Apparatus and method for recognizing user's posture in horse-riding simulator

Patent metadata

Field	Value
Publication number	US-9420963-B2
Application number	US-201414244301-A
Country	US
Kind code	B2
Filing date	Apr 3, 2014
Priority date	Aug 16, 2013
Publication date	Aug 23, 2016
Grant date	Aug 23, 2016

Abstract

An apparatus for recognizing a user's posture in a horse-riding simulator, the apparatus comprising: a standard posture model generation module configured to find out a standard posture model by selecting feature points from an expert database, and generate the standard posture model; and a posture recognizing module configured to obtain a user's posture from the horse-riding simulator, recognize a user's horse-riding posture by matching the obtained user's posture with the standard posture model generated in the standard posture model generation module, and suggest a standard posture model appropriate for a user's level.

First claim

What is claimed is:

1. An apparatus for recognizing a user's posture in a horse-riding simulator, the apparatus comprising:
a standard posture model generation module configured to find out a standard posture model by selecting horse-riding posture feature points from an expert database, and generate the standard posture model; and
a posture recognizing module configured to obtain a user's posture from the horse-riding simulator, recognize a user's horse-riding posture by matching the obtained user's posture with the standard posture model generated in the standard posture model generation module, and suggest a standard posture model appropriate for a user's level.

2. The apparatus of claim 1 , wherein the standard posture model generation module comprises:
a posture image input unit configured to input horse-riding posture images of an expert;
a feature point selecting unit configured to extract positions of a body from the horse-riding posture images of an expert inputted from the posture image input unit and select horse-riding posture feature points in the posture image;
a posture image normalizing unit configured to normalize the horse-riding posture images of an expert inputted from the posture image collecting input unit;
an average position and deviation extracting unit configured to calculate an average position of the horse-riding posture feature points inputted from the feature point selection unit and extract a deviation of the posture image from the average position;
a posture deviation pattern analyzing unit configured to analyze a posture deviation and change pattern from various posture images produced depending on a user's level or walking pattern of a horse; and
a posture model generation unit configured to generate the standard posture model by applying the posture deviation and change pattern analyzed in the posture deviation pattern analyzing unit.

3. The apparatus of claim 2 , wherein the posture image collection inputting unit is configured to input the posture images of an expert from the expert database that stores the horse-riding posture images of an expert.

4. The apparatus of claim 2 , wherein the feature point selecting unit extracts positions of shoulder, elbow, hands or feet from the horse-riding images and selects characteristics of the posture image.

5. The apparatus of claim 1 , wherein the posture recognizing module comprises:
an image acquisition unit configured to acquire a user's posture image;
a posture detecting unit configured to separate the posture area from the user's posture image obtained from the image acquisition unit and detect the user's posture;
a user's posture characteristic extracting unit configured to extract a body position of a user from the posture area and detect a posture characteristic of the user;
a characteristic normalizing unit configured to normalize the posture characteristic of the user detected from the user's posture characteristic detecting unit; and
a recognizing unit configured to recognize the user's horse-riding posture by matching the user's posture with the standard posture model.

6. The apparatus of claim 5 , wherein the posture detecting unit is configured to detect the user's posture by minimizing a surrounding environment effect for the user's posture image obtained from the image acquisition unit and correctly separating only the posture area of the user.

7. The apparatus of claim 5 , wherein the user's posture characteristic extracting unit is configured to detect the posture characteristic of the user by extracting positions of shoulder, elbow, hands or feet from the posture area of the user.

8. The apparatus of claim 5 , wherein the recognizing unit is further configured to suggest a standard posture model appropriate for a user's level.

9. The apparatus of claim 5 , wherein the image acquisition unit is configured to obtain the user's posture image through a vision sensor.

10. A method for recognizing a user's posture in a horse-riding simulator, the method comprising:
finding out a standard posture model by selecting horse-riding posture feature points from an expert database to generate the standard posture model; and
obtaining a user's posture from the horse-riding simulator to recognize a user's horse-riding posture by matching the obtained user's posture with the standard posture model.

11. The method of claim 10 , wherein said generating the standard posture model comprises:
inputting horse-riding posture images of an expert;
extracting positions of a body from the horse-riding posture images of an expert to select horse-riding posture feature points in the posture image;
normalizing the horse-riding posture images of an expert;
calculating an average position of the horse-riding posture feature points and extracting a deviation of the posture image from the average position;
analyzing a posture deviation and change pattern from various posture images produced depending on a user's level or walking pattern of a horse; and
generating a standard posture model by applying the analyzed posture deviation and change pattern.

12. The method of claim 11 , wherein the horse-riding posture image is inputted from an expert database that stores the horse-riding posture images of an expert.

13. The method of claim 11 , wherein the horse-riding posture feature points are selected by extracting positions of shoulder, elbow, hands or feet from the horse-riding image.

14. The method of claim 10 , wherein said recognizing the user's horse-riding posture comprises:
acquiring a user's posture image;
separating a user's posture area from the user's posture image obtained and detecting the user's posture;
detecting a posture characteristic of the user by extracting a body position of the user from the posture area of the user;
normalizing the detected posture characteristic of the user; and
recognizing the user's horse-riding posture by matching the user's posture with the standard posture model.

15. The method of claim 14 , wherein said detecting the posture characteristic of the user comprises extracting positions of shoulder, elbow, hands or feet from the user's posture area to detect the posture characteristic.

16. The method of claim 14 , further comprising:
suggesting a standard posture model appropriate for a user's level.

Assignees

• Electronics & Telecommunications Res Inst

Inventors

• Kim Kye Kyung
• Kang Sang Seung
• Chi Suyoung
• Lee Dong-Jin
• Chung Yun Koo
• Han Mun Sung
• Kim Jae Hong
• Park Jong-Hyun

Classifications

• A61B2503/10

  Athletes · CPC title

• A61B5/1128 Primary

  using image analysis (A61B5/1127 takes precedence) · CPC title

• A61B5/1121

  Determining geometric values, e.g. centre of rotation or angular range of movement · CPC title

• A63G15/00

  Rocking horses on runners adapted for progressive movement · CPC title

• G09B9/00

  Simulators for teaching or training purposes (for the use of weapons F41; computing aspects G06; {protocols for games, networked simulations or virtual reality H04L67/131}) · CPC title