Method and apparatus for autonomously recognizing at least one object in an image

• US 9,710,708 B1
• Filed: 03/24/2014
• Issued: 07/18/2017
• Est. Priority Date: 03/24/2014
• Status: Active Grant

• Current Assignee: Vecna Robotics, Inc.
• Original Assignee: Vecna Technologies, Inc
• Inventors: Checka, Neal
• Primary Examiner(s): Azarian, Seyed
• Application Number:
  US14/222,864
• Time in Patent Office: 1,212 Days
• Field of Search: 382100, 382103, 382106-107, 382118, 382154-155, 382159, 382162, 382168, 382173, 382181, 382199, 382203, 382206, 382209, 382219, 382232, 382254, 382274, 382276, 382286-291, 382305, 382321, 4555561, 3405681, 3405731, 348152
• CPC Class Codes: G06V 10/7788 the supervisor being a human, e.g. interactive learning with a human teacher
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G06V 40/20 Movements or behaviour, e.g. gesture recognition recognition of facial expressions G06V40/16

Abstract

A method and apparatus for autonomously recognizing at least one object in an image is provided. The method includes obtaining an image and providing an image analyzer for autonomously analyzing the image and autonomously identifying at least one item of information regarding at least one object in the image. The at least one item of information regarding the at least one object may be identified without regard to the at least one object'"'"'s appearance, location, or orientation in the image and without requiring any outside intervention or human input, so that information regarding the at least one object, such as the at least one object'"'"'s actions, movements, and locations may be obtained from the image.

Claims

1. A robot apparatus for autonomously determining activities with regard to an object in a video image, comprising:

an image capturing element for capturing a video image, the video image including one or more activities being undertaken by the object;

a video display that provides the video image captured by the image capturing element for viewing by an operator;

a user interface that is readily accessible by the operator at least portions of which are visually displayed on at least a portion of the video display as the captured video is running and being observed by the operator, the user interface providing the operator with the ability to create a library of video snippets contained within the robot apparatus for future comparison to a synthetic dataset of the video image being captured by the image capturing element, each snippet representing at least one motion being undertaken by the object of interest, the user interface enabling the operator to tag selected portions of the video to create and label the desired snippet at least for future counting and categorization; and

an image analyzer at least including hardware and software that autonomously analyzes the video image and autonomously identifies other future portions of the video image and other video images that are substantially similar to at least one video snippet from the library of previously created video snippets contained within the robot apparatus and tracks the number of times each snippet occurs in the video images.

2. The apparatus as defined in claim 1, wherein the image analyzer analyzes the video image in real time.

3. The apparatus as defined in claim 1, wherein the video display is a touch screen style display and the user interface provides accessibility via the operator touching the screen.