Á¤º¸°úÇÐȸ ³í¹®Áö B : ¼ÒÇÁÆ®¿þ¾î ¹× ÀÀ¿ë
ÇѱÛÁ¦¸ñ(Korean Title) |
½Ç½Ã°£ ÀÀ¿ëÀ» À§ÇÑ ¿þÀÌºí¸´ º¯È¯ ±â¹ÝÀÇ ¾ó±¼ °ËÃâ |
¿µ¹®Á¦¸ñ(English Title) |
Wavelet Transform-based Face Detection for Real-time Applications |
ÀúÀÚ(Author) |
¼ÛÇØÁø
°íº´Ã¶
º¯Çý¶õ
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 30 NO. 09 PP. 0829 ~ 0842 (2003. 10) |
Çѱ۳»¿ë (Korean Abstract) |
ÃÖ±Ù È»ó ȸÀÇ, È»ó ÀüÈ, ¸ð¹ÙÀÏ È¯°æ¿¡¼ÀÇ È»ó Åë½Å, ¾ó±¼ ÀνÄÀ» ÀÌ¿ëÇÑ º¸¾È ½Ã½ºÅÛ µîÀÇ »ó¾÷È¿¡ ÈûÀÔ¾î ºñµð¿À¿¡¼ÀÇ ¾ó±¼ °ËÃâ ¹× ÃßÀû ±â¼úÀº ´«ºÎ½Å ¹ßÀüÀ» ÀÌ·èÇÏ¿´´Ù. ¶ÇÇÑ, ¾ó±¼ ¿ä¼Ò °ËÃâÀº ¿ä¼Ò ±× ÀÚü»Ó ¾Æ´Ï¶ó Á¤È®ÇÑ ¾ó±¼ ¿µ¿ª °ËÃâÀ» À§ÇÑ Çʼö ´Ü°è·Î¼ Áß¿äÇÑ ¿¬±¸ ÁÖÁ¦°¡ µÇ°í ÀÖ´Ù. ±×·¯³ª ¿µ»ó¿¡ ³ªÅ¸³ º¹ÀâÇÑ ¹è°æ°ú Ä«¸Þ¶ó Á¶ÀÛ ¹× Á¶¸í¿¡ ÀÇÇÑ »ö»ó ¿Ö°î ±×¸®°í ´Ù¾çÇÑ Á¶¸í Á¶°Ç µîÀº ¾ó±¼ °ËÃâ ¹× ÃßÀû, ¿ä¼Ò °ËÃâ¿¡ ÀÖ¾î ¿©ÀüÈ÷ Å« Àå¾Ö°¡ µÇ°í ÀÖ´Ù. ÀÌ¿¡ µû¶ó, º» ³í¹®¿¡¼´Â ½Ç½Ã°£ È»ó Åë½ÅÀ» À§ÇÑ »õ·Î¿î ¾ó±¼ ¿µ¿ª °ËÃâ ¹× ÃßÀû ¾Ë°í¸®Áò°ú °ËÃâµÈ ¾ó±¼ ¿µ¿ª¿¡¼ È¿°úÀûÀ¸·Î ´« ¿µ¿ªÀ» °ËÃâÇÒ ¼ö ÀÖ´Â ¾Ë°í¸®ÁòÀ» Á¦¾ÈÇÑ´Ù. Á¦¾ÈÇÏ´Â ¾ó±¼ °ËÃâ ¾Ë°í¸®ÁòÀº º¹ÀâÇÑ ¹è°æ°ú ´Ù¾çÇÑ Á¶¸í Á¶°Ç¿¡ °ü°è¾øÀÌ ¾ó±¼À» °ËÃâÇÏ°í ÃßÀûÇϱâ À§ÇØ ¿þÀÌºí¸´ º¯È¯µÈ ¼¼ Á¾·ùÀÇ ºÎ ¿µ¿ªÀ» ÀÌ¿ëÇÏ¿© ¾ó±¼ ÇüÆÇÀ» »ý¼ºÇÏ°í ¿þÀÌºí¸´ º¯È¯µÈ ÀÔ·Â ¿µ»ó°úÀÇ À¯»çµµ¸¦ ÃøÁ¤ÇÏ¿© ¾ó±¼À» °ËÃâÇÑ´Ù. ƯÈ÷ ´Ù¾çÇÑ Á¶¸í Á¶°ÇÀ» ±Øº¹Çϱâ À§ÇØ ÃÖ¼Ò-ÃÖ´ë Á¤±ÔÈ¿Í È÷½ºÅä±×·¥ ÆòȰȸ¦ È¥ÇÕ Àû¿ëÇÏ¿© ¸Å¿ì ¹à°Å³ª, ¸Å¿ì ¾îµÎ¿î ¿µ»ó¿¡¼ÀÇ ¾ó±¼ ¿À °ËÃâ ¹× ³õħÀ» ÁÙÀÏ ¼ö ÀÖ¾úÀ¸¸ç ¼¼ °¡Áö Å©±âÀÇ ¾ó±¼ ÇüÆÇÀ» ÀÌ¿ëÇÔÀ¸·Î½á ÀÔ·Â ¿µ»ó¿¡ Á¸ÀçÇÏ´Â ´Ù¾çÇÑ Å©±âÀÇ ¾ó±¼µµ °ËÃâÇÒ ¼ö ÀÖ¾ú´Ù. ¶ÇÇÑ È¿°úÀûÀÎ ¾ó±¼ ÃßÀû ¾Ë°í¸®ÁòÀ» ÅëÇØ ´ÙÀ½ ÇÁ·¹ÀÓ¿¡¼ÀÇ ¾ó±¼ À§Ä¡¸¦ ¿¹ÃøÇÏ°í ¿¹ÃøµÈ ¾ó±¼ À§Ä¡¸¦ Áß½ÉÀ¸·Î Ž»ö ¿µ¿ªÀ» Á¤ÇØ ÇüÆÇ Á¤ÇÕÀ» ¼öÇàÇÔÀ¸·Î½á ¾ó±¼ °ËÃâ·üÀ» ³ôÀÌ¸é¼ ¼öÇà ½Ã°£µµ ´ÜÃà½Ãų ¼ö ÀÖ¾ú´Ù. ¼öÁ÷, ¼öÆò ¹æÇâ Åõ¿µÀ» ÀÌ¿ëÇÑ ÇÕ¸®ÀûÀÎ ´« °ËÃâ ¾Ë°í¸®ÁòÀº ¾îµÎ¿î Á¶¸íÀ̳ª ºÎÁ¤È®ÇÑ ¾ó±¼ ¿µ¿ª¿¡¼µµ ¸¸Á·½º·¯¿î °á°ú¸¦ º¸¿©ÁÖ¾ú´Ù. |
¿µ¹®³»¿ë (English Abstract) |
In this paper, we propose the new face detection and tracking method based on template matching for real-time applications such as, teleconference, telecommunication, front stage of surveillance system using face recognition, and video-phone applications. Since the main purpose of paper is to track a face regardless of various environments, we use template-based face tracking method. To generate robust face templates, we apply wavelet transform to the average face image and extract three types of wavelet template from transformed low-resolution average face. However template matching is generally sensitive to the change of illumination conditions, we apply Min-max normalization with histogram equalization according to the variation of intensity. Tracking method is also applied to reduce the computation time and predict precise face candidate region. Finally, facial components are also detected and from the relative distance of two eyes, we estimate the size of facial ellipse. |
Å°¿öµå(Keyword) |
¾ó±¼ °ËÃâ
¿þÀÌºí¸´ ¾ó±¼ ÇüÆÇ
ÇüÆÇ Á¤ÇÕ
¾ó±¼ ÃßÀû
´« °ËÃâ
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|