Object region extraction system, method and program

Patent metadata

Field	Value
Publication number	US-9299000-B2
Application number	US-201113806400-A
Country	US
Kind code	B2
Filing date	Jun 16, 2011
Priority date	Jun 23, 2010
Publication date	Mar 29, 2016
Grant date	Mar 29, 2016

Abstract

An object region extraction system, an object region extraction method, and an object region extraction program,
including an effectiveness evaluation quantity acquisition unit to acquire an evaluation quantity which increases or decreases when an angle formed by two back projections selected from back projections of one point of an object approach a right angle, wherein the back projections cross each other at a position in three-dimensional space and obtained from respective images, which include the one point and captured by respective m-number of image acquisition units, where m is greater than one, and an object region determination unit to, when the evaluation quantity is equal to or larger than or equal to or smaller than a predetermined value, determine the position to be a region occupied by the object, and output the determination result.

First claim

The invention claimed is:

1. An object region extraction system comprising:
an effectiveness evaluation quantity acquisition unit for acquiring an evaluation quantity which increases (or decreases) with coming closer to right angle of an angle made by two back projections selected from back projections of one point of an object, wherein said back projections are crossing each other at a position in three-dimensional space and obtained from respective images, which are including said one point and captured by respective ones of m-number of image acquisition units (m: a plural number);
an object region determination unit which, when said evaluation quantity is equal to or larger than (or, equal to or smaller than) a predetermined value, determining said position to be a region occupied by said object, and outputting said determination result; and
an object likelihood acquisition unit for selecting said m-number of image acquisition units each of which captured an image enabling back projection of said one point, from n-number of image acquisition units (n: equal to or larger than said number m) capable of capturing said position, and calculates a ratio of said number m to said number n as an object likelihood; wherein
said object region determination unit for determining said position to be a region occupied by said object if each of said evaluation quantity and said object likelihood is equal to or larger than a respective predetermined value or both of them are equal to or larger than said predetermined values.

2. The object region extraction system according to claim 1 , wherein said effectiveness evaluation quantity acquisition units calculates said evaluation quantity using said angle closest to right angle among said angles made by every pair of said back projections taken from said m-number of back projections.

3. The object region extraction system according to claim 2 , wherein said effectiveness evaluation quantity acquisition unit calculates said evaluation quantity for said position only if the object likelihood for said position is equal to or larger than a predetermined value.

4. The object region extraction system according to claim 3 , wherein, said object region determination unit calculates a posterior probability of said object's existence at said position and that of said object's non-existence at said position are calculated from said object likelihood and said evaluation quantity for said position, and also from a prior probability of said object's existence at said position and that of said object's non-existence at said position, which are determined in advance with respect to said position, and then compares both of the posterior probabilities, and then determines whether or not said position is a region occupied by said object is determined.

5. A non-transitory storage medium storing an object region extraction program which causes a computer to execute:
an effectiveness evaluation quantity acquisition step of acquiring an evaluation quantity which increases (or decreases) with the coming closer to aright angle of an angle made by two back projections selected from back projections of one point of an object, wherein said back projections are crossing each other at a position in three-dimensional space and obtained from respective images, which are including said one point and captured by respective ones of m-number of image acquisition units (m: a plural number);
an object region determination step of, when said evaluation quantity is equal to or larger than (or, equal to or smaller than) a predetermined value, determining said position to be a region occupied by said object, and outputting said determination result;
an object likelihood acquisition step of selecting said m-number of image acquisition units each of which captured an image enabling back projection of said one point, from among n-number of image acquisition units (n: equal to or larger than said number m) capable of capturing said position, and calculating a ratio of said number m to said number n as an object likelihood; and
said object region determination step of determining said position to be a region occupied by said object if each of said evaluation quantity and said object likelihood is equal to or larger than respective predetermined values or both of them are equal to or larger than said predetermined values.

6. An object region extraction method comprising:
acquiring an evaluation quantity which increases (or decreases) with the coming closer to a right angle of an angle made by two back projections selected from back projections of one point of an object, wherein said back projections are crossing each other at a position in three -dimensional space and obtained from respective images, which are including said one point and captured by respective ones of m-number of image acquisition units (m: a plural number);
determining said position to be a region occupied by said object when said evaluation quantity is equal to or larger than (or, equal to or smaller than) a predetermined value, and outputting said determination result;
selecting said m-number of image acquisition units each of which captured an image enabling back projection of said one point, from among n-number of image acquisition units (n: equal to or larger than said number m) capable of capturing said position, and calculating a ratio of said number m to said number n as an object likelihood; and
determining said position to be a region occupied by said object if each of said evaluation quantity and said object likelihood is equal to or larger than respective predetermined values or both of them are equal to or larger than said predetermined values.

7. An object region extraction system comprising:
an effectiveness evaluation quantity acquisition unit for acquiring an evaluation quantity which increases (or decreases) with coming closer to right angle of an angle made by two back projections selected from back projections of one point of an object, wherein said back projections are crossing each other at a position in three-dimensional space and obtained from respective images, which are including said one point and captured by respective ones of m-number of image acquisition units (m: a plural number);
an object region determination unit for, when said evaluation quantity is equal to or larger than (or, equal to or smaller than) a predetermined value, determining said position to be a region occupied by said object, and outputting said determination result; and
an object likelihood acquisition unit for selecting said m-number of image acquisition units each of which captured an image enabling back projection of said one point, from n-number of image acquisition units (n: equal to or larger than said number m) capable of capturing said position, and calculates a ratio of said number m to said number n as an object likelihood wherein
said object region determination unit for determining said position to be a region occupied by said object if each of said evaluation quantity and said object likelihood is equal to or larger than a respective predetermined value or both of them are equal to or larger than said predetermined values.

Assignees

• Ikeda Hiroo
• Nec Corp

Inventors

• Ikeda Hiroo

Classifications

• H04N13/20 Primary

  Image signal generators · CPC title

• G06T7/11

  Region-based segmentation · CPC title

• H04N13/02

  Electricity · mapped topic

• G06T7/0081

  Physics · mapped topic

• G06K9/46 Primary

  Physics · mapped topic