FOCUS
ENCODING and decoding images
The TPX3CAM is an advanced intensified hybrid Complementary Metal-Oxide-Semiconductor( CMOS) sensor designed for single-photon detection. This sensor is based on technology originally developed for particle detection at CERN. An optically sensitive silicon sensor is bonded to a specialized fast readout chip. This combination results in a sensor featuring 256 × 256 pixels, each measuring 55μm × 55μm, with exceptionally high quantum efficiency. To achieve single-photon sensitivity, the incoming signal must be amplified. This is accomplished by placing a photon intensifier in front of the sensor. The intensifier consists of a photocathode that converts the incoming light into an electron beam. This electron beam is then amplified using a multichannel plate( MCP) and subsequently converted back into photons by a scintillator. The intensifier has a quantum efficiency of 20 % at 810nm. The photon flash produced during the amplifying process hits more than one pixel on the camera, but centroid algorithms resolve this by identifying the central pixel based on spatial distribution and timing. This sensor distinguishes itself from others on the market with its event-based operating mode, in contrast to the frame-based approach used by most other cameras.
The high-speed readout chip records each pixel individually, assigning a time tag to each photon that strikes the sensor with sufficient energy. Each time tag corresponds to a detection event. The detection process operates as follows: each pixel has its own electrical threshold Vth, and only pixels that exceed this threshold trigger the readout process and time measurement. The chip records the Time of Arrival( TOA) with a precision of 1.56 ns and the Time Over Threshold( TOT) with a precision of 25 ns. These two time tags are subsequently used in the centroid process, enabling single-photon sensitivity. This new technology makes it straightforward to detect temporal coincidences between photons. In post-processing, a temporal correlation window of width ΔT is applied across the entire dataset. Within this window, all pixel positions corresponding to detection events are examined. If at least two events occur within the same temporal window, the pixel positions of these events are identified and recorded. By repeating this process for all time intervals in the dataset, a spatial coincidence map is constructed. This map highlights the positions of pixels where coincident detections occurred, enabling the recovery of spatial correlations between photons with high precision.
Figure 4. a, Tpx3Cam camera setup configuration( the computer is not represented). b, The incoming SPDC photons interact with the intensifier resulting in a flash of photons. Those are detected by the CMOS sensor and the electronic signal is recorded by the fast readout electronics embedded in the sensor. Post processing enable the correlation measurement between the pixels.
time-stamping cameras, such as the Tpx3Cam( see box), are extremely promising. They benefit from a much higher temporal resolution, typically 1-10 ns, higher frame rates- up to 8.10 5 frames per second- and can operate in an event-based detection mode, which considerably simplifies output data processing. Integrated into our experiment, the Tpx3cam enables, for example, the acquisition of correlation images in under 10 seconds with quality comparable to those obtained with the EMCCD, representing a 1000- fold reduction in acquisition time.
Preliminary results obtained with this camera are shown in Figure 3.
These new camera technologies will facilitate the use of our encoding-decoding imaging scheme and gradually lead it towards potential applications. For example, promising avenues we are exploring include the possibility of transmitting images in an absolutely secure manner, bridging the fields of quantum key distribution and imaging. Another key direction is improving image transmission through complex or aberrating media, such as atmospheric turbulence, fog, or biological tissues, where quantum light demonstrates greater robustness than classical signals.
CONCLUSION As seen through the example described in this paper, correlation-based quantum imaging approaches- a subcategory of optical quantum imaging and sensing schemes- represent a new form of imaging with significant potential. However, it is also evident that substantial technological efforts are still required
46 www. photoniques. com I Photoniques 131