Image sensing device
Patent Information
- Application Number
- CN202511526972.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2025-04-25
- Filing Date
- 2025-10-24
- Publication Date
- 2026-10-27
Smart Images

Figure CN122994637A_ABST
Abstract
Description
Technical Field
[0001] The techniques and embodiments disclosed herein generally relate to image sensing apparatus capable of generating parallax images. Background Technology
[0002] Generally, CMOS image sensing devices, implemented using complementary metal-oxide-semiconductor (CMOS) technology, capture optical images by utilizing the light-sensitive semiconductor material's properties. With the development of the automotive, medical, computer, and communications industries, the demand for high-performance image sensing devices is constantly increasing in various fields such as smartphones, digital cameras, game consoles, the Internet of Things (IoT), robotics, security cameras, and medical miniature cameras.
[0003] For devices used to photograph a target object (e.g., a camera), accurately focusing on the target object to capture a sharp image (e.g., a still image) or video (e.g., a moving image) is crucial. To improve autofocus accuracy, it is necessary to reduce the amount of parallax between parallax images. Summary of the Invention
[0004] Various embodiments of this disclosure relate to an image sensing device capable of reducing the amount of parallax between a high-sensitivity image and a low-sensitivity image in an image sensor comprising high-sensitivity pixels and low-sensitivity pixels, thereby generating a high dynamic range (HDR) image.
[0005] According to embodiments of this disclosure, an image sensing device may include: a plurality of pixel blocks, each of the plurality of pixel blocks including a plurality of pixel groups arranged in a Bayer pattern; a merging circuit configured to merge pixels included in the plurality of pixel groups; and a phase difference adjuster configured to generate a left image and a right image based on images received from the plurality of pixel blocks and the merging circuit, and configured to adjust a baseline length between the left image and the right image. Each of the plurality of pixel groups may include a first pixel having a first photosensitivity and a plurality of second pixels having a second photosensitivity. The phase difference adjuster may include a baseline adjuster configured to adjust the baseline length between first pixels in different pixel blocks located within the plurality of pixel blocks.
[0006] It should be understood that the above overview and the following detailed description of this disclosure are illustrative and descriptive, and are intended to provide a further description of the claimed embodiments of this disclosure. Attached Figure Description
[0007] The above and other features and advantages of this disclosure will become apparent from the following detailed description when considered in conjunction with the accompanying drawings.
[0008] Figure 1 This is a block diagram illustrating an imaging apparatus based on some embodiments of the present disclosure.
[0009] Figure 2 This is a diagram illustrating a case where the length of the baseline between pixels (i.e., the baseline length) is not adjusted according to the arrangement pattern of pixels in a pixel group, based on some embodiments of this disclosure.
[0010] Figure 3 This illustrates some embodiments based on this disclosure. Figure 1 The block diagram shown is of an image sensing device.
[0011] Figure 4 These are pixel blocks illustrating some embodiments based on this disclosure. Figure 3 The diagram shows the structure of a continuously arranged pixel array.
[0012] Figure 5 This illustrates some embodiments based on this disclosure. Figure 4 A diagram showing the structure of a pixel sub-block in the pixel array.
[0013] Figures 6A to 6D This illustrates some embodiments based on this disclosure. Figure 4 and Figure 5 A diagram showing a structure in which pixel blocks are arranged continuously.
[0014] Figures 7A to 7D This is an illustration of some embodiments based on the present disclosure for description. Figure 3 A diagram showing the operation of offsetting pixel groups in the offset setting circuit.
[0015] Figure 8 This illustrates some embodiments based on this disclosure. Figure 3 A diagram illustrating the operation of the phase difference adjuster.
[0016] Figures 9A to 9C This illustrates some embodiments based on this disclosure. Figure 3 A diagram illustrating the operation of the sign regulator. Detailed Implementation
[0017] Embodiments of this disclosure provide an image sensing apparatus capable of generating parallax images that can be used in a configuration to substantially solve one or more technical or engineering problems and mitigate limitations or drawbacks encountered in some imaging devices in the art. Some embodiments of this disclosure relate to an image sensing apparatus capable of reducing the amount of parallax between a high-sensitivity image and a low-sensitivity image in an image sensor comprising high-sensitivity pixels and low-sensitivity pixels, thereby generating a high dynamic range (HDR) image. Recognizing the aforementioned problems, the image sensing apparatus according to embodiments of this disclosure can improve autofocus accuracy by reducing the amount of parallax between high-sensitivity images and low-sensitivity images.
[0018] Referring now to some embodiments of the present disclosure as illustrated in the accompanying drawings. Wherever possible, the same reference numerals will be used throughout the drawings to refer to the same or similar parts. While embodiments of the present disclosure are readily adaptable to various modifications and alternatives, specific embodiments are shown in the drawings. However, these embodiments should not be construed as limiting oneself to the particular embodiments described herein.
[0019] Various embodiments will be described below with reference to the accompanying drawings. However, it should be understood that the embodiments are not limited to the specific embodiments, but include various modifications, equivalents, and / or substitutions of the embodiments. The embodiments of this disclosure can provide various advantageous effects that can be directly or indirectly recognized by those skilled in the art.
[0020] Figure 1 This is a block diagram illustrating an imaging apparatus 10 based on some embodiments of the present disclosure.
[0021] Reference Figure 1 The imaging device 10 may include an image sensing device 100, a sensing controller 200, and a parallax calculator 300.
[0022] Here, the image sensing device 100 can capture an object and convert the optical signal of the captured object into an electrical signal to generate a parallax image (PIMG). For example, the image sensing device 100 may include a plurality of pixels that capture an object and convert the optical signal obtained by capturing the object into an electrical signal (pixel signal), and an analog-to-digital converter (ADC) may be used to convert the pixel signal output from the plurality of pixels into a digital signal to generate a parallax image (PIMG).
[0023] Here, the term "parallax" (or disparity) can refer to the difference in the position or orientation of an object depending on the viewing position. A parallax image can be an image in which the calculated disparity value for each pixel in a reference image is used as the pixel value for each pixel.
[0024] Multiple pixels can include high-sensitivity pixels and low-sensitivity pixels for high dynamic range (HDR) imaging.
