WO1999048281A1 - Cmos integration sensor with fully differential column readout circuit for light adaptive imaging - Google Patents

Cmos integration sensor with fully differential column readout circuit for light adaptive imaging Download PDF

Info

Publication number
WO1999048281A1
WO1999048281A1 PCT/US1999/005830 US9905830W WO9948281A1 WO 1999048281 A1 WO1999048281 A1 WO 1999048281A1 US 9905830 W US9905830 W US 9905830W WO 9948281 A1 WO9948281 A1 WO 9948281A1
Authority
WO
WIPO (PCT)
Prior art keywords
column
imager
row
pixel
pixels
Prior art date
Application number
PCT/US1999/005830
Other languages
French (fr)
Inventor
Bedabrata Pain
Zhimin Zhou
Eric R. Fossum
Original Assignee
California Institute Of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to US7817298P priority Critical
Priority to US60/078,172 priority
Application filed by California Institute Of Technology filed Critical California Institute Of Technology
Publication of WO1999048281A1 publication Critical patent/WO1999048281A1/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00Details of television systems
    • H04N5/30Transforming light or analogous information into electric information
    • H04N5/335Transforming light or analogous information into electric information using solid-state image sensors [SSIS]
    • H04N5/369SSIS architecture; Circuitry associated therewith
    • H04N5/374Addressed sensors, e.g. MOS or CMOS sensors
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00Details of television systems
    • H04N5/30Transforming light or analogous information into electric information
    • H04N5/335Transforming light or analogous information into electric information using solid-state image sensors [SSIS]
    • H04N5/341Extracting pixel data from an image sensor by controlling scanning circuits, e.g. by modifying the number of pixels having been sampled or to be sampled
    • H04N5/347Extracting pixel data from an image sensor by controlling scanning circuits, e.g. by modifying the number of pixels having been sampled or to be sampled by combining or binning pixels in SSIS
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00Details of television systems
    • H04N5/30Transforming light or analogous information into electric information
    • H04N5/335Transforming light or analogous information into electric information using solid-state image sensors [SSIS]
    • H04N5/357Noise processing, e.g. detecting, correcting, reducing or removing noise
    • H04N5/3575Noise processing, e.g. detecting, correcting, reducing or removing noise involving a correlated sampling function, e.g. correlated double or triple sampling
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N5/00Details of television systems
    • H04N5/30Transforming light or analogous information into electric information
    • H04N5/335Transforming light or analogous information into electric information using solid-state image sensors [SSIS]
    • H04N5/369SSIS architecture; Circuitry associated therewith
    • H04N5/378Readout circuits, e.g. correlated double sampling [CDS] circuits, output amplifiers or A/D converters

Abstract

An imager (10) that is better suited for low-light detection capability. In accordance with a preferred embodiment, the imager may be easily configured to provide an imager (10) having multi-resolution capability where SNR can be adjusted for optimum low-level detectibility. Multi-resolution signal processing functionality is provided on-chip to achieve high speed imaging, employs an improved pixel binning approach with fully differential circuits situated so that all extraneous and pick-up noise is eliminated. The current implementation requires no frame transfer memory, thereby reducing chip size. The reduction in area enables larger area format light adaptive imager implementations.

Description

1 -

CMOS INTEGRATION SENSOR WITH FULLY DIFFERENTIAL OT.TTM READOUT CIRCUIT FOR LIGHT ADAPTIVE IMAGING

.Statement as to Federally Sponsored Research

The invention described herein was made in the performance of work under a NASA contract, and is subject to the provisions of Public Law 96-517 (35 USC 202) in which the Contractor has elected to retain title.

Field of the Invention

The present disclosure is directed to active pixel sensors, and more particularly to multi-resolution active pixel sensor array i agers for light adaptive imaging applications .

Background

The CMOS active pixel sensor ("APS") has permitted the realization of high performance products. Each pixel has an active amplifier that buffers the photosignal. A column-parallel bus readout architecture is often used. In this architecture, the columns are connected to individual signal processing modules, which include, for example, A to D converters, and double sampling elements.

A constant challenge in smart imager technology continues to be how to enhance signal to noise ratio ("SNR") under low illumination conditions.

One way to do this is to trade spatial resolution for SNR by summing neighborhood pixels (pixel binning) . A CMOS imager that averages signals from a neighborhood of pixels has been demonstrated in "Programmable Multiresolution CMOS Active Pixel Sensor", in Solid-state - 2 -

Sensor Arrays & CCD Camera, Proc. SPIE vol. 2654, pp. 72- 81, 1996, by Panicacci, et al.

A CMOS imager with frame memory and pixel binning has been demonstrated in a reference titled, "Frame- transfer CMOS Active Pixel Sensor with Pixel Binning' , special issue on Solid-State Image Sensors, IEEE Trans. On Electron Devices, vol. 44 (10), pp. 1759-1763, 1997, authored by Pain, Zhou and Fossum.

