Image enhancement using semantic components

Patent metadata

Field	Value
Publication number	US-9460490-B2
Application number	US-201514956986-A
Country	US
Kind code	B2
Filing date	Dec 2, 2015
Priority date	Nov 30, 2013
Publication date	Oct 4, 2016
Grant date	Oct 4, 2016

Abstract

A system for determining a high resolution output image that includes receiving a low resolution image and determining an intermediate high resolution image. The system detects semantic features based upon the input image and selects corresponding semantic components from a database based upon the detected semantic features. The first intermediate high resolution image is modified based upon information from the corresponding semantic components to determine the high resolution output image.

First claim

We claim:

1. A method, using an image enhancement system, for determining a high resolution output image comprising:
(a) receiving a low resolution image having a first resolution;
(b) determining a first intermediate high resolution image having a second resolution based upon said low resolution image;
(c) determining a second intermediate high resolution image having a third resolution based upon said low resolution image;
(d) detecting semantic features of said second intermediate high resolution image and selecting corresponding semantic components from a database based upon said detected semantic features;
(e) modifying said first intermediate high resolution image based upon information from said corresponding semantic components to determine said high resolution output image having a fourth resolution.

2. The method of claim 1 wherein said determining said first intermediate high resolution image is based upon a dictionary technique.

3. The method of claim 1 wherein said detecting semantic features is based upon a key point extraction technique.

4. The method of claim 1 wherein said detected semantic features is used to select first semantic components.

5. The method of claim 4 where said corresponding semantic components is based upon said first semantic components.

6. The method of claim 5 wherein said first semantic components have a first semantic resolution, said corresponding semantic components have a second semantic resolution.

7. The method of claim 6 wherein said database includes information representative of said first semantic components and said corresponding semantic components.

8. The method of claim 7 wherein said database includes information related to different variants of the same semantic content.

9. The method of claim 6 wherein a multi-dimensional scaling operation is used to determine said corresponding semantic components.

10. The method of claim 1 wherein said information from said corresponding semantic components is edge based information.

11. The method of claim 10 wherein said information is combined for a plurality of corresponding semantic components.

12. The method of claim 1 wherein modifying said first intermediate high resolution image is based upon reducing an error function.

Assignees

• Sharp Laboratories America Inc

Inventors

• Choudhury Anustup Kumar
• Segall Christopher A

Classifications

• G06T3/4053 Primary

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

• G06T5/50

  using two or more images, e.g. averaging or subtraction · CPC title

• G06F18/21

  Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation · CPC title

• G06K9/48

  Physics · mapped topic

• G06K9/46

  Physics · mapped topic