[0025] In this context, each of the high-sensitivity pixels can refer to a pixel that exhibits a relatively large increase in response due to an increase in incident light intensity. Here, the term "response" can refer to the pixel signal generated by the high-sensitivity pixel that detects the incident light intensity. Each of the high-sensitivity pixels can refer to a pixel that has relatively high photosensitivity to incident light. Each of the low-sensitivity pixels can refer to a pixel that exhibits a relatively small increase in response due to an increase in incident light intensity. Each of the low-sensitivity pixels can refer to a pixel that has relatively low photosensitivity to incident light.
[0026] The high-sensitivity pixels and low-sensitivity pixels can have sensing conditions that are changed based on the control signal (SCON) of the sensing controller 200. Signals generated by some of the high-sensitivity pixels or low-sensitivity pixels can be used as signals for autofocus.
[0027] For example, the image sensing device 100 can generate a parallax image (PIMG) for autofocus. The parallax image (PIMG) can include a high-sensitivity parallax image generated using signals from some of the high-sensitivity pixels, or a low-sensitivity parallax image generated using pixel signals from some of the low-sensitivity pixels. See below for further details. Figures 3 to 9C The configuration and operation of the image sensing device 100 are described in more detail.
[0028] Furthermore, the sensing controller 200 can generate a control signal (SCON) according to preset logic. The sensing controller 200 can control at least one of the exposure time, conversion gain, and analog gain of the image sensing device 100 based on the control signal (SCON). For example, the sensing controller 200 may include position information (i.e., position information within the pixel array) for preset high-sensitivity pixels and preset low-sensitivity pixels. The sensing controller 200 can control at least one of the exposure time, conversion gain, and analog gain of the high-sensitivity pixels and low-sensitivity pixels based on the position information to allow the image sensing device 100 to generate HDR (High Dynamic Range) images.
[0029] The parallax calculator 300 can use the parallax image (PIMG) received from the image sensing device 100 to calculate the horizontal parallax (i.e., first-direction parallax) or vertical parallax (i.e., second-direction parallax) of the image sensing device 100. Here, the parallax calculator 300 can calculate the horizontal or vertical parallax by comparing high-sensitivity parallax images with each other or by comparing low-sensitivity parallax images with each other.
[0030] For example, the disparity calculator 300 can calculate horizontal disparity by comparing at least two high-sensitivity disparity images (e.g., a left high-sensitivity disparity image and a right high-sensitivity disparity image) in the horizontal direction. The disparity calculator 300 can also calculate horizontal disparity by comparing at least two low-sensitivity disparity images (e.g., a left low-sensitivity disparity image and a right low-sensitivity disparity image) in the horizontal direction.
[0031] In some implementations, the parallax calculator 300 can calculate vertical parallax by comparing at least two high-sensitivity parallax images (e.g., a top high-sensitivity parallax image and a bottom high-sensitivity parallax image) in the vertical direction with each other. The parallax calculator 300 can also calculate vertical parallax by comparing at least two low-sensitivity parallax images (e.g., a top low-sensitivity parallax image and a bottom low-sensitivity parallax image) in the vertical direction with each other.
[0032] Figure 2 This diagram illustrates a case where the length of the baseline between pixels (hereinafter referred to as the baseline length) is not adjusted according to the pixel arrangement pattern within a pixel group (PG) based on some embodiments of this disclosure. Figure 2 As an example, a pixel group (PG) consisting of four green (G) pixels arranged in a (2×2) matrix structure will be described.
[0033] The image sensing device 100 may include a phase difference detection autofocus (PDAF) function that automatically focuses by detecting the phase difference between adjacent pixels. The phase difference detection autofocus method uses the phase difference between two or more measurement points at different distances to measure the offset direction and amount from the center image obtained by the image sensing device 100. Furthermore, to improve the PDAF function, the image sensing device 100 uses a structure in which multiple pixels of the same color are arranged adjacent to each other, and a microlens (ML) is applied to the multiple pixels.
[0034] For example, in a pixel group (PG), among the four green (G) pixels, "TL" can represent the pixel located in the top left position, "TR" can represent the pixel located in the top right position, "BL" can represent the pixel located in the bottom left position, and "BR" can represent the pixel located in the bottom right position. TL and TR pixels can be adjacent to each other in a first direction (i.e., the horizontal direction). BL and BR pixels can be adjacent to each other in a first direction (i.e., the horizontal direction). TL and BL pixels can be adjacent to each other in a second direction (i.e., the vertical direction). TR and BR pixels can be adjacent to each other in a second direction (i.e., the vertical direction). The four pixels (TL, TR, BL, BR) included in a pixel group (PG) can share a microlens (ML).
[0035] However, due to parallax differences caused by factors such as astigmatism of lenses, image sensing devices with structures in which a single microlens (ML) is applied to multiple pixels may struggle to perform autofocus accurately. Specifically, the parallax calculator 300 can compare high-sensitivity parallax images with each other, or it can compare low-sensitivity parallax images with each other, enabling the parallax calculator 300 to calculate horizontal or vertical parallax. However, the baseline lengths between high-sensitivity parallax images and between low-sensitivity parallax images may change, making accurate parallax calculation impossible.
[0036] For example, a left parallax image can be generated using the top-left low-sensitivity pixel (TL) located on the top-left side of a pixel group (PG), and a right parallax image can be generated using the top-right low-sensitivity pixel (TR) located on the top-right side of a pixel group (PG). Here, the length between the center of the low-sensitivity pixel (TL) and the center of the low-sensitivity pixel (TR) can be defined as "BL1".
[0037] In addition, with " At least three high-sensitivity pixels (HP1) arranged in a “┘” shape can be merged to generate a left parallax image, and at least three high-sensitivity pixels (HP2) arranged in a “┘” shape can be merged to generate a right parallax image. Here, the baseline length between the center of the high-sensitivity pixel (HP1) and the center of the high-sensitivity pixel (HP2) can be defined as “BL2”.
[0038] However, the baseline length (BL2) between high-sensitivity pixels (HP1, HP2) and the baseline length (BL1) between low-sensitivity pixels (TL, TR) can be different from each other. That is, the baseline length (BL2) between high-sensitivity pixels (HP1, HP2) can be shorter than the baseline length (BL1) between low-sensitivity pixels (TL, TR). Therefore, the amount of disparity in the high-sensitivity disparity image can be reduced, resulting in a larger amount of disparity between the high-sensitivity disparity image and the low-sensitivity disparity image, which may prevent autofocus from being performed correctly.
[0039] Therefore, embodiments of this disclosure can adjust the baseline length of the low-sensitivity image (or high-sensitivity image) to match the baseline of the high-sensitivity image (or low-sensitivity image). Thus, embodiments can improve autofocus accuracy by reducing the amount of disparity between the high-sensitivity parallax image and the low-sensitivity parallax image. Reference will be made below to... Figures 3 to 9C The detailed operation of the image sensing apparatus according to embodiments of the present disclosure is described in more detail.
[0040] Figure 3 This illustrates some embodiments based on this disclosure. Figure 1 Block diagram of the image sensing device 100 shown.
[0041] Reference Figure 3 The image sensing device 100 may include a pixel array 110, a merging circuit 120, a phase difference adjuster 130, and a parallax image generator 140. Figure 1 The components of the image sensing device 100 shown are discussed by way of example only, and embodiments of this disclosure include many other variations, substitutions, modifications, alterations, and alterations. In this disclosure, the term "pixel" may be used to refer to an image sensing pixel configured to detect incident light to generate an electrical signal carrying an image of the incident light.
[0042] Pixel array 110 may include a plurality of pixels arranged in multiple rows and multiple columns. In one embodiment, the plurality of pixels may be arranged in a two-dimensional (2D) pixel array including rows and columns. In another embodiment, the plurality of pixels may be arranged in a three-dimensional (3D) pixel array. The plurality of pixels may be grouped into pixel groups.
[0043] Multiple pixels can convert optical signals into electrical signals on a pixel-by-pixel or pixel-group basis, and can output pixel signals (PS). For example, pixel array 110 may include high-sensitivity pixels and low-sensitivity pixels. The pixel signal (PS) may include a high-sensitivity pixel signal detected by the high-sensitivity pixels and a low-sensitivity pixel signal detected by the low-sensitivity pixels. See below for further details. Figures 4 to 6D The configuration of pixel array 110 is described in more detail.
[0044] The merging circuit 120 can perform the merging of pixel signals (PS) received from the pixel array 110, convert the merged pixel signals into digital signals, and generate a merged image (BIMG). Here, the term "merging" can refer to the process of compressing data by discarding some data acquired from all pixels or by adding some data acquired from all pixels to other data when the desired image resolution can be achieved using only some of the pixels of the image sensor.
[0045] In some implementations, the merging circuit 120 can generate a merged image (BIMG) by summing the pixel signals (PS) received from the pixel array 110 and converting the summed pixel signals (PS) into a digital signal. Here, the term "summing" can refer to the process of summing (or averaging) the pixel values (or pixel signals) of pixels.
[0046] In this disclosure, for ease of description and better understanding, the merging circuit 120 will be described as performing the merging of pixel signals (PS) as an example. For example, the merging circuit 120 can generate a high-sensitivity image by merging at least three high-sensitivity pixel signals to adjust the amount of parallax. In some embodiments, to adjust the amount of parallax, the merging circuit 120 can generate a low-sensitivity image by merging at least three low-sensitivity pixel signals to generate an HDR image. Here, the high-sensitivity image and the low-sensitivity image can be included in the aforementioned merged image (BIMG). In this disclosure, the case in which the merging circuit 120 generates a merged image (BIMG) by merging at least three high-sensitivity pixel signals will be described.
[0047] Phase difference adjuster 130 can generate a phase difference image (PDIMG) by adjusting the phase difference between a high-sensitivity image and a low-sensitivity image. In this case, phase difference adjuster 130 may include offset setting circuitry 131, baseline adjuster 132, and sign adjuster 133.
[0048] The offset setting circuit 131 can set the offset differently based on the position of the pixel. For example, the offset setting circuit 131 can set the position of a low-sensitivity pixel as a reference position. The offset setting circuit 131 can also set the offset differently based on the distance of other pixels from the reference pixel position. (Referring to this below...) Figures 7A to 8 The operation of the offset setting circuit 131 is described in more detail.
[0049] Baseline adjuster 132 can adjust the baseline length of a high-sensitivity image or a low-sensitivity image to generate a phase difference image (PDIMG). For example, baseline adjuster 132 can obtain the baseline length of the left image and the baseline length of the right image. Furthermore, baseline adjuster 132 can adjust the baseline length in the direction of the image with the shorter baseline length between the left and right images. Reference will be made below. Figure 8 The detailed configuration and operation of the baseline adjuster 132 are described in more detail.
[0050] When the sign used as a reference value for disparity calculation in the array structure of the phase difference image (PDIMG) is inverted, the sign adjuster 133 can adjust the sign. (Referring to the following text...) Figures 9A to 9C The detailed configuration and operation of the sign regulator 133 are described in more detail.
[0051] The parallax image generator 140 can generate a parallax image (PIMG) by aligning the phase of a phase difference image (PDIMG). Here, the parallax image (PIMG) can include a high-sensitivity parallax image and a low-sensitivity parallax image.
[0052] For example, the parallax image generator 140 can generate at least two high-sensitivity parallax images in the horizontal direction (e.g., a left high-sensitivity parallax image and a right high-sensitivity parallax image). The parallax image generator 140 can also generate at least two low-sensitivity parallax images in the horizontal direction (i.e., a left low-sensitivity parallax image and a right low-sensitivity parallax image).
[0053] In some implementations, the parallax image generator 140 may generate at least two high-sensitivity parallax images in the vertical direction (e.g., a top high-sensitivity parallax image and a bottom high-sensitivity parallax image). The parallax image generator 140 may also generate at least two low-sensitivity parallax images in the vertical direction (e.g., a top low-sensitivity parallax image and a bottom low-sensitivity parallax image).
[0054] Figure 4 These are pixel blocks illustrating some embodiments based on this disclosure. Figure 3 The diagram shows the structure of a continuously arranged pixel array.
[0055] Reference Figure 4 The pixel array 110 may include a plurality of pixel blocks (PXB1 to PXB4). The plurality of pixel blocks (PXB1 to PXB4) may be arranged adjacent to each other in a first direction (e.g., the X-axis direction). In some embodiments, the plurality of pixel blocks (PXB1 to PXB4) may be arranged adjacent to each other in a second direction (e.g., the Y-axis direction).
