|Publication number||USRE42739 E1|
|Application number||US 12/717,188|
|Publication date||Sep 27, 2011|
|Filing date||Dec 13, 2001|
|Priority date||Dec 19, 2000|
|Also published as||WO2002051128A2, WO2002051128A3|
|Publication number||12717188, 717188, PCT/2001/1773, PCT/CA/1/001773, PCT/CA/1/01773, PCT/CA/2001/001773, PCT/CA/2001/01773, PCT/CA1/001773, PCT/CA1/01773, PCT/CA1001773, PCT/CA101773, PCT/CA2001/001773, PCT/CA2001/01773, PCT/CA2001001773, PCT/CA200101773, US RE42739 E1, US RE42739E1, US-E1-RE42739, USRE42739 E1, USRE42739E1|
|Original Assignee||Harusaki Technologies, Llc|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (16), Referenced by (4), Classifications (10), Legal Events (4)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application is a 371 application of PCT application PCT/CA01/01773 filed Dec. 13, 2001, which claims the benefit of U.S. provisional application 60/256,336 filed Dec. 19, 2000. This application was published in English on Jun. 27, 2002 as International Publication Number WO 02/51128 A2.
The invention relates generally to CMOS imagers, and more particularly to the processing of column level signals in an imager.
Currently, CMOS imagers are well known in the art and are implemented in many different applications. One of the primary areas that CMOS imagers can be found is in portable applications such as digital cameras that use battery power. It is therefore very desirable to have a CMOS imager that is powered with a low supply voltage, for example one volt (1V).
Having a voltage this low, however, leads to a number of problems. For one, the quantization values of an imager become very small. For example, a typical CMOS imager might use 8 bits of resolution and have a supply voltage of 3.3 volts. This leads to a quantization level of 3.3V/256=12.9 mV. If an imager is implemented with a one volt supply and 8 bits of resolution then the quantization level is 1V/256=3.9 mV. The proportion of noise relative to this low quantization value becomes very large and has a greater impact on the output.
Another problem involves the pixel structure that is used in most CMOS imagers, namely using 3 n-type MOS transistors with a photodiode. The pixel consists of a precharge transistor, a source follower amplifying transistor and an access transistor. When the pixel is precharged to the supply voltage, one threshold voltage, Vt, is lost through the precharge transistor. Once the pixel is exposed to light and the output is being read out to the column, another Vt is lost through the source follower amplifying transistor. A typical Vt for MOS transistors is 0.07 V, leaving a maximum voltage output to the column of 0.86V (1V-2*0.07V), further limiting the quantization levels.
Quantizing the output from the pixels into digital data for the Digital Signal Processor (DSP) may be carried out in a number of known ways. Correlated Double Sampling (CDS) quantizes the output of the column signal twice, once at the beginning of the integration period and once at the end. From these two samples the DSP calculates the slope of the pixel output to determine the brightness of the pixel. This method is highly susceptible to noise and as a result of the high noise to quantization level ratio can be very inaccurate in a low voltage CMOS imager. Also, by using only two points to calculate the slope of the pixel output there is no accounting for a saturated pixel. Any pixels that saturate at different times in the integration period are all calculated as having the same slope, and thus all those pixels are considered to have the same value.
Successive approximation is another method of converting the analog output from the pixel into a digital signal. In this system the output at the end of the integration period is successively compared to different threshold levels, each representing one bit of resolution of the final output. The thresholds are created through capacitors that must be sized in precise ratios (C, ½C, ¼C, etc.) which is both difficult to implement and also occupies a large amount of space. Also, the capacitors on every column across the array must be matched with one another for consistent results, which is not always attainable due to process impurities. The problem of noise is a large factor in successive approximation, as both the small quantization levels to noise ratio as well as capacitance noise, which is larger in small capacitors, will accumulate to yield inaccurate results. Successive approximation also does not directly account for the problem of threshold voltage loss through the pixel transistors. Finally, successive approximation takes a single sample at the end of the integration period and therefore, has no way of differentiating between pixels that saturate at different times throughout the integration period.
Therefore, there is a need for an efficient method and apparatus that is suited to low voltage imagers for processing the pixel output signals at the column level.