Summary The present disclosure is directed to an improved pixel-binning imager. In accordance with a preferred embodiment, the imager may be easily configured to provide an imager having multi-resolution capability where SNR can be adjusted for optimum low-level detectibility.

Further in accordance with the preferred implementation, multi-resolution signal processing functionality is provided on-chip to achieve high speed imaging, as well as low power consumption.

An imager architecture described preferably has an improved pixel binning approach with fully differential circuits situated so that all extraneous and pick-up noise is eliminated. Unlike the frame-transfer APS with pixel binning, the current implementation minimizes the necessary memory, thereby reducing chip size. The reduction in area enables larger area format light adaptive imager implementations.

Brief Description of the Drawings

Fig. 1 is a block diagram of an APS imager with on-chip variable resolution in accordance with the present invention. - 3 -

Fig. 2 is a more detailed schematic diagram of the APS imager shown in Fig.l.

FIG. 3 shows an exemplary timing diagram for generating two adjacent 3x2 kernels using the imager of the present invention.

Fig. 4 graphically shows expected results of signal detectibility as a function of kernel size achievable by the imager of the present invention,

Description of the Preferred Embodiments Fig. 1 is a block diagram of an APS imager 10 provided with on-chip variable resolution. Imager 10 is a variable resolution CMOS active pixel imager that enhances SNR at low illumination level. In the preferred embodiment, imager 10 is an x-row by y-column photosensitive array 15 of pixels 17. The imager includes a controller that is programmable to read out any sized kernel 18, where a kernel is an n-row by m- column block of pixels. Each kernel 18 represents the summed value of all the pixel values in its region. In the illustrative example, the kernel 18 is a 3-row by 2- column sized region in array 15.

Imager 10 has a sensor row decoder 20 at the side of array 15. An entire row of pixels is selected for readout at each time. Each pixel 17 is preferably a photogate-type active pixel as shown in Fig. 2, with an in-chip buffer circuit MP2 that is controlled by a photogate transfer signal (TX) enabling readout of integrated charge by lowering the charge barrier. A reset signal (RSTP) and select signal (row) enable the buffered pixel signal to drive the associated column output line, all in a conventional manner. - 4 -

A column integration array 30 is connected in parallel to the row outputs. An individual column integrator 35 is located at the bottom of the array, associated with each column. Each integrator 35 is associated with a row of pixels. These collectively feed associated column memory 40, which is constituted by y- columns of associated column memory circuits 45, e.g. capacitors. Figure 2 shows details of the single column integrator 35 being coupled to its associated column memory capacitor circuit 45. The signal output from the pixel 17, is connected to the column integrator portion 35 in a differential manner. Two totally different paths to the differential opamp A are shown. One path is through the signal transistor MS, controlled by the control signal PHIS. This leads the signal through capacitor CMS, where it is again controlled through second transistor MMS . A totally parallel path for the reset signal goes through transistor MR, CMR, and MMR. To enable correlated double sampling, a crowbar circuit CB is also provided. The opamp A is connected as a fully differential switched capacitors integrators. Column memory capacitor circuit 45 uses capacitors CLS and CLR and switches MC9 and MC10 for signal and reset levels, respectively. It should be appreciated that an imager 10 with for example 512 more columns of pixels than another imager will possess 512 more column integrators and column memory capacitor circuit pairs 50. Each pair 50 serves a particular column. In the exemplary embodiment to be described in greater detail below, imager 10 is read out one row at a time. Therefore, at a given time, one row of pixels is sampled simultaneously at the bottom of the column, using appropriate column decoder select logic 60 shown in block - 5 - diagrammatic form in Fig. 1. Each pixel 16 across a sampled row is first sampled on the capacitors CMR and CMS inside associated column integrator 35 connected to a particular column. The signal PHIS is pulsed to bring the signal level onto the capacitor CMS, and the signal PHIR is momentarily pulsed to bring the reset level onto the capacitor CMR. These levels are then coupled to the opamp A by pulsing the RDC signals. The opamp differentially integrates using the integrating capacitors CIS and CIR. Double-sampling is carried out using the crowbar switch CB as described later herein. This process continues until all of the rows in a given kernel are summed. The size of the column integrator scales linearly with the number of columns. For instance, assuming a 1024x1024 imager with a 10 μm pixel, there are 1024 column-integrators 35 and the width of each column- integrator is 10 μm. The total width of column integrator assembly 30 is then 10x1024 μm = 10 cm. The length (height) of CIA 30 is of course fixed at around 2 mm.

As explained above, each column memory capacitor circuit 45 is constituted by capacitor pair CLS and CLR. There are only as many capacitors as there are columns (y-columns) in the array 15.

Unlike the architecture in a frame-transfer type APS imager, this system requires only as many integrators as there are lines, and hence provides significant advantages over previous systems that required a memory for the entire frame of lines and rows.