[0056] Pixel blocks (PXB1 to PXB4) can be arranged adjacent to each other in an (M×N) matrix structure, where "M" and "N" are integers greater than or equal to 2 and can be the same or different from each other. For example, pixel block (PXB1) may include four pixel sub-blocks (PXSB1 to PXSB4) arranged in a (2×2) matrix structure. Pixel block (PXB2) may include four pixel sub-blocks (PXSB5 to PXSB8) arranged in a (2×2) matrix structure. Pixel block (PXB3) may include four pixel sub-blocks (PXSB9 to PXSB12) arranged in a (2×2) matrix structure. Pixel block (PXB4) may include four pixel sub-blocks (PXSB13 to PXSB16) arranged in a (2×2) matrix structure.
[0057] The following will refer to Figures 5 to 6D A more detailed description of the configuration of the pixel sub-block (PXSB) is provided.
[0058] Figure 5 This illustrates some embodiments based on this disclosure. Figure 4 A diagram showing the structure of a pixel sub-block in the pixel array.
[0059] Reference Figure 5Each pixel subblock (PXSB) may include multiple pixel groups (PGR, PGGr, PGGb, PGB) arranged adjacent to each other. Each pixel group (PGR, PGGb, PGGr, PGB) may include multiple pixels (PX) that share a microlens (ML) and have a color filter of the same color.
[0060] The pixels (PX) of each pixel group (PGR, PGGb, PGGr, PGB) can be arranged adjacent to each other in an (M×N) matrix structure, where "M" and "N" are integers greater than or equal to 2 and can be the same or different from each other. For example, a pixel group (PGR) may include four red (R) pixels (PX) arranged in a (2×2) matrix structure while sharing a microlens (ML). Each of the pixel groups (PGGr, PGGb) may include four green (Gr, Gb) pixels (PX) arranged in a (2×2) matrix structure while sharing a microlens (ML). A pixel group (PGB) may include four blue (B) pixels (PX) arranged in a (2×2) matrix structure while sharing a microlens (ML). Pixel groups (PGR, PGGr, PGGb, PGB) may be arranged in a Bayer pattern.
[0061] In addition to the aforementioned pixel groups, it is understood that the remaining pixel groups in the pixel sub-block (PXSB) are also arranged in the same pattern. Furthermore, the arrangement pattern of the pixel groups (PGR, PGR, PGRb, PGB) and the number of pixel groups included in each pixel sub-block (PXSB) are not limited to this and can be varied as needed.
[0062] Each pixel (PX) may include a photosensitive element that converts incident light into an electrical signal corresponding to the incident light. For example, the photosensitive element may be implemented as a photodiode, a phototransistor, a grating, a pinned photodiode, or a combination thereof.
[0063] Figures 6A to 6D This illustrates some embodiments based on this disclosure. Figure 4 and Figure 5 A diagram showing a structure in which pixel blocks are arranged continuously.
[0064] Reference Figures 6A to 6D Each of the pixel sub-blocks (PXSB1 to PXSB16) may include a plurality of pixel groups (PGs) arranged adjacent to each other in a first direction (e.g., the X-axis direction) and a second direction (e.g., the Y-axis direction). Figures 6A to 6D Four of the multiple pixel groups (PGs) shown can correspond to Figure 5 The pixel sub-block (PXSB) shown.
[0065] exist Figures 6A to 6DIn the implementation of this disclosure, for ease of description and better understanding, only some green (G) pixels in the pixel sub-blocks (PXSB1 to PXSB16) are numbered as pixel groups (PG1 to P16).
[0066] For example, sub-block (PXSB1) may include pixel group (PG1), sub-block (PXSB2) may include pixel group (PG3), sub-block (PXSB3) may include pixel group (PG2), and sub-block (PXSB4) may include pixel group (PG4). Sub-block (PXSB5) may include pixel group (PG5), sub-block (PXSB6) may include pixel group (PG7), sub-block (PXSB7) may include pixel group (PG6), and sub-block (PXSB8) may include pixel group (PG8).
[0067] Sub-block (PXSB9) may include pixel group (PG9), sub-block (PXSB10) may include pixel group (PG11), sub-block (PXSB11) may include pixel group (PG10), and sub-block (PXSB12) may include pixel group (PG12). Sub-block (PXSB13) may include pixel group (PG13), sub-block (PXSB14) may include pixel group (PG15), sub-block (PXSB15) may include pixel group (PG14), and sub-block (PXSB16) may include pixel group (PG16).
[0068] Although green (G) has been disclosed as an example color in the embodiments of this disclosure, the scope or spirit of this disclosure is not limited thereto, and it should be noted that the embodiments of this disclosure can also be applied to blue (B) pixels or red (R) pixels.
[0069] Multiple pixel groups (PG1 to PG16) may include high-sensitivity pixels (GH01, GH10, GH11, GH40, GH41, GH51, GH04, GH14, GH15, GH44, GH45, GH54, GH09, GH18, GH19, GH48, GH49, GH59, GH012, GH112, GH113, GH412, GH413, GH512, GH017, GH116, GH117, GH416, GH417, GH517, GH020, GH1) for creating HDR images. 20, GH121, GH420, GH421, GH520, GH025, GH124, GH125, GH424, GH425, GH525, GH028, GH128, GH129, GH428, GH429, GH528) and low-sensitivity pixels (GL00, GL50, GL05, GL55, GL08, GL58, GL013, GL513, GL016, GL516, GL021, GL521, GL024, GL524, GL029, GL529) used to create HDR images. For example, the pixels (PX) included in each pixel group (PG1 to PG16) can correspond to either high-sensitivity pixels or low-sensitivity pixels. Figures 6A to 6D Of the pixels (PX) shown, some pixels (PX) indicated by "H" may correspond to high-sensitivity pixels. Figures 6A to 6D Of the pixels (PX) shown, some pixels (PX) indicated by "L" may correspond to low-sensitivity pixels.
[0070] High-sensitivity or low-sensitivity pixels of pixel group (PG1 to PG16) can be used to generate a parallax image in a first direction (e.g., horizontal direction). High-sensitivity or low-sensitivity pixels of pixel group (PG1 to PG16) can be used to generate a parallax image in a second direction (e.g., vertical direction).
[0071] In some implementations, each of the pixel groups (PG1 to PG16) may include three high-sensitivity pixels and one low-sensitivity pixel. Within each pixel block (PXB1 to PXB4), the arrangement patterns of high-sensitivity and low-sensitivity pixels among adjacent pixel groups (PG1 to PG16) in a first direction may be symmetrical to each other in that first direction. Furthermore, the arrangement patterns of high-sensitivity and low-sensitivity pixels among adjacent pixel groups (PG1 to PG16) in a second direction may be symmetrical to each other in that second direction.
