Image processing apparatus

• US 8,351,737 B2
• Filed: 11/10/2009
• Issued: 01/08/2013
• Est. Priority Date: 11/11/2008
• Status: Active Grant

• Current Assignee: Seiko Epson Corporation (Seiko Group)
• Original Assignee: Seiko Epson Corporation (Seiko Group)
• Inventors: Shingai, Kosuke
• Primary Examiner(s): Koziol, Stephen R
• Application Number:
  US12/615,982
• Publication Number:
  US 20100119174A1
• Time in Patent Office: 1,155 Days
• Field of Search: 382/254, 382/275
• US Class Current: 382/275
• CPC Class Codes: H04N 1/00328 with an apparatus processing optically-read information
• CPC Class Codes: H04N 1/00336 with an apparatus performing pattern recognition, e.g. of a face or a geographic feature image or video recognition or understanding of scenes G06V20/00
• CPC Class Codes: H04N 1/32128 attached to the image data, e.g. file header, transmitted message header, information on the same page or in the same computer file as the image for information embedded in the image H04N1/32144
• CPC Class Codes: H04N 1/40062 Discrimination between different image types, e.g. two-tone, continuous tone
• CPC Class Codes: H04N 2201/3232 of a page, copy or picture number
• CPC Class Codes: H04N 2201/3243 of type information, e.g. handwritten or text document
• CPC Class Codes: H04N 2201/3251 where the modified version of the image is relating to a person or face
• CPC Class Codes: H04N 2201/3253 Position information, e.g. geographical position at time of capture, GPS data

Abstract

Provided is an image processing apparatus for performing a correction process with respect to object image data, the image processing apparatus including: a first determining unit that determines whether facial image data including a face of a person, which is extracted from the object image data, is attached to relevant object image data; a second determining unit that determines whether reduction image data of the object image data is attached to the relevant object image data; a selecting unit that selects the facial image data or the reduction image data as selection image data; a correction parameter setting unit that extracts a feature amount from the selection image data and sets a correction parameter for image correction based on the feature amount; and a correction processing unit that corrects the object image data based on the correction parameter.

Claims

1. An image processing apparatus for performing a correction process with respect to object image data, the image processing apparatus comprising:

a first determining unit that determines whether facial image data including a face of a person, which is extracted from the object image data, is attached to the object image data;

a second determining unit that determines whether reduction image data of the object image data is attached to the object image data;

a selecting unit that selects the facial image data or the reduction image data as selection image data;

a correction parameter setting unit that extracts a feature amount from the selection image data and sets a correction parameter for image correction based on the feature amount; and

a correction processing unit that corrects the object image data based on the correction parameter,

wherein the selecting unit selects the facial image data as the selection image data when the first determining unit determines that the facial image data is attached to the object image data, and

the selecting unit selects the reduction image data as the selection image data when the first determining unit determines that the facial image data is not attached to the object image data and the second determining unit determines that the reduction image data is attached to the object image data.

2. The image processing apparatus according to claim 1, wherein, when the facial image data is attached to the object image data, the selecting unit determines whether a ratio of a size of the facial image data with respect to a size of the object image data is equal to or larger than a predetermined first ratio, and selects the facial image data as the selection image data only when the ratio is equal to or larger than the first ratio.

3. The image processing apparatus according to claim 2, wherein the selecting unit determines whether the ratio of the size of the facial image data with respect to the size of the object image data is equal to or larger than a predetermined second ratio greater than the first ratio when facial image data of the first ratio or more is attached to the object image data, selects only the facial image data as the selection image data when the ratio is equal to or larger than the second ratio, determines whether the reduction image data of the object image data is attached to the object image data when the ratio is not equal to or larger than the second ratio, and selects the facial image data of the first ratio or more and the reduction image data as the selection image data when the reduction image data of the object image data is attached to the object image data.

