Sampling and readout of an image sensor

FIELD OF THE INVENTION

0001The present invention relates generally to electronic image sensors for use in digital cameras and other image capture devices, and more particularly to sampling and readout techniques for use with an electronic image sensor.

BACKGROUND OF THE INVENTION

0002A typical electronic image sensor comprises a number of light sensitive picture elements ("pixels") arranged in a two-dimensional array. Such an image sensor may be configured to produce a color image by forming a color filter array (CFA) over the pixels. One commonly used type of CFA pattern is the Bayer pattern, disclosed in <patcit id="pcit0001" dnum="US3971065A"><text>U.S. Patent No. 3,971,065</text></patcit>, entitled "Color Imaging Array,". The Bayer CFA pattern provides each pixel with color photoresponse exhibiting a predominant sensitivity to one of three designated portions of the visible spectrum. The three designated portions may be, for example, red, green and blue, or cyan, magenta and yellow. A given CFA pattern is generally characterized by a minimal repeating unit in the form of a subarray of contiguous pixels that acts as a basic building block for the pattern. Multiple copies of the minimal repeating unit are juxtaposed to form the complete pattern.

0003An image captured using an image sensor with a Bayer CFA pattern has only one color value at each pixel. Therefore, in order to produce a full color image, the missing color values at each pixel are interpolated from the color values of nearby pixels. Numerous such interpolation techniques are known in the art. See, for example, <patcit id="pcit0002" dnum="US5652621A"><text>U.S. Patent No. 5,652,621</text></patcit>, entitled "Adaptive Color Plane Interpolation in Single Sensor Color Electronic Camera,".

0004In accordance with conventional practice, sampling and readout of the pixels of an image sensor having a Bayer CFA pattern will generally sample all the pixels in a given row into column circuits, and then the entire row of pixels would be read from the column circuits in a single operation. The sampling and readout proceeds in this manner in order to preserve the pixel order of the Bayer CFA pattern in the readout data.

0005<patcit id="pcit0003" dnum="US20070024931A" dnum-type="L"><text>U.S. Patent Application Publication No. 2007/0024931</text></patcit>, entitled "Image Sensor with Improved Light Sensitivity," discloses improved CFA patterns, including patterns which provide certain of the pixels with a panchromatic photoresponse. Such patterns are also generally referred to herein as "sparse" CFA patterns. A panchromatic photoresponse has a wider spectral sensitivity than those spectral sensitivities represented in the selected set of color photoresponses and may, for example, have high sensitivity across substantially the entire visible spectrum. Image sensors configured with the improved CFA patterns exhibit greater light sensitivity and are thus well-suited for use in applications involving low scene lighting, short exposure time, small aperture, or other restrictions on the amount of light reaching the image sensor.

0006<patcit id="pcit0004" dnum="US6734906B1"><text>US6734906 B1</text></patcit> discloses an image sensor comprising: an array of pixels arranged in rows and columns, the columns being separated into groups, wherein each group comprises two columns and is associated with a common output line shared by said two columns, wherein a sparse color filter array pattern is configured over the array of pixels and the sparse color filter array pattern has a minimal repeating unit comprising sixteen pixels arranged in four cells of 2x2 pixels ; and sampling and readout circuitry comprising, for each group of two columns in the pixel array, a first column circuit and a second column circuit, wherein for each group of columns, the sampling and read-out circuitry is configured to sample a pixel through the common output line into one of the column circuits associated with said group and to read out a previously sampled pixel from the other column circuit associated with said group.

0007<patcit id="pcit0005" dnum="EP0954032A"><text>EP0954032 A</text></patcit> discloses an image sensing apparatus having a plurality of unit cells, each including a plurality of photoelectric conversion elements and a common circuit shared by the plurality of photoelectric conversion elements.

0008Conventional sampling and readout techniques such as those described above in the context of the Bayer CFA pattern can be inefficient when applied to sparse CFA patterns. Accordingly, a need exists for improved sampling and readout techniques for use with sparse CFA patterns.

SUMMARY OF THE INVENTION

0009Illustrative embodiments of the invention provide sampling and readout techniques that are particularly well-suited for use with image sensors having sparse CFA patterns.

0010Accordingly, the present invention provides an image sensor as set out in Claim 1. Further aspects of the present invention are set out in the remaining claims.

