The present invention relates generally to semiconductor imagers. More specifically, the present invention relates to methods and apparatus for a semiconductor imager capable of imaging over a high dynamic range of brightness levels.
The pixel reading circuit 150 includes a reset signal sample and hold (SH) circuit 151 for sampling and holding the reset signal Vrst. The pixel reading circuit 150 also includes a photo signal sample and hold circuit 152 for sampling and holding the photo signal Vsig. The sample and hold circuits 151, 152 are respectively triggered when control signals SHR, SHS are asserted to sample and hold the reset Vrst and photo Vsig signals. The pixel reading circuit further includes a differential amplifier 160, a bias circuit 170, and nodes C and D. As illustrated, column line 125 couples the output of the pixel at node B to the input of the pixel reading circuit at node C. The output of the differential amplifier 160 is presented at node D.
The pixel 100 is operated by asserting the ROW control signal to cause the row select transistor 112 to conduct. The RESET control signal is asserted to cause a reset voltage from node A (e.g., VAAPIX) to be applied to charge storage node E to cause the pixel 100 to output a reset signal Vrst through transistors 111 and 112. The SHR control signal is asserted to cause the sample and hold circuit 151 to sample and hold the reset signal Vrst. The RESET and SHR control signals are then deasserted. The light sensitive element 101 is exposed to light during a charge integration period, i.e., an exposure period. During the exposure, the light sensitive element 101 produces and stores charge related to the amount of incident light. Charge accumulates at node P based on the intensity of the incident light and the length of the integration period. At the end of the integration period, the accumulated charge is transferred to storage node E by transistor 114 by asserting the TX control signal to cause the pixel 100 to output a photo signal Vsig through transistors 111 and 112. The SHS control signal is asserted to cause the sample and hold circuit 152 to sample and hold the photo signal Vsig. The TX and SHS control signals are then deasserted. Both the reset signal Vrst and the photo signal Vsig are output at node B, albeit at different times. The bias circuit 170 biases the column line 125 to provide a stable signal path for the reset and photo signals Vrst, Vsig.
As noted, the reset signal Vrst is sampled and held by the reset signal circuit 151, while the photo signal Vsig is sampled and held by circuit 152. The sampled and held photo and reset signals are supplied as inputs to differential amplifier 160, which generates the pixel output signal Vpixel=(Vrst−Vsig). As the reset Vrst and photo Vsig signals are produced during the same image frame, this is known as correlated double sampling (CDS) which minimizes the effect of kTC noise associated with the pixel 100 and the pixel reading circuit 150. The resulting pixel output signal Vpixel is output at node D.
Typically, the imager 200 is operated on a rolling shutter basis, in which the rows of pixels are sequentially selected for read out, where each pixel in a selected row outputs its reset Vrst and photo Vsig signals at the same time to a corresponding set of pixel reading circuits. That is, a row of pixels from the array 201 is selected by the control circuit 250 by sending a row address from the control circuit 250 to the row decoder 220. The row decoder 220 decodes the row address and operates the row driver 210. The row driver 210 asserts the ROW control signal on a line coupled to the row select transistor 112 of each pixel in the selected row.
The assertion of the ROW control signal causes the row select transistor 112 of each pixel 100 in the selected row to conduct. The column driver 260 includes the same number of pixel reading circuit 150 as the number of pixels 100 in each row of the pixel array 201, so that each pixel in a selected row is associated with its own pixel reading circuit 150. This permits each pixel 100 in the selected row to output its reset Vrst and photo Vsig signals at node B, which is received at node C of the associated pixel reading circuit 150. Each pixel reading circuit 150 outputs a corresponding pixel output signal Vpixel at node D. The control circuit 250 operates the column decoder 260 to cause the column driver circuit 270 to select a column from the selected pixels. The output from node D of the pixel reading circuit 150 associated with the pixel in the selected column is routed via node D′ to an analog to digital converter 280, which converts the output to a digital value.
