Imaging device, control device, and spiking neural network
AssigneeSony Semiconductor Solutions Corporation
InventorsTAKESHI OYAKAWA, SUSUMU HOGYOKU
Life of this patent
Estimating the term…
Abstract
Saturation of an event output is suppressed while a decrease in sensitivity is suppressed. An imaging device includes: a light receiving section that is provided in pixels arranged in a matrix in a row direction and a column direction and outputs a pulse on the basis of incidence of photons; a counter that is provided in the pixels and counts the pulse output from the light receiving section; and a comparator that is provided in the pixels and outputs an event on the basis of a comparison result between a count value by the counter and a counter threshold. The comparator may output the event when the count value exceeds the counter threshold. The comparator may reset the counter when the count value exceeds the counter threshold.
Claims20
An imaging device comprising:
- a light receiving section that is provided in pixels arranged in a matrix in a row direction and a column direction and outputs a pulse on a basis of incidence of photons;
- a counter that is provided in the pixels and counts the pulse output from the light receiving section;
- and
- a comparator that is provided in the pixels and outputs an event on a basis of a comparison result between a count value by the counter and a counter threshold.
A control device comprising
- a control section that receives, as an input of an event, a comparison result between a count value of a pulse output on a basis of incidence of a photon and a counter threshold, and controls the counter threshold on a basis of an output rate of the event.
A spiking neural network comprising:
- a plurality of first spiking neurons, each receiving an input of a pulse generated on a basis of incidence of photons at spatial positions different from each other, and firing on a basis of an input rate of the pulse;
- and
- a second spiking neuron to which firing of the plurality of first spiking neurons is respectively input and that fires on a basis of an input rate of firing of the first spiking neurons.
Dependencies are inferred from the claim text.