In the case of a 1024x1024 imager, only 1024 CLSs (and CLRs) are necessary. Pixel binning is still employed but accomplished without use of an entire stored - 6 - frame of pixel values. This is possible, because in full resolution mode, once a row of information makes its way through the CIA block 30 and is sampled in CLS and CLR, the information is immediately read out through the output block by successively turning on CS for successive columns .

CS is essentially a column select signal generated by column select logic 60 that causes the charges stored in CLS and CLR in a selected column to be available to a common global output integrator 70. Global output integrator 70, explained in greater detail below, is a fully differential charge-to-voltage conversion transimpedance amplifier (TIA) . Data from each row of the kernel is read out before the next row is sampled on CMS and CMR.

In accordance with a preferred implementation, pixel averaging happens in the row direction first. For instance, assuming a 3x4 kernel selected size, values (signal & reset levels) from three different rows (same column) are successively sampled on CMS and CMR, respectively. Every sampling is followed by an integration. After three cycles, capacitors CIS and CIR hold the accumulated result from the three rows of the kernel. In order to accumulate the signal from these three rows, column-integrators are reset only every third cycle — i.e. RSTC is closed not in every cycle (since that would erase that data from CIS and CIR) , but every third cycle, allowing the capacitors to add to the previous signal. The entire process happens in column parallel fashion, generating summation in the row direction. In order to generate the 3x4 kernel, integration in the column direction needs to be carried out. This is done during readout by simultaneously - 7 - closing four adjacent CS signals. Each closure of CS causes the accumulated row values from four neighboring columns to be summed together in the global output integrator 70. It should be readily appreciated that associated column integration switches are embedded in the column select logic 60. An alternative method is to progressively close CS switches, and then successively integrate the adjacent column values in COS and COR. The only difference then is that RSTO, the reset switch in the output stage, is closed every fourth cycle (for a 3x4 kernel) , instead of every cycle, as might typically be expected.

The kernels can be placed anywhere in the array, and are selected through row and column decoders. Hence, the size and direction of kernels is fully user programmable. In a preferred implementation, the rows and columns are selected in sequential order, however this is easily user definable, and any order could be selected. Random access is a matter of which row and column decoders are selected. Programming the size of the summation kernels is essentially determined by switching.

A timing diagram for generating two adjacent 3x2 kernels is shown in Fig. 3.

During pixel readout, column integrators 35 are reset by pulsing RSTC and RSTC1 high. At that point amplifier A offsets are stored on the capacitors CIS and CIR. When Vcm is the common-mode voltage, during the reset phase, the amplifier A offsets force the input of the amplifiers to be: Vcm + Voffs, and Vcm + Vo£fr, respectively, where identifiers r and s refer to the signal and reset side, the two branches of amplifier A. For reference purposes, signal side shall be at the top and reset side at the bottom (see Fig. 2) . During reset, the potentials across the capacitors are: V+ - vcm - Voffs, and V+ - Vcπι - Voffr, respectively. In the next phase, RSTCIB is turned high, connecting the capacitors (CIS and CIR) across the amplifier A. Since the amplifier input nodes remain at the previous levels (i.e., V^ + Voffs, and Vcm + Voffr, respectively) , voltages at the output of amplifier A become V+, and are independent of Voffs and Voffr, indicating the outputs are free from offset. Offset elimination is extremely critical, since one or more columns are summed together. In presence of offset that will vary from one column to another, this will result in an unacceptably high fixed pattern noise in the multi- resolution output.

Following the offset correction phase, ROW 1 is selected. The reset and the signal levels from the pixels in that row are sampled by enabling PHIR and PHIS respectively as shown in Fig. 3. ROW 1 is the first row of the kernel, and can be located anywhere in the pixel array. The signals, sampled on CMR and CMS respectively, are VR and VS. The integration is completed by closing the rowbar CB thereby averaging the contents of the two capacitors CMS and CMR. This causes amplifier A outputs to become:

v„(D=vor(o)-^.^S.

where numbers in parenthesis denote values at the end of a cycle. Identifier 0 is reserved for the reset cycle. As a result of this operation, a signal of (VR - Vs) is added differentially to both sides of the amplifier branches, - 9 - causing the voltage on CIS to go up by 0.5* (VR - Vs) and the voltage on CIR to go down by the same amount. As can be seen from the timing diagram, this process is carried out three (3) times to generate summation for pixels from a given column and three successive rows.

Following the row summing, LDC is pulsed low, sampling the row accumulation signals onto CLS and CLR respectively. At this point the column integrators are reset, preparing them for the next cycle of kernel summing. In order to read two kernels, each of 3x2 size, the successive column selects (CSl through CS4) are pulsed high. RSTO is used to reset the global amplifier (A0+ and A0-) . In order to sum 2 columns, RSTO is pulsed high before CSl and CS2 are pulsed, ensuring one kernel of 3x2 size is ready for readout. Following this, RSTO is pulsed again to prepare for generation of the second kernel sum, which proceeds along the same lines, except that CS3 and CS4 are pulsed successively, instead of CSl and CS2.

