Person recognition apparatus and person recognition method

US9530078B2 · US · B2

Patent metadata
FieldValue
Publication numberUS-9530078-B2
Application numberUS-201414200270-A
CountryUS
Kind codeB2
Filing dateMar 7, 2014
Priority dateMar 18, 2013
Publication dateDec 27, 2016
Grant dateDec 27, 2016

How to read this patent

A practical reading order for non-experts. Skip the full description unless you need deep technical detail.

  1. Title

    What the patent document calls the invention.

  2. Abstract

    A short plain-language summary of the technical disclosure.

  3. Assignees and inventors

    Who owns or filed the patent and who is credited as inventor.

  4. Key dates

    Filing, priority, publication, and grant dates set the timeline.

  5. First independent claim

    The legal scope of protection — read this for what is actually claimed.

  6. CPC / IPC classifications

    Technology tags used to group this patent with similar filings.

  7. Citations and related patents

    Prior art links and similar publications in this corpus.

Abstract

Official abstract text for this publication.

A person recognition apparatus is disclosed that includes an image input unit, a face detection unit in which a face is expressed from the inputted image data, as a score which takes a value in accordance with facial likeness, a facial feature point detection unit, a feature extraction unit, a feature data administrative unit, a person identification unit to calculate similarity between the amount calculated by the feature extraction unit and the amount stored in the feature data administrative unit, a number of candidates calculation unit which displays the images stored in descending order of the similarity, and calculates a score from the face detection unit and the facial feature point detection unit, a candidate confirmation unit in which images displayed in descending order of the similarity are subjected to visual inspection.

First claim

Opening claim text (preview).

What is claimed is: 1. A person recognition apparatus, comprising: an input device to input image data; a face detection device configured to: detect a face area in which a face of a person is expressed from the inputted image data, and to output a score with a large value when a face is detected, and a small value when a face is not detected, based on a brightness value of each pixel of the inputted image; detect and output a size of the face in a longitudinal direction and a lateral direction; detect and output a yaw direction, a pitch direction, and a roll direction of the face; a number of candidates calculation device configured to: adaptively determine a threshold value relating to a calculation of the number of candidates for each input, based on the value of the detected longitudinal direction, lateral direction, yaw direction, pitch direction, and roll direction of the face; display the facial images stored in the feature data administrative device in descending order of the similarity calculated by the person identification device; and calculate a number of facial images to be displayed in accordance with the threshold value, the score outputted from the face detection device, and the yaw direction, pitch direction, and roll direction outputted from the facial feature point detection device; and a candidate confirmation device configured to: display retrieval candidates of the facial images in descending order of the similarity and subject said facial images to visual inspection by a user, and recalculate the number of the facial images displayed by the number of candidates calculation device in accordance with an increase or decrease of the number of candidates to be confirmed. 2. The apparatus of claim 1 , wherein: the second feature amount includes the facial image of the person; and an output device is configured to display and output a facial image corresponding to the face area detected from the inputted image data, and the facial images of the persons of the number corresponding to the number of candidates from a higher rank of the persons extracted in descending order of the similarity. 3. The apparatus of claim 2 , further comprising: a selection device configured to select a prescribed facial image out of the facial images of the persons of the number corresponding to the number of candidates; wherein an output device displays and outputs the facial image corresponding to the face area detected from the inputted image data, and the facial image selected by the selection device, side by side. 4. The apparatus of claim 2 , wherein: an output device is configured to display and output the facial images of the persons of the number corresponding to the number of candidates in the order of age of the extracted person, side by side. 5. The apparatus of claim 2 , wherein: an output device is configured to display and output the facial images of the persons of the number corresponding to the number of candidates in the order of the face direction of the extracted person, side by side. 6. The apparatus of claim 1 , further comprising: an acceptance device configured to accept an input of a distribution number; wherein an output device is configured to distribute person information of the persons of the number corresponding to the number of candidates from a higher rank of the persons extracted in descending order of the similarity, by the inputted distribution number, and outputs the distributed person information. 7. A person recognition method, comprising: inputting image data; detecting a face area in which a face of a person is expressed from the inputted image data, and outputting a score with a large value when a face is detected, and a small value when a face is not detected, based on a brightness value of each pixel of the inputted image; detecting and outputting a size of the face in a longitudinal direction and a lateral direction; detecting and outputting a yaw direction, a pitch direction, and a roll direction of the face; adaptively determine a threshold value relating to a calculation of the number of candidates for each input, based on the value of the detected longitudinal direction, lateral direction, yaw direction, pitch direction, and roll direction of the face; displaying the facial images stored in the feature data administrative device in descending order of the calculated similarity, and calculating a number of facial images to be displayed in accordance with the threshold value, the outputted score, and the outputted yaw direction, pitch direction, and roll direction of the face; subjecting to visual inspection by a user retrieval candidates of the facial images displayed in descending order of the similarity, and recalculating the number of the facial images to be displayed in accordance with increase or decrease of the number of candidates to be confirmed.

Assignees

Inventors

Classifications

  • User interactive design; Environments; Toolboxes · CPC title

  • Software arrangements specially adapted for pattern recognition, e.g. user interfaces or toolboxes therefor · CPC title

  • G06V40/172Primary

    Classification, e.g. identification · CPC title

  • Physics · mapped topic

  • G06K9/6253Primary

    Physics · mapped topic

Patent family

Related publications grouped by family.

External sources

Frequently asked questions

Answers are generated from the same data shown on this page.

What does patent US9530078B2 cover?
A person recognition apparatus is disclosed that includes an image input unit, a face detection unit in which a face is expressed from the inputted image data, as a score which takes a value in accordance with facial likeness, a facial feature point detection unit, a feature extraction unit, a feature data administrative unit, a person identification unit to calculate similarity between the amo…
Who is the assignee on this patent?
Toshiba Kk
What technology area does this patent fall under?
Primary CPC classification G06V40/172. Mapped technology areas include Physics.
When was this patent published?
Publication date Tue Dec 27 2016 00:00:00 GMT+0000 (Coordinated Universal Time) (B2). Legal status and post-grant events are not shown on this page.
What related patents are in patentsdb?
We list 8 related publications on this page (citations in our corpus or others sharing the same primary CPC).