Image processing apparatus, control method thereof, and storage medium

Patent metadata

Field	Value
Publication number	US-10313649-B2
Application number	US-201715719747-A
Country	US
Kind code	B2
Filing date	Sep 29, 2017
Priority date	Oct 20, 2016
Publication date	Jun 4, 2019
Grant date	Jun 4, 2019

Abstract

An image processing apparatus comprises: an image generation unit configured to generate one or more reduced images based on an input image, the input image and the reduced images forming hierarchized images; a gain map generation unit configured to generate a gain map for each hierarchical level by applying a first tone characteristic to each of the hierarchized images and also further applying a second tone characteristic to at least one of the reduced images; and a composing unit configured to compose the created gain maps to generate a composed gain map, wherein the gain map generation unit applies, for each subject region, the first tone characteristic and the second tone characteristic to the at least one of the reduced images, and wherein the first tone characteristic and the second tone characteristic are mutually different tone characteristics.

First claim

What is claimed is:

1. An image processing apparatus comprising:
at least one processor or circuit configured to function as following units:
an image generation unit configured to generate one or more reduced images that are reduced in a stepwise fashion based on an input image, the input image and the reduced images forming hierarchized images;
a gain map generation unit configured to generate a gain map for each hierarchical level that represents a gain at each pixel position of an image, by applying a first tone characteristic to each of the hierarchized images and also further applying a second tone characteristic to at least one of the one or more reduced images; and
a composing unit configured to compose gain maps generated for each hierarchical level to generate a composed gain map that represents gain applied to each pixel position of the input image,

wherein the gain map generation unit applies, for each subject region, the first tone characteristic and the second tone characteristic to the at least one of the one or more reduced images, and
wherein the first tone characteristic and the second tone characteristic are mutually different tone characteristics associated with a respective subject region included in the input image.

2. The apparatus according to claim 1 , the at least one processor or circuit further configured to function as a discrimination unit configured to discriminate a predetermined subject region from the input image,
wherein the gain map generation unit, based on a discrimination result by the discrimination unit, applies, for each subject region, the first tone characteristic and the second tone characteristic to the at least one of the one or more reduced images.

3. The apparatus according to claim 2 , wherein the second tone characteristic is a tone characteristic that is associated with the predetermined subject region included in the input image, and the first tone characteristic is a tone characteristic that is associated with a region that is different to the predetermined subject region.

4. The apparatus according to claim 2 , wherein the gain map generation unit differentiates a number of the one or more reduced images to which the first tone characteristic and the second tone characteristic are applied in accordance with an accuracy of region discrimination by the discrimination unit.

5. The apparatus according to claim 2 , wherein the gain map generation unit applies the first tone characteristic and the second tone characteristic to the one or more reduced images for which a degree of reduction decreases as accuracy of region discrimination by the discrimination unit increases.

6. The apparatus according to claim 2 , wherein, for the first tone characteristic and the second tone characteristic, a difference in gain applied to a range of luminance signals that are not in common with the predetermined subject region and a region that is not the predetermined subject region is smaller than a difference in gain applied to a range of luminance signals that are in common with the predetermined subject region and the region that is not the predetermined subject region.

7. The apparatus according to claim 1 , wherein the gain map generation unit includes a most reduced image in the at least one of the one or more reduced images.

8. The apparatus according to claim 2 , wherein the discrimination unit discriminates from the input image a second predetermined subject region that differs from the predetermined subject region, and
wherein the gain map generation unit, based on the discrimination result by the discrimination unit, further applies a third tone characteristic that is different to the first tone characteristic and the second tone characteristic and is associated with the second predetermined subject region to the at least one of the one or more reduced images.

9. An image processing apparatus, comprising:
at least one processor or circuit configured to function as following units:
an image reduction unit configured to generate one or more reduced images by reducing an input image and generate an image group that includes the input image and the one or more reduced images; and
a gain map generation unit configured to respectively generate a gain map that indicates a gain value at each position in an image by applying a predetermined tone characteristic to each image in the image group, wherein the gain map generation unit generates the gain map by applying, to the image out of the image group for which a resolution is relatively lowest, tone characteristics for which a characteristic differs for each subject region included in the image;
a gain map composing unit configured to generate a composed gain map by composing a plurality of gain maps generated based on each image of the image group; and
an image conversion unit configured to perform a tone conversion by using the composed gain map on the input image.

10. The apparatus according to claim 9 , the at least one processor or circuit further configured to function as a discrimination unit configured to discriminate a predetermined subject region from the image,
wherein the gain map generation unit generates the gain map by applying the tone characteristics for which the characteristic differs for each subject region included in the image, based on a discrimination result by the discrimination unit.

11. The apparatus according to claim 10 , wherein the gain map generation unit differentiates a number of reduced images to which different tone characteristics are applied for each subject region included in an image in accordance with an accuracy of region discrimination by the discrimination unit.

12. The apparatus according to claim 10 , wherein the gain map generation unit increases a number of reduced images to which different tone characteristics are applied for each subject region included in an image as an accuracy of region discrimination by the discrimination unit increases.

13. A control method of an image processing apparatus, the method comprising:
generating one or more reduced images that are reduced in a stepwise fashion based on an input image, the input image and the reduced images forming hierarchized images;
generating a gain map for each hierarchical level that represents a gain at each pixel position of an image, by applying a first tone characteristic to each of the hierarchized images and also further applying a second tone characteristic to at least one of the one or more reduced images; and
composing gain maps generated for each hierarchical level to generate a composed gain map that represents gain applied to each pixel position of the input image,
wherein, for each subject region, the first tone characteristic and the second tone characteristic are applied to the at least one of the one or more reduced images, and
wherein the first tone characteristic and the second tone characteristic are mutually different tone characteristics associated with a respective subject region included in the input image.

14. A control method of an image processing apparatus, the method comprising:
generating at least one or more reduced images by reducing an input image and generate an image group that includes the input image and the at least one or more reduced images;
respectively generating a gain map that indicates a gain value at each position in an image by applying a predetermined a tone characteristic to each image in the image group, wherein the gain map is generated by applying, to an image out of the image group for which a resolution is relatively low, tone characteristics for which a characteristic differs for e

Assignees

• Canon Kk

Inventors

• Kimura Naoto

Classifications

• H04N23/76

  by influencing the image signals · CPC title

• H04N5/23229

  Electricity · mapped topic

• H04N5/208

  for compensating for attenuation of high frequency components, e.g. crispening, aperture distortion correction · CPC title

• H04N1/6027 Primary

  Correction or control of colour gradation or colour contrast (H04N1/6058 takes precedence) · CPC title

• G06T2207/20016

  Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform · CPC title