[0072] For example, the high-sensitivity pixels (GH01, GH10, GH11, GH09, GH18, GH19, GH017, GH116, GH117, GH025, GH124, GH125) in the pixel group (PG1, PG5, PG9, PG13) can have an arrangement pattern formed in the shape of a "┘". The high-sensitivity pixels (GH04, GH14, GH15, GH012, GH112, GH113, GH020, GH120, GH121, GH028, GH128, GH129) in the pixel group (PG3, PG7, PG11, PG15) adjacent to the pixel group (PG1, PG5, PG9, PG13) in the first direction can have an arrangement pattern formed in the shape of an "L".
[0073] In the second direction, the high-sensitivity pixels (GH40, GH41, GH51, GH48, GH49, GH59, GH416, GH417, GH517, GH424, GH425, GH525) in the pixel group (PG2, PG6, PG10, PG14) adjacent to the pixel group (PG1, PG5, PG9, PG13) can have a configuration of " The high-sensitivity pixels (GH44, GH45, GH54, GH412, GH413, GH512, GH420, GH421, GH520, GH428, GH429, GH528) in the pixel group (PG4, PG8, PG12, PG16) adjacent to the pixel group (PG2, PG6, PG10, PG14) in the first direction can have an arrangement pattern formed in the shape of "┌".
[0074] The pixel array 110 according to this embodiment may include pixels (PX) having different photosensitivity for HDR implementation. Pixels (PX) with the same photosensitivity may be arranged symmetrically with each other in adjacent pixel groups (PG1 to PG16) to achieve autofocus. Here, a symmetrical arrangement pattern in a first direction may mean that the center of the corresponding arrangement pattern is located on the same series of arrangement patterns in the first direction, and a symmetrical arrangement pattern in a second direction may mean that the center of the corresponding arrangement pattern is located on the same series of arrangement patterns in the second direction.
[0075] like Figures 6A to 6D The pixel groups (PG1 to PG16) arranged as shown can be formed at various positions within the pixel array 110, and the number of pixel groups (PG1 to PG16) is not limited thereto. Furthermore, in Figures 6A to 6D In this arrangement, the positions of high-sensitivity pixels and low-sensitivity pixels can be interchanged. For example, low-sensitivity pixels can be formed in... Figures 6A to 6DThe high-sensitivity pixel is located at the position shown, and the high-sensitivity pixel can be formed in... Figures 6A to 6D The low-sensitivity pixel is shown at its location. Furthermore, within the range where the arrangement patterns of the high-sensitivity pixels in the pixel group (PG1 to PG16) are symmetrical to each other in the first and second directions, the arrangement patterns of the high-sensitivity pixels in the pixel group (PG1 to PG16) can be changed to different shapes.
[0076] Figures 7A to 7D This is an illustration of some embodiments based on the present disclosure for description. Figure 3 A diagram showing the operation of offsetting pixel groups in the offset setting circuit.
[0077] Reference Figure 7A The offset setting circuit 131 can set the offset based on the position of the pixel blocks (PXB1 to PXB4).
[0078] The offset setting circuit 131 can set the offset not only using the low-sensitivity pixels (GL00) of pixel group (PG1) in pixel block (PXB1), but also using the low-sensitivity pixels (GL50) of pixel group (PG2) located closest to pixel group (PG1) in the second direction (i.e., the vertical direction). For example, the offset setting circuit 131 can set the offset by summing the pixel values of the low-sensitivity pixels (GL00) of pixel group (PG1) and the low-sensitivity pixels (GL50) of pixel group (PG2) arranged in the left column (L1) of pixel block (PXB1). Here, the term "pixel value" can refer to the aforementioned pixel signal. And the term "summing" can refer to the process of adding (or averaging) the pixel values (or pixel signals) of pixels. For example, the offset (OS) of the left column (L1) can be defined as "1L".
[0079] The offset setting circuit 131 can set the offset not only using the low-sensitivity pixels (GL05) of pixel group (PG3) in pixel block (PXB1), but also using the low-sensitivity pixels (GL55) of pixel group (PG4) located closest to pixel group (PG3) in the second direction (i.e., the vertical direction). For example, the offset setting circuit 131 can set the offset by summing the pixel values of the low-sensitivity pixels (GL05) of pixel group (PG3) and the low-sensitivity pixels (GL55) of pixel group (PG4) arranged in the right column (R2) of pixel block (PXB1). For example, the offset (OS) of the right column (R2) can be defined as "1R".
[0080] Reference Figure 7BThe offset setting circuit 131 can set the offset not only using the low-sensitivity pixels (GL08) of pixel group (PG5) in pixel block (PXB2), but also using the low-sensitivity pixels (GL58) of pixel group (PG6) located closest to pixel group (PG6) in the second direction (i.e., the vertical direction). For example, the offset setting circuit 131 can set the offset by summing the pixel values of the low-sensitivity pixels (GL08) of pixel group (PG5) and the low-sensitivity pixels (GL58) of pixel group (PG6) arranged in the left column (L3) of pixel block (PXB2). For example, the offset (OS) of the left column (L3) can be defined as "2L".
[0081] The offset setting circuit 131 can set the offset not only using the low-sensitivity pixels (GL013) of the pixel group (PG7) in the pixel block (PXB2), but also using the low-sensitivity pixels (GL513) of the pixel group (PG8) located closest to the pixel group (PG7) in the second direction (i.e., the vertical direction). For example, the offset setting circuit 131 can set the offset by summing the pixel values of the low-sensitivity pixels (GL013) of the pixel group (PG7) and the low-sensitivity pixels (GL513) of the pixel group (PG8) arranged in the right column (L4) of the pixel block (PXB2). For example, the offset (OS) of the right column (R4) can be defined as "2R".
[0082] Reference Figure 7C and Figure 7D It is understood that the offset setting circuit 131 can set the offset (OS) in the pixel block (PXB3, PXB4) in the same way as described above.
[0083] Reference Figure 7C The offset setting circuit 131 can set the offset by summing the pixel values of the low-sensitivity pixels (GL016, GL516) of the pixel group (PG9, PG10) arranged in the left column (L5) of the pixel block (PXB3). For example, the offset (OS) of the left column (L5) can be defined as "3L". The offset setting circuit 131 can also set the offset by summing the pixel values of the low-sensitivity pixels (GL021, GL521) of the pixel group (PG11, PG12) arranged in the right column (R6) of the pixel block (PXB3). For example, the offset (OS) of the right column (R6) can be defined as "3R".
