Back to results
US 9,432,639 B2Granted
Security video detection of personal distress and gesture commands
AssigneeHoneywell International Inc.
InventorsEric Oh, David S. Zakrewski, Mi Suen Lee
Life of this patent
Estimating the term…
Abstract
A method and apparatus for operating a security system are provided. The method includes the steps of the security system monitoring a secured area, a video camera obtaining images of the secured area, detecting at least one person within the obtained images and identifying a predetermined context associated with the identified at least one person within the obtained images.
Claims18
A method comprising:
- a security system monitoring a secured area;
- a video camera obtaining images of the secured area;
- detecting at least one authorized person and at least one other person within the obtained images;
- analyzing video frames of the obtained images for specific contexts involving the detected at least one authorized person or at least one other person that require further action, wherein the analyzing does not operate by detecting locomotion or movement of humans across one of the obtained images and, instead, the analyzing operates to detect specific acts performed by or a specific condition of the detected at least one authorized person or at least one other person;
- and
- identifying a predetermined context associated with the detected at least one authorized person or at least one other person within the analyzed video frames requiring further action of the obtained images.
3 more elements
An apparatus comprising:
- means for monitoring a secured area;
- means for obtaining a sequence of images of the secured area;
- means for identifying at least one authorized person and at least one other person within the obtained sequence of images;
- a context detection processor that analyzes video frames of the obtained sequence of images for specific contexts involving the identified at least one authorized person or at least one other person that require, further action, wherein the context detection processor does not operate by detecting locomotion or movement of humans across one of the obtained sequence of images and, instead, the context detection processor operates to detect specific acts performed by or a specific condition of the identified at least one authorized person or at least one other person;
- and
- means for detecting a predetermined action performed by or associated with the identified at least one authorized person or at least one other person within the analyzed video frames requiring further action of the obtained sequence of images.
3 more elements
An apparatus comprising:
- a security system that monitors a secured area;
- a video camera that obtains images proximate the secured area;
- a programmed processor that identifies at least one authorized person and at least one other person within the obtained images;
- a context detection processor that analyzes video frames of the obtained images for specific contexts involving the identified at least one authorized person or at least one other person that require further action, wherein the context detection processor does not operate by detecting locomotion or movement of humans across one of the obtained images and, instead, the context detection processor operates to detect specific acts performed by or a specific condition of the identified at least one authorized person or at least one other person;
- a programmed processor that detects a predetermined context associated with the identified at least one authorized person or at least one other person within the analyzed video frames requiring further action of the obtained images;
- and
- a programmed processor that performs a predefined action in response to the detected predetermined context.
4 more elements
Dependencies are inferred from the claim text.