The invention is directed to a method and apparatus for processing pixel output signals from column lines in an imager having an array of pixels in rows and columns. The output signals on each column line are converted to digital signals using a sigma-delta type analog-to-digital converter and the digital signals are fed to a digital signal processor.
In accordance with another aspect of the invention, the sigma-delta type converter is monitored to stop sampling of a pixel with the detection of pixel saturation. With the sigma-delta type converter providing an output of mainly logical ones for a high input signal level and an output of mainly zeros for a low input signal level, detecting pixel saturation is carried out by counting a predetermined number of consecutive zeros.
In accordance with a further aspect of the invention, the sigma-delta type converter output is sent to the digital signal processor with the detection of the saturation of a pixel.
In accordance with another aspect of the invention, sets of a predetermined number of sigma-delta type converter output samples are condensed by a decimator into binary numbers of predetermined bit length. In addition, the number of sigma-delta type converter output samples in a set wherein the pixel has saturated is determined using a further counter and this number is fed to the decimator. Alternately, the determined number may be fed to the digital signal processor.
In accordance with yet another aspect of this invention, the next pixel in a column is controlled to be read with the saturation of the previous pixel, and the next row of pixels may be controlled to be read with the saturation of the pixels in the previous row.
In accordance with a further aspect of this invention, the outputs of the decimators may be multiplexed to provide one or more inputs to the digital signal processor. A single multiplexer may be used providing one input line to the digital signal processor, or a number of multiplexers may be used providing a number of input lines to the digital signal processor.
Other aspects and advantages of the invention, as well as the structure and operation of various embodiments of the invention, will become apparent to those ordinarily skilled in the art upon review of the following description of the invention in conjunction with the accompanying drawings.
The invention will be described with reference to the accompanying drawings, wherein:
The column level sigma-delta analog-to-digital converter 208 is able to solve the various problems associated with having a low voltage CMOS imager 200. The sigma-delta converter 208 overcomes the problems of noise through oversampling the output of the pixels 202. More samples yields more accurate results through line-of-best-fit approximations on the output of the converter 208. The line-of-best-fit approximation also allows the sigma-delta converter 208 to indirectly account for the threshold voltage Vt drops across the transistors in the pixel.
In accordance with the present invention the sigma-delta converter 208 is capable of accounting for saturating pixels 202. The integration time tint of the imager 200 corresponds to a set number of outputs from the sigma-delta converter 208 depending on the sampling frequency. If a pixel 202 does not saturate, then the sigma-delta converter 208 will output the set number of samples and the digital signal processor (DSP) will calculate the correct output. In order to account for a pixel 202 that saturates, the output of the sigma-delta converter 208 is monitored and will stop sampling when pixel 202 saturation has been detected. The DSP will then calculate the line-of-best-fit knowing that it has received less values and correspondingly calculates the result.
The invention may be implemented using column level sigma-delta converters 308 having inputs connected to the analog column output lines 304 for pixels 302 with each column level sigma-delta converter 308 having a simple counter 309 associated with it, as shown in
There are a number of benefits that are created as a result of implementing this type of column level sigma-delta converter 308. The first and most important of which is that it compensates for the various problems associated with having a low supply voltage Vsupply. The method used to compensate for pixels 302 saturating also leads to additional benefits. With dark pixels 302, the sigma-delta converter 308 will simply read out the pixel 302 with the predetermined number of samples during the integration period resulting in excellent resolution. On the other hand, if the pixel 302 is exposed to bright light, the pixel 302 saturates and the sampling period is less, allowing the DSP to process and output the data faster. Also by accommodating for pixels 302 to saturate, this invention acts as an integrated auto-exposure feature.