4. The image processing apparatus according to claim 1, wherein the selecting unit determines whether area information, which indicates an area occupied by the facial image data in the object image data, is attached to the facial image data when the facial image data is attached to the object image data, determines whether an area of a predetermined ratio or more specified by the area information is included in a predetermined main area serving as a main area in the object image data when the area information is attached to the facial image data, and selects the facial image data as the selection image data only when the area of the predetermined ratio or more is included in the predetermined main area.

5. The image processing apparatus according to claim 1, wherein the selecting unit determines whether additional information, which indicates that the object image data is not an image employing a person as a subject, is attached to the object image data, determines whether the reduction image data of the object image data is attached to the object image data, without determining whether the facial image data is attached to the object image data, when the additional information is attached to the object image data, and selects the reduction image data as the selection image data when the additional information is attached to the object image data.

6. The image processing apparatus according to claim 1, wherein, when the facial image data is not attached to the object image data, the selecting unit determines whether area information, which indicates an area occupied by a face of a person in the object image data, is attached to the object image data, and, when the facial image data is attached to the object image data, the selecting unit extracts an area specified by the area information from the object image data as partial image data, and regards the partial image data as the facial image data to determine that the facial image data is attached to the object image data.

7. An image processing method for performing a correction process with respect to object image data, the image processing method comprising:

a first determining step of determining whether facial image data including a face of a person, which is extracted from the object image data, is attached to the object image data;

a second determining step of determining whether reduction image data of the object image data is attached to the object image data;

selecting the facial image data or the reduction image data as selection image data;

extracting a feature amount from the selection image data and setting a correction parameter for image correction based on the feature amount; and

correcting the object image data based on the correction parameter,

wherein selecting the selection image data comprises;

selecting the facial image data as the selection image data when the first determining step determines that the facial image data is attached to the object image data, and

selecting the reduction image data as the selection image data when the first determining step determines that the facial image data is not attached to the object image data and the second determining step determines that the reduction image data is attached to the object image data.

8. The image processing method according to claim 7, wherein, when the facial image data is attached to the object image data, the selecting the selection image data further comprises determining whether a ratio of a size of the facial image data with respect to a size of the object image data is equal to or larger than a predetermined first ratio, and selecting the facial image data as the selection image data only when the ratio is equal to or larger than the first ratio.

9. The image processing method according to claim 8, wherein the selecting the selection image data further comprises determining whether the ratio of the size of the facial image data with respect to the size of the object image data is equal to or larger than a predetermined second ratio greater than the first ratio when facial image data of the first ratio or more is attached to the object image data, selecting only the facial image data as the selection image data when the ratio is equal to or larger than the second ratio, determining whether the reduction image data of the object image data is attached to the object image data when the ratio is not equal to or larger than the second ratio, and selecting the facial image data of the first ratio or more and the reduction image data as the selection image data when the reduction image data of the object image data is attached to the object image data.

10. The image processing method according to claim 7, wherein the selecting the selection image data further comprises determining whether area information, which indicates an area occupied by the facial image data in the object image data, is attached to the facial image data when the facial image data is attached to the object image data, determining whether an area of a predetermined ratio or more specified by the area information is included in a predetermined main area serving as a main area in the object image data when the area information is attached to the facial image data, and selecting the facial image data as the selection image data only when the area of the predetermined ratio or more is included in the predetermined main area.

11. The image processing method according to claim 7, wherein the selecting the selection image data further comprises determining whether additional information, which indicates that the object image data is not an image employing a person as a subject, is attached to the object image data, determining whether the reduction image data of the object image data is attached to the object image data, without determining whether the facial image data is attached to the object image data, when the additional information is attached to the object image data, and selecting the reduction image data as the selection image data when the additional information is attached to the object image data.

12. The image processing method according to claim 7, wherein, when the facial image data is not attached to the object image data, the selecting the selection image data further comprises determining whether area information, which indicates an area occupied by a face of a person in the object image data, is attached to the object image data, and, when the facial image data is attached to the object image data, the selecting the selection image data further comprises extracting an area specified by the area information from the object image data as partial image data, and regarding the partial image data as the facial image data to determine that the facial image data is attached to the object image data.