Each column integrator 35 generally described above includes a fully differential switched-capacitor integrator, a pair of column memory capacitors, CLR and CLS, and the MOS switches (MS, MR, CB, MMS, MMR, MCI, MC2, MC3, MC4, MC7 and MC8) needed for the integration operation. The sample and hold capacitors, CMR and CMS, for the pixel reset and signal levels serve as the input capacitors for each column integrator 35. The column memory capacitors, CLS and CLR, are the input capacitors for the global output integrator 70.

The global output integrator 70 uses two matched single-ended two-stage opamps . The illustrative implementation is configured to drive 30 pF and 1 MW load at above 8 Mpixels/sec required for 30 frames/sec readout - 10 - of a 512 x 512 element array. Amplifier A is a folded cascade opamp with switched capacitor common mode feedback circuit. Its operation is set for much lower speed due to the column parallel readout. The designed 2 MHZ unit gain frequency and 60 dB DC gain are sufficient for column parallel integrator settling with better than 9-bit accuracy. The amplifier design is optimized, in a conventional manner, to use minimum transistor size and lowest bias current.

For an n x m (n columns and m rows) kernel summation readout, signals from m rows of the sensor pixel are integrated by the column integrators 35 one row at a time. The reset and signal levels of each row are first sampled on the S/H capacitors CMS and CMR as the integrators 35 are reset. They are then differentially integrated on the integrating capacitors CIS and CIR. This process continues until all the rows in a given kernel are summed. The reset level pertains to the output of the pixel (at the input to the COLUMN block) when it is reset (when RSTP is turned on) , and the signal level pertains to that the output of the pixel after the signal charges have been dumped in the sense node.

The integrated signals are sampled and accumulated on the column memory capacitors CLS and CLR. After the row summation is completed, every n consecutive columns are integrated after each reset of the global integrator 70. The summed signals from n x kernels are read out serially from the output of the global integrator 70. The summation kernel size is programmable according to the illumination condition. - 11 -

By using a square kernel size of n x n, the S/N enhancement is τ/~n . At low illumination, S/N enhancement is greater than n since the circuit read noise dominates in the imager noise.

The column-wise high residual fixed pattern noise (FPN) is mostly caused by the column opamp offset. In the fully differential readout, the offset is first sampled on the feedback capacitors as the integrator is auto-zeroed. To first order, it is compensated at each step of signal integration. Clock feedthrough appears as common mode pulse to the integrator and does not contribute to FPN. Residual FPN is due to the capacitor ratio mismatch on the two sides of the integrator and is given by,

(1)

where m is the number of row summation, Da is the mismatch in capacitor ratio and Vc is the common mode voltage. The temporal read noise consists of noise from the pixel, the detector shot noise, noise associated with switching (kTC noise) and noise from the opamps . The output referred noise for n x m kernel summation can be approximated by,

(y^ a n^ 2kT a^lma1 + a + β + 2m(\ + a)β + mg 2 N}

'U 2 )

CMR = CMS = CM; Cm = CB = ; Cm » . * €L; C0R - Cos = C0; a = ^- = ^- ;

C where Λ = — ¥-

" r L g is the conversion gain measured in volts/electrons; and is the average number of electrons per pixel during a single exposure. The noise - 12 - voltage at full resolution readout is estimated to be about 320 MV for 125 frames/sec image readout rate, which is very close to the measured value. Column-wise fixed pattern noise is caused by the mismatch between the two branches of the amplifier, caused by threshold voltage mismatches. This causes the two differential outputs to be unbalanced around the common-mode level (the ideal average of the output signals) . This is corrected by sampling the unbalance in CIS and CIR during reset (RSTC is high) .

In another preferred embodiment, a 128 x 128 prototype sensor was implemented by using a 1.2 μm single poly, double metal, n-well process with linear capacitor option. The sensor pixel size was 24 μm x 24 μm with an optical fill factor of 29%. The column circuit was laid out in the 24 mm column pitch and had a total length of about 0.9 mm. The total chip area was about 4.7 mm x 5.2 mm. The fabricated parts were tested up to 125 frames/sec. The tested readout speed was limited by the capability of the pulse generator and the data acquisition board used in the test bed. The characterization results are summarized in the following table.

- 13 -

Table 1 - Summary of the test results.

Imager format 128 x 128

Integrator linearity: better than 8 bit out of 1.8 V swing Sensor saturation: 1.2 V

Temporal Noise: 303 μV r.m.s.

Dynamic range: 72 dB (disregarding FPN)

Conversion gain: 8.3μV/e

Power consumption: 24 W @ 125 frames/sec FPN: 6 mV

Dark current: 54 mV/sec (0.6 nA/cm2

The sensor demonstrates 1.2 V saturation signal, 72 dB dynamic range and 8.3 mV/e- conversion gain. The FPN is about 6 mV (0.5% saturation), read noise 300 MV and dark current 0.6 nA/cm2. More than 40% of the total 24 mW power is consumed by the global integrator 70 opamps due to the required driving capability.

