Television – Camera – system and detail – Solid-state image sensor
Reexamination Certificate
1999-08-13
2004-02-24
Christensen, Andrew (Department: 2615)
Television
Camera, system and detail
Solid-state image sensor
C348S263000, C348S267000, C257S229000
Reexamination Certificate
active
06697114
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to MOS active pixel sensors. More particularly, the present invention relates to a MOS active pixel sensor having a triple slope light-to-output-voltage transfer gain characteristic.
2. The Prior Art
Integrated image sensors are known in the art. Such sensors have been fabricated from charge-coupled devices (CCDs) and as bipolar and MOS image sensors.
In the CCD imager art, on-chip frame storage capability has been employed previously. It has been motivated by the need to shift sensed charges out during a video frame time without letting them be contaminated by further exposure as the charges travel across the CCD array. Two storage techniques are commonly employed in the CCD imager art. According to the first technique, a second separate on-chip CCD array is provided under a light shield, and the entire image is quickly shifted along one dimension into the storage array, since shifting in one dimension is fast enough to avoid significant contamination. According to the second technique, line-storage CCDs are provided between the lines of sensors, with local light shielding. In the CCD art, techniques have been developed for preventing leakage and contamination due to minority carrier diffusion and leakage, but these techniques are not applicable to the CMOS sensor array because the silicon fabrication processes are different.
CMOS image sensors and image sensor circuitry may be organized in a manner similar to that which is disclosed in co-pending application Ser. No. 08/969,383, filed Nov. 13, 1997. Individual pixel sensors may be designed in a number of different ways. One possible pixel sensor design comprises a photodiode having its anode connected to a fixed voltage potential such as ground. The cathode of the photodiode is connectable to an amplifier. The cathode of the photodiode is also connectable to a reference potential via a reset switch so that the photodiode is reverse biased. The output of the amplifier is attached to a row-select switch, which is connected to a row select line and a column line.
The pixel sensor is first reset by turning on the reset switch. Then the reset switch is then turned off so that integration of photocurrent from the photodiode can begin. The current from the photodiode is integrated on the amplifier input node capacitance to form a voltage signal. At the appropriate time, the voltage on the row select line is raised, which activates the row-select switches in each pixel sensor in the row. This allows the amplifier to drive column line. The column line then leads down to more circuitry that will typically amplify and store the signal, and then convert the signal into digital form for inclusion in a digital pixel stream.
Another possible pixel sensor design provides for storage of the signal within the pixel sensor and is thus referred to as a storage pixel sensor. A storage pixel sensor comprises a photodiode having its anode connected to a fixed voltage potential such as ground. The cathode of the photodiode is connectable to a storage capacitor via a transfer switch. The storage capacitor has a first plate connected to the transfer switch and a second plate connected to a fixed potential such as ground.- The cathode of the photodiode is also connectable to a reference potential via a reset switch so that the photodiode is reverse biased. An amplifier has its input connected to the storage capacitor. The output of the amplifier is attached to a row-select switch, which is connected to a row select line and a column line.
The storage pixel sensor is first reset by turning on both reset switch and transfer switch. Then the reset switch is turned off so that integration of photocurrent from the photodiode can begin. Since the transfer switch is turned on, the capacitance of the storage capacitor adds to the capacitance of the photodiode during integration, thereby increasing the charge capacity and therefore intensity range of the storage-pixel sensor. This also reduces variation in the pixel output due to capacitance fluctuations since gate oxide capacitance from which the storage capacitor is formed is better controlled than junction capacitance of the photodiode.
When the integration is complete (determined by exposure control circuitry, not shown), the transfer switch is turned off, isolating the voltage level corresponding to the integrated photocharge on the storage capacitor. Shortly thereafter, the photodiode itself is reset to the reference voltage by again turning on the reset switch. This action will prevent the photodiode from continuing to integrate during the readout process and possibly overflowing excess charge into the substrate which could affect the integrity of the signal on the storage element.
After the reset switch is turned back on, the readout process can begin. At the appropriate time, voltage on the row select line is raised, which activates the row-select switches in each pixel sensor in the row. This allows the current from the amplifier to travel to the column line. The column line is coupled to more circuitry that will typically amplify the signal, and then convert the signal into digital form for inclusion in a digital pixel stream.
One problem encountered with prior-art imagers is a limitation on the dynamic range of images that can be captured by the array. Images that contain both low-light-level pixels and high-light-level pixels could be improved if the dynamic range of the imager could be increased.
In an active pixel sensor, the sensitivity of measuring charges generated by photons can be described as a charge-to-voltage gain or light-to-output-voltage transfer gain. Typically, in a prior art active pixel sensor, this gain is accounted for by two factors. A first factor is the reciprocal of the capacitance of the charge accumulation node in the sensor where photocharge accumulates to change a potential (a reciprocal capacitance represents units of volts per coulomb). A second factor is the gain of the readout amplifier, typically less than one using a source follower. Voltage dependence of the photodiode capacitance and other capacitances, and nonlinearities of the readout amplifier transistor can make the gain vary with level, so that the overall transfer curve may be somewhat nonlinear. A nonlinearity in which higher light intensities give lower gains is said to be compressive. A significant degree of compressive nonlinearity can have a beneficial effect on the signal-to-noise ratio of the image at low light levels, and can thereby enhance the usable dynamic range of the imager.
It is therefore an object of the present invention to provide a pixel sensor and an array of pixel sensors that overcome some of the shortcomings of the prior art.
A further object of the present invention is to provide a storage pixel sensor and an imaging array of storage pixel sensors that includes image level compression.
Another object of the present invention is to provide a storage pixel sensor and an imaging array of storage pixel sensors that includes multi-breakpoint image level compression.
BRIEF DESCRIPTION OF THE INVENTION
According to the present invention, a storage pixel sensor having built-in compression is disclosed. The pixel sensor of the present invention has a first light-to-output-voltage transfer gain up to a first light integration threshold, a second light-to-output-voltage transfer gain lower than the first light-to-output-voltage transfer gain up to a second light integration threshold, and a third light-to-output-voltage transfer gain less than the second light-to-output-voltage transfer gain after the second light integration threshold. The storage pixel of the present invention may be referred to herein as a triple-slope active pixel sensor and has a larger dynamic range than pixel sensors without this feature.
A triple-slope MOS active pixel sensor disposed on a semiconductor substrate comprises a first capacitive storage element having a first terminal connected to a fixed potential and a second terminal a
Christensen Andrew
Foveon, Inc.
Genco Brian
Sierra Patent Group Ltd.
LandOfFree
Triple slope pixel sensor and arry does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Triple slope pixel sensor and arry, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Triple slope pixel sensor and arry will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3346336