The analog to digital converter 280 is coupled to an imager processor 290. The image processor operates upon the digital values produced by the analog to digital converter 280 and outputs a processed digital value to the output circuit 295. If the imager 200 is a color imager, it is important that the digital values pixel output signals Vpixel, and thus the corresponding digital values produced by the analog to digitail converter 280, are linear functions of the incident light. The output circuit 295 is used to send the processed digital value to other devices (not illustrated). The other devices may be, for example, an internal storage device, a removable memory device, a display screen, or an external interface.
It is desirable for an imager 200 to be capable of imaging in the presence of a high level of dynamic range in the brightness of the light incident upon the pixels 100. It is desirable for the imaging to be performed as a linear function of incident light. However, for any given integration period, in the presence of incident light having a high level of brightness, there is a danger of overexposure due to the light sensitive element 101 producing too many photo generated charges. Conversely, in the presence of incident light having a low level of brightness, there is a danger of underexposure due to the light sensitive element 101 producing too few photo generated charges. Accordingly, there is a need and desire for an imager capable of linearly operating over a high dynamic range.
Exemplary embodiments of the method and apparatus of the present invention provide an imager with a rolling shutter which operates each pixel using both a short integration period and a long integration period. Thus, at any given point in time, different rows of the imager may be at different stages of either a short integration period or a long integration period. Analog pixel signals of both integration periods are read by a pixel reading circuit, which provides parallel signal paths for both integration periods and generates a short integration pixel output signal and a long integration pixel output signal. Both pixel output signals are digitized to produce corresponding digital values. Both digital values are respectively linear functions of the incident light.
The foregoing and other advantages and features of the invention will become more apparent from the detailed description of exemplary embodiments of the invention given below with reference to the accompanying drawings, in which:
Now referring to the drawings, where like reference numerals designate like elements, there is shown in
The imager 400 includes a new column driver 260′. The new column driver 260′ includes a plurality of the pixel output circuits 150′ illustrated in
Node D1 is coupled to a SI analog to digital converter 280a, while node D2 is coupled to a LI analog to digital converter 280b. The SI and LI analog to digital converter 280a, 280b may each be identical to the analog digital converter 280 (
The image processor 290′ is coupled to both analog to digital converters 280a, 280b and may accept digital values produced by both analog to digital converters 280a, 280b for further processing. The processed result is supplied to an output circuit 295, which may be coupled to another device via node D″, as previously discussed in connection with
Now also referring to
At an initial time t0a, the ROW control signal corresponding to the selected row is asserted high and the SHR, SHS1, SHS2 control signal are asserted low.
At time t0b, the RESET and TX control signals are also asserted high.
At time t1a, the TX control signal is asserted low for each pixel in the selected row.
At time t1b, the RESET control signal is asserted low for each pixel in the selected row. With both the TX and RESET control signals asserted low, photo generated charges begin to accumulate at node P of each pixel in the selected row, thereby starting the short integration period SI at time t1a.
At time t1c, the ROW control signal is asserted low.
At time t2a, the ROW control signal is asserted high.
At time t2b, the TX control signal is asserted high. This causes the transfer transistor 114 to conduct. Photo generated charges are transferred from the photodiode 101 via accumulation node P to charge storage node E. The transferred photo generated charges reduce the voltage level of charge node E, which is coupled to the gate of the source follower transistor 111.
At time t3a, the TX control signal for the selected row is asserted low. New photo generated charges now accumulate at accumulation node P and are not transferred to the charge storage node E because the transfer transistor 114 is now non-conducting. The voltage at the charge storage node E is now not affected by new photo generated charges. This marks the end of the short integration period SI, and starts a long integration period LI.
At time t3b, the ROW control signal is asserted low.
At time t4, the ROW control signal of the selected row is asserted high. This causes the row select transistor 112 to conduct. The photo signal corresponding to the short integration period is output at node B. Simultaneously, the SHS1 control signal is asserted high. The photo signal associated with the short integration period Vsig1 is output at node B, and sampled and held by the SI sample and hold circuit 152a.
At time t5, the SHS1 control signal is asserted low, and the RESET control signal is asserted high. This sets the charge storage node E to a predetermined voltage and the reset signal Vrst is output at node B. Simultaneously, the SHR control signal is asserted high, causing sample and hold circuit 151 to sample and hold the reset signal.
At time t6a, the RESET control signal is asserted low. At time t6b, the SHR control signal is asserted low. The reset signal Vrst is no longer output at node B.
