Subject identification systems and methods

Patent metadata

Field	Value
Publication number	US-11295150-B2
Application number	US-201816645823-A
Country	US
Kind code	B2
Filing date	Sep 7, 2018
Priority date	Sep 13, 2017
Publication date	Apr 5, 2022
Grant date	Apr 5, 2022

Abstract

Disclosed techniques relate to identifying subjects in digital images. In various embodiments, a first camera (276, 376, 476, 976) may acquire (1002) digital images (420) depicting a subject in an area. Portion(s) of the digital image(s) that depict a face of the subject may be detected (1004) as detected face image(s). Features of the detected face image(s) may be compared (1006) with features of subject reference templates associated with subject(s) in a subject reference database (412). The subject reference database stores subject reference templates that are associated with multiple subjects and that are generated based on data received from a second camera (256, 456). Based on the comparing, an identity of the subject may be determined (1014) and the features of the detected face image(s) may be stored (1016) in the subject reference database in association with the subject, e.g., as an auxiliary template feature vector for later use.

First claim

What is claimed is:

1. A method implemented by one or more processors, the method comprising:
acquiring, using a first camera, one or more digital images that depict a subject in an area;
detecting, as one or more detected face images associated with the subject, one or more portions of the one or more digital images that depict a face of the subject;
comparing features of the one or more detected face images with features of subject reference templates associated with one or more subjects in a subject reference database, wherein the comparing comprises:
determining a first distance between a face feature vector generated from one or more of the detected face images and a first template feature vector generated from a first subject reference template associated with a first subject in the subject reference database; and
determining a second distance between the face feature vector and a second template feature vector generated from a subject reference template associated with a second subject in the subject reference database;

wherein the subject reference database stores subject reference templates related to a plurality of subjects, and wherein the subject reference templates stored in the subject reference database are generated and stored based on data contemporaneously received from a second camera;
wherein, for each given subject reference template, the given subject reference template is associated with an identity of one of the plurality of subjects contemporaneously with the given subject reference template being generated; and
based on the comparing:
determining an identity of the subject, and
storing the features of the one or more detected face images in the subject reference database in association with the subject as an auxiliary template feature vector for later use, wherein the storing is based on a relationship between the first and second distances.

2. The method of claim 1 , wherein the comparing comprises:
applying the one or more detected face images as input across a neural network to generate one or more corresponding face feature vectors; and
applying at least some of the subject reference templates as input across the neural network to generate a plurality of corresponding template feature vectors.

3. The method of claim 2 , wherein the neural network comprises a convolutional neural network.

4. The method of claim 1 , wherein the second camera is located in a registration or triage area.

5. The method of claim 4 , wherein the first camera is located in a waiting room or general ward.

6. The method of claim 1 , further comprising:
acquiring, using the first camera, one or more subsequent digital images that depict a second subject in the area;
detecting, as one or more detected face images associated with the second subject, one or more portions of the one or more subsequent digital images that depict a face of the second subject; and
comparing features of the one or more detected face images associated with the second subject with features of auxiliary subject reference templates associated with one or more subjects in the subject reference database, wherein the auxiliary subject reference templates are generated based on digital images captured by the first camera to determine an identity of the second subject.

7. The method of claim 1 , wherein the auxiliary template feature vector replaces an existing template feature vector associated with the subject.

8. A system comprising one or more processors and memory operably coupled with the one or more processors, wherein the memory stores instructions that, in response to execution of the instructions by one or more processors, cause the one or more processors to:
acquire, using a first camera, one or more digital images that depict a subject in an area;
detect, as one or more detected face images associated with the subject, one or more portions of the one or more digital images that depict a face of the subject;
compare features of the one or more detected face images with features of subject reference templates associated with one or more subjects in a subject reference database, wherein the instructions to compare include instructions to:
determine a first distance between a face feature vector generated from one or more of the detected face images and a first template feature vector generated from a first subject reference template associated with a first subject in the subject reference database; and
determine a second distance between the face feature vector and a second template feature vector generated from a subject reference template associated with a second subject in the subject reference database,

wherein the subject reference database stores subject reference templates associated with a plurality of subjects, and wherein the subject reference templates stored in the subject reference database are generated and stored based on data contemporaneously received from a second camera;
wherein, for each given subject reference template, the given subject reference template is associated with an identity of one of the plurality of subjects contemporaneously with the given subject reference template being generated; and
based on the comparison:
determine an identity of the subject, and
store the features of the one or more detected face images in the subject reference database in association with the subject as an auxiliary template feature vector, wherein the storage is based on a relationship between the first and second distances.

9. The system of claim 8 , wherein the comparison comprises:
applying the one or more detected face images as input across a neural network to generate one or more corresponding face feature vectors; and
applying at least some of the subject reference templates as input across the neural network to generate a plurality of corresponding template feature vectors.

10. The system of claim 9 , wherein the neural network comprises a convolutional neural network.

11. The system of claim 8 , wherein the second camera is located in a registration or triage area.

12. The system of claim 11 , wherein the first camera is located in a waiting room or general ward.

13. The system of claim 8 , further comprising instructions to:
acquire, using the first camera, one or more subsequent digital images that depict a second subject in the area;
detect, as one or more detected face images associated with the second subject, one or more portions of the one or more subsequent digital images that depict a face of the second subject; and
compare features of the one or more detected face images associated with the second subject with features of auxiliary subject reference templates associated with one or more subjects in the subject reference database, wherein the auxiliary subject reference templates are generated based on digital images captured by the first camera to determine an identity of the second subject.

14. The system of claim 8 , wherein the auxiliary template feature vector replaces an existing template feature vector associated with the subject.

15. At least one non-transitory computer-readable medium comprising instructions that, in response to execution of the instructions by one or more processors, cause the one or more processors to perform the following operations:
acquiring, using a first camera, one or more digital images that depict a subject in an area;
detecting, as one or more detected face images associated with the subject, one or more portions of the one or more digital images that depict a face of the subject;
comparing features of the one or more detected face

Assignees

• Koninklijke Philips Nv

Inventors

• Wang Haibo
• Van Zon Cornelis Conradus Adrianus Maria

Classifications

• G06T7/248 Primary

  involving reference images or patches · CPC title

• G06V40/16

  Human faces, e.g. facial parts, sketches or expressions · CPC title

• G06V10/772

  Determining representative reference patterns, e.g. averaging or distorting patterns; Generating dictionaries · CPC title

• G06V20/52

  Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69) · CPC title

• G06V10/82

  using neural networks · CPC title