BRIEF DESCRIPTION OF THE DRAWINGS

0011The above and other objects, features, and advantages of the present invention will become more apparent when taken in conjunction with the following description and drawings wherein identical reference numerals have been used, where possible, to designate identical features that are common to the figures, and wherein: <ul id="ul0001" list-style="none" compact="compact"><li><figref idref="f0001">FIG. 1</figref> is a block diagram of a digital camera having an image sensor with a sparse CFA pattern and incorporating sampling and readout circuitry in accordance with an illustrative embodiment of the invention;</li><li><figref idref="f0002">FIG. 2</figref> is a schematic diagram of one possible implementation of a portion of a pixel array in the image sensor of the digital camera of <figref idref="f0001">FIG. 1</figref>;</li><li><figref idref="f0003">FIG. 3</figref> shows one possible implementation of sampling and readout circuitry comprising column circuits for sampling and readout of the pixel array of <figref idref="f0002">FIG. 2</figref> ; and</li><li><figref idref="f0004 f0005 f0006 f0007 f0008 f0009 f0010 f0011 f0012 f0013 f0014 f0015 f0016">FIGS. 4 through 13</figref> show examples of sampling and readout operations applied to the pixel array of <figref idref="f0002">FIG. 2</figref>, wherein <figref idref="f0005 f0006 f0011">FIGS. 5A-B and 9A</figref> are embodiments of the invention and <figref idref="f0007 f0008">FIGS. 6A-B</figref>, <figref idref="f0010">FIG. 8</figref>, <figref idref="f0012">FIG. 9B</figref> and <figref idref="f0013 f0016">FIGs. 10-13</figref> are examples useful for understanding the invention.</li></ul>

DETAILED DESCRIPTION OF THE INVENTION

0012The present invention will be illustrated herein in conjunction with particular embodiments of image capture devices, image sensors, and associated sampling and readout techniques. It should be understood, however, that these illustrative arrangements are presented by way of example only, and should not be viewed as limiting the scope of the invention in any way. Those skilled in the art will recognize that the disclosed arrangements can be adapted in a straightforward manner for use with a wide variety of other types of image capture devices, image sensors, and associated sampling and readout techniques.

0013<figref idref="f0001">FIG. 1</figref> shows a digital camera in which sampling and readout techniques are implemented in an illustrative embodiment of the invention. In the digital camera, light 10 from a subject scene is input to an imaging stage 11. The imaging stage 11 comprises lens 12, neutral density (ND) filter 13, iris 14 and shutter 18. The light 10 is focused by lens 12 to form an image on an image sensor 20. The amount of light reaching the image sensor 20 is regulated by iris 14, ND filter 13 and the time that shutter 18 is open. Image sensor 20 converts the incident light to an electrical signal for each pixel. The image sensor 20 may be, for example, a charge coupled device (CCD) type or active pixel sensor (APS) type image sensor, although other types of image sensors may be used in implementing the invention. APS type image sensors fabricated using a complementary metal-oxide-semiconductor (CMOS) process are often referred to as CMOS image sensors. The image sensor 20 generally has a two-dimensional array of color and panchromatic pixels configured in accordance with a sparse CFA pattern. Examples of sparse CFA patterns that may be used with the image sensor 20 include those described in the above-cited <patcit id="pcit0006" dnum="US20070024931A" dnum-type="L"><text>U.S. Patent Application Publication No. 2007/0024931</text></patcit>, although other CFA patterns may be used in other embodiments of the invention.

0014An analog signal from image sensor 20 is processed by analog signal processor 22 and applied to analog to digital (A/D) converter 24. Timing generator 26 produces various clocking signals to select particular rows and columns of the pixel array for processing, and synchronizes the operation of analog signal processor 22 and A/D converter 24. The image sensor 20, analog signal processor 22, A/D converter 24, and timing generator 26 collectively form an image sensor stage 28 of the digital camera. The components of image sensor stage 28 may comprise separately fabricated integrated circuits, or they may be fabricated as a single integrated circuit as is commonly done with CMOS image sensors. The A/D converter 24 outputs a stream of digital pixel values that are supplied via a bus 30 to a memory 32 associated with a digital signal processor (DSP) 36. Memory 32 may comprise any type of memory, such as, for example, synchronous dynamic random access memory (SDRAM). The bus 30 provides a pathway for address and data signals and connects DSP 36 to memory 32 and A/D converter 24.

