IP Library › Granted Patent US 11,875,545
Granted Patent B2
US 11,875,545 · App. 16/961,538 · Granted Jan 16, 2024

Method and device for processing asynchronous signals generated by an event-based light sensor

Inventor: Francesco Galluppi (Paris, FR)
Assignee: GENSIGHT BIOLOGICS
G06V10/145A61N1/36128G06V10/44A61N1/36046
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,875,545
App. No.
16/961,538
Granted
Jan 16, 2024
Kind
B2
Abstract

A light sensor has a matrix of sensing elements each providing a respective event-based signal sequence including events asynchronously produced as a function of variations of light incident on the sensing element and light level values respectively associated with at least some of the events. The method generates an image comprising pixels corresponding spatially to the sensing elements of the matrix, and determines a set of edge pixels among the pixels of the image. Pixels are selected in response to events included in the event-based signal sequences and for each selected pixel a local contrast measure is evaluated to decide whether or not the selected pixel belongs to the set of edge pixels.

Claims (34)

1. A method for processing asynchronous signals generated by a light sensor, the light sensor having a matrix of sensing elements, the method comprising:

receiving, from each sensing element of the matrix, a respective event-based signal sequence including events asynchronously produced as a function of variations of light incident on the sensing element and light level values respectively associated with at least some of the events;

generating an image comprising pixels corresponding spatially to the sensing elements of the matrix; and

repeatedly determining a set of edge pixels among the pixels of the image,

wherein generating the image comprises updating each pixel of the image based on a light level value associated with a most recent event in the event-based signal sequence received from the sensing element corresponding to said pixel of the image,

wherein determining the set of edge pixels comprises:

selecting pixels of the image, wherein the selected pixels comprise pixels of the image corresponding to sensing elements from which the received event-based signal sequences include at least one event after a previous determination of the set of edge pixels, and pixels of the previously determined set of edge pixels;

evaluating respective local contrast measures with respect to the selected pixels of the image; and

deciding whether or not the selected pixels belong to the set of edge pixels based on the evaluated local contrast measures.

2. The method of claim 1 , wherein the selected pixels comprise each pixel of the image corresponding to a sensing element from which the received event-based signal sequence includes at least one event after the previous determination of the set of edge pixels.

3. The method of claim 2 , wherein the selected pixels further include each pixel of the previously determined set of edge pixels.

4. The method of claim 2 , wherein the selected pixels consist only of the pixels selected in response to events included in the respective event-based signal sequences received from the sensing elements.

5. The method of claim 1 , wherein the set of edge pixels is determined periodically, with a period in a range of 1 to 100 milliseconds, preferably in a range of 10 to 100 milliseconds.

6. The method of claim 1 , further comprising outputting a contour representation as a matrix of time-varying bits,

wherein each of the time-varying bits of the contour representation corresponds spatially to a respective pixel of the image and has a first value when the respective pixel of the image belongs to the set of edge pixels, and a second value when the respective pixel of the image does not belong to the set of edge pixels.

7. The method of claim 6 , wherein the contour representation is used to control a light modulator.

8. The method of claim 6 , wherein the contour representation is transmitted to a retinal implant for stimulation of retina cells.

9. A signal processing unit comprising:

an interface for connecting to a light sensor having a matrix of sensing elements and receiving, from each sensing element of the matrix, a respective event-based signal sequence including events asynchronously produced as a function of variations of light incident on the sensing element and light level values respectively associated with at least some of the events; and

a processor for generating an image comprising pixels corresponding spatially to the sensing elements of the matrix, and repeatedly determining a set of edge pixels among the pixels of the image,

wherein the processor is configured to update a respective pixel value of each pixel of the image based on a light level value associated with a most recent event in the event-based signal sequence received from the sensing element corresponding to said pixel of the image,

wherein the processor is configured to determine the set of edge pixels by:

selecting pixels of the image, wherein the selected pixels comprise pixels of the image corresponding to sensing elements from which the received event-based signal sequences include at least one event after a previous determination of the set of edge pixels, and pixels of the previously determined set of edge pixels;

evaluating respective local contrast measures with respect to the selected pixels of the image; and

deciding whether or not the selected pixels belong to the set of edge pixels based on the evaluated local contrast measures.

10. A computer program product comprising stored instructions to be executed in a processor associated with a light sensor having a matrix of sensing elements, wherein execution of the instructions by the processor controls steps of:

receiving, from each sensing element of the matrix, a respective event-based signal sequence including events asynchronously produced as a function of variations of light incident on the sensing element and light level values respectively associated with at least some of the events;

generating an image comprising pixels corresponding spatially to the sensing elements of the matrix; and

repeatedly determining a set of edge pixels among the pixels of the image,

wherein generating the image comprises updating each pixel of the image based on a light level value associated with a most recent event in the event-based signal sequence received from the sensing element corresponding to said pixel of the image,

wherein determining the set of edge pixels comprises:

selecting pixels of the image, wherein the selected pixels comprise pixels of the image corresponding to sensing elements from which the received event-based signal sequences include at least one event after a previous determination of the set of edge pixels, and pixels of the previously determined set of edge pixels;

evaluating respective local contrast measures with respect to the selected pixels of the image; and

deciding whether or not the selected pixels belong to the set of edge pixels based on the evaluated local contrast measures.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 15, 2020
From: GALLUPPI, FRANCESCO
To: GENSIGHT BIOLOGICS
Reel/Frame 053219/0632 →
Priority Claims (1)
EP 18305020 · Jan 11, 2018 · regional
Continuity (1)
Related Publication 20210383146A1 · Dec 9, 2021