Image processing apparatus and method

Patent metadata

Field	Value
Publication number	US-9361667-B2
Application number	US-201113818027-A
Country	US
Kind code	B2
Filing date	Aug 26, 2011
Priority date	Aug 27, 2010
Publication date	Jun 7, 2016
Grant date	Jun 7, 2016

Abstract

The present technology relates to an image processing apparatus and an image processing method, and particularly to an image processing apparatus and an image processing method capable of suppressing an increase in a load on a subject and obtaining a captured image of the subject with higher image quality.
An imaging unit reduces a light amount and performs a plurality of imagings of the fundus of the eye so as to generate a plurality of fundus images. A biological information alignment processing unit aligns the fundus images by using biological information of a subject. A super-resolution processing unit superimposes an aligned input image on a previous super-resolution result image so as to generate a new super-resolution result image. The super-resolution processing unit stores or outputs the super-resolution result image in a storage unit or from an output unit, and supplies the super-resolution result image to a super-resolution result image buffer so as to be stored. The present technology may be applied to, for example, an image processing apparatus.

First claim

The invention claimed is:

1. An image processing apparatus comprising:
a superimposition unit configured to superimpose a plurality of captured images on each other so as to generate a superimposed image, wherein each captured image is obtained by an imaging unit and is of a fundus irradiated with a first amount of light and has a first quality, and wherein the superimposed image has a second quality which is higher than the first quality;
a grayscale correction parameter setting unit configured to set a grayscale correction parameter for the superimposed image according to biological information of the fundus in the superimposed image generated by the superimposition unit; and
an image correction unit configured to generate an image brighter than each said captured image on the basis of the grayscale correction parameter set by the grayscale correction parameter setting unit and the superimposed image generated by the superimposition unit.

2. The image processing apparatus according to claim 1 , further comprising:
a high resolution unit configured to increase a resolution of the image generated by the image correction unit.

3. The image processing apparatus according to claim 2 , wherein the high resolution unit is configured to increase the resolution of the image generated by the image correction unit, according to the biological information of the fundus.

4. An image processing method of an image processing apparatus, comprising:
causing a superimposition unit to superimpose a plurality of captured images on each other so as to generate a superimposed image, wherein each captured image is obtained by an imaging unit and is of a fundus irradiated with a first amount of light and has a first quality, and wherein the superimposed image has a second quality which is higher than the first quality;
causing a grayscale correction parameter setting unit to set a grayscale correction parameter for the superimposed image according to biological information of the fundus in the superimposed image generated; and
causing an image correction unit to generate an image brighter than each said captured image on the basis of the set grayscale correction parameter and the superimposed image generated.

5. A non-transitory computer readable medium having stored thereon a a set of computer-executable instructions which when executed causes a computer to function as:
a superimposition unit configured to superimpose a plurality of captured images on each other so as to generate a superimposed image;
a grayscale correction parameter setting unit configured to set a grayscale correction parameter for the superimposed image according to biological information of a fundus in the superimposed image generated by the superimposition unit; and
an image correction unit configured to generate an image brighter than each said captured image on the basis of the grayscale correction parameter set by the grayscale correction parameter setting unit and the superimposed image generated by the superimposition unit.

6. An apparatus comprising:
a processor configured to execute instructions to:
superimpose a plurality of captured images on each other so as to generate a superimposed image, wherein each captured image is obtained by an imaging unit and is of a fundus irradiated with a first amount of light and has a first quality, and wherein the superimposed image has a second quality which is higher than the first quality;
set a grayscale correction parameter for the superimposed image according to biological information of the fundus in the generated superimposed image; and
generate an image brighter than each said captured image on the basis of the set grayscale correction parameter and the superimposed image generated; and
a memory that records therein the instructions which can be executed by a computer.

7. An image processing apparatus comprising:
an alignment information generation unit configured to align a plurality of captured images obtained by an imaging unit which images a fundus irradiated with a small amount of light, so as to generate alignment information;
a blurring removal processing unit configured to estimate and remove blurring of the captured images aligned by the alignment information generation unit, so as to generate blurring removed images; and
a superimposition unit configured to superimpose the blurring removed images generated by the blurring removal processing unit on each other on the basis of the alignment information generated by the alignment information generation unit, so as to generate a superimposed image,
wherein the alignment information generation unit is configured to generate the alignment information by performing the alignment such that a position of the fundus in the captured image obtained by the imaging unit is adjusted to be at a position of the fundus in the superimposed image generated at a previous time, and
wherein the superimposition unit is configured to superimpose the blurring removed images on the superimposed image generated at the previous time one by one on the basis of the alignment information generated through the alignment by the alignment information generation unit, so as to generate the fundus image which is obtained by superimposing the plurality of captured images and has a dynamic range wider than that of the captured image.

8. An image processing apparatus comprising:
a superimposition unit configured to superimpose a plurality of captured images on each other obtained by an imaging unit which images a subject; and
a high resolution unit configured to increase a resolution of a captured image which is generated by the superimposition unit superimposing the plurality of captured images, according to biological information of the subject,
wherein the high resolution unit includes:
a super-resolution processing portion configured to perform a super-resolution process of improving the resolution of the captured image through learning about the captured image according to each of a plurality of kinds of learning dictionaries, so as to generate a plurality of super-resolution result images which are captured images with a high resolution;
a part image evaluation portion configured to evaluate each super-resolution result image which is generated through the super-resolution process by the super-resolution processing portion by using each learning dictionary, for each part as a living body included in the super-resolution result image, in a method according to a feature of an image of each part; and
an image selection portion configured to select the optimal super-resolution result image from the plurality of super-resolution result images on the basis of the evaluation by the part image evaluation portion.

9. The image processing apparatus according to claim 8 , wherein the high resolution unit further includes:
a part recognition portion configured to analyze the super-resolution result image and recognize a part as a living body included in the super-resolution result image, and
wherein the part image evaluation portion is configured to evaluate each super-resolution result image for each part recognized by the part recognition portion, in the method according to the feature of the image of each part.

10. An image processing method of an image processing apparatus, comprising:
causing a superimposition unit to superimpose a plurality of captured images on each other obtained by an imaging unit which images a subject;
causing a high resolution unit to increase a resolution of a captured image which is generated by superimposing the plurality of captured images, according to biological information of the subject;
causing a super-resolution processing portion

Assignees

• Aisaka Kazuki
• Kobayashi Seiji
• Kishimoto Takuya
• Ootsuki Tomoyuki
• Sony Corp

Inventors

• Aisaka Kazuki
• Kobayashi Seiji
• Kishimoto Takuya
• Ootsuki Tomoyuki

Classifications

• G06T5/001 Primary

  Physics · mapped topic

• G06T3/4053

  based on super-resolution, i.e. the output image resolution being higher than the sensor resolution · CPC title

• G06T2207/20208

  High dynamic range [HDR] image processing · CPC title

• G06T7/0038

  Physics · mapped topic

• G06T5/002

  Physics · mapped topic