Image processing method, image processing device, and image processing program

• US 8,170,364 B2
• Filed: 04/17/2006
• Issued: 05/01/2012
• Est. Priority Date: 05/13/2005
• Status: Expired due to Fees

• Current Assignee: Konica Minolta OPTO Incorporated (Konica Minolta Inc.)
• Original Assignee: Konica Minolta OPTO Incorporated (Konica Minolta Inc.)
• Inventors: Nakajima, Takeshi, Ito, Tsukasa, Takano, Hiroaki, Sato, Daisuke
• Primary Examiner(s): Couso, Yon
• Application Number:
  US11/919,949
• Publication Number:
  US 20090116043A1
• Time in Patent Office: 2,206 Days
• Field of Search: 382/274, 382/128, 382/162, 382/167, 382/282, 358/1.9, 358/518, 358/521, 348/234
• US Class Current: 382/274
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 5/40 using histogram techniques
• CPC Class Codes: G06T 5/92 based on global image properties
• CPC Class Codes: G06V 40/161 Detection; Localisation; Normalisation
• CPC Class Codes: H04N 1/407 Control or modification of tonal gradation or of extreme levels, e.g. background level
• CPC Class Codes: H04N 1/46 Colour picture communication systems
• CPC Class Codes: H04N 1/60 Colour correction or control H04N1/54 takes precedence

Abstract

An image processing method includes the steps of: obtaining image data formed by signals of a plurality of pixels expressing a color image; discriminating a photographic scene of an image from the image data; determining a gradation correction method based on a result of discriminating a photographic scene; determining a first gradation correction amount based on a result of discriminating a photographic scene; detecting a face area in an image of the image data and calculating a face detection reliability; determining a second gradation correction amount based on an average luminance of the face area; determining a third gradation correction amount by weighted averaging the first gradation correction amount and the second gradation correction amount based on the face detection reliability; and applying a gradation correction process to the image data by using the third gradation correction amount and the gradation method.

Claims

1. An image processing apparatus comprising:

an obtaining unit for obtaining image data formed by signal of a plurality of pixels expressing a color image;

a scene discrimination unit for discriminating a photographic scene from the image data;

a gradation correction method decision unit for determining a gradation correction method based on a result of discriminating a photographic scene;

a first gradation correction amount decision unit for determining a first gradation correction amount based on a result of discriminating a photographic scene;

a face detection unit for detecting a face area in an image of the image data and calculating a face detection reliability;

a second gradation correction amount decision unit for determining a second gradation correction amount based on an average luminance of the face area detected by the face detection unit;

a third gradation correction amount decision unit for weighted averaging the first gradation correction amount and the second gradation correction amount based on the face detection reliability, and determining a third gradation correction amount; and

a gradation correction processing unit for applying a gradation correction process to the image data by using the third gradation correction amount and the gradation method,

wherein the scene discrimination unit comprises;

a first occupancy rate calculation unit for dividing the image data into areas specified by combinations of predetermined brightness and hue, and calculating the first occupancy rate expressing a rate of the divided area against entire image data for the each divided area;

a first index calculation unit for calculating a first and a second indexes by multiplying predetermined first and second coefficients, which are different each other and set in advance, to the first occupancy rate;

a second occupancy calculation unit for dividing the image data into predetermined areas specified by combinations of predetermined distances from an external edge of a screen and brightness of the image data, and calculating second occupancy rates expressing the occupancy ratios of the predetermined areas against the entire image data for respective divided areas;

a second index calculation unit for calculating a third index by multiplying a predetermined third coefficient to the second occupancy rate;

a third index calculation unit for calculating a fourth index by multiplying a predetermined fourth coefficient at least to an average luminance of skin color in a center portion of a screen of the image data; and

a discrimination unit for discriminating a photographic scene of the image data based on the first, second, third and fourth indexes, which have been calculated.

2. An image processing apparatus comprising:

an obtaining unit for obtaining a first image data formed by signals of a plurality of pixels expressing a color image;

an image reduction unit for forming a second image data by reducing an image size of the first image data;

a scene discrimination unit for discriminating a photographic scene of an image from the second image data;

a gradation correction method decision unit for determining a gradation correction method based on a result of discriminating a photographic scene;

a first gradation correction amount decision unit for determining a first gradation correction amount based on a result of discriminating a photographic scene;

a face detection unit for detecting a face area in an image of the second image data and calculating a face detection reliability;

a second gradation correction amount decision unit for determining a second gradation correction amount based on an average luminance of the face area detected by the face detection unit;

a third gradation correction amount decision unit for determining a third gradation correction amount by weighted averaging the first gradation correction amount and the second gradation correction amount based on the face detection reliability; and

a gradation correction processing unit for applying a gradation correction process to the first image data by using the third gradation correction amount and the gradation method,

wherein the scene discrimination unit comprises;

a first occupancy rate calculation unit for dividing the second image data into areas specified by combinations of predetermined brightness and hue, and calculating the first occupancy rate expressing a rate of the divided area against entire second image data for the each divided area

a first index calculation unit for calculating a first and a second indexes by multiplying predetermined first and second coefficients, which are different each other and set in advance, to the first occupancy rate;

a second occupancy rate calculation unit for dividing the second image data into predetermined areas specified by combinations of predetermined distances from an external edge of a screen and brightness of the second image data, and calculating second occupancy rates expressing the occupancy ratios of the predetermined areas against the entire second image data for respective divided areas;

a second index calculation unit for calculating a third index by multiplying a predetermined third coefficient to the second occupancy rate;

a third index calculation unit for calculating a fourth index by multiplying a predetermined fourth coefficient at least to an average luminance of skin color in a center portion of a screen of the second image data; and

a discrimination unit for discriminating a photographic scene of the second image data based on the first, second, third and fourth indexes, which have been calculated.

3. A computer readable medium recorded thereon an image processing program to allow computer to realize a gradation correction processing function comprising:

obtaining image data formed by signals of a plurality of pixels expressing a color image;

discriminating a photographic scene of an image from the image data;

determining a gradation correction method based on a result of the step of discriminating a photographic scene,

determining a first gradation correction amount based on a result of the step of discriminating a photographic scene;

detecting a face by detecting a face area in an image of the image data and calculating a face detection reliability;

determining a second gradation correction amount based on an average luminance of the face area detected in the step of detecting a face;

determining a third gradation correction amount by weighted averaging the first gradation correction amount and the second gradation correction amount based on the face detection reliability; and

applying a gradation correction process to the image data by using the third gradation correction amount and the gradation method,

wherein the function of discriminating a photographic scene comprises;

calculating a first occupancy rate by dividing the image data into areas specified by combinations of predetermined brightness and hue, and calculating the first occupancy rate expressing a rate of the divided area against entire image data for the each divided area;

calculating a first and a second indexes by multiplying predetermined first and second coefficients, which are different each other and set in advance, to the first occupancy rate;

calculating second occupancy rates by dividing the image data into predetermined areas specified by combinations of predetermined distances from an external edge of a screen and brightness of the image data, and calculating second occupancy rates expressing the occupancy ratios of the predetermined areas against the entire image data for respective divided areas;

calculating a third index by multiplying a predetermined third coefficient to the second occupancy rate;

calculating a fourth index by multiplying a predetermined fourth coefficient at least to an average luminance of skin color in a center portion of a screen of the image data; and

discriminating a photographic scene of the image data based on the first, second, third and fourth indexes, which have been calculated.