Dictionary data registration apparatus and dictionary data registration method

• US 8,385,608 B2
• Filed: 09/28/2009
• Issued: 02/26/2013
• Est. Priority Date: 03/29/2007
• Status: Expired due to Fees

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Aoki, Yasuhiro
• Primary Examiner(s): Bali, Vikkram
• Application Number:
  US12/568,341
• Publication Number:
  US 20100067751A1
• Time in Patent Office: 1,247 Days
• Field of Search: None
• US Class Current: 382/118
• CPC Class Codes: G06F 18/28 Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries
• CPC Class Codes: G06V 10/772 Determining representative reference patterns, e.g. averaging or distorting patterns; Generating dictionaries
• CPC Class Codes: G06V 40/172 Classification, e.g. identification

Abstract

A face data registration apparatus is an apparatus to register dictionary data for walker authentication by use of a face image. The face data registration apparatus causes a posture data collection unit to collect posture variation data from a plurality of face images obtained by photographing a face of a registrant whose posture (face direction) is changed, further causes a walking data collection unit to collect walking data from a face image obtained by photographing a face of the registrant who actually walks along a passage, causes a synthesizing unit to create dictionary data by synthesizing posture variation data and walking data and registers the created dictionary data as dictionary data used to authenticate a person who walks in the passage by face collation into a dictionary data unit.

Claims

1. A dictionary data registration apparatus, comprising:

a first image input unit that inputs a plurality of images obtained by photographing a face of a registrant in various postures;

a first face detection unit that detects a plurality of face images from the plurality of images inputted by the first image input unit;

a first data collection unit that collects a plurality of face data items based on the plurality of face images detected by the first face detection unit;

a second image input unit that inputs a plurality of images obtained by photographing the face of the registrant while walking;

a second face detection unit that detects a plurality of face images from the plurality of images inputted by the second image input unit;

a second data collection unit that collects a plurality of face data items based on the plurality of face images detected by the second face detection unit;

a synthesizing unit that synthesizes the face data items collected by the first data collection unit and the face data items collected by the second data collection unit; and

a storage unit that stores the synthesized face data items as dictionary data of the registrant.

2. The dictionary data registration apparatus according to claim 1, further comprising a posture determination unit that determines a posture in the face image detected by the first face detection unit and wherein the first data collection unit collects face data obtained from the face image that is determined to be a desired posture by the posture determination unit.

3. The dictionary data registration apparatus according to claim 1, wherein the first data collection unit collects a plurality of face data items obtained from a plurality of face images in a plurality of images obtained by photographing the face of the registrant in various postures who exists at a preset point, and the second data collection unit collects a plurality of face data items obtained from a plurality of face images in a plurality of images obtained by photographing the registrant who is walking in a peripheral portion of the preset point.

4. The dictionary data registration apparatus according to claim 3, further comprising a position determination unit that determines a position of the registrant and wherein the second data collection unit collects face data obtained from a plurality of images obtained by photographing the face of the registrant who is determined to be walking in the peripheral portion of the preset point by the position determination unit.

5. The dictionary data registration apparatus according to claim 1, further comprising a display unit that displays guidance that urges the registrant to change the posture.

6. The dictionary data registration apparatus according to claim 1, wherein the first and second image input units sequentially inputs images from a camera that successively photographs images in the passage along which a to-be-authenticated person walks.

7. The dictionary data registration apparatus according to claim 1, wherein the first and second image input units inputs image data from a recording medium in which images containing a face of a registrant are recorded.

8. A dictionary data registration apparatus, comprising:

a first image input unit that inputs a plurality of images obtained by photographing a face of a registrant in various postures;

a first face detection unit that detects a plurality of face images from the plurality of images inputted by the first image input unit;

a first data collection unit that collects a plurality of face data items based on the plurality of face images detected by the first face detection unit;

a second image input unit that inputs a plurality of images obtained by photographing the face of the registrant while an illumination condition is varied;

a second face detection unit that detects a plurality of face images from the plurality of images inputted by the second image input unit;

a second data collection unit that collects a plurality of face data items based on the plurality of face images detected by the second face detection unit;

a synthesizing unit that synthesizes the face data items collected by the first data collection unit and the face data items collected by the second data collection unit; and

a storage unit that stores the synthesized face data items as dictionary data of the registrant.

9. A dictionary data registration method, comprising:

inputting, via a first input unit, a plurality of images obtained by photographing a face of a registrant in various postures;

detecting, via a first detection unit, a plurality of face images from the plurality of images inputted by the first input unit;

collecting, via a first collection unit, a plurality of face data items based on the plurality of face images detected by the first detection unit;

inputting, via a second input unit, a plurality of images obtained by photographing the face of the registrant while walking;

detecting, via a second detection unit, a plurality of face images from the plurality of images inputted by the second input unit;

collecting, via a second collection unit, a plurality of face data items based on the plurality of face images detected by the second detection unit;

synthesizing the face data items collected by the first collection unit and the face data items collected by the second collection unit; and

storing the synthesized face data items as dictionary data of the registrant into a storage unit.

10. A dictionary data registration method, comprising:

inputting, via a first input unit, a plurality of images obtained by photographing a face of a registrant in various postures;

detecting, via a first detection unit, a plurality of face images from the plurality of images inputted by the first input unit;

collecting, via a first collection unit, a plurality of face data items based on the plurality of face images detected by the first detection unit;

inputting, via a second input unit, a plurality of images obtained by photographing the face of the registrant while varying an illumination condition;

detecting, via a second detection unit, a plurality of face images from the plurality of images inputted by the second input unit;

collecting, via a second collection unit, a plurality of face data items based on the plurality of face images detected by the second detection unit;

synthesizing the face data items collected by the first collection unit and the face data items collected by the second collection unit; and

storing the synthesized face data items as dictionary data of the registrant into a storage unit.