polarizer, wherein the transverse axis is defined as an axis of symmetry of the symmetrical polarization filter, wherein the axis of symmetry is: coplanar with the first polarizer and the second polarizer, and configured to pass through a center point of the entire symmetrical polarization filter, wherein the center point is configured at an equidistant point between the opposite edges of the symmetrical polarization filter, and wherein the transverse axis is the axis of symmetry shared by the mirror symmetry across the axis of symmetry of the symmetrical polarization filter between the plurality of apertures of the transversely adjacent polarizers of all polarizers in the filter.

2. A symmetrical polarization filter as claimed in claim 1, wherein two of the first plurality of apertures are offset by a gap and substantially aligned along a non-zero polarization angle relative to a vertical axis orthogonal to the horizontal axis.

3. A polarization filter symmetrical to claim 2, wherein the polarization angle is rotated approximately 22.5 degrees clockwise relative to one of the vertical axes.

4. A symmetrical polarization filter as claimed in claim 2, further comprising: A third polarizer is laterally adjacent to the first polarizer and defines a third plurality of apertures, which are offset by the spacing and substantially aligned at an angle different from the polarization angle.

5. A symmetrical polarization filter as claimed in claim 4, wherein the angular orientation includes a rotational polarization angle of about 45 degrees or about -45 degrees relative to the polarization angle.

6. The symmetrical polarization filter of claim 4, further comprising: A fourth polarizer is diagonally adjacent to the first polarizer, vertically adjacent to the third polarizer, and laterally adjacent to the second polarizer. The fourth polarizer defines a fourth plurality of apertures, which are separated by the spacing and are mirror-symmetrical about the horizontal axis to the third plurality of apertures of the third polarizer.

7. A symmetrical polarized filter as described in claim 1, wherein: The first polarizer is characterized by a first extinction ratio; the second polarizer is characterized by a second extinction ratio; and the first extinction ratio and the second extinction ratio are substantially equal.

8. A symmetrical polarized filter as described in claim 1, wherein: The first polarizer is disposed over a first photodiode of an image sensor; the second polarizer is disposed over a second photodiode of the image sensor; the first plurality of apertures cover a first portion of the first photodiode; the second plurality of apertures cover a second portion of the second photodiode; and the first portion and the second portion are substantially equal in area.

9. A symmetrical polarization filter as claimed in claim 1, wherein one of the first plurality of pores is overlaid on a first center of one of the first polarizers.

10. An imaging system comprising: An image sensor includes a plurality of photodiodes configured as a photodiode array, the plurality of photodiodes being disposed on a substrate to form a plurality of polarized pixels. The polarized pixels include: a symmetrical polarization filter comprising: a first polarizer defining a first plurality of apertures; and a second polarizer adjacent to the first polarizer, the second polarizer defining a second plurality of apertures, wherein the first plurality of apertures are mirror-symmetrical with respect to a horizontal axis of the symmetrical polarization filter between the first and second polarizers, wherein the horizontal axis is defined as an axis of symmetry of the symmetrical polarization filter, wherein the axis of symmetry is coplanar with the first and second polarizers and is configured to pass through a central point of the entire symmetrical polarization filter, wherein the central point is configured at an equidistant point between opposite edges of the symmetrical polarization filter. Furthermore, the horizontal axis is the plurality of pores between the lateral adjacent polarizers of all polarizers in the filter, and the axis of symmetry of the symmetrical polarization filter is the axis of symmetry shared by the mirror symmetry.

11. The imaging system of claim 10, wherein two of the first plurality of apertures are offset by a gap and substantially aligned along a non-zero polarization angle relative to a vertical axis of the symmetrical polarization filter, the vertical axis being orthogonal to the horizontal axis.

12. The imaging system of claim 11, wherein the polarization angle is rotated approximately 22.5 degrees clockwise relative to one of the vertical axes.

13. The imaging system of claim 11, further comprising: A third polarizer is laterally adjacent to the first polarizer and defines a third plurality of apertures, which are offset by the spacing and substantially aligned at an angle different from the polarization angle.

14. The imaging system of claim 13, wherein the angular orientation includes a rotational polarization angle of about 45 degrees or about -45 degrees relative to the polarization angle.

15. The imaging system of claim 13, further comprising: A fourth polarizer is diagonally adjacent to the first polarizer, vertically adjacent to the third polarizer, and laterally adjacent to the second polarizer. The fourth polarizer defines a fourth plurality of apertures, which are separated by the spacing and are mirror-symmetrical about the horizontal axis to the third plurality of apertures of the third polarizer.

16. The imaging system of claim 10, wherein: The first polarizer is characterized by a first extinction ratio; the second polarizer is characterized by a second extinction ratio; and the first extinction ratio and the second extinction ratio are substantially equal.

17. The imaging system of claim 10, wherein: The first polarizer is disposed over one of the plurality of photodiodes; the second polarizer is disposed over one of the plurality of photodiodes; the first plurality of apertures cover a first portion of one of the first photodiodes; the second plurality of apertures cover a second portion of one of the second photodiodes; and the first portion and the second portion are substantially equal in area.

18. The imaging system of claim 10, wherein one of the first plurality of apertures is overlaid on a first center of one of the first polarizers, and wherein one of the second plurality of apertures is overlaid on a second center of one of the second polarizers.

19. The imaging system of claim 10, further comprising: A controller coupled to the image sensor and logic, when executed by the controller, causes the imaging system to perform operations including: reading out an electrical signal generated in response to incident light, the electrical signal being generated by the plurality of photodiodes.

20. The imaging system of claim 19, further comprising additional logic, which, when executed by the controller, causes the imaging system to perform further operations including: determining, at least in part, a focus state of the imaging system as part of phase-detection autofocus (PDAF) based on the electrical signal.

21. An image sensor comprising a plurality of image pixels, each image pixel comprising a plurality of sub-pixels and one or more groups of polarization pixels, wherein: The plurality of sub-pixels includes at least one blue sub-pixel, at least one green sub-pixel, and at least one red sub-pixel. Each of the plurality of sub-pixels is laterally or diagonally adjacent to one of the one or more groups of polarized pixels. Each of the one or more groups of polarized pixels includes: a first polarized pixel and a second polarized pixel, and a symmetrical polarization filter, which includes: a first polarizer defining a first plurality of apertures, the first polarizer being disposed over the first polarized pixel; and a second polarizer laterally adjacent to the first polarizer, defining a second plurality of apertures, the second polarizer being disposed over the second polarized pixel. The first plurality of apertures are mirror-symmetrical with respect to the horizontal axis of the symmetrical polarization filter between the first polarizer and the second polarizer, wherein the horizontal axis is defined as the axis of symmetry of the symmetrical polarization filter, wherein the axis of symmetry is coplanar with the first polarizer and the second polarizer, and is configured to pass through the center point of the entire symmetrical polarization filter, wherein the center point is configured at an equidistant point between the opposite edges of the symmetrical polarization filter, and wherein the horizontal axis is the axis of symmetry shared by the mirror symmetry of the symmetrical polarization filter across the axis of symmetry of the symmetrical polarization filter between the plurality of apertures of the laterally adjacent polarizers of all polarizers in the filter.

Images