[0084] Reference Figure 7DThe offset setting circuit 131 can set the offset by summing the pixel values of the low-sensitivity pixels (GL024, GL524) of the pixel group (PG13, PG14) arranged in the left column (L7) of the pixel block (PXB4). For example, the offset (OS) of the left column (L7) can be defined as "4L". The offset setting circuit 131 can also set the offset by summing the pixel values of the low-sensitivity pixels (GL029, GL529) of the pixel group (PG15, PG16) arranged in the right column (R8) of the pixel block (PXB4). For example, the offset (OS) of the right column (R8) can be defined as "4R".
[0085] As described above, the corresponding offsets xL and xR (where x is a natural number greater than 1) of the left column L (2x-1) and right column R (2x) arranged in the pixel block (PXBx) can be defined as the relative distances with respect to the reference column. That is, the left column L1 and right column R2 arranged in the leftmost pixel block (PXB1) are defined as having offsets of 1L and 1R (see [reference]). Figure 7A ).
[0086] In this disclosure, for ease of description, the position of the pixel block (PXB1) located at the leftmost position in the first direction is set as a reference position. Therefore, since the offsets of the left column L1 and right column R2 arranged in pixel block (PXB1) are set to "1L" and "1R", the offset value can be defined as "1". Furthermore, the offsets of the left column L3 and right column R4 arranged in the pixel block (PXB2) positioned most adjacent (closest) to pixel block (PXB1) are set to "2L" and "2R", so the offset value can be defined as "2". Additionally, the offsets of the left column L5 and right column R6 arranged in the pixel block (PXB3) positioned second adjacent to pixel block (PXB1) are set to "3L" and "3R", so the offset value can be defined as "3". The offsets of the left column L7 and right column R8 in the pixel block (PXB4) that is the third adjacent (relatively farthest) to the pixel block (PXB1) are set to "4L" and "4R", so the offset value can be defined as "4".
[0087] As described above, the offset value of pixel block PXBx (where x is a natural number) can be defined based on the offsets xL and xR of the left and right columns arranged in pixel block PXBx.
[0088] That is, it can be seen that as the distance between the target pixel block and the reference pixel block (PXB1) located at the reference position becomes shorter, the offset value of the corresponding pixel block is set to a smaller offset value, and as the distance between the target pixel block and the reference pixel block (PXB1) becomes longer, the offset value of the corresponding pixel block is set to a higher offset value.
[0089] Figure 8 This illustrates some embodiments based on this disclosure. Figure 3 A diagram illustrating the operation of the phase difference adjuster 130.
[0090] Referring to the above Figures 7A to 7D and Figure 8 The phase difference adjuster 130 can generate a left image using low-sensitivity pixels arranged in the left column of different pixel blocks (PXB1 to PXB4). Furthermore, the phase difference adjuster 130 can generate a right image using low-sensitivity pixels arranged in the right column of different pixel blocks (PXB1 to PXB4).
[0091] In order to adjust the phase difference according to the difference in baseline length, the phase difference adjuster 130 can blend the right image when generating the left image, or it can blend the left image when generating the right image.
[0092] The process of generating the left image via the phase difference adjuster 130 will be described in more detail below (LIC). The phase difference adjuster 130 can use not only the low-sensitivity pixels (GL00, GL50, GL016, GL516) arranged in the left column (L1, L5) of the two pixel blocks (PXB1, PXB3) arranged on both sides of the pixel block (PXB2), but also the low-sensitivity pixels (GL13, GL513) arranged in the right column (R4) of the pixel block (PKB2) arranged between the two pixel blocks (PXB1, PXB3) to generate the left image.
[0093] That is, the phase difference adjuster 130 can generate a left image using a first value (S1) obtained by summing the pixel values of low-sensitivity pixels (GL00, GL50) arranged in the left column (L1) of pixel block (PXB1), a second value (S2) obtained by summing the pixel values of low-sensitivity pixels (GL013, GL513) arranged in the right column (R4) of pixel block (PXB2), and a third value (S3) obtained by summing the pixel values of low-sensitivity pixels (GL016, GL516) arranged in the left column (L5) of pixel block (PXB3). In other words, the phase difference adjuster 130 can generate a left image by summing the first to third values (S1, S2, S3).
[0094] The process of generating the right image (RIC) by the phase difference adjuster 130 will be described in more detail below. The phase difference adjuster 130 can generate the right image not only using the low-sensitivity pixels (GL013, GL513, GL029, GL529) in the right column (R4, R8) of the two pixel blocks (PXB2, PXB4) arranged on both sides of the pixel block (PXB3), but also using the low-sensitivity pixels (GL016, GL516) arranged in the left column (L5) of the pixel block (PKB3) arranged between the two pixel blocks (PXB2, PXB4).
[0095] That is, the phase difference adjuster 130 can generate a right image by using a fourth value (S4) obtained by summing the pixel values of low-sensitivity pixels (GL013, GL513) arranged in the right column (R4) of the pixel block (PXB2), a fifth value (S5) obtained by summing the pixel values of low-sensitivity pixels (GL016, GL516) arranged in the left column (L5) of the pixel block (PXB3), and a sixth value (S6) obtained by summing the pixel values of low-sensitivity pixels (GL029, GL529) arranged in the right column (R8) of the pixel block (PXB4). In other words, the phase difference adjuster 130 can generate a right image by summing the fourth to sixth values (S4, S5, S6).
[0096] For ease of description, embodiments of this disclosure have described the generation of a left image using three values (S1 to S3) and a right image using three values (S4 to S6), but embodiments of this disclosure are not limited thereto. In another embodiment, the left image can be generated by further using other pixel groups to the left of the pixel group where the low-sensitivity pixels are located, or the right image can be generated by further using other pixel groups to the right of the pixel group where the low-sensitivity pixels are located.
[0097] During the process of generating the left image (LIC), the offset setting circuit 131 can determine the offset (OS1) using the following formula 1.
[0098] [Formula 1]
[0099] (1+2+3) / 3 = 2
[0100] Refer to formula 1 and Figures 7A to 7D The offset values "1" of pixel block (PXB1), "2" of pixel block (PXB2), and "3" of pixel block (PXB3) are summed, and the summed value "6" is divided by "3" (i.e., the number of pixel blocks), so that the offset (OS1) can become "2".