0015The DSP 36 is one of a plurality of processing elements of the digital camera that are indicated as collectively comprising a processing stage 38. The other processing elements of the processing stage 38 include exposure controller 40 and system controller 50. Although this partitioning of digital camera functional control among multiple processing elements is typical, these elements may be combined in various ways without affecting the functional operation of the camera and the application of the present invention. A given one of the processing elements of processing stage 38 can comprise one or more DSP devices, microcontrollers, programmable logic devices, or other digital logic circuits. Although a combination of three separate processing elements is shown in the figure, alternative embodiments may combine the functionality of two or more of these elements into a single processor, controller or other processing element. Sampling and readout techniques as described herein may be implemented at least in part in the form of software that is executed by one or more such processing elements.

0016The exposure controller 40 is responsive to an indication of an amount of light available in the scene, as determined by brightness sensor 42, and provides appropriate control signals to the ND filter 13, iris 14 and shutter 18 of the imaging stage 11.

0017The system controller 50 is coupled via a bus 52 to DSP 36 and to program memory 54, system memory 56, host interface 57 and memory card interface 60. The system controller 50 controls the overall operation of the digital camera based on one or more software programs stored in program memory 54, which may comprise Flash electrically erasable programmable read-only memory (EEPROM) or other nonvolatile memory. This memory is also used to store image sensor calibration data, user setting selections and other data which must be preserved when the camera is turned off. System controller 50 controls the sequence of image capture by directing exposure controller 40 to operate the lens 12, ND filter 13, iris 14, and shutter 18 as previously described, directing the timing generator 26 to operate the image sensor 20 and associated elements, and directing DSP 36 to process the captured image data.

0018In the illustrated embodiment, DSP 36 manipulates the digital image data in its memory 32 according to one or more software programs stored in program memory 54 and copied to memory 32 for execution during image capture. After an image is captured and processed, the resulting image file stored in memory 32 may be, for example, transferred via host interface 57 to an external host computer, transferred via memory card interface 60 and memory card socket 62 to removable memory card 64, or displayed for the user on an image display 65. The image display 65 is typically an active matrix color liquid crystal display (LCD), although other types of displays may be used.

0019The digital camera further comprises a user control and status interface 68 including a viewfinder display 70, an exposure display 72, user inputs 74 and status display 76. These elements may be controlled by a combination of software programs executed on exposure controller 40 and system controller 50. The user inputs 74 typically include some combination of buttons, rocker switches, joysticks, rotary dials or touchscreens. Exposure controller 40 operates light metering, exposure mode, autofocus and other exposure functions. The system controller 50 manages a graphical user interface (GUI) presented on one or more of the displays, e.g., on image display 65. The GUI typically includes menus for making various option selections and review modes for examining captured images.

0020Processed images may be copied to a display buffer in system memory 56 and continuously read out via video encoder 80 to produce a video signal. This signal may be output directly from the camera for display on an external monitor, or processed by display controller 82 and presented on image display 65.

0021It is to be appreciated that the digital camera as shown in <figref idref="f0001">FIG. 1</figref> may comprise additional or alternative elements of a type known to those skilled in the art. Elements not specifically shown or described herein may be selected from those known in the art. As noted previously, the present invention may be implemented in a wide variety of other types of digital cameras or image capture devices. For example, the present invention can be implemented in imaging applications involving mobile phones and automotive vehicles. Also, as mentioned above, certain aspects of the embodiments described herein may be implemented at least in part in the form of software executed by one or more processing elements of an image capture device. Such software can be implemented in a straightforward manner given the teachings provided herein, as will be appreciated by those skilled in the art.

0022The image sensor 20 shown in <figref idref="f0001">FIG. 1</figref> includes a two-dimensional array of light sensitive pixels fabricated on a silicon substrate that provide a way of converting incoming light at each pixel into an electrical signal that is measured. As the sensor is exposed to light, free electrons are generated and captured within the electronic structure at each pixel. Capturing these free electrons for some period of time and then measuring the number of electrons captured, or measuring the rate at which free electrons are generated, allows the light level at each pixel to be measured. In the former case, accumulated charge is shifted out of the array of pixels to a charge to voltage measurement circuit as in a CCD type image sensor, or the area close to each pixel contains elements of a charge to voltage measurement circuit as in an APS type or CMOS image sensor.

0023Whenever general reference is made to an image sensor in the following description, it is understood to be representative of the image sensor 20 from <figref idref="f0001">FIG. 1</figref>. It is further understood that all examples and their equivalents of image sensor architectures and pixel patterns of the present invention disclosed in this specification are used for image sensor 20.

