Radiant energy – Infrared-to-visible imaging
Reexamination Certificate
1999-10-13
2002-05-07
Dang, Hung Xuan (Department: 2873)
Radiant energy
Infrared-to-visible imaging
C250S581000, C250S584000
Reexamination Certificate
active
06384413
ABSTRACT:
BACKGROUND
The present disclosure relates, in general, to image sensors and, in particular, to focal plane infrared readout circuits.
In general, image sensors find applications in a wide variety of fields, including machine vision, robotics, guidance and navigation, automotive applications, and consumer products. Imaging systems that operate in the infrared (IR) wavelength region are required for a number of space-based or airborne applications such as monitoring global atmospheric temperature profiles, relative humidity profiles, cloud characteristics and the distribution of minor constituents in the atmosphere. Such IR imaging systems also can be used for fire prevention and control, for enhanced visibility in foggy conditions, and for spectroscopic applications.
In general, focal planes used in imaging applications typically operate with a relatively long exposure or integration time to improve the signal-to-noise ratio. However, for airborne applications, using a long exposure time represents a trade off between tolerable motion blur and the required signal-to-noise ratio. The required signal-to-noise ratio for airborne spectrometers is usually quite strict because the signal collected by each photodetector is reduced by the number of spectral channels.
Time-delay-integration (TDI) is one technique for achieving a low noise level by increasing the effective integration time without introducing motion blur. However, the use of a mechanical scanning arrangement for imaging spectrometry applications precludes the use of TDI techniques for enhancing the signal-to-noise ratio. Instead, the size of the photodetectors is increased. The increased detector size, however, results in a higher detector capacitance.
FIG. 1
illustrates a photodetector represented schematically by a current source in parallel with a variable resistor and a capacitor. A large detector resistance (R
d
) allows the direct current (dc) component of the injection efficiency to be nearly unity. A high detector capacitance (C
d
) reduces the cut-off frequency for the injection efficiency, causing problems for circuit operation at high speeds or short exposure times. In particular, a large detector capacitance prevents the detector node from being charged or discharged at high speed, causing the current signal injected into the readout circuit to lag behind the detector current. Retention of charges on the photodetector caused by a large response time constant produces an output that depends on the amount of detector current from the previous frame. That can result in the appearance of a residual, or “ghost,” image. The presence of such residual images can cause significant errors in the estimation of ground and atmospheric constituents which are based on the ratios of detector outputs. Therefore, achieving a low noise readout at short exposure times for detectors with a large capacitance poses a major challenge for focal plane multiplexer design.
SUMMARY
In general, according to one aspect, an integrated circuit for reading out signals from a sensor includes a buffered direct injection input circuit including a differential amplifier with an injection transistor coupled between an input and output of the differential amplifier to provide active feedback. The differential amplifier can include a pair of input transistors and a pair of cascode transistors, as well as a current mirror load.
The integrated circuit can be used in various applications including infrared imagers, such as spectrometers, that include multiple infrared photodetectors and readout circuits for reading out signals from the photodetectors. Each readout circuit can include a buffered direct injection input circuit.
Various implementations include one or more of the following features. The cascode transistors can be arranged to reduce an input-to-output capacitance of the differential amplifier. For example, each cascode transistor can have a respective gate and source/drain regions. The gates of the cascode transistors can be electrically coupled to a common voltage and a source/drain region of each cascode transistor can be coupled to a respective source/drain region of one of the input transistors.
Each readout circuit can include an integration capacitor so that, during operation, photocurrent from an associated one the sensors is injected onto the integration capacitor through the injection transistor. Additionally, a feedback capacitor can be coupled between the output of the differential amplifier and the input transistor to which the output of the associated one of the sensors is coupled. The differential amplifier can have an effective input capacitance less than a capacitance of the integration capacitor. Furthermore, the integration capacitor can be reset by applying a reset voltage to a gate of a transistor in parallel with the integration capacitor.
A screen transistor can be coupled between the injection transistor and the integration capacitor. During operation, the screen transistor is preferably biased in saturation.
In another aspect, a method of using an imager includes detecting radiation with infrared sensors having respective capacitances at least as high as about 10 picofarads and injecting photocurrent from each sensor into an associated readout circuit having a buffered direct injection input circuit and an input impedance that is smaller than an impedance of the sensor. The photocurrent injected into each readout circuit is sampled at least once every 100 microseconds.
Various implementations include one or more of the following advantages. Photocurrent can be injected onto the integration capacitor with high injection efficiency at high speed. A high speed, low noise, wide dynamic range linear infrared multiplexer array for reading out infrared detectors with large capacitances can be achieved even when short exposure times are used. The effect of image lag can, therefore, be reduced.
To achieve high-speed BDI operation, the amplifier output resistance can be kept low to increase the amplifier cut-off frequency. A relatively high gain can be achieved by using transistors with large size to increase the transconductance and by using FETs with long device length to reduce short channel effects and provide a large output resistance. The cascade configuration can help reduce input-to-output capacitance.
Additionally, the feedback capacitor can help reduce overshoots during rising transitions. The screen FET can help reduce errors caused by channel length modulation.
Other features and advantages will be readily apparent from the following description, accompanying drawings and the claims.
REFERENCES:
patent: 5515003 (1996-05-01), Kimura
Niblack et al., Multiplexed focal plane array upgrade for the A Viris Westrick, 04/93, SPIE vol. 1946 Infrared Detectors and instrumentation.
Blessinger, Comparative study of linear multiplexer designs for a remote sensing application, 04/94, SPIE vol. 2226 Infrared Readout Electronics.
Mendis et al., Low-light-level image sensor with on-chip signal processing, 04/93, SPIE vol. 1952.
Bluzer et al., Current readout of infrared detectors, 03/87, Optical Engineering vol. 26 No. 3.
Nelson et al., General noise processes in hybrid infrared focal plane arrays, 11/91, Optical Engineering vol. 30 No. 11.
California Institute of Technology
Dang Hung Xuan
Fish & Richardson PC
LandOfFree
Focal plane infrared readout circuit does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Focal plane infrared readout circuit, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Focal plane infrared readout circuit will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2858691