Á¤º¸°úÇÐȸ ³í¹®Áö B : ¼ÒÇÁÆ®¿þ¾î ¹× ÀÀ¿ë
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
¸ðµ¨¿¡ ±â¹ÝÇÑ À̵¿ ¹°Ã¼ÀÇ ÃßÀû |
¿µ¹®Á¦¸ñ(English Title) |
Model-based Tracking of Moving Object |
ÀúÀÚ(Author) |
Àå´ë½Ä
±è°è¿µ
ÃÖÇüÀÏ
Daesik Jang
Gyeyoung Kim
Hyungil Choi
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 23 NO. 12 PP. 1298 ~ 1308 (1996. 12) |
Çѱ۳»¿ë (Korean Abstract) |
º» ³í¹®¿¡¼´Â Ä«¸Þ¶óÀÇ ½Ã°è(FOV: Field Of View) ¾ÈÀ¸·Î ÁøÀÔÇÏ´Â ¹°Ã¼¸¦ °¨ÁöÇÏ°í °¨ÁöµÈ ¹°Ã¼¸¦ ÃßÀûÇÏ¿© ¿µ»ó ÇÁ·¹ÀÓÀÇ Áß¾Ó¿¡ À§Ä¡Çϵµ·Ï ÇÏ´Â ½Ç½Ã°£ ÃßÀû ½Ã½ºÅÛÀ» Á¦¾ÈÇÑ´Ù. º» ³í¹®¿¡¼´Â ¿òÁ÷ÀÌ´Â ¹°Ã¼¸¦ °¨ÁöÇÏ°í ÃßÀûÇϱâ À§ÇÑ ±âº»ÀûÀÎ Àü·«À¸·Î ¸ðµ¨ Á¤ÇÕ ±â¹ýÀ» ÀÌ¿ëÇÏ¸ç ¸ñÇ¥ ¹°Ã¼ÀÇ ¸ð¾çÀ̳ª ¸í¾ÏÀÇ º¯È¿¡ ÀûÀÀÀûÀ¸·Î ÃßÀûÀ» ¼öÇàÇϱ⠿ìÇØ ¸ðµ¨ÀÌ ÃßÀû Áß¿¡ µ¿ÀûÀ¸·Î º¯ÈÇÒ ¼ö ÀÖµµ·Ï Çã¿ëÇÑ´Ù. ¶ÇÇÑ Ä®¸¸ ÇÊÅ͸¦ ÀÌ¿ëÇÏ¿© ¹°Ã¼ÀÇ À̵¿ ±ËÀûÀ» Ä®¸¸ ÇÊÅÍÀÇ »óÅ Àμöµé¿¡ ÀúÀåÇÑ´Ù. ±×¸®°í Ä®¸¸ ÇÊÅÍÀÇ ¿¹ÃøµÈ »óÅ ÀμöµéÀº ¸ðµ¨ Á¤ÇÕ°ú ¼º¸ÀåÄ¡ Á¦¾î¸¦ È¿À²ÀûÀ¸·Î ¼öÇàÇÏ°í Ž»ö ¿µ¿ªÀ» Á¦ÇÑÇϴµ¥ ÀÌ¿ëµÈ´Ù. ½ÇÇè °á°ú´Â ¸ðµ¨ Á¤ÇÕÀ» ÅëÇØ ¸ñÇ¥ ¹°Ã¼¸¦ °°ÇÇÏ°Ô ÃßÀûÇÒ ¼ö ÀÖÀ½À» º¸¿© ÁØ´Ù. ¶ÇÇÑ ½ÇÇèÀ» ÅëÇØ ¿¹ÃøµÈ Ä®¸¸ ÇÊÅÍÀÇ ¿¹Ãø°ªµéÀÌ ¼º¸ÀåÄ¡¸¦ Á¦¾îÇÏ°í ¸ðµ¨ Á¤ÇÕÀ» À§ÇØ Å½»ö ¿µ¿ªÀ» ã´Âµ¥ Á¤È®ÇÏ°Ô ÀÌ¿ëµÇ´Â °ÍÀ» ¾Ë ¼ö ÀÖ´Ù.
|
¿µ¹®³»¿ë (English Abstract) |
This paper describes a real-time tracking system which detects an object entering into the field of view of camera and executes the tracking of detected object by controlling a servo device so that a target object always lies at the center of image frame. In order to detect and track a moving object, we basically apply a model matching strategy. We allow the model to vary dynamically during the tracking process so that it can assimilate the variations of shape and intensities of target object. We also utilize Kalman filter so that a tracking history can be encoded into state parameters of Kalman filter. The estimated state parameters of Kalman filter will then be used to reduce search areas for model matching and to control servo device.
Experimental results show that the model adaptation allows robust tracking of target object in dynamic environments. These experiments also confirm that the predicted values of Kalman filter are very accurate in controlling the servo device and finding out the search areas for model matching. This paper concludes with the theoretical bound within which a tracking system can follow the movement of target object.
|
Å°¿öµå(Keyword) |
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|