0024In the context of an image sensor, a pixel refers to a discrete light sensing area and charge shifting or charge measurement circuitry associated with the light sensing area. In the context of a digital color image, the term pixel commonly refers to a particular location in the image having associated color values.

0025<figref idref="f0002">FIG. 2</figref> shows a portion of a pixel array 100 in the image sensor 20 of the digital camera of <figref idref="f0001">FIG. 1</figref> in an illustrative embodiment. Each pixel 102 of the pixel array 100 comprises a photodiode 104 and a transfer gate (TG) 106. Additional circuitry 108 comprising multiple transistors is shared by a subarray of four pixels arranged in a 2x2 block. The 2x2 pixel block is one example of what is more generally referred to herein as a "cell." There is a different set of additional circuitry 108 associated with each of the 2x2 pixel cells. The configuration of pixel array 100 in this embodiment is referred to as a four transistor, four shared (4T4S) arrangement, as four of the pixels and their associated four TGs share the additional circuitry 108. Adjacent each of the pixels 102 in the array 100 is an indicator of the corresponding color, which may be red (R), blue (B), green (G) or panchromatic (P), in accordance with a designated sparse CFA pattern of the image sensor 20. The particular sparse CFA pattern used in the illustrative embodiments described herein is a panchromatic checkerboard pattern disclosed in the above-cited <patcit id="pcit0007" dnum="US20070024931A" dnum-type="L"><text>U.S. Patent Application Publication No. 2007/0024931</text></patcit>, although numerous other CFA patterns may be used.

0026The portion of the array 100 shown in <figref idref="f0002">FIG. 2</figref> includes four rows of eight pixels each, with the two upper rows of this portion being referred to herein as a blue/green row pair, and the two lower rows being referred to herein as a red/green row pair. The minimal repeating unit in this particular CFA pattern is a subarray of 16 contiguous pixels comprising the left half or right half of the portion of the pixel array 100 as shown in <figref idref="f0002">FIG. 2</figref>. Thus, the minimal repeating unit comprises 16 pixels arranged in four four-pixel cells as follows: <ul id="ul0002" list-style="none" compact="compact"><li>ZPYP</li><li>PZPY</li><li>YPXP</li><li>PYPX</li></ul> where P represents one of the panchromatic pixels and X, Y and Z represent respective color pixels. In this particular embodiment, X, Y and Z are red, green and blue, respectively. Alternatively, X, Y and Z may be individually selected in a different manner from red, green and blue, or may be individually selected from another set of colors, such as cyan, magenta and yellow. Patterns with other minimal repeating units, such as minimal repeating units of at least twelve pixels as described in the above-cited <patcit id="pcit0008" dnum="US20070024931A" dnum-type="L"><text>U.S. Patent Application Publication No. 2007/0024931</text></patcit>, may be used.

0027The columns in the portion of the pixel array 100 shown in <figref idref="f0002">FIG. 2</figref> are separated into groups, with each group comprising two of the columns and sharing a common output. For example, the pixels in the first two columns at the left side of the array share a common output denoted PixCol_0/1. Similarly, the pixels in the next two columns of the array share a common output denoted PixCol_2/3. The remaining two pairs of columns share the respective common outputs denoted PixCol_4/5 and PixCol_6/7. Each pixel in a given one of the 2x2 pixel cells is connectable to its shared common output via the additional circuitry 108 associated with that cell. The additional circuitry 108 in this embodiment comprises three N-type metal-oxide-semiconductor (NMOS) transistors interconnected as shown. In other embodiments, different types of column groupings, pixel cells and shared outputs, as well as circuitry for connecting the pixel cells to the shared outputs, may be used.

0028Elements of the pixel array 100 in <figref idref="f0002">FIG. 2</figref> are coupled to supply voltage Vdd and substrate voltage Vsub as shown. Control signals applied to the pixel array include TG control signals TG_P0, TG_C1, TG_C2 and TG_P3, as well as additional control signals including row select (RS) signals and reset gate (RG) signals. The subscripts N+1 and N associated with these signals refer to respective upper and lower row pairs in the array.

0029As will be described in greater detail below, the pixel array 100 is advantageously configured to permit binning of same-color pixels and binning of panchromatic pixels. The term "binning" as used herein generally refers to simultaneously connecting two or more pixels from the same pixel cell to the same common output prior to sampling that output.

