Face detection with temperature and distance validation

Patent metadata

Field	Value
Publication number	US-10169671-B2
Application number	US-201715435503-A
Country	US
Kind code	B2
Filing date	Feb 17, 2017
Priority date	Feb 17, 2017
Publication date	Jan 1, 2019
Grant date	Jan 1, 2019

Abstract

An electronic device incorporates features that are dependent on finding a face by image processing of an image taken by an image capturing device. To avoid false positives and false negatives due to poor focus, exposure, or spoofing with a picture, the electronic device validates, by distance and/or temperature, a candidate face within a captured image. Distance information is used to scale the candidate face to an actual size for comparison against biometric data on a range of sizes of a human face. Detected temperature is compared against biometric data on the temperature of a human face sensed in the infrared spectrum. Confidence value for face detection is increased or decreased in relation to the validation by size/temperature. For a validated candidate having a confidence value above a threshold, the electronic device can adjust camera controls of an image capturing device or enable a face recognition security component.

First claim

What is claimed is:

1. A method comprising:
receiving an image captured by an image capturing device;
processing the image for characteristics associated with a presence of at least one candidate face with a spatial dimension; and
in response to receiving, from the image characteristics, an initial confidence value indicating the presence of a candidate face:
detecting at least one of (i) a distance from the image capturing device to the candidate face and (ii) a temperature of the candidate face;
determining, based on the detected at least one of the distance to and the temperature of the candidate face and using known biometric characteristics of a human face, whether the candidate face is a validated candidate face;
in response to determining that the candidate face is a validated candidate face, increasing a confidence value assigned to the candidate face to a higher confidence value assigned as a final confidence value;
in response to determining that the candidate face is not a validated candidate face, decreasing the confidence value assigned to the candidate face to a lower confidence value assigned as the final confidence value;
determining whether the final confidence value is above a threshold value; and
in response to the final confidence value being above the threshold value, performing at least one action selected from among a group comprising at least one of (i) adjusting camera image controls and (ii) enabling a face recognition security component.

2. The method of claim 1 , further comprising:
determining whether the image contains the candidate face with insufficient exposure or focus for processing by the face recognition security component of the image capturing device; and
in response to the candidate face not having sufficient exposure or focus for processing by the face recognition security component:
autonomously adjusting at least one of an exposure and a focus of image controls of the image capturing device to enhance image capture of the candidate face;
receiving an enhanced image of the candidate face; and
providing the enhanced image of the candidate face to the face recognition security component.

3. The method of claim 2 , further comprising:
in response to the face recognition security component recognizing the candidate face as belonging to an approved user, unlocking a user interface of the image capturing device; and
in response to the face recognition security component not recognizing the candidate face as belonging to an approved user while the user interface is unlocked, locking the user interface of the image capturing device.

4. The method of claim 1 , wherein:
detecting at least one of: a distance from the image capturing device to the candidate face; and a temperature of the candidate face comprises detecting both: the distance from the image capturing device to the candidate face; and the temperature of the candidate face; and
determining whether the candidate face is a validated candidate face comprises:
determining whether the spatial dimension of the candidate face at the detected distance is within a biometric size range of a human face to indicate a size validated candidate face; and
determining whether the detected temperature is within a biometric temperature range of a human face to indicate a temperature validated candidate face; and

the method further comprises:
in response to a failure to validate either of: the size and the temperature of the candidate face, decreasing the confidence value assigned to the candidate face to or below the lower confidence value; and
in response to validation of both: the size and the temperature of the candidate face, increasing the confidence value assigned to the candidate face to at least the higher confidence value.

5. The method of claim 1 , further comprising:
detecting the temperature of the candidate face, the temperature sensed by an infrared sensor comprising a lens that is focused to a nominal user position from the image capturing device;
determining whether a position of the temperature of the candidate face sensed by the infrared sensor is at the nominal user position that coincides with a location of the candidate face;
in response to determining that the position coincides with the location of the candidate face, determining whether the sensed temperature is within the biometric temperature range of a human face to indicate the validated candidate face; and
in response to determining that the temperature sensed does not coincide with the position of the candidate face, obtaining a temperature within the candidate face by one of: waiting for the candidate face to move to the nominal user position, refocusing the lens of the infrared sensor to detect the temperature of the candidate face, and selecting a temperature value at a position within an array of temperature values sensed by the infrared sensor that corresponds to a location of the candidate face.

6. The method of claim 1 , further comprising:
detecting the temperature by an infrared sensor comprising a lens that is focused to a nominal user position from the image capturing device; and
displaying, on a user interface of the image capturing device, an indication of the nominal user position and the detected temperature.

7. The method of claim 6 , further comprising:
presenting an affordance on the user interface to enable selection of one of: a front side image capturing device on a user-interface side of the image capturing device and a backside image capturing device;
in response, at least in part, to detecting user selection of the front side image capturing device, displaying the detected temperature of the candidate face; and
in response to detecting user selection of the backside image capturing device, displaying a detected temperature sensed from an object in a field of view of the backside image capturing device.

8. The method of claim 1 , further comprising:
receiving the spatial dimension comprising an image width and an image height of the candidate face;
determining an actual width and an actual height of the candidate face respectively based on the image width and image height of the spatial dimension at the detected distance; and
determining whether the actual width and the actual height are both within a biometric size range of a human face to indicate the validated candidate face.

9. The method of claim 1 , wherein:
increasing the confidence value comprises setting a flag for face detection to true;
decreasing the confidence value comprises setting the flag for face detection to false; and
determining whether the confidence value exceeds the threshold value comprises determining whether the flag is set to true.

10. The method of claim 1 , wherein, in response to not receiving the initial confidence value indicating the presence of a candidate face having a spatial dimension, disabling a sensor that detects one of: the distance from the image capturing device to the candidate face and the temperature of the candidate face.

11. An electronic device, comprising:
an image capturing device having camera controls to adjust focus and exposure of a viewing area;
a memory having stored thereon a face detection module which performs image processing of an image taken by the image capturing device and provides a confidence value and spatial dimension of each candidate face detected in the image;
at least one sensor from among a group comprising a distance-measuring sensor and an infrared sensor, the at least one sensor oriented to detect a corresponding one of a distance to or a temperature of a candidate face detected by the image capturing device; and
a processor in comm

Assignees

• Motorola Mobility Llc

Inventors

• Matsimanis Peter A
• Manavella Andrea
• Mercurio Cesare

Classifications

• G06V40/40

  Spoof detection, e.g. liveness detection · CPC title

• G06V40/166 Primary

  using acquisition arrangements · CPC title

• H04N23/611

  where the recognised objects include parts of the human body · CPC title

• H04N23/60

  Control of cameras or camera modules · CPC title

• G06V10/993

  Evaluation of the quality of the acquired pattern · CPC title