[0101] During the process of generating the right image (RIC), the offset setting circuit 131 can determine the offset (OS2) using the following formula 2.
[0102] [Equation 2]
[0103] (2+3+4) / 3 = 3
[0104] Refer to formula 2 and Figures 7A to 7D The offset values "2" of pixel block (PXB2), "3" of pixel block (PXB3), and "4" of pixel block (PXB4) are summed, and the summed value "9" is divided by "3" (i.e., the number of pixel blocks), so that the offset (OS2) can be changed to "3".
[0105] Therefore, since the offset (OS2) when generating the right image is "3" and the offset (OS1) when generating the left image is "2", the offset setting circuit 131 can set the value "1" corresponding to the difference between the offsets (OS2, OS1) to the offset value corresponding to the phase difference between the right and left images.
[0106] During the process of generating the left image (LIC), the baseline adjuster 132 can use Equation 3 below to calculate the baseline length.
[0107] [Formula 3]
[0108] (0+1+0) / 3 = 1 / 3
[0109] During the process of generating the right image (RIC), the baseline adjuster can use Equation 4 below to calculate the baseline length.
[0110] [Formula 4]
[0111] (1+0+1) / 3 = 2 / 3
[0112] As can be seen from Equations 3 and 4 above, the baseline adjuster 132 can assign the digit "0" to the left image and the digit "1" to the right image to distinguish between the left and right images. The blend of the left and right images can be their average. Then, the baseline length of the left image can become "1 / 3", and the baseline length of the right image can become "2 / 3". The right image has a longer baseline length than the left image.
[0113] The baseline adjuster 132 can adjust the baseline length in the direction of the image with the shorter baseline length between the left and right images. Therefore, the baseline adjuster 132 can adjust the baseline length of the right image to "1 / 3" in the same way as the baseline length of the left image.
[0114] Figures 9A to 9C This illustrates some embodiments based on this disclosure. Figure 3 A diagram illustrating the operation of the sign regulator.
[0115] Reference Figures 9A to 9C The description will include a pixel group (PG) comprising four green (G) pixels arranged in a (2×2) matrix structure. To expand the dynamic range of the image sensing device 100, the number of high-sensitivity pixels in the pixel group (PG) can be much greater than the number of low-sensitivity pixels. For example, the pixel group (PG) may include four pixels (TL, TR, BL, BR) arranged in a (2×2) matrix structure. The three pixels represented by "H" in the pixel group (PG) can be high-sensitivity pixels, while the pixel represented by "L" can be a low-sensitivity pixel. Figures 9A to 9C In the implementation method, the terms "and" will be omitted in this document. Figure 2 The descriptions overlap, therefore, for the sake of brevity, the redundant descriptions will be omitted.
[0116] Reference Figure 9A The length BL1 between the center of the low-sensitivity pixel (TL) in pixel group (PG11) and the center of the low-sensitivity pixel (TR) in pixel group (PG12) (i.e., the baseline length) is set to "1". The baseline direction between the low-sensitivity pixel (TL) located on the upper left side of pixel group (PG11) and the low-sensitivity pixel (TR) located on the upper right side of pixel group (PG12) is indicated by an arrow. That is, the baseline direction of the low-sensitivity pixel can be from the low-sensitivity pixel (TL) to the right of the low-sensitivity pixel (TR).
[0117] Reference Figure 9B The merging circuit 120 can merge at least three high-sensitivity pixels (HP1) arranged in a "┘" shape in pixel group (PG11). Furthermore, the merging circuit 120 can merge at least three high-sensitivity pixels (HP2) arranged in an "L" shape in pixel group (PG12). Here, the length between the center of the high-sensitivity pixel (HP1) and the center of the high-sensitivity pixel (HP2) (i.e., the baseline length BL2) is set to "-1 / 3".
[0118] The baseline direction between the high-sensitivity pixels (HP1) of pixel group (PG11) and the high-sensitivity pixels (HP2) of pixel group (PG12) is determined by... Figure 9B The arrows shown indicate that the baseline direction of the high-sensitivity pixel can be from the high-sensitivity pixel (HP1) to the left of the high-sensitivity pixel (HP2).
[0119] However, because the baseline length (BL1) between low-sensitivity pixels (TL, TR) has a positive (+) value and the baseline length (BL2) between high-sensitivity pixels (HP1, HP2) has a negative (-) value, the baseline direction (i.e., the baseline orientation) may be reversed. When the baseline sign is reversed, disparity calculation errors may occur when calculating disparity later.
[0120] Therefore, as Figure 9C As shown, the sign adjuster 133 of this disclosure can move the effective position of the high-sensitivity pixel (HP1) of pixel group (PG11) to the right, and can move the effective position of the high-sensitivity pixel (HP2) of pixel group (PG12) to the left, so as to... Figure 9B Compared to the direction of the high-sensitivity pixels (HP1, HP2), the baseline direction can be reversed. Here, moving the effective position of the pixel can mean adjusting the operator during the process of summing the pixel values of the pixels so as to reverse the baseline direction.
[0121] In this case, since the baseline between high-sensitivity pixels (HP1, HP2) also has a positive (+) value in the same way as the baseline between low-sensitivity pixels (TL, TR), parallax calculation errors can be prevented.
[0122] A phase difference image (PDIMG) generated based on the offset determined by the offset setting circuit 131, the baseline length adjusted by the baseline adjuster 132, and the baseline direction adjusted by the sign adjuster 133 can be sent to the parallax image generator 140.
[0123] As can be clearly seen from the above description, the image sensing device according to the embodiments of the present disclosure can improve autofocus accuracy by reducing the amount of parallax between high-sensitivity images and low-sensitivity images.
[0124] The embodiments disclosed herein can provide various beneficial effects that can be directly or indirectly recognized by those skilled in the art.
[0125] While several illustrative embodiments have been described, it should be understood that modifications and enhancements to the disclosed embodiments and other embodiments can be designed based on the descriptions and / or illustrations herein. Therefore, the scope of this disclosure should not be limited to the above-described embodiments, but should include their equivalents. Furthermore, embodiments can be combined to form additional embodiments.
[0126] Cross-reference to related applications
[0127] This patent application claims priority and benefit to Korean Patent Application No. 10-2025-0054886, filed on April 25, 2025, the disclosure of which is incorporated herein by reference in its entirety.