0030<figref idref="f0003">FIG. 3</figref> illustrates exemplary sampling and readout circuitry comprising column circuits 110 that are configured for sampling and readout of particular column pairs and are located at the bottom of the pixel array 100. In this example, there is one column output for every two columns of pixels. The column circuits 110 comprise sample-and-hold switches 112, column circuit capacitors 114, and column output switches 116. The signals SHx_E and SHx_O that control the switches 112 provide sample-and-hold for even (E) and odd (O) pixel columns, where x = R for reset and x = S for signal. The sampling and readout circuitry in this embodiment further comprises top column circuits, configured in a manner similar to bottom column circuits 110 but not shown in the figure, at the top of the pixel array 100. Top and bottom column circuits configured in the manner illustrated in <figref idref="f0003">FIG. 3</figref> will be denoted more specifically herein, for example, in conjunction with <figref idref="f0004">FIG. 4</figref>, as top and bottom column circuits 110T and 110B, respectively.

0031The shared common outputs PixCol_0/1 through PixCol_14/15 as shown in <figref idref="f0003">FIG. 3</figref> are coupled to respective column circuits in the set of column circuits 110. Generally, each such shared common output is coupled to two distinct column circuits in the present embodiment, with one of the column circuits being a top column circuit and another of the column circuits being a bottom column circuit. Thus, in this embodiment, the two column circuits associated with a given group of two columns are on opposite sides of the pixel array 100. Again, only those column circuits at the bottom of the pixel array are shown in <figref idref="f0003">FIG. 3</figref>. In other embodiments, the two shared column circuits may both be on the same side of the pixel array.

0032Other signals shown in the <figref idref="f0003">FIG. 3</figref> diagram include those denoted voltage output signal (VSIG) and voltage output reset (VRST), as well as decode signals DEC_K through DEC_K+7, with one of the decode signals associated with each column circuit. Each column circuit in this embodiment comprises a pair of capacitors and the associated switches.

0033The term "sampling and readout circuitry" as used herein is intended to be construed generally, so as to encompass, for example, not only column circuits such as those shown in <figref idref="f0003">FIG. 3</figref>, but also other related switching elements associated with the pixel array 100, such as additional circuitry 108. Thus, sampling and readout circuitry in a given embodiment may encompass switches or other elements that are formed integrally with the pixel array. Sampling and readout circuitry is generally implemented within the sensor array 20, for example, as in the case of a CMOS image sensor.

0034Examples of sampling and readout operations will now be described in greater detail with reference to the diagrams of <figref idref="f0004 f0005 f0006 f0007 f0008 f0009 f0010 f0011 f0012 f0013 f0014 f0015 f0016">FIGS. 4 through 13</figref>, wherein <figref idref="f0005 f0011">FIGS. 5A-B and 9A</figref> are embodiments of the invention and <figref idref="f0007 f0008">FIGS. 6A-B</figref>, <figref idref="f0010">FIG. 8</figref>, <figref idref="f0012">FIG. 9B</figref> and <figref idref="f0013 f0016">FIGS. 10-13</figref> are examples useful for understanding the invention. Each of the examples shown in <figref idref="f0004 f0005 f0006 f0007 f0008 f0009 f0010 f0011 f0012 f0013 f0014 f0015 f0016">FIGS. 4 through 13</figref> utilizes the same four-row portion of pixel array 100 that is shown in <figref idref="f0002">FIG. 2</figref>. Each of the examples to be described may be viewed as a particular operating mode of a given set of sampling and readout circuitry implemented in the digital camera of <figref idref="f0001">FIG. 1</figref>.

0035<figref idref="f0004">FIG. 4</figref> generally illustrates the diagrammatical conventions that will be utilized in describing the sampling and readout operations for the remaining figures. A portion of the pixel array 100 is shown in the figure, configured in accordance with the above-noted panchromatic checkerboard CFA pattern. As in <figref idref="f0002">FIG. 2</figref>, the particular portion of the array that is shown in <figref idref="f0004">FIG. 4</figref> comprises four rows of eight pixels each, although it is to be understood that sampling and readout operations for the remaining portions of the array are carried out in a similar manner. The pixel array is assumed to be configured in the 4T4S arrangement described in conjunction with <figref idref="f0002">FIG. 2</figref>, and its associated sampling and readout circuitry is assumed to be configured as shown in <figref idref="f0003">FIG. 3</figref>. The text at the left of the array ("C2") identifies one or more TGs that are being activated in conjunction with the sampling operation for a given 2x2 subarray of shared pixels. The "TG Arrangement" diagram on the left side of the figure shows the relationship between TGs and pixels in the 2x2 subarray, with TGs denoted C1 and C2 representing two pixels of the same color and TGs denoted P0 and P3 representing two panchromatic pixels.

