Back to results
US 11,080,891 B2Granted
Method and system for detecting dangerous situation
AssigneeSamsung Electronics Co., Ltd.
InventorsJoo-young Kim, Joon-ho Kim, Do-jun Yang, Hyun-woo Lee, Jong-hee Han
Life of this patent
Estimating the term…
Abstract
A method of detecting a dangerous situation includes obtaining a first image including a first object capable of generating a movement, by using a dynamic vision sensor (DVS); detecting, from the first image, a second image including the first object in a predefined pose; and determining whether a situation of the first object is the dangerous situation by analyzing the second image.
Claims17
A method of detecting a dangerous situation, the method comprising:
- obtaining static images each comprising a first object capable of generating a movement, by using a dynamic vision sensor (DVS);
- analyzing the static images one by one, wherein the analyzing of the static images one by one comprises:
- analyzing a first static image from among the static images by determining whether a pose of the first object in the first static image corresponds to a predefined pose, and
- in response to determining that the pose of the first object in the first static image does not correspond to the predefined pose, proceeding to analyze a next static image from among the static images;
- based on only when it is determined that the pose of the first object in the first static image corresponds to the predefined pose, obtaining information about a first movement variation of the first object by analyzing a plurality of images from among the static images, the plurality of images including the first static image;
- and
- determining whether a situation of the first object is the dangerous situation based on the information about the first movement variation of the first object.
4 more elements
A device for detecting a dangerous situation, the device comprising:
- a dynamic vision sensor (DVS) configured to obtain static images each comprising a first object capable of generating a movement;
- and
- a processor configured to:
- analyze the static images one by one, wherein the analyzing of the static images one by one comprises:
- analyzing a first static image from among the static images by determining whether a pose of the first object in the first static image corresponds to a predefined pose, and
- in response to determining that the pose of the first object in the first static image does not correspond to the predefined pose, proceeding to analyze a next static image from among the static images;
- only when it is determined that the pose of the first object in the first static image corresponds to the predefined pose, obtain information about a first movement variation of the first object by analyzing a plurality of images from among the static images, the plurality of images including the first static image, and
- determine whether a situation of the first object is the dangerous situation based on the information about the first movement variation of the first object.
5 more elements
A non-transitory computer-readable storage medium storing instructions for causing a computer to:
- obtain static images comprising a first object capable of generating a movement, by using a dynamic vision sensor (DVS);
- analyze the static images one by one, wherein the analyzing of the static images one by one comprises:
- analyzing a first static image from among the static images by determining whether a pose of the first object in the first static image corresponds to a predefined pose, and
- in response to determining that the pose of the first object in the first static image does not correspond to the predefined pose, proceeding to analyze a next static image from among the static images;
- only when it is determined that the pose of the first object in the first static image corresponds to the predefined pose, obtain information about a first movement variation of the first object by analyzing a plurality of images from among the static images, the plurality of images including the first static image;
- and
- determine whether a situation of the first object is a dangerous situation based on the information about the first movement variation of the first object.
4 more elements
Dependencies are inferred from the claim text.