Claims
1. An image sensing device, the image sensing device comprising: Multiple pixel blocks, each of which comprises multiple groups of pixels arranged in a Bayer pattern; A merging circuit that merges pixels included in the plurality of pixel groups; as well as A phase difference adjuster generates left and right images based on images received from the plurality of pixel blocks and the merging circuit, and adjusts the baseline length between the left and right images. in, Each of the plurality of pixel groups includes a first pixel having a first photosensitivity and a plurality of second pixels having a second photosensitivity; and The phase difference adjuster includes a baseline adjuster that adjusts the baseline length between the first pixels in different pixel blocks among the plurality of pixel blocks.
2. The image sensing device according to claim 1, wherein, The first pixel and the plurality of second pixels share a microlens in each of the plurality of pixel groups.
3. The image sensing device according to claim 1, wherein, The merging circuit merges the plurality of second pixels.
4. The image sensing device according to claim 1, in, Each of the plurality of pixel blocks includes: Multiple first pixel groups; A plurality of second pixel groups, wherein the plurality of second pixel groups are arranged adjacent to the first pixel group in a first direction; A plurality of third pixel groups, the plurality of third pixel groups being arranged adjacent to the first pixel group in a second direction; and A plurality of fourth pixel groups are arranged adjacent to the third pixel group in the first direction, and in, The arrangement pattern of the first pixel group is symmetrical to the arrangement pattern of the second pixel group in the first direction; and The arrangement pattern of the third pixel group is symmetrical to the arrangement pattern of the fourth pixel group in the first direction.
5. The image sensing device according to claim 4, wherein, Each of the first pixel group arranges the first pixel and the plurality of second pixels in a (2×2) matrix structure, and The first pixel is located on the upper left side of each first pixel group.
6. The image sensing device according to claim 4, wherein, Each of the second pixel group causes the first pixel and the plurality of second pixels to be arranged in a (2×2) matrix structure, and The first pixel is located to the upper right of each second pixel group.
7. The image sensing device according to claim 4, wherein, Each of the third pixel group causes the first pixel and the plurality of second pixels to be arranged in a (2×2) matrix structure, and The first pixel is located on the lower left side of each third pixel group.
8. The image sensing device according to claim 4, wherein, Each of the fourth pixel group causes the first pixel and the plurality of second pixels to be arranged in a (2×2) matrix structure, and The first pixel is located on the lower right side of each fourth pixel group.
9. The image sensing device according to claim 1, the image sensing device further comprising an offset setting circuit, the offset setting circuit setting an offset differently according to the position of the first pixel in the plurality of pixel blocks.
10. The image sensing device according to claim 9, wherein, The offset setting circuit causes the offset value to be set to a higher offset as the distance to the individual pixel block used as a reference among the plurality of pixel blocks increases.
11. The image sensing device according to claim 9, wherein, The offset setting circuit sets the offset value by using the pixel values of at least two first pixels included in different pixel groups in each of the plurality of pixel blocks.
12. The image sensing device according to claim 11, wherein, The offset setting circuit sets the left offset value by calculating the sum of the pixel values of two first pixels that are adjacent to each other in the second direction on the left column of each of the plurality of pixel groups; and The offset setting circuit sets the right offset value by calculating the sum of the pixel values of two first pixels that are adjacent to each other in the second direction on the right column of each of the plurality of pixel groups.
13. The image sensing device according to claim 1, wherein, The plurality of pixel blocks include: First pixel block; A second pixel block is arranged adjacent to the first pixel block in a first direction; A third pixel block, which is arranged adjacent to the second pixel block in the first direction; and A fourth pixel block, which is arranged adjacent to the third pixel block in the first direction.
14. The image sensing device according to claim 13, wherein, The phase difference adjuster: The first value is calculated by summing the pixel values of the two first pixels located in the first left column in each of the pixel groups included in the first pixel block; The second value is calculated by summing the pixel values of the two first pixels located in the first right column in each of the pixel groups included in the second pixel block; The third value is calculated by summing the pixel values of the two first pixels located in the second left column of each pixel group included in the third pixel block; and The left image is generated by summing the first value, the second value, and the third value.
15. The image sensing device according to claim 13, wherein, The phase difference adjuster: The fourth value is calculated by summing the pixel values of the two first pixels located in the second right column of each pixel group included in the second pixel block; The fifth value is calculated by summing the pixel values of the two first pixels located in the third left column of each pixel group included in the third pixel block; and The sixth value is calculated by summing the pixel values of the two first pixels located in the third right column of each pixel group included in the fourth pixel block; and The right image is generated by summing the fourth, fifth, and sixth values.
16. The image sensing device according to claim 13, wherein, The baseline adjuster: The baseline length of the left image is calculated by summing the left image of the first pixel block, the right image of the second pixel block, and the left image of the third pixel block. and The baseline length of the right image is calculated by summing the right image of the second pixel block, the left image of the third pixel block, and the right image of the fourth pixel block.
17. The image sensing device according to claim 16, wherein, The baseline adjuster: The baseline length is calculated by assigning a first value to the left image and a second value to the right image; and The baseline length is adjusted in the direction of the image with the shorter baseline length between the left and right images.
18. The image sensing device according to claim 1, wherein, The phase difference adjuster further includes a sign adjuster that adjusts the baseline direction between the plurality of second pixels included in different pixel groups to be the same as the baseline direction between the first pixels.
19. The image sensing device according to claim 18, wherein, The first pixel group in the plurality of pixel groups arranges the first pixel and the plurality of second pixels in a (2×2) matrix structure, and the first pixel is located on the upper left side of each first pixel group; The second pixel group in the plurality of pixel groups arranges the first pixel and the plurality of second pixels in a (2×2) matrix structure, with the first pixel located on the upper right side of each second pixel group; and The baseline adjuster: The length between the center of the first pixel included in the first pixel group and the center of the first pixel included in the second pixel group is calculated as the first baseline length; and The length between the center of the second pixel included in the first pixel group and the center of the second pixel included in the second pixel group is calculated as the second baseline length.
20. The image sensing device according to claim 19, wherein, The sign adjuster reverses the baseline direction by adjusting the operator when summing the plurality of second pixels included in the first pixel group and the plurality of second pixels included in the second pixel group.
Citation Information
Patent Citations
Display device and head mount display
KR1020250054886A