Figure 4 shows the detailed measurement for signal (mV) and SNR(dB) enhancement as the kernel size is increased from 1 x 1 to 2 x 8 at constant illumination and exposure time. The output signal linearity over 1.2 V range indicates good accuracy of the row and column summation. An 11-dB SNR improvement is achieved, as expected by the theoretical prediction from Equation 2. In sum, a multi-resolution APS for light adaptive imaging applications has been demonstrated by successfully integrating fully differential opamp based - 14 - integrator circuits. Good uniformity and low readout noise was achieved. Enhancement of SNR at low light level was demonstrated to have been achieved by programmable multi-resolution readout at constant frame rate .

The fully differential implementation presented above suppresses clock feedthrough and all other sources of common mode noise, particularly substrate coupling and capacitive coupling noise. Substrate coupling noise arises from variations on power, ground and signal lines, and its magnitude can be significant due to reduction in substrate resistance in nearly all advanced sub-micron CMOS process technologies. In the differential topology presented, the effects of these noise phenomena are suppressed by ensuring that only the difference signal can pass through the circuit. For instance, when signal is sampled on CMS and CMR, charge feedthrough from the switches MS and MR are bound to happen. However, since the switch and capacitance sizes are the same, the feedthrough voltage (Vf) will be nearly the same in both capacitors, especially for small signals where the signals sampled are close to each other. Because the circuit amplifies the difference signal, the common-mode feedthrough (Vf) ) does not affect the circuit operation. In accordance with the presently disclosed implementation, low-noise is achieved by using fully differential circuits so that all extraneous and pick-up noise is eliminated. Low-noise is critical for light-adaptive imaging, since without low noise, pixel binning will not produce sqrt(N) improvement in SNR for low-light level imaging. It should be appreciated that SNR enhancement for an n x n pixel is not sqrt(n) but n, since n2 pixels are involved. - 15 -

The imager disclosed herein has extended low-light detection capability, achieved by trading spatial resolution for increased SNR. Unlike the prior art multiresolution chip which produces averaged output and is therefore not suitable for low-light-level signal detection, imager 10 is well adapted for low-light-level signal detection. In the prior art imager, 6 dB attenuation to the signal was calculated which significantly impairs low-light-level signal detection. Imager 10 does not suffer from such attenuation.

When compared to the prior art frame-transfer APS imager with pixel-binning, the current implementation requires no frame buffer memory, thereby reducing the chip size by as much as three fold or more, while preserving pixel-binning capability and reducing read noise. The reduction in area enables large format light adaptive imager implementation without chip size posing limitations, as in previous imager architectures. In addition, unlike CCD or off-chip summation solutions, the current approach provides orders of magnitude lower power due to use of CMOS imaging technology and low-power analog signal processing circuits . Particularly when compared to known off-chip summation approaches, the preferred implementation provides high speed low-light level data due to reduction of data volume through pixel binning.

Although only a few embodiments have been described in detail below, those having ordinary skill in the art would certainly understand that many modifications are possible in the preferred embodiment without departing from the teachings thereof.

All such modifications are intended to be encompassed by the following claims.

Claims

- 16 - What is claimed is:
1. An active pixel sensor (APS) imager comprising: a semiconductor substrate; a pixel sensor array of (x) rows and (y) columns of pixels, formed on said substrate; a neighborhood selector, formed on said substrate, selecting a kernel of n by m pixels in said array; a column integrator array, formed on said substrate, defined by a plurality of column integrators, said column integrators coupled to corresponding columns of pixels and operating to differentially integrate pixel signals on each selected row for each pixel block; a column memory array, formed on said substrate, defined by an array of differentially integrated capacitor circuits, said capacitor circuits summing the differentially integrated pixel signals from the column integrators to generate, at the end of a row summing cycle, a row summed charge signal indicative of said n pixels in said row; and a global output integrator, formed on said substrate, receiving said row summed charge signal for each of said m columns, and generating a pixel block summed output by summing said (m) signals.
2. The imager of claim 1, wherein said pixel sensor array includes a plurality of pixels, each of which include a photosensor element, and a buffer.
3. The APS imager of claim 1, wherein said neighborhood selector is variable, enabling multiple resolutions of mode imager. - 17 -
4. The APS imager of claim 1, wherein each column integrator is a differential circuit including a pair of integrating capacitors configured to substantially eliminate extraneous and pick-up noise.
5. The APS imager of claim 4, wherein each capacitor circuit is a fully differential circuit including a pair of summing capacitors.
6. The APS imager of claim 4, wherein each global output integrator is a fully differential circuit including a pair of opamps .
7. The APS imager of claim 1, wherein each pixel block is a square block.
8. The APS imager of claim 1, further comprising column select logic and row select logic generating addressing and timing control signals on the basis of a presently selected resolution mode.
9. The APS imager of claim 1, further comprising column select logic and row select logic generating addressing and timing control signals on the basis of a presently selected pixel block size.
10. A method of operating a pixel sensor, comprising: obtaining a pixel sensor imager including a pixel sensor array of (x) rows and (y) columns of pixels; defining a sub block of pixels, of (n) rows and (m) columns of neighboring row and column pixels, where n is less than x and m is less than y; - 18 - differentially integrating pixel signals on each of said n rows, using a column integrator array defined by an array of column integrators, said column integrators coupled to said columns of pixels; summing, said n rows, using a differential integrator, to generate end of a row summing cycle, an (n) row summed charge signal; and generating a summed output, using a global output integrator, by summing (m) of the n-row charge signals at the end of a column summing cycle.
11. The method of claim 10, further comprising generating addressing and timing control signals on the basis of a presently selected resolution mode.
12. The APS imager of claim 10, further comprising column select logic and row select logic generating addressing and timing control signals on the basis of a presently selected pixel block size.
PCT/US1999/005830 1998-03-16 1999-03-16 Cmos integration sensor with fully differential column readout circuit for light adaptive imaging WO1999048281A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US7817298P true 1998-03-16 1998-03-16
US60/078,172 1998-03-16