0036Vertical arrows coming out of the pixel array 100 show the sampling operations into either the top column circuits or the bottom column circuits. Note that the sampling arrows reflect the fact that there is only one pixel output line for every two columns, but each pixel output line can be sampled by top and bottom column circuits. The text at the left of the vertical arrows ("Sample Even" or "Sample Odd") indicates which column circuits are being affected by the sampling operation.

0037Top and bottom column circuits 110T and 110B are shown by the rectangles at the top and bottom, respectively, of the pixel array 100. The nomenclature in the rectangles indicates T or B for top or bottom and includes a number for readout address. Rightward pointing arrows coming out of the column circuits indicate which column circuits are being read out. The column circuits are read in sequence. The text at the left of the rightward pointing arrows ("Read Odd" or "Read Even") indicates which column circuit addresses are to be read, where, as noted above, the addresses are denoted by the numbers in the column circuit rectangles.

0038The term "line" is used to refer to a readout unit, that is, a set of pixels that are read out in a given readout operation. Thus, "Line 0" as shown in <figref idref="f0004">FIG. 4</figref> represents a readout unit given by the sequence B0, T1, B2, and T3. In this figure, only odd addresses are being read out of the top column circuits 110T while only even addresses are being read out of the bottom column circuits 110B. Also, it should be noted that Line 0 provides readout of half as many pixels (4 in this case) as are contained in a given row in the array (8 in this case). Although concurrent sampling and readout operations are shown in this figure, such an arrangement is not a requirement of the invention.

0039<figref idref="f0005">FIGS. 5A</figref> and <figref idref="f0006">5B</figref> show two different sequences of lines, denoted Option 1 and Option 2, for sampling and readout of the four-row portion of the pixel array 100 at full resolution and without binning of colors. It is apparent that eight lines, denoted Line 0 through Line 7, are required to read out the four rows in each of the two different options.

0040<figref idref="f0007">FIGS. 6A</figref> and <figref idref="f0008">6B</figref> show two different sequences of lines, again denoted Option 1 and Option 2, for sampling and readout of the four-row portion of the pixel array 100 at full resolution with binning of colors to improve signal-to-noise ratio (SNR) and to improve frame rate. Note that six lines are required to read out the four rows in this example.

0041In both of the <figref idref="f0007 f0008">FIG. 6</figref> options, the first four lines are the same. However, in Option 2, the sampling and readout order for the panchromatic pixels in lines 4 and 5 is the opposite of that used in Option 1. This second option ensures that the panchromatic pixels in the same relative positions in each 2x2 pixel cell are read using the same output channel, regardless of whether a given such panchromatic pixel is from the blue/green row pair or the green/red row pair. For example, with reference to <figref idref="f0007">FIG. 6A</figref>, it can be seen that output channel T0 is used in Line 1 to read out a lower left panchromatic pixel 600 in one of the 2x2 pixel cells, while output channel B0 is used in Line 4 to read out a lower left panchromatic pixel 602 in another 2x2 pixel cell. Thus, in <figref idref="f0007">FIG. 6A</figref>, panchromatic pixels in the same relative position in different 2x2 pixel cells are read out using different column circuits. The <figref idref="f0008">FIG. 6B</figref> sampling and readout arrangement addresses this issue. For example, it can be seen from this figure that lower left panchromatic pixels 600 and 602 are now both read out using column circuit T0, in Line 1 and Line 5, respectively. Such an arrangement may be preferred in embodiments in which there are differences between the top and bottom column circuits that would otherwise require gain correction or another type of adjustment.

0042<figref idref="f0009">FIG. 7</figref> shows the relationship between the rows of pixels in the four-row portion of the pixel array 100 (left diagram) and the output lines that are read out of the array for full resolution readout using both non-binned color pixels (middle diagram) and binned color pixels (right diagram). The middle and right diagrams correspond respectively to the diagrams "Full Resolution Non-Binned Color (Option 2)" in <figref idref="f0006">FIG. 5B</figref> and "Full Resolution Binned Color (Option 2)" in <figref idref="f0008">FIG. 6B</figref> . Readout is from bottom to top in the four-row portion of the array, with the pixels being denoted by their color R, B, G or P followed by a number <i>ij</i> in which the first digit <i>i</i> denotes the row (<i>i</i> = 0, 1, 2 or 3) and the second digit <i>j</i> denotes the column (<i>j</i> = 0, 1, 2, ... 7).

