Human facial detection and recognition system

• US 10,169,644 B2
• Filed: 10/24/2014
• Issued: 01/01/2019
• Est. Priority Date: 10/24/2013
• Status: Active Grant

• Current Assignee: Blue Line Security Solutions, LLC
• Original Assignee: Blue Line Security Solutions, LLC
• Inventors: Silva, Marcos
• Primary Examiner(s): Tran, Thai Q
• Assistant Examiner(s): Braniff, Christopher
• Application Number:
  US15/031,010
• Publication Number:
  US 20160275342A1
• Time in Patent Office: 1,530 Days
• Field of Search: None
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06V 10/446 using Haar-like filters, e.g. using integral image techniques
• CPC Class Codes: G06V 10/7553 based on shape, e.g. active shape models [ASM]
• CPC Class Codes: G06V 40/165 using facial parts and geometric relationships
• CPC Class Codes: G06V 40/166 using acquisition arrangements
• CPC Class Codes: G06V 40/168 Feature extraction; Face representation
• CPC Class Codes: G06V 40/172 Classification, e.g. identification

Abstract

Aspects of the present disclosure provide an image-based face detection and recognition system that processes and/or analyzes portions of an image using “image strips” and cascading classifiers to detect faces and/or various facial features, such an eye, nose, mouth, cheekbone, jaw line, etc.

Claims

1. A method for detecting faces comprising:

detecting, using one or more computing devices, at least a portion of a face from at least one image of a sequence of images by applying a plurality of image strips to the sequence of images, the plurality of image strips together forming a particular curvature and a particular pattern, each image strip of the plurality of image strips having a same width, and determining at least one of an edge-like feature of the face and a ridge-like feature of the face by;

generating, using the one or more computing devices, a mapping of facial points corresponding to the at least one image by;

generating a plurality of points on an image plane corresponding to the at least one image and determining a distance between each point to the plurality of points, each point corresponding to the at least one of the edge-like feature of the face and the ridge-like feature of the face;

projecting the plurality of points from the image plane to a surface of a face model;

mirroring the plurality of points using facial symmetry;

identifying, using the one or more computing devices, a match for the mapping of facial points by comparing the distance between each point to the plurality of points and the mapping of facial points to a recognized image stored in a database; and

triggering, using the one or more computing devices, an alert of the match and transmitting the alert for display to a user computer device.

2. The method of claim 1 further comprising receiving high definition video from images from at least one high-definition camera;

and

capturing snapshots of the high definition video to generate the sequence of images.

3. The method of claim 1 wherein detecting the at least a portion of the face comprises applying a cascade classifier to the sequence of images.

4. The method of claim 1, further comprising compressing the sequence of images by re-using identical image data captured in respective images of the sequence of images.

5. The method of claim 1, further comprising storing an indication in a database when the match is identified and providing the alert to the user computing device.

6. A system for detecting faces comprising:

memory storing computer-readable instructions for detecting faces; and

at least one processor to execute the computer-readable instructions to;

detect at least a portion of a face from at least one image of a sequence of images to apply a plurality of image strips to the sequence of images, the plurality of image strips together forming a particular curvature and a particular pattern, each image strip of the plurality of image strips having a same width, and determine at least one of an edge-like feature of the face and a ridge-like feature of the face to;

generate a mapping of facial points corresponding to the at least one image by;

generating a plurality of points on an image plane corresponding to the at least one image and determining a distance between each point to the plurality of points, each point corresponding to the at least one of the edge-like feature of the face and the ridge-like feature of the face;

projecting the plurality of points from the image plane to a surface of a mean shape;

identify a match for the mapping of facial points by comparing the distance between each point to the plurality of points and the mapping of facial points to an image stored in a database; and

trigger an alert of the match and transmit the alert for display to a user computer device.

7. The system of claim 6, the at least one processor further to receive high definition video from images from at least one high-definition camera;

and

capture snapshots of the high definition video to generate the sequence of images.

8. The system of claim 6, wherein the at least one processor to detect the at least a portion of the face comprises applying a cascade classifier to the sequence of images.

9. The system of claim 6, the at least one processor further to compress the sequence of images by re-using identical image data captured in respective images of the sequence of images.

10. The system of claim 6, the at least one processor further to store an indication in a database when the match is identified and providing the alert to the user computing device.

11. A non-transitory computer readable medium encoded with instructions stored thereon for detecting faces, the instructions when executed by a processor, causing the processor to perform operations comprising:

detecting at least a portion of a face from at least one image of a sequence of images;

applying a plurality of image strips to the sequence of images, the plurality of image strips together forming a particular curvature and a particular pattern, each image strip of the plurality of image strips having a same width, and determining at least one of an edge-like feature of the face and a ridge-like feature of the face;

generating a mapping of facial points corresponding to the at least one image by;

generating a plurality of points on an image plane corresponding to the at least one image and determining a distance between each point to the plurality of points, each point corresponding to the at least one of the edge-like feature of the face and the ridge-like feature of the face;

projecting the plurality of points from the image plane to a surface of a mean shape;

mirroring the points using facial symmetry;

identifying a match for the mapping of facial points by comparing the distance between each point to the plurality of points and the mapping of facial points to a recognized image stored in a database; and

triggering an alert of the match and transmitting the alert for display to a user computer device.

12. The non-transitory computer readable medium of claim 11, the operations further comprising receiving high definition video from images from at least one high-definition camera;

and

capturing snapshots of the high definition video to generate the sequence of images.

14. The non-transitory computer readable medium of claim 11, the operations further comprising compressing the sequence of images by re-using identical image data captured in respective images of the sequence of images.

15. The non-transitory computer readable medium of claim 11, the operations further comprising storing an indication in a database when the match is identified and providing the alert to the user computing device.

13. The non-transitory computer readable medium 13, wherein the detecting the at least a portion of the face comprises applying a cascade classifier to the sequence of images.