Method and device for image processing

Patent metadata

Field	Value
Publication number	US-2018249072-A1
Application number	US-201715443012-A
Country	US
Kind code	A1
Filing date	Feb 27, 2017
Priority date	Feb 27, 2017
Publication date	Aug 30, 2018
Grant date	—

Abstract

A method and a device, such as an imaging device, for image processing for automatic detection of focus area are disclosed herein. In accordance with an embodiment, the device for image processing is the imaging device that is configured to extract a plurality of object features of a plurality of objects in a field-of-view (FOV) of the imaging device. A plurality of confidence maps are generated based on the extracted plurality of object features. The imaging device is further configured to determine a focus area corresponding to the FOV, based on the generated plurality of confidence maps by use of a pre-determined rule.

First claim

What is claimed is:

1 . A method for image processing, said method comprising:
extracting, by an imaging device, a plurality of object features of a plurality of objects in a field-of-view (FOV) of said imaging device; generating, by said imaging device, a plurality of confidence maps based on said extracted plurality of object features; and determining, by said imaging device, a focus area corresponding to said FOV based on said generated plurality of confidence maps by use of a pre-determined rule.

2 . The method according to claim 1 , further comprising capturing, by said imaging device, an image of said FOV, based on said determined focus area.

3 . The method according to claim 1 , further comprising setting, by said imaging device, said determined focus area corresponding to a center portion of said FOV when said pre-determined rule specifies that a depth-of-field (DOF) value of said FOV exceeds a pre-determined DOF threshold value.

4 . The method according to claim 3 , wherein said DOF value is determined based on setting parameters of said imaging device and depth information of said FOV.

5 . The method according to claim 1 , further comprising detecting, by said imaging device, said plurality of objects in said FOV, wherein said plurality of objects correspond to one or more human beings.

6 . The method according to claim 5 , further comprising setting, by said imaging device, said determined focus area corresponding to an object among said plurality of objects in said FOV when said pre-determined rule specifies that a motion value of said object is highest as compared with other motion values of other objects in said FOV.

7 . The method according to claim 5 , further comprising setting, by said imaging device, said determined focus area corresponding to an object among said plurality of objects in said FOV when said pre-determined rule specifies that motion values of said plurality of objects are equal and a depth value of said object is smallest as compared with other depth values of other objects in said FOV.

8 . The method according to claim 5 , further comprising setting, by said imaging device, said determined focus area corresponding to a front view of a facial portion of an object among said plurality of objects in said FOV when said pre-determined rule specifies that motion values of said plurality of objects are less than a pre-determined motion threshold value and a depth value of said object is smallest as compared with other depth values of facial portions of other objects in said FOV.

9 . The method according to claim 5 , further comprising setting, by said imaging device, said determined focus area corresponding to an object among said plurality of objects in said FOV when said pre-determined rule specifies that facial portions of said plurality of objects are not detected, wherein a human being with a smallest depth value in said FOV is focused in said FOV when a presence of said one or more human beings in said FOV is detected independent of facial detection of said one or more human beings, and wherein a non-human object with a smallest depth value is focused when said one or more human beings are absent in said FOV.

10 . The method according to claim 1 , further comprising setting, by said imaging device, said determined focus area corresponding to an object among said plurality of objects in said FOV when said pre-determined rule specifies that a depth value of said object is smallest as compared with other depth values of other objects in said FOV.

11 . The method according to claim 1 , wherein said pre-determined rule is based on one or more of: a lens setting, depth information, and/or motion information of said plurality of objects in said FOV of said imaging device.

12 . The method according to claim 1 , further comprising generating, by said imaging device, a final confidence map.

13 . The method according to claim 12 , wherein said determination of said focus area is within said final confidence map of said FOV.

14 . An imaging device comprising:
one or more circuits being configured to: extract a plurality of object features of a plurality of objects in a field-of-view (FOV) of said imaging device; generate a plurality of confidence maps based on said extracted plurality of object features; and determine a focus area corresponding to said FOV based on said generated plurality of confidence maps by use of a pre-determined rule.

15 . The imaging device according to claim 14 , wherein said one or more circuits are configured to capture, an image of said FOV, based on said determined focus area.

16 . The imaging device according to claim 15 , wherein said one or more circuits are configured to set said determined focus area corresponding to a center portion of said FOV when said pre-determined rule specifies that a depth-of-field (DOF) value of said FOV exceeds a pre-determined DOF threshold value.

17 . The imaging device according to claim 16 , wherein said DOF value is determined based on setting parameters of said imaging device and depth information of said FOV.

18 . The imaging device according to claim 14 , wherein said FOV comprises a plurality of objects, wherein said plurality of objects are one or more human beings.

19 . The imaging device according to claim 18 , wherein said one or more circuits are configured to set said determined focus area corresponding to an object among said plurality of objects in said FOV when said pre-determined rule specifies that a motion value of said object is highest as compared with other motion values of other objects in said FOV.

20 . The imaging device according to claim 18 , wherein said one or more circuits are configured to set said determined focus area corresponding to an object among said plurality of objects in said FOV when said pre-determined rule specifies that motion values of said plurality of objects are equal and a depth value of said object is smallest as compared with other depth values of other objects in said FOV.

21 . The imaging device according to claim 18 , wherein said one or more circuits are configured to set said determined focus area corresponding to a front view of a facial portion of an object among said plurality of objects in said FOV when said pre-determined rule specifies that motion values of said plurality of objects are less than a pre-determined motion threshold value and a depth value of said object is smallest as compared with other depth values of facial portions of other objects in said FOV.

22 . The imaging device according to claim 18 , wherein said one or more circuits are configured to set said determined focus area corresponding to an object among said plurality of objects in said FOV when said pre-determined rule specifies that facial portions of said plurality of objects are not detected, wherein a human being with a smallest depth value in said FOV is focused in said FOV when a presence of said one or more human beings in said FOV is detected independent of facial detection of said one or more human beings, and wherein a non-human object with a smallest depth value is focused when said one or more human beings are absent in said FOV.

23 . The imaging device according to claim 14 , wherein said one or more circuits are configured to set said determined focus area corresponding to an object among said plurality of objects in said FOV when said pre-determined rule specifies that a depth value of said

Assignees

• Sony Corp

Inventors

• Li Pingshan

Classifications

• G06V40/161 Primary

  Detection; Localisation; Normalisation · CPC title

• H04N23/671

  in combination with active ranging signals, e.g. using light or sound signals emitted toward objects · CPC title

• H04N23/611

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

• H04N23/675

  comprising setting of focusing regions · CPC title

• G06V10/255

  Detecting or recognising potential candidate objects based on visual cues, e.g. shapes · CPC title