Low-light CCD needs no intensification

One of the challenges in low-light-level (L3) imaging is to reduce the effects of dark-current shot noise and readout noise. While slow-scan imaging systems use cooled detectors to reduce the dark current and dark-current noise, they read out data slowly to reduce system bandwidth and readout noise. For long integration times, these sensors are cooled to cryogenic temperatures to give noise of a few electrons and produce high-quality images.

Using an electric field between electrodes on the output and multiplication registers on a CCD results in impact ionization of charge carriers. This causes amplification of the charge signal, allowing the CCD to image low-light-level images without image intensification.

For TV-rate CCDs, however, L3 performance is not as good as from intensified CCDs. These devices amplify the signal before the noise from the CCD sensor is introduced by using a microchannel-plate (MCP) intensifier coupled to the CCD. Although sensitivity and resolution of image intensifiers have improved over the last few years, this approach has several disadvantages that include poor gain of the MCP, low fixed-pattern signal-to-noise ratio, and reduced contrast in the resulting image.

At the SPIE Annual Meeting (Aug. 2000; San Diego, CA), Marconi Applied Technologies (Elmsford, NY) introduced a novel low-light-level CCD technology, dubbed L3CCD, that allows low-light-level imaging to be performed with CCDs without image intensifiers. Although there are no current pinouts available as the final format of the sensor has not been approved, the front-illuminated, 576-pixel/line, 625-line device has been built into the company's analog L3 camera.

In a typical CCD, the signal charge is collected in an array of pixels. Following an integration period, charge is transferred to an output register. This is then applied to a charge-detection circuit to produce a voltage that represents the signal. The L3CCD patented by Marconi operates in a similar manner (see figure). First, charge is accumulated in pixels of an image area, subsequently transferred to a storage section, and then read out to an output register.

To achieve L3 imaging, Marconi has chosen to multiply the charge carriers within the CCD image area during the transfer process as charge is clocked from the pixel to output registers. This is accomplished by establishing an electric field between pairs of control elements, which in a conventional CCD would be used to collect and move charge through the CCD output registers.

Signal charge in the output register is transferred to the multiplication register by drive pulses applied to two electrodes to produce charge transfer from the CCD. Charge carriers are then accelerated by the large field between the two electrodes so that additional carriers are generated through impact ionization. This results in signal amplification and a CCD that is capable of imaging performance at light levels down to overcast starlight.

By removing the intensifier altogether, modulation transfer function is improved, and image halo and scintillation-effects become available. Unlike intensified systems, the CCD65 sensor is solid-state and is not damaged by overexposure. The device uses advanced inverted-mode operation to reduce dark-current noise and a Peltier cooler to reduce dark current, dark-current shot noise, and fixed-pattern noise.