Back to results
US 2003/0068084 A1Published application
Image processing method
AssigneeFuji Photo Film Co., Ltd.
InventorsNaoto Kinjo, Naoki Takaoka, Jun Enomoto
Life of this patent
Estimating the term…
Abstract
Each crest-shape distribution region is sectioned in accordance with a value of a color obtained from extracted elements such as hue, saturation and lightness. A characteristic amount at each of a plurality of positions of an image including a designated eye is calculated, on the basis of values of obtained characteristic amounts, an image of a pupil portion is extracted, and red-eye correction processing is carried out.
Claims53
An image processing method comprising the steps of:
- making an image region, which includes an eye region having discoloration and which has been designated in advance, to be an xy plane, and obtaining an image characteristic amount for each pixel by using any one of hue, saturation and lightness or a combination of two or more of the hue, the saturation and the lightness;
- setting a three-dimensional xyz space on whose z axis are plotted the image characteristic amounts, and sectioning the xy plane of each region whose z axis values are distributed in a crest-shape with respect to the spread of the xy plane;
- and
- determining a discoloration region of a pupil in accordance with any one of or a combination of two or more of information about the shape, information about the position, information about the surface area and a statistical image characteristic amount on the xy plane of each of the sectional regions, and correcting the region determined to be a discoloration region to an image of a normal eye.
1 more element
An image processing method comprising the steps of:
- making an image region, which includes an eye region having discoloration and which has been designated in advance, to be an xy plane, and obtaining an image characteristic amount for each pixel by using anyone of hue, saturation and lightness or a combination of two or more of the hue, the saturation and the lightness;
- setting a three-dimensional xyz space on whose z axis are plotted the image characteristic amounts, and sectioning the xy plane of each region whose z axis values are distributed in a crest-shape with respect to the spread of the xy plane;
- and
- determining a discoloration region of a pupil in accordance with any one of or a combination of two or more of information about the shape, information about the position, information about the surface area and a statistical image characteristic amount on the xy plane of each of the sectional regions, and performing correction including a process for having gradation such that the lightness and/or saturation is gradually lowered from the periphery of the determined discoloration region of the pupil to the central portion so as to correct the determined discoloration region of the pupil to an image of a normal eye.
1 more element
An image processing method comprising the steps of:
- making an image region, which includes an eye region having discoloration and which has been designated in advance, to be an xy plane, and obtaining an image characteristic amount for each pixel by using anyone of hue, saturation and lightness or a combination of two or more of the hue, the saturation and the lightness;
- setting a three-dimensional xyz space on whose z axis are plotted the image characteristic amounts, and sectioning the xy plane of each region whose z axis values are distributed in a crest-shape with respect to the spread of the xy plane;
- and
- determining a discoloration region of a pupil in accordance with any one of or a combination of two or more of information about the shape, information about the position, information about the surface area and a statistical image characteristic amount on the xy plane of each of the sectional regions, and adjusting the position having the highest lightness in the determined discoloration region to be a catch light position, and performing correction including a process for forming a catch light pattern at the catch light position, so as to correct the determined discoloration region of the pupil to an image of a normal eye.
1 more element
An image processing method comprising the steps of:
- making an image region, which includes an eye region having discoloration and which has been designated in advance, to be an xy plane, and obtaining an image characteristic amount for each pixel by using anyone of hue, saturation and lightness or a combination of two or more of the hue, the saturation and the lightness;
- setting a three-dimensional xyz space on whose z axis are plotted the image characteristic amounts, and sectioning the xy plane of each region whose z axis values are distributed in a crest-shape with respect to the spread of the xy plane;
- and
- determining a discoloration region of a pupil in accordance with any one of or a combination of two or more of information about the shape, information about the position, information about the surface area and a statistical image characteristic amount on the xy plane of each of the sectional regions and enlarging/reducing normal color tone pupils cut out from a normal pupil region to have a size which coincides with the size of the determined discoloration region, and performing correction including a process for pasting the normal color tone pupils to the pupil region determined to be a discoloration region, as to correct the discoloration region of the pupil to an image of a normal eye.
1 more element
An image processing method applied to a digital photographic system for making a photographic print by subjecting digital image data to image processing, said image processing method comprising the steps of:
- selecting an image which is to be corrected, and an image including a characteristic portion corresponding to a portion to be corrected of the image to be corrected;
- designating a region including the portion to be corrected of the image to be corrected, and a region including the characteristic portion of the selected image;
- and
- using the characteristic portion of the selected image to correct the portion to be corrected of the image to be corrected.
1 more element
Dependencies are inferred from the claim text.