Patent
Marker generating and marker detecting system, method and program
العنوان: | Marker generating and marker detecting system, method and program |
---|---|
Patent Number: | 8,655,076 |
تاريخ النشر: | February 18, 2014 |
Appl. No: | 12/524027 |
Application Filed: | January 23, 2008 |
مستخلص: | A marker generating system is characterized in having a special feature extracting element that extracts a portion, as a special feature, including a distinctive pattern in a video image not including a marker; a unique special feature selecting element that, based on the extracted special feature, selects a special feature of an image, as a unique special feature, that does not appear on the video image; and a marker generating element that generates a marker based on the unique special feature. |
Inventors: | Nakajima, Noboru (Tokyo, JP) |
Assignees: | NEC Corporation (Tokyo, JP) |
Claim: | 1. A marker generating system, comprising: feature extracting unit for extracting as a feature a segment containing a distinctive pattern in a background video image not containing a marker in an environment in which detection of a marker is performed; feature storage unit for calculating and storing a frequency distribution of said extracted features; unique feature selector for selecting as a unique feature a portion having a frequency equal to or smaller than a predefined value from said frequency distribution of the features; and marker generator for generating a marker based on said unique feature. |
Claim: | 2. The marker generating system according to claim 1 , further comprising: invariant feature converter for generating an invariant feature that is invariant against variation of said background video image from said extracted feature, wherein said unique feature selector selects a portion that is not said invariant feature as a unique feature. |
Claim: | 3. The marker generating system according to claim 2 , wherein said invariant feature converter generates a geometrical invariant feature from said feature. |
Claim: | 4. The marker generating system according to claim 2 , wherein said invariant feature converter generates an object-color invariant feature from said feature. |
Claim: | 5. The marker generating system according to claim 2 , wherein said invariant feature converter generates a texture invariant feature from said feature. |
Claim: | 6. The marker generating system according to claim 2 , wherein said invariant feature converter receives a multi-dimensional feature as an input and generates a multi-dimensional invariant feature. |
Claim: | 7. The marker generating system according to claim 2 , wherein said invariant feature converter generates an invariant feature from a combination of any one or more of a geometrical invariant feature, an object-color invariant feature, a texture invariant feature, and a multi-dimensional invariant feature thereof. |
Claim: | 8. The marker generating system according to claim 2 , wherein said unique feature selector calculates a frequency distribution of said invariant features, and selects as a unique feature a portion having a frequency equal to or smaller than a predefined threshold from the calculated frequency distribution of the invariant features. |
Claim: | 9. The marker generating system according to claim 8 , wherein said unique feature selector controls the number of unique features to be selected by modifying said threshold. |
Claim: | 10. The marker generating system according to claim 2 , further comprising: video image input controller for providing start and end times at which a video image is input to said video image input unit, and a time at which imaging is performed. |
Claim: | 11. The marker generating system according to claim 10 , wherein, in response to a signal from said video image input controller, said video image input unit performs imaging, whereupon a marker is generated. |
Claim: | 12. The marker generating system according to claim 10 , wherein, in response to a signal from said video image input controller, said video image input unit performs imaging, whereupon said invariant feature converter accumulates said invariant feature accordingly. |
Claim: | 13. The marker generating system according to claim 1 , wherein said unique feature selector is configured to increase/decrease the number of unique features to be selected. |
Claim: | 14. The marker generating system according to claim 1 , further comprising: video image input unit for inputting said video image. |
Claim: | 15. The marker generating system according to claim 14 , wherein said video image input unit is capable of panning, tilting, zooming or moving. |
Claim: | 16. A marker generating method performed by an information processing device including at least a processor and a memory storage device, the method comprising: via the processor, extracting as a feature a segment containing a distinctive pattern in a background video image not containing a marker in an environment in which detection of a marker is performed; via the processor, calculating a frequency distribution of said extracted features; via the processor, selecting as a unique feature a portion having a frequency equal to or smaller than a predefined value from said frequency distribution of the features; and generating a marker based on said unique feature. |
Claim: | 17. The marker generating method according to claim 16 , further comprising: via the processor, generating an invariant feature from said extracted feature; and via the processor, selecting as a unique feature a portion that is not said invariant feature. |
Claim: | 18. The marker generating method according to claim 17 , wherein said invariant feature is a geometrical invariant feature generated from said feature. |
Claim: | 19. The marker generating method according to claim 17 , wherein said invariant feature is an object-color invariant feature generated from said feature. |
Claim: | 20. The marker generating method according to claim 17 , wherein said invariant feature is a texture invariant feature generated from said feature. |
Claim: | 21. The marker generating method according to claim 17 , wherein said invariant feature is a combination of any one or more of a geometrical invariant feature, an object-color invariant feature, a texture invariant feature, and a multi-dimensional invariant feature thereof. |
Claim: | 22. The marker generating method according to claim 17 , further comprising: via the processor, calculating a frequency distribution of said invariant features, and selecting, via the processor, as a unique feature a portion having a frequency equal to or smaller than a predefined threshold from the calculated frequency distribution of the invariant features. |
Claim: | 23. The marker generating method according to claim 22 , further comprising: via the processor, controlling the number of unique features to be selected by modifying said threshold. |
Claim: | 24. The marker generating method according to claim 16 , further comprising: via the processor, generating a multi-dimensional invariant feature from a multi-dimensional feature. |
Claim: | 25. The marker generating method according to claim 16 , further comprising: via the processor, increasing/decreasing the number of unique features to be selected. |
Claim: | 26. The marker generating method according to claim 16 , further comprising: via the processor, starting marker generation processing as imaging of the video image is started. |
Claim: | 27. The marker generating method according to claim 16 , further comprising: via the processor, accumulating an invariant feature as imaging of the video image is started. |
Claim: | 28. A non-transitory computer-readable storage medium in which a program is stored, said program sufficient to cause an information processing apparatus to execute the processing of: extracting as a feature a segment containing a distinctive pattern in a background video image not containing a marker in an environment in which detection of a marker is performed; calculating a frequency distribution of said extracted features; selecting as a unique feature a portion having a frequency equal to or smaller than a predefined value from said frequency distribution of the features; and generating a marker based on said unique feature. |
Current U.S. Class: | 382/190 |
Patent References Cited: | 5343303 August 1994 Delmas 5412487 May 1995 Nishimura et al. 7110591 September 2006 Neubauer et al. 7447331 November 2008 Brown et al. 7519218 April 2009 Takemoto et al. 7536045 May 2009 Alessi et al. 7979432 July 2011 Momosaki et al. 8009305 August 2011 Ishikawa et al. 2004/0151383 August 2004 Alessi et al. 2006/0047693 March 2006 Kojima et al. 2006/0280333 December 2006 Ikeda et al. 2012/0024953 February 2012 Fan et al. 1585455 February 2005 07-254037 October 1995 11-075095 March 1999 2000-207568 July 2000 2003-223639 August 2003 2003-256783 September 2003 2005-293141 October 2005 2005-293579 October 2005 2006-190110 July 2006 2006-350578 December 2006 |
Other References: | International Search Report—PCT/JP2008/050857—Feb. 26, 2008. cited by applicant Chinese Official Action—200880002920.X—Aug. 20, 2012. cited by applicant |
Primary Examiner: | Desire, Gregory M |
Attorney, Agent or Firm: | Young & Thompson |
رقم الانضمام: | edspgr.08655076 |
قاعدة البيانات: | USPTO Patent Grants |
الوصف غير متاح. |