At time t7, the TX and SHS2 control signals are asserted high. This causes the transfer transistor 114 to conduct, which transfers charges accumulated at the accumulation node P during the long integration period LI to the charge storage node E. This decreases the voltage at the charge storage node, and thus, the voltage at the gae of the source follower transistor 111. The photo signal associated with the long integration period Vsig2 is output at node B and sampled and held by the sample and hold circuit 152b.
At time t8a, the TX control signal is asserted low.
At time t8b, the SHS2 control signal is asserted low. The photo signal associated with the long integration period Vsig2 is no longer output at node B.
By now the sample and hold circuits 151, 152a, 152b respectively hold the reset signal, the short integration photo signal Vsig1, and the long integration period photo signal Vsig2.
At time t9, the SI differential amplifier 160a now generates a short integration pixel output signal Vpixel1 at node D1 using the reset signal Vrst and the SI photo signal Vsig1. This signal Vpixel1 is not a true correlated double sampled signal as the SI photo signal Vsig1 and reset signal Vrst were not sampled back-to-back. The LI differential amplifier 160b also generates the LI pixel output signal Vpixel2 at node D2 using the reset signal Vrst and the long integration photo signal Vsig2. This signal Vpixel2 is a true correlated double sampled signal.
At time t10, the output of the Vpixel1 and Vpixel2 signal has completed.
The two pixel output signal Vpixel1, Vpixel2 are respectively supplied to analog to digital converters 280a, 280b to produce corresponding digital values, which can then be processed by image processor 290′, and output to another device via output circuit 295. Because the SI and LI pixel output signals Vpixel1, Vpixel2 were not fused together in the analog domain, each signal Vpixel1, Vpixel2 is a (different) linear function of the incident light, and each signal is therefore compatible with conventional color processing algorithms.
The present invention therefore provides for a pixel reading circuit 150′ which samples and holds a plurality of photo signals taken over different integration periods. The SI pixel output signal is suitable if the incident light is of a high brightness level. Although this signal is not generated using correlated double sampled signal, kTC noise should be of little concern as a high brightness environment would have a good signal to noise ratio (i.e., the signal level will be high in comparison to the kTC noise level at high levels of brightness). The LI pixel output signal is suitable if the incident light is of a low brightness level, as the signal to noise ratio of this signal is optimized by using correlated double sampling, which minimizes the affect of kTC noise. The present invention also provides for parallel processing and output of the pixel output signal formed from the different integration period in order. The pixel output signals are not fused or combined until after digitization. This permits each pixel output signal Vpixel1, Vpixel2 to have be linear functions of incident light, which is advantageous for color processing by the image processor 290′. The image processor 290′ may, for example, operate the digital value(s) corresponding to either or both pixel output signals Vpixel1, Vpixel2.
Referring now to
The image processor 290′ takes the digital output from analog to digital converters 280a and 280b for each pixel and digitally processes the signals to produce a representation of the image having a higher dynamic range than that created by the image processor 290 (
System 700, for example a camera system, generally comprises a central processing unit (CPU) 702, such as a microprocessor, that communicates with an input/output (I/O) device 706 over a bus 720. Imaging device 400 also communicates with the CPU 702 over the bus 720. The system 700 also includes random access memory (RAM) 704, and can include removable memory 714, such as flash memory, which also communicate with the CPU 702 over the bus 720. The imager 400 may be combined with a processor, such as a CPU, digital signal processor, or microprocessor, with or without memory storage on a single integrated circuit or on a different chip than the processor.
It should be appreciated that other embodiments of the invention include a method of manufacturing the imager 400 of the invention as illustrated in
While the invention has been described in detail in connection with the exemplary embodiments, it should be understood that the invention is not limited to the above disclosed embodiments. Rather, the invention can be modified to incorporate any number of variations, alternations, substitutions, or equivalent arrangements not heretofore described, but which are commensurate with the spirit and scope of the invention. Accordingly, the invention is not limited by the foregoing description or drawings, but is only limited by the scope of the appended claims.
Number | Date | Country | |
---|---|---|---|
Parent | 10927591 | Aug 2004 | US |
Child | 12155823 | US |