Implementing a sigma-delta analog-to-digital converter 208 at the column level in this manner also leads to a number of further embodiments. The first embodiment is illustrated in
In a further embodiment as illustrated in
Similarly, as shown in
While the invention has been described according to what is presently considered to be the most practical and preferred embodiments, it must be understood that the invention is not limited to the disclosed embodiments. Those ordinarily skilled in the art will understand that various modifications and equivalent structures and functions may be made without departing from the spirit and scope of the invention as defined in the claims. Therefore, the invention as defined in the claims must be accorded the broadest possible interpretation so as to encompass all such modifications and equivalent structures and functions.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5134487 *||Nov 5, 1990||Jul 28, 1992||Canon Kabushiki Kaisha||Using common circuitry for different signals|
|US5248971 *||May 19, 1992||Sep 28, 1993||Mandl William J||Method and apparatus for multiplexed oversampled analog to digital modulation|
|US5264940 *||Oct 4, 1991||Nov 23, 1993||Olympus Optical Co., Ltd.||Image sensing apparatus having exposure level and dynamic range control circuit|
|US5461425 *||Feb 15, 1994||Oct 24, 1995||Stanford University||CMOS image sensor with pixel level A/D conversion|
|US5886659 *||Aug 21, 1997||Mar 23, 1999||California Institute Of Technology||On-focal-plane analog-to-digital conversion for current-mode imaging devices|
|US5955899 *||Jan 27, 1997||Sep 21, 1999||Intel Corporation||Compact comparator|
|US6104844 *||Oct 12, 1999||Aug 15, 2000||Siemens Aktiengesellschaft||Image sensor having a multiplicity of pixel sensor regions|
|US6353324 *||Nov 5, 1999||Mar 5, 2002||Bridge Semiconductor Corporation||Electronic circuit|
|US6473122 *||Dec 6, 1999||Oct 29, 2002||Hemanth G. Kanekal||Method and apparatus to capture high resolution images using low resolution sensors and optical spatial image sampling|
|US6606120 *||Apr 12, 1999||Aug 12, 2003||Foveon, Inc.||Multiple storage node full color active pixel sensors|
|US6665013 *||May 3, 1999||Dec 16, 2003||California Institute Of Technology||Active pixel sensor having intra-pixel charge transfer with analog-to-digital converter|
|US6678039 *||May 23, 2002||Jan 13, 2004||Canesta, Inc.||Method and system to enhance dynamic range conversion useable with CMOS three-dimensional imaging|
|US6765619 *||Apr 4, 2000||Jul 20, 2004||Pixim, Inc.||Method and apparatus for optimizing exposure time in image acquisitions|
|US6831684 *||Aug 15, 2000||Dec 14, 2004||Pixim, Inc.||Circuit and method for pixel rearrangement in a digital pixel sensor readout|
|US7339621 *||Dec 13, 2001||Mar 4, 2008||Psion Teklogix Systems, Inc.||Imager output signal processing|
|EP0912043A2||Oct 26, 1998||Apr 28, 1999||Texas Instruments Incorporated||Improvements in or relating to image sensing devices|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8477056 *||Jun 1, 2010||Jul 2, 2013||Infinera Corporation||Method, system, and apparatus for interpolating an output of an analog-to-digital converter|
|US9140795 *||Sep 20, 2012||Sep 22, 2015||Mesa Imaging Ag||Time of flight sensor with subframe compression and method|
|US20110291865 *||Dec 1, 2011||Han Henry Sun||Method, system, and apparatus for interpolating an output of an analog-to-digital converter|
|US20130070132 *||Sep 20, 2012||Mar 21, 2013||Mesa Imaging Ag||Time of Flight Sensor with Subframe Compression and Method|
|U.S. Classification||348/308, 348/294|
|International Classification||H04N3/14, H04N5/335, H04N5/378, H04N5/353|
|Cooperative Classification||H04N5/3535, H04N5/378|
|European Classification||H04N5/378, H04N5/353B|
|Jan 18, 2011||AS||Assignment|
Owner name: HARUSAKI TECHNOLOGIES, LLC, DELAWARE
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:PSION TEKLOGIX SYSTEMS INC.;REEL/FRAME:025654/0826
Effective date: 20070917
|Mar 30, 2011||AS||Assignment|
Owner name: HARUSAKI TECHNOLOGIES, LLC, DELAWARE
Free format text: CORRECTIVE ASSIGNMENT TO CORRECT THE ADDRESS OF THE ASSIGNEE, PREVIOUSLY RECORDED ON REEL 025654 FRAME 0826. ASSIGNOR(S) HEREBY CONFIRMS THE ASSIGNMENT;ASSIGNOR:PSION TEKLOGIX SYSTEMS INC.;REEL/FRAME:026074/0126
Effective date: 20070917
|Jun 19, 2012||CC||Certificate of correction|
|Aug 25, 2015||FPAY||Fee payment|
Year of fee payment: 8