Image feature extraction method and system

Patent metadata

Field	Value
Publication number	US-9681047-B2
Application number	US-201514814160-A
Country	US
Kind code	B2
Filing date	Jul 30, 2015
Priority date	Aug 7, 2014
Publication date	Jun 13, 2017
Grant date	Jun 13, 2017

Abstract

Disclosed is an image feature extraction method including a step of defining a combination of at least two kinds of solid angles along at least two directions, of an input spherical image; a step of determining respective values of the combination of the at least two kinds of solid angles, so that surface areas of spherical crowns of spherical segments, which are obtained by dividing the spherical image by the respective values of the combination of the at least two kinds solid angles, have a same value; and a step of generating, by utilizing the respective values of the combination of the at least two kinds of solid angles, an image feature template so as to conduct image feature extraction.

First claim

What is claimed is:

1. An image feature extraction method, comprising:
a definition step of defining a combination of at least two kinds of solid angles along at least two directions, of an input spherical image;
a determination step of determining respective values of the combination of the at least two kinds of solid angles, so that surface areas of spherical crowns of spherical segments, which are obtained by dividing the spherical image by the respective values of the combination of the at least two kinds of solid angles, having the same value; and
a generation step of generating, by utilizing the respective values of the combination of the at least two kinds of solid angles, an image feature template so as to conduct image feature extraction.

2. The method according to claim 1 , further comprising:
a conversion step of converting a panoramic image so as to obtain the spherical image.

3. The method according to claim 2 , wherein:
the panoramic image is an equirectangular image, and the equirectangular image is converted into the spherical image by utilizing

r
⁢

:

⁢

[

-
π

,
π

]

×

[

-

π
2

,

π
2

]

→

R
2

(

λ
,
ϕ

)

↦

(

cos
⁡

(
λ
)

⁢

cos
⁡

(
ϕ
)

,

sin
⁡

(
λ
)

⁢

cos
⁡

(
ϕ
)

,

sin
⁡

(
ϕ
)

)

,

wherein, λ stands for a degree of longitude of the equirectangular image; φ stands for a degree of latitude of the equirectangular image; r stands for mapping from the equirectangular image to the spherical image; and R stands for a radius of the spherical image.

4. The method according to claim 1 , wherein:
the at least two directions of the combination of the at least two kinds of solid angles include at least two among directions along X, Y, and Z axes of the spherical image.

5. The method according to claim 4 , wherein, the definition step includes:
utilizing

θ
x

=

2
⁢

π
⁡

(

1
-

cos
⁡

(

arccos
⁡

(

cos
⁢

⁢
ϕ
⁢

⁢

cos
⁡

(

π
2

-
λ

)

)

)

)

Assignees

• He Na
• Wang Gang
• Shi Zhongchao
• Liu Dianchao
• Lu Yaojie
• Ricoh Co Ltd

Inventors

• He Na
• Wang Gang
• Shi Zhongchao
• Liu Dianchao
• Lu Yaojie

Classifications

• H04N23/698

  for achieving an enlarged field of view, e.g. panoramic image capture · CPC title

• H04N5/2628 Primary

  Alteration of picture size, shape, position or orientation, e.g. zooming, rotation, rolling, perspective, translation · CPC title

• G06V10/446

  using Haar-like filters, e.g. using integral image techniques · CPC title

• G06K9/4614

  Physics · mapped topic

• H04N5/23238 Primary

  Electricity · mapped topic