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
EP99912623A EP1086578A4 (en) 1998-03-16 1999-03-16 Cmos integration sensor with fully differential column readout circuit for light adaptive imaging
AU30961/99A AU3096199A (en) 1998-03-16 1999-03-16 Cmos integration sensor with fully differential column readout circuit for lightadaptive imaging
JP2000537366A JP3708824B2 (en) 1998-03-16 1999-03-16 CMOS integrated sensor with fully differential column readout circuit for optical adaptive imaging

Publications (1)

Publication Number Publication Date
WO1999048281A1 true WO1999048281A1 (en) 1999-09-23

Family

ID=22142387

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1999/005830 WO1999048281A1 (en) 1998-03-16 1999-03-16 Cmos integration sensor with fully differential column readout circuit for light adaptive imaging

Country Status (4)

Country Link
EP (1) EP1086578A4 (en)
JP (1) JP3708824B2 (en)
AU (1) AU3096199A (en)
WO (1) WO1999048281A1 (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2803067A1 (en) * 1999-12-23 2001-06-29 Intermec Scanner Technology Ct Optoelectronics device and method code entry by means of a two-dimensional sensor useful OPTIMIZED dimension
EP1117248A2 (en) * 1999-10-02 2001-07-18 Philips Electronics N.V. Method of reading the detectors of a sensor matrix and sensor device
WO2002025934A2 (en) * 2000-09-25 2002-03-28 Sensovation Ag Image sensor device, apparatus and method for optical measurements
WO2003036940A1 (en) * 2001-10-24 2003-05-01 Foveon, Inc. Aggregation of active pixel sensor signals
WO2005050977A1 (en) 2003-11-13 2005-06-02 Micron Technology, Inc. Pixel signal binning and interpolation in column circuits of a sensor circuit
US7268814B1 (en) * 1999-10-05 2007-09-11 California Institute Of Technology Time-delayed-integration imaging with active pixel sensors
WO2008016504A1 (en) * 2006-07-31 2008-02-07 Hewlett-Packard Development Company, L.P. Adaptive binning method and apparatus
WO2009008580A1 (en) * 2007-07-10 2009-01-15 Samsung Electronics Co, . Ltd. Imaging apparatus and method of improving sensitivity of the same
US7808537B2 (en) 2006-09-07 2010-10-05 Canon Kabushiki Kaisha Photoelectric conversion apparatus with fully differential amplifier
US9270868B2 (en) 2005-03-15 2016-02-23 Hewlett-Packard Development Company, L.P. Charge coupled device
WO2017133736A3 (en) * 2016-02-07 2017-09-28 Tichawa Vision Gmbh Method for line-by-line image scanning
US10308702B2 (en) 2000-06-06 2019-06-04 Bristol-Myers Squibb Comapny BSL2v2c2-Ig polypeptides

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6423961B1 (en) * 2000-01-24 2002-07-23 Motorola, Inc. Pixel readout switched capacitor buffer circuit and method therefor
CN104516135B (en) * 2015-01-21 2017-06-06 京东方科技集团股份有限公司 A kind of display methods of display panel, display device and display device

Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5402171A (en) * 1992-09-11 1995-03-28 Kabushiki Kaisha Toshiba Electronic still camera with improved picture resolution by image shifting in a parallelogram arrangement
US5434620A (en) * 1991-02-22 1995-07-18 Nippondenso Co., Ltd. Image sensor
US5471515A (en) * 1994-01-28 1995-11-28 California Institute Of Technology Active pixel sensor with intra-pixel charge transfer
US5841126A (en) * 1994-01-28 1998-11-24 California Institute Of Technology CMOS active pixel sensor type imaging system on a chip
US5877715A (en) * 1997-06-12 1999-03-02 International Business Machines Corporation Correlated double sampling with up/down counter
US5883668A (en) * 1995-12-05 1999-03-16 Olympus Optical Co., Ltd. Solid-state image pickup apparatus
US5892540A (en) * 1996-06-13 1999-04-06 Rockwell International Corporation Low noise amplifier for passive pixel CMOS imager
US5896173A (en) * 1995-07-07 1999-04-20 Siemens Aktiengesellschaft Image detector with selective readout of pixels in groups or individually
US5900623A (en) * 1997-08-11 1999-05-04 Chrontel, Inc. Active pixel sensor using CMOS technology with reverse biased photodiodes
US5909026A (en) * 1996-11-12 1999-06-01 California Institute Of Technology Integrated sensor with frame memory and programmable resolution for light adaptive imaging
US5917547A (en) * 1997-07-21 1999-06-29 Foveonics, Inc. Two-stage amplifier for active pixel sensor cell array for reducing fixed pattern noise in the array output
US5920345A (en) * 1997-06-02 1999-07-06 Sarnoff Corporation CMOS image sensor with improved fill factor

Patent Citations (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5434620A (en) * 1991-02-22 1995-07-18 Nippondenso Co., Ltd. Image sensor
US5402171A (en) * 1992-09-11 1995-03-28 Kabushiki Kaisha Toshiba Electronic still camera with improved picture resolution by image shifting in a parallelogram arrangement
US5471515A (en) * 1994-01-28 1995-11-28 California Institute Of Technology Active pixel sensor with intra-pixel charge transfer
US5841126A (en) * 1994-01-28 1998-11-24 California Institute Of Technology CMOS active pixel sensor type imaging system on a chip
US5896173A (en) * 1995-07-07 1999-04-20 Siemens Aktiengesellschaft Image detector with selective readout of pixels in groups or individually
US5883668A (en) * 1995-12-05 1999-03-16 Olympus Optical Co., Ltd. Solid-state image pickup apparatus
US5892540A (en) * 1996-06-13 1999-04-06 Rockwell International Corporation Low noise amplifier for passive pixel CMOS imager
US5909026A (en) * 1996-11-12 1999-06-01 California Institute Of Technology Integrated sensor with frame memory and programmable resolution for light adaptive imaging
US5920345A (en) * 1997-06-02 1999-07-06 Sarnoff Corporation CMOS image sensor with improved fill factor
US5877715A (en) * 1997-06-12 1999-03-02 International Business Machines Corporation Correlated double sampling with up/down counter
US5917547A (en) * 1997-07-21 1999-06-29 Foveonics, Inc. Two-stage amplifier for active pixel sensor cell array for reducing fixed pattern noise in the array output
US5900623A (en) * 1997-08-11 1999-05-04 Chrontel, Inc. Active pixel sensor using CMOS technology with reverse biased photodiodes

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP1086578A4 *

Cited By (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1117248A2 (en) * 1999-10-02 2001-07-18 Philips Electronics N.V. Method of reading the detectors of a sensor matrix and sensor device
EP1117248A3 (en) * 1999-10-02 2001-09-12 Philips Electronics N.V. Method of reading the detectors of a sensor matrix and sensor device
US7268814B1 (en) * 1999-10-05 2007-09-11 California Institute Of Technology Time-delayed-integration imaging with active pixel sensors
US7143941B2 (en) 1999-12-23 2006-12-05 Intermec Ip Corp. Optoelectronic device and process for acquiring symbols, such as bar codes, using a two-dimensional sensor
EP1115085A1 (en) * 1999-12-23 2001-07-11 Intermec Scanner Technology Center Barcode reader with a two dimensional sensor and adjustable scanning
US7413125B2 (en) 1999-12-23 2008-08-19 Intermec Ip Corp. Optoelectronic device and process for acquiring symbols, such as bar codes, using a two-dimensional sensor
FR2803067A1 (en) * 1999-12-23 2001-06-29 Intermec Scanner Technology Ct Optoelectronics device and method code entry by means of a two-dimensional sensor useful OPTIMIZED dimension
US6732930B2 (en) 1999-12-23 2004-05-11 Intermec Ip Corp. Optoelectronic device and process for acquiring symbols, such as bar codes, using a two-dimensional sensor
US10308702B2 (en) 2000-06-06 2019-06-04 Bristol-Myers Squibb Comapny BSL2v2c2-Ig polypeptides
WO2002025934A2 (en) * 2000-09-25 2002-03-28 Sensovation Ag Image sensor device, apparatus and method for optical measurements
WO2002025934A3 (en) * 2000-09-25 2002-12-12 Paul Anthony Hing Image sensor device, apparatus and method for optical measurements
US7839450B2 (en) 2000-09-25 2010-11-23 Sensovation Ag Image sensor device, apparatus and method for optical measurements
US6794627B2 (en) 2001-10-24 2004-09-21 Foveon, Inc. Aggregation of active pixel sensor signals
WO2003036940A1 (en) * 2001-10-24 2003-05-01 Foveon, Inc. Aggregation of active pixel sensor signals
US7319218B2 (en) 2003-11-13 2008-01-15 Micron Technology, Inc. Method and apparatus for pixel signal binning and interpolation in column circuits of a sensor circuit
WO2005050977A1 (en) 2003-11-13 2005-06-02 Micron Technology, Inc. Pixel signal binning and interpolation in column circuits of a sensor circuit
US7154075B2 (en) 2003-11-13 2006-12-26 Micron Technology, Inc. Method and apparatus for pixel signal binning and interpolation in column circuits of a sensor circuit
US9270868B2 (en) 2005-03-15 2016-02-23 Hewlett-Packard Development Company, L.P. Charge coupled device
WO2008016504A1 (en) * 2006-07-31 2008-02-07 Hewlett-Packard Development Company, L.P. Adaptive binning method and apparatus
US7808537B2 (en) 2006-09-07 2010-10-05 Canon Kabushiki Kaisha Photoelectric conversion apparatus with fully differential amplifier
WO2009008580A1 (en) * 2007-07-10 2009-01-15 Samsung Electronics Co, . Ltd. Imaging apparatus and method of improving sensitivity of the same
WO2017133736A3 (en) * 2016-02-07 2017-09-28 Tichawa Vision Gmbh Method for line-by-line image scanning

