• Àüü
  • ÀüÀÚ/Àü±â
  • Åë½Å
  • ÄÄÇ»ÅÍ
´Ý±â

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹Á¤º¸Åë½ÅÇÐȸ ³í¹®Áö (Journal of the Korea Institute of Information and Communication Engineering)

Çѱ¹Á¤º¸Åë½ÅÇÐȸ ³í¹®Áö (Journal of the Korea Institute of Information and Communication Engineering)

Current Result Document : 13 / 13 ÀÌÀü°Ç ÀÌÀü°Ç

ÇѱÛÁ¦¸ñ(Korean Title) Convex-hullÀ» ÀÌ¿ëÇÑ ±âÇÏÇÐÀû Ư¡ ±â¹ÝÀÇ ¼Õ ¸ð¾ç ÀÎ½Ä ±â¹ý
¿µ¹®Á¦¸ñ(English Title) Hand shape recognition based on geometric feature using the convex-hull
ÀúÀÚ(Author) ÃÖÀαԠ  À¯Áö»ó   In-kyu Choi   Jisang Yoo  
¿ø¹®¼ö·Ïó(Citation) VOL 18 NO. 08 PP. 1931 ~ 1940 (2014. 08)
Çѱ۳»¿ë
(Korean Abstract)
º» ³í¹®¿¡¼­´Â Å°³ØÆ®(Kinect) ½Ã½ºÅÛ¿¡¼­ ȹµæÇÑ ±íÀÌ ¿µ»óÀ¸·ÎºÎÅÍ convex-hullÀ» ÀÌ¿ëÇÑ ±âÇÏÇÐÀû Ư¡ ±â¹ÝÀÇ ¼Õ ¸ð¾ç ÀÎ½Ä ±â¹ýÀ» Á¦¾ÈÇÑ´Ù. Å°³ØÆ® ½Ã½ºÅÛÀº ±íÀÌ ¿µ»ó°ú »ç¿ëÀÚÀÇ °ñ°Ý Á¤º¸¸¦ Á¦°øÇÏ´Â Ä«¸Þ¶ó·Î ¼Õ ¿µ¿ª °ËÃâ¿¡ À¯¿ëÇÏ°Ô È°¿ëÇÒ ¼ö ÀÖ´Ù. Á¦¾ÈÇÏ´Â ±â¹ý¿¡¼­´Â Å°³ØÆ®·Î ȹµæÇÑ ±íÀÌ ¿µ»ó¿¡¼­ ¼Õ ¿µ¿ªÀ» °ËÃâÇÏ°í, ÀÌ ¼Õ ¿µ¿ªÀÇ convex-hullÀ» ±¸ÇÑ´Ù. ¼Õ ¸ð¾ç¿¡ µû¶ó¼­ º¯ÇÏ´Â convex-hull¿¡¼­ ÀâÀ½À¸·Î »ý±ä °æ°èÁ¡ ¹× ÀνĿ¡ ºÒÇÊ¿äÇÑ °æ°èÁ¡À» ÀÏ·ÃÀÇ ±â¹ýÀ» ÅëÇØ Á¦°ÅÇÑ´Ù. Ãß·ÁÁø °æ°èÁ¡À» ÅëÇØ À籸¼ºµÈ convex-hullÀ» ƯÁ¤ ´Ù°¢ÇüÀ¸·Î ÆÇ´ÜÇÏ°í, ÀÌ ´Ù°¢ÇüÀÇ ³»°¢ÀÇ ÇÕÀ» ÀÌ¿ëÇÏ¿© ¼Õ ¸ð¾çÀ» ÀνÄÇÏ°Ô µÈ´Ù. ½ÇÇèÀ» ÅëÇØ Á¦¾ÈÇÏ´Â ±â¹ýÀÌ ÀνÄÇÏ°íÀÚ ÇÏ´Â ¸ðµ¨¿¡ ´ëÇÏ¿© ³ôÀº ÀνķüÀ» º¸¿©Áشٴ °ÍÀ» È®ÀÎÇÏ¿´°í, ´Ü¼øÈ÷ ƯÁ¤ ¹æÇâÀ¸·Î °íÁ¤µÈ ¼Õ ¸ð¾ç»Ó¸¸ ¾Æ´Ï¶ó °°Àº ¸ð¾çÀ̳ª ¹æÇâÀÌ Æ²¾îÁø ¼Õ ¸ð¾ç¿¡ ´ëÇؼ­µµ ¿ì¼öÇÑ ÀÎ½Ä ¼º´ÉÀ» È®ÀÎÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
In this paper, we propose a new hand shape recognition algorithm based on the geometric features using the convex-hull from the depth image acquired by Kinect system. Kinect is a camera providing a depth image and user¡¯s skeleton information and used for detecting hand region. In the proposed algorithm, hand region is detected in a depth image acquired by Kinect and convex-hull of the region is found. Boundary points caused by noise and unnecessary points for recognition are eliminated in the convex-hull that changes depending on hand shape. Hand shape is recognized by the sum of internal angle of a polygon that is matched with convex-hull reconstructed with selected boundary points. Through experiments, we confirm that proposed algorithm shows high recognition rate not only for five models but also those cases rotated.
Å°¿öµå(Keyword) º¼·Ï¼±Ã¼   ±íÀÌ¿µ»ó   ¼Õ ¸ð¾ç ÀνĠ  °æ°èÁ¡ Á¦°Å   Convex-hull   depth image   hand shape recognition   boundary point removal  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå