Smart quantum light detector

A light detection system for detecting light from a classified type of light source, comprising:

• a light detector configured to detect individual photons from light reflected or scattered by an object illuminated by light from a light source;
• and
• a processing system configured to communicate with said light detector to receive signals to be processed,
• wherein said light detector is configured to detect the individual photons for a measurement time period to provide a times series of individual photon events, and
• wherein said processing system is configured to:
• segment said time series into a plurality of time bins;
• determine a number of detected photons within each time bin of said plurality of time bins to provide a time series of photon counts per time bin;
• determine a probability distribution P(n) from said time series of photon counts per time bin, said probability distribution providing a probability of detection of n photons, wherein n=0, 1, 2, . . . , nmax;
• input each of values of P(n) as a nmax+1 component of a feature vector into a single neuron neural network, said single neuron neural network having been previously trained on a plurality of light source types;
• and
• provide as output a classifier that has a value that identifies a light source type,
• wherein an average number of photons in said plurality of time bins is less than one photon, and
• wherein the processor is configured to identify a type of said light source based on the detected individual photons.

An optical imaging system for forming images from a classified type of light source, comprising:

• a plurality of light detectors arranged in a patterned array, each of the plurality of light detectors configured to detect individual photons from light reflected or scattered by an object illuminated by light from a light source;
• and
• a processing system configured to communicate with said plurality light detectors to receive signals to be processed to provide an image from said classified type of light source,
• wherein each of said plurality of light detectors is configured to detect individual photons for a measurement time period to provide a corresponding times series of individual photon events, and
• wherein said processing system is configured, for each of said plurality of light detectors, to:
• segment each said time series into a plurality of time bins;
• determine a number of detected photons within each time bin of said plurality of time bins to provide a corresponding time series of photon counts per time bin;
• determine a probability distribution P(n) from each said time series of photon counts per time bin, said probability distribution providing a probability of detection of n photons, wherein n=0, 1, 2, . . . , nmax;
• input each of values of P(n) as a nmax+1 component of a feature vector into a single neuron neural network, said single neuron neural network having been previously trained on a plurality of light source types;
• and
• provide as output a classifier that has a value that identifies a light source type,
• wherein an average number of photons in said plurality of time bins is less than one photon,
• wherein the processing system is configured to identify a type of said light source based on the detected individual photons.