• 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 :

ÇѱÛÁ¦¸ñ(Korean Title) ÀÓº£µðµå ½Ã½ºÅÛÀ» À§ÇÑ °í¼ÓÀÇ ¼Õµ¿ÀÛ ÀÎ½Ä ¾Ë°í¸®Áò
¿µ¹®Á¦¸ñ(English Title) Fast Hand-Gesture Recognition Algorithm For Embedded System
ÀúÀÚ(Author) Ȳµ¿Çö   Àå°æ½Ä   Dong-Hyun Hwang   Kyung-Sik Jang  
¿ø¹®¼ö·Ïó(Citation) VOL 21 NO. 07 PP. 1349 ~ 1354 (2017. 07)
Çѱ۳»¿ë
(Korean Abstract)
º» ³í¹®¿¡¼­´Â ÀÓº£µðµå ½Ã½ºÅÛ¿¡ È°¿ëÇÒ ¼ö ÀÖ´Â °í¼ÓÀÇ ¼Õµ¿ÀÛ ÀÎ½Ä ¾Ë°í¸®ÁòÀ» Á¦¾ÈÇÑ´Ù. ±âÁ¸ÀÇ ¼Õµ¿ÀÛ ÀÎ½Ä ¾Ë°í¸®ÁòÀº ¼ÕÀÇ À±°û¼±À» ±¸¼ºÇÏ´Â ¸ðµç Á¡À» ÃßÃâÇÏ´Â À±°û¼± ÃßÀû °úÁ¤ÀÇ °è»êº¹Àâµµ°¡ ³ô±â ¶§¹®¿¡ ÀÓº£µðµå ½Ã½ºÅÛ, ¸ð¹ÙÀÏ µð¹ÙÀ̽º¿Í °°Àº Àú¼º´ÉÀÇ ½Ã½ºÅÛ¿¡¼­ÀÇ È°¿ë¿¡ ¾î·Á¿òÀÌ ÀÖ¾ú´Ù. Á¦¾ÈÇÏ´Â ¾Ë°í¸®ÁòÀº À±°û¼± ÃßÀû ¾Ë°í¸®ÁòÀ» »ç¿ëÇÏ´Â ´ë½Å µ¿½É¿ø ÃßÀûÀ» ÀÀ¿ëÇÏ¿© Ãß»óÈ­µÈ ¼Õ°¡¶ôÀÇ À±°û¼±À» ÃßÁ¤ÇÑ ´ÙÀ½ Ư¡À» ÃßÃâÇÏ¿© ¼Õµ¿ÀÛÀ» ºÐ·ùÇÑ´Ù. Á¦¾ÈµÈ ¾Ë°í¸®ÁòÀº Æò±Õ ÀνķüÀº 95%ÀÌ°í Æò±Õ ¼öÇà½Ã°£Àº 1.29ms·Î¼­ ±âÁ¸ÀÇ À±°û¼± ÃßÀû ¹æ½ÄÀ» »ç¿ëÇÏ´Â ¾Ë°í¸®Áò¿¡ ºñÇØ ÃÖ´ë 44%ÀÇ ¼º´ÉÇâ»óÀ» º¸¿´°í ÀÓº£µðµå ½Ã½ºÅÛ, ¸ð¹ÙÀÏ µð¹ÙÀ̽º¿Í °°Àº Àú¼º´ÉÀÇ ½Ã½ºÅÛ¿¡¼­ÀÇ È°¿ë°¡´É¼ºÀ» È®ÀÎÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
In this paper, we propose a fast hand-gesture recognition algorithm for embedded system. Existing hand-gesture recognition algorithm has a difficulty to use in a low performance system such as embedded systems and mobile devices because of high computational complexity of contour tracing method that extracts all points of hand contour. Instead of using algorithms based on contour tracing, the proposed algorithm uses concentric-circle tracing method to estimate the abstracted contour of fingers, then classify hand-gestures by extracting features. The proposed algorithm has an average recognition rate of 95% and an average execution time of 1.29ms, which shows a maximum performance improvement of 44% compared with algorithm using the existing contour tracing method. It is confirmed that the algorithm can be used in a low performance system such as embedded systems and mobile devices.
Å°¿öµå(Keyword) ¼Õµ¿ÀÛÀνĠ  Àΰ£-ÄÄÇ»ÅÍ »óÈ£ÀÛ¿ë   ÄÄÇ»ÅÍ ºñÀü   ÀÓº£µðµå ½Ã½ºÅÛ   Hand Gesture Recognition   Human-Computer Interaction   Computer Vision   Embedded System  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå