# US 2009/0290798 A1 · Image search method and device

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2009/0290798 A1
- Document ID: US20090290798A1
- Status: Published application
- Kind: A1
- Application number: 12/065,445
- Filed: 2006-08-31
- Published: 2009-11-26
- Earliest claimed: 2005-08-31
- Assignees: Toyota Jidosha Kabushiki Kaisha; Czech Technical University in Prague
- Inventors: Katsuhiro Sakai; Ondrej Chum; Jiri Matas
- Web page: https://patentgrep.com/patent/US20090290798A1

## CPC

- G06F 16/5854: {using shape and object relationship}
- G06V 10/462: {Salient features, e.g. scale invariant feature transforms [SIFT]}
- G06V 10/757: {Matching configurations of points or features}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

An image search method that is robust and fast (with computational complexity of logarithmic order relative to the number of models). The image search method including: extracting a plurality of specific regions possessing such a property that a shape can be normalized regardless of an affine transformation thereof, as affine-invariant regions from one or more learning images; calculating, with respect to a reference affine-invariant region, other neighboring affine-invariant regions as a set; deforming the neighboring affine-invariant regions by a transformation to normalize the shape of the reference affine-invariant region; and outputting the deformed shapes of the neighboring affine-invariant regions, together with combination of the reference affine-invariant region and the neighboring affine-invariant regions.

## Claims

### Claim at position 1

1: An image search method comprising:
an extraction step of extracting a plurality of specific regions possessing such a property that a shape can be normalized regardless of an affine transformation thereof, as affine-invariant regions from one or more learning images;
a calculation step of calculating, with respect to a reference affine-invariant region, other neighboring affine-invariant regions as a set;
a deformation step of deforming the neighboring affine-invariant regions by a transformation to normalize the shape of the reference affine-invariant region; and
an output step of outputting the deformed shapes of the neighboring affine-invariant regions, together with combination of the reference affine-invariant region and the neighboring affine-invariant regions.

### Claim at position 2 (depends on 1)

2: The image search method according to claim 1, wherein the affine-invariant regions are set as local coordinate systems.

### Claim at position 3 (depends on 1)

3: The image search method according to claim 1, further comprising:
an input step of inputting an object image as a search object;
an object deformation step of extracting a plurality of specific regions possessing such a property that a shape can be normalized regardless of an affine transformation thereof, as affine-invariant regions from the object image, calculating, with respect to a reference affine-invariant region, other neighboring affine-invariant regions as a set, and deforming the neighboring affine-invariant regions by a transformation to normalize the shape of the reference affine-invariant region; and
a search step of finding correspondence between the affine-invariant regions extracted from the learning images and the affine-invariant regions extracted from the object image, based on combination of the reference affine-invariant region and the neighboring affine-invariant regions in the object image, the deformed shapes of the neighboring affine-invariant regions, and an output result outputted by the output step.

### Claim at position 4

4: An image search apparatus comprising:
extraction means for extracting a plurality of specific regions possessing such a property that a shape can be normalized regardless of an affine transformation thereof, as affine-invariant regions from one or more learning images;
calculation means for calculating, with respect to a reference affine-invariant region, other neighboring affine-invariant regions as a set;
deformation means for deforming the neighboring affine-invariant regions by a transformation to normalize the shape of the reference affine-invariant region; and
output means for outputting the deformed shapes of the neighboring aft−me-invariant regions, together with combination of the reference affine-invariant region and the neighboring affine-invariant regions.

### Claim at position 5 (depends on 4)

5: The image search apparatus according to claim 4, wherein the affine-invariant regions are set as local coordinate systems.

### Claim at position 6 (depends on 4)

6: The image search apparatus according to claim 4, further comprising:
input means for inputting an object image as a search object;
object deformation means for extracting a plurality of specific regions possessing such a property that a shape can be normalized regardless of an affine transformation thereof, as affine-invariant regions from the object image, calculating, with respect to a reference affine-invariant region, other neighboring affine-invariant regions as a set, and deforming the neighboring affine-invariant regions by a transformation to normalize the shape of the reference affine-invariant region; and
search means for finding correspondence between the affine-invariant regions extracted from the learning images and the affine-invariant regions extracted from the object image, based on combination of the reference affine-invariant region and the neighboring affine-invariant regions in the object image, the deformed shapes of the neighboring affine-invariant regions, and an output result outputted by the output means.