0043As mentioned previously, the term "line" is used to refer to a readout unit. It is apparent from <figref idref="f0009">FIG. 7</figref> that two lines are required to read out one full resolution row when using non-binned color (middle diagram), with one line comprising all the color pixels from the row and the other line comprising all the panchromatic pixels from the row. Similarly, three lines are required to read out two full resolution rows when using binned color (right diagram), with one line comprising the binned color pixels from the two rows, one line comprising all the panchromatic pixels from one row, and the remaining line comprising all the panchromatic pixels from the other row.

0044With reference now to <figref idref="f0010">FIG. 8</figref>, a sequence of lines is shown for sampling and readout of the four-row portion of the pixel array 100 at full resolution with binning of colors and double sampling. This exemplary oversampling arrangement provides further SNR improvement at the cost of reduced frame rate. Although the figure shows the sampling and readout for only the green/red row pair (i.e., the lower two rows of the four-row portion of the array), it is to be understood that the green/blue row pair (i.e., the upper two rows of the four-row portion of the array) is read out identically to the red/green row pair. Also, sampling and readout are non-concurrent in this example, that is, occur at separate times.

0045<figref idref="f0011">FIGS. 9A</figref> and <figref idref="f0012">9B</figref> show two different sequences of lines, denoted Option 1 and Option 2, for sampling and readout of the four-row portion of the pixel array 100 at one-quarter resolution and with separate binning of the colors and panchromatic pixels to improve SNR and to improve frame rate. Four lines, denoted Line 0 through Line 3, are required to read out the four rows in each of the two different options. In Option 1 as shown in <figref idref="f0011">FIG. 9A</figref>, green pixels are read out using both top and bottom column circuits. As previously indicated, this type of situation may be undesirable in implementations in which there are differences between the top and bottom column circuits. The sampling and readout arrangement shown in <figref idref="f0012">FIG. 9B</figref> addresses this issue by ensuring that all green pixels are read out using column circuits on one side of the array, namely, the top column circuits in this example. This eliminates any differences in the green response that would otherwise be attributable to the green pixel pairs being read out of column circuits on both top and bottom of the array.

0046<figref idref="f0013">FIG. 10</figref> shows a sequence of lines for sampling and readout of the four-row portion of the pixel array 100 at one-quarter resolution, with separate binning of the colors and panchromatic pixels, and with double sampling to provide further SNR improvement at the cost of reduced frame rate. As in the <figref idref="f0010">FIG. 8</figref> example, the sampling and readout in <figref idref="f0013">FIG. 10</figref> is shown for only the green/red row pair, but it is to be understood that the green/blue row pair is read out identically to the red/green row pair. Also, sampling and readout are again non-concurrent in this example.

0047<figref idref="f0014">FIG. 11</figref> shows three different line sequences for quarter-resolution video sampling and readout of the four-row portion of pixel array 100. All of these sequences provide output consistent with that which would be provided if a Bayer CFA pattern were used. The differences between the three sequences relate to how many panchromatic pixels are included with the color pixels in the binned result. Including panchromatic pixels with the color pixels increases photographic speed but decreases color saturation. The top diagram in <figref idref="f0014">FIG. 11</figref> shows the color only case, the middle diagram shows a medium desaturation case using one panchromatic pixel for each pair of same-color pixels, and the bottom diagram shows a higher desaturation case using two panchromatic pixels for each pair of same-color pixels.

0048Turning now to <figref idref="f0015">FIG. 12</figref>, another sequence of lines is shown for quarter-resolution video sampling and readout of the four-row portion of pixel array 100. In this sequence, double sampling is used to provide SNR improvement at the cost of reduced frame rate. Similar sequences can be used for desaturated video in which one or two panchromatic pixels are binned with each pair of same-color pixels, as in the <figref idref="f0014">FIG. 11</figref> examples.

0049<figref idref="f0016">FIG. 13</figref> shows two different sequences of lines for sampling and readout of the four-row portion of the pixel array 100 that are particularly useful in fast autofocus applications. The upper diagram shows an arrangement with binning of panchromatic pixels, while the lower diagram shows an arrangement with binning of color pixels and output consistent with a Bayer CFA pattern. In both cases, there is aggressive vertical subsampling (i.e., taking 1 out of every 3 row pairs) that reduces the amount of data to read and increases the frame rate. The relatively high horizontal sampling frequency in both cases allows good determination of focus. The panchromatic case will have the best photographic sensitivity, an important consideration given that the high frame rate will limit the exposure time. The Bayer output case will allow the data to be used for both preview and video purposes as well as for determination of focus.

