Method and an apparatus for selecting a pixel's mask

Patent metadata

Field	Value
Publication number	US-12505564-B2
Application number	US-202318156061-A
Country	US
Kind code	B2
Filing date	Jan 18, 2023
Priority date	Jan 18, 2023
Publication date	Dec 23, 2025
Grant date	Dec 23, 2025

Abstract

A computational platform and a method for use in a process of matching pairs of pixels, wherein each of the members of a pixel's pair belong to another image captured by a different image capturing sensor, and wherein the computational platform comprises at least one processor configured to carry out a process of matching pairs of pixels based on selecting a pixel mask to be used by selecting neighboring pixels of a given pixel from among all of its neighboring pixels, will be used for the matching process of said given pixel with the other member of its pixel's pair.

First claim

The invention claimed is:

1 . A computational platform for use in a process of deriving depth from stereo images, the platform comprising: a) two image capturing sensors; b) a pattern projector configured to emit a pre-defined pattern into a common field of view of the two image capturing sensors; and c) at least one processor configured to enable projecting the pre-defined pattern on a plane and to capture the image of the projected pre-defined pattern and to carry out a process of matching pairs of pixels belonging to the captured images, wherein said matching process is based on selecting for each given pixel, a pixel mask defined by neighboring pixels to be used in matching said given pixel with its counterpart in the other image from among a plurality of directional pixel masks, and wherein selecting each given pixel's mask from among the plurality of directional pixel masks is based on identifying a direction for which the neighboring pixels comprised in the selected directional pixel mask have a minimal signal variance.

2 . The computational platform of claim 1 , wherein the at least one processor is further configured to affect a geometrical concatenation of all pixel masks for which the ratio between the mask standard deviation and the minimal value of the minimal signal variance, is less than a value of a pre-defined threshold.

3 . A method for deriving depth from stereo images, by a computational platform that comprises: a) two image capturing sensors; b) a pattern projector configured to emit a pre-defined pattern into a common field of view of the two image capturing sensors; and c) at least one processor configured to enable projecting the pre-defined pattern on a plane and to capture the image of the projected pre-defined pattern and wherein the method comprises matching pairs of pixels belonging to two images captured by a respective one of the two image capturing sensors, by selecting for each given pixel, a pixel mask defined by neighboring pixels in matching said given pixel with its counterpart in the other image, from a plurality of directional pixel masks, and wherein selecting each given pixel's mask from among the plurality of directional pixel masks is based on identifying a direction for which the neighboring pixels comprised in the selected directional pixel mask have a minimal signal variance.

4 . The method of claim 3 , further comprising a step of affecting a geometrical concatenation of all pixel masks for which the ratio between the mask standard deviation and the minimal value of the minimal signal variance, is less than a value of a pre-defined threshold.

5 . An image capturing apparatus comprising the computational platform as claimed in claim 1 .

Assignees

• Inuitive Ltd

Inventors

• Elikhis Edward
• Peretz Dor

Classifications

• G06F17/10

  Complex mathematical operations {(function generation by table look-up G06F1/03; evaluation of elementary functions by calculation G06F7/544)} · CPC title

• G06T2207/30196

  Human being; Person · CPC title

• G06T2207/10012

  Stereo images · CPC title

• G06T7/593

  from stereo images · CPC title

• G06T7/33 Primary

  using feature-based methods · CPC title