Also Published As

Publication number Publication date
EP1086578A4 (en) 2001-09-26
AU3096199A (en) 1999-10-11
JP3708824B2 (en) 2005-10-19
JP2002507863A (en) 2002-03-12
EP1086578A1 (en) 2001-03-28

Similar Documents

Publication Publication Date Title
US6166367A (en) Programmable analog arithmetic circuit for imaging sensor
JP5114493B2 (en) Method for generating an image in electronic form, image element for image sensor for image generation and image sensor
EP1078509B1 (en) Active pixel sensor with capacitorless correlated double sampling
KR100555609B1 (en) Cmos image sensor with improved fill factor
US6067113A (en) Offset noise cancellation in array image sensors
KR100555608B1 (en) Cmos image sensor with reduced fixed pattern noise
US5631704A (en) Active pixel sensor and imaging system having differential mode
US5461425A (en) CMOS image sensor with pixel level A/D conversion
US5909026A (en) Integrated sensor with frame memory and programmable resolution for light adaptive imaging
US7245321B2 (en) Readout circuit with gain and analog-to-digital conversion for image sensor
US7402789B2 (en) Methods for pixel binning in an image sensor
US7542086B2 (en) Solid-state image-sensing device
US6518910B2 (en) Signal processing apparatus having an analog/digital conversion function
US20050128327A1 (en) Device and method for image sensing
US7154078B2 (en) Differential column readout scheme for CMOS APS pixels
JP2006020171A (en) Differential comparator, analog/digital converter, imaging apparatus
KR20110109891A (en) Solid-state image pickup apparatus, signal processing method for a solid-state image pickup apparatus, and electronic apparatus
KR100572206B1 (en) dc offset and gain correction method and system for image sensors Cmos
EP1271930B1 (en) Image sensing apparatus capable of outputting image with converted resolution, its control method, and image sensing system
US6844896B2 (en) Modification of column fixed pattern column noise in solid state image sensors
EP2109306B1 (en) Pixel array with global shutter
JP2965777B2 (en) The solid-state imaging device
EP0933928A2 (en) Active pixel sensor read-out channel
US5886659A (en) On-focal-plane analog-to-digital conversion for current-mode imaging devices
US20030137594A1 (en) Image sensing apparatus, signal detection apparatus, and signal accumulation apparatus

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): AE AL AM AT AU AZ BA BB BG BR BY CA CH CN CU CZ DE DK EE ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MD MG MK MN MW MX NO NZ PL PT RO RU SD SE SG SI SK SL TJ TM TR TT UA UG UZ VN YU ZA ZW

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): GH GM KE LS MW SD SL SZ UG ZW AM AZ BY KG KZ MD RU TJ TM AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE BF BJ CF CG CI CM GA GN GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
NENP Non-entry into the national phase in:

Ref country code: KR

ENP Entry into the national phase in:

Ref country code: JP

Ref document number: 2000 537366

Kind code of ref document: A

Format of ref document f/p: F

WWE Wipo information: entry into national phase

Ref document number: 1999912623

Country of ref document: EP

REG Reference to national code

Ref country code: DE

Ref legal event code: 8642

WWP Wipo information: published in national office

Ref document number: 1999912623

Country of ref document: EP

WWW Wipo information: withdrawn in national office

Ref document number: 1999912623

Country of ref document: EP