0050The sampling and readout arrangements of the illustrative embodiments advantageously allow the color pixels from a given row pair to be read out separately from the panchromatic pixels. The exemplary 2x2 pixel arrangement of <figref idref="f0002">FIG. 2</figref> in which two columns of pixels share a common output means that two sampling operations are required to access all the pixels from a given row. With a sparse CFA pattern, such as the panchromatic checkerboard pattern used in the illustrative embodiments above, improved performance is obtained by reading out the color pixels separately from the panchromatic pixels so as to avoid having to separate them later in memory. Furthermore, the above-described techniques allow the sampling of half a row of pixels (e.g., either the panchromatic pixels or the color pixels) concurrently with the reading out of a previously sampled half row of pixels, effectively eliminating the overhead of the sampling time. Another advantage is that the described techniques allow a pair of same-color pixels or a pair of panchromatic pixels to be combined in the charge domain prior to readout, thereby improving SNR and reducing readout time. For example, certain of the techniques require only three sampling and readout operations for each row pair, two for the panchromatic pixels and one for the binned color pixels.

0051The illustrative examples make it clear that the sampling and readout techniques described herein are highly flexible, and provide improved performance in a wide variety of applications, including operating modes involving double sampling (i.e., oversampling to improve SNR at the expense of readout time), video and autofocus. These and other additional operating modes may be made selectable in a given image capture device depending on capture conditions, end user preference or other factors.

0052The invention has been described in detail with particular reference to certain illustrative embodiments thereof, but it will be understood that variations and modifications can be effected within the scope of the invention as set forth in the appended claims.

PARTS LIST

0053<dl id="dl0001" compact="compact"><dt>10 :</dt><dd>light from subject scene</dd><dt>11 :</dt><dd>imaging stage</dd><dt>12 :</dt><dd>lens</dd><dt>13 :</dt><dd>neutral density filter</dd><dt>14:</dt><dd>iris</dd><dt>18 :</dt><dd>shutter</dd><dt>20 :</dt><dd>image sensor</dd><dt>22 :</dt><dd>analog signal processor</dd><dt>24 :</dt><dd>analog to digital (A/D) converter</dd><dt>26 :</dt><dd>timing generator</dd><dt>28 :</dt><dd>image sensor stage</dd><dt>30 :</dt><dd>digital signal processor (DSP) bus</dd><dt>32 :</dt><dd>digital signal processor (DSP) memory</dd><dt>36 :</dt><dd>digital signal processor (DSP)</dd><dt>38 :</dt><dd>processing stage</dd><dt>40 :</dt><dd>exposure controller</dd><dt>42 :</dt><dd>brightness sensor</dd><dt>50 :</dt><dd>system controller</dd><dt>52:</dt><dd>bus</dd><dt>54 :</dt><dd>program memory</dd><dt>56 :</dt><dd>system memory</dd><dt>57 :</dt><dd>host interface</dd><dt>60 :</dt><dd>memory card interface</dd><dt>62 :</dt><dd>memory card socket</dd><dt>64 :</dt><dd>memory card</dd><dt>65 :</dt><dd>image display</dd><dt>68 :</dt><dd>user control and status interface</dd><dt>70 :</dt><dd>viewfinder display</dd><dt>72 :</dt><dd>exposure display</dd><dt>74 :</dt><dd>user inputs</dd><dt>76 :</dt><dd>status display</dd><dt>80 :</dt><dd>video encoder</dd><dt>82 :</dt><dd>display controller</dd><dt>100 :</dt><dd>pixel array</dd><dt>102 :</dt><dd>pixel</dd><dt>104 :</dt><dd>photodiode</dd><dt>106 :</dt><dd>transfer gate (TG)</dd><dt>108 :</dt><dd>additional circuitry</dd><dt>110 :</dt><dd>column circuits</dd><dt>110T :</dt><dd>top column circuits</dd><dt>110B :</dt><dd>bottom column circuits</dd><dt>112 :</dt><dd>sample-and-hold switches</dd><dt>114 :</dt><dd>column circuit capacitors</dd><dt>116 :</dt><dd>column output switches</dd><dt>600 :</dt><dd>lower-left panchromatic pixel</dd><dt>602 :</dt><dd>lower-left panchromatic pixel</dd></dl>