Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
±íÀÌ ¿µ»ó ±â¹Ý ÇÊÅ͸¦ Àû¿ëÇÑ È¿°úÀûÀΠȣ¸ð±×·¡ÇÇ ÃßÁ¤ ¹æ¹ý |
¿µ¹®Á¦¸ñ(English Title) |
A Method for Effective Homography Estimation Applying a Depth Image-Based Filter |
ÀúÀÚ(Author) |
ÁÖ¿ëÁØ
È«¸í´ö
À±Àdzç
°í½ÂÇö
Á¶±Ù½Ä
Yong-Joon Joo
Myung-Duk Hong
Ui-Nyoung Yoon
Seung-Hyun Go
Geun-Sik Jo
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 08 NO. 02 PP. 0061 ~ 0066 (2019. 02) |
Çѱ۳»¿ë (Korean Abstract) |
Áõ°Çö½ÇÀº Ä«¸Þ¶ó·Î ÃÔ¿µÇÏ°í ÀÖ´Â ¿µ»ó¿¡ °¡»óÀÇ °´Ã¼¸¦ ½Ç½Ã°£À¸·Î ÇÕ¼ºÇÏ¿© °¡»óÀÇ °´Ã¼°¡ Çö½Ç¿¡ Á¸ÀçÇÏ´Â °Íó·³ º¸ÀÌ°Ô ÇÏ´Â ±â¼úÀÌ´Ù. Áõ°Çö½Ç¿¡¼ Çö½Ç¿¡ Á¸ÀçÇÏ´Â ¹°Ã¼¿¡ °¡»óÀÇ ¹°Ã¼¸¦ Áõ°Çϱâ À§Çؼ´Â Çö½Ç¿¡ Á¸ÀçÇÏ´Â ¹°Ã¼ÀÇ À§Ä¡¿Í ¹æÇâÀ» Á¤È®ÇÏ°Ô ÃßÁ¤ÇØ¾ß Çϴµ¥, ÀÌ ¶§ È°¿ëµÇ´Â ±â¼úÀÌ ¿µ»óÀÇ È£¸ð±×·¡ÇÇ(Homography) ÀÌ´Ù. ÀÌ·¯ÇÑ È£¸ð±×·¡ÇÇ´Â ¿µ»óÀÇ Æ¯Â¡Á¡ ÁÂÇ¥¿¡ RANSAC ¾Ë°í¸®ÁòÀ» Àû¿ëÇÏ¿© ÃßÁ¤ÇÒ ¼ö Àִµ¥, RANSAC ¾Ë°í¸®ÁòÀ» ÀÌ¿ëÇÑ È£¸ð±×·¡ÇÇ ÃßÁ¤ ¹æ½ÄÀº È£¸ð±×·¡ÇǸ¦ ÃßÁ¤ÇÏ°íÀÚÇÏ´Â ¹°Ã¼ ÀÌ¿ÜÀÇ ¹è°æ¿¡ Ư¡Á¡ÀÌ ¸¹À» °æ¿ì Á¤È®ÇÑ È£¸ð±×·¡ÇǸ¦ ÃßÁ¤ÇÒ ¼ö ¾ø´Ù´Â ¹®Á¦Á¡ÀÌ Á¸ÀçÇß´Ù. º» ³í¹®¿¡¼´Â È£¸ð±×·¡ÇǸ¦ ÃßÁ¤ÇÏ°íÀÚÇÏ´Â ¹°Ã¼°¡ °¡±îÀÌ¿¡ ÀÖ°í ¹è°æÀÌ »ó´ëÀûÀ¸·Î ¸Ö¸® À§Ä¡ÇØÀÖÀ» ¶§ ¿µ»ó °¢ Çȼ¿ÀÇ °Å¸® °ªÀ» ¾Ë ¼ö ÀÖ´Â ±íÀÌ ¿µ»óÀ» È°¿ëÇÏ¸é ¹°Ã¼¿Í ¹è°æÀ» ½±°Ô ºÐ¸®ÇÒ ¼ö ÀÖ´Ù´Â Á¡À» ÀÌ¿ëÇÏ¿© ¹è°æÀÇ Æ¯Â¡Á¡À» ÇÊÅ͸µÇÏ´Â ¹æ¹ýÀ» Á¦¾ÈÇÑ´Ù. À̸¦ À§ÇÏ¿© º» ³í¹®¿¡¼´Â Èæ¹éÁ¶ ¿µ»óÀÎ ±íÀÌ ¿µ»óÀ» Otsu ¾Ë°í¸®ÁòÀ» ÀÌ¿ëÇÏ¿© »ç¿ëÀÚ¿Í °Å¸®°¡ °¡±î¿î ¿µ¿ª°ú °Å¸®°¡ ¸Õ ¿µ¿ªÀ¸·Î ÀÌÁøÈÇÏ°í, RGB ¿µ»ó¿¡¼ ÃßÃâµÈ Ư¡Á¡ Áß¿¡¼ °Å¸®°¡ ¸Õ ¿µ¿ª¿¡ À§Ä¡ÇÑ Æ¯Â¡Á¡À» Á¦°ÅÇÔÀ¸·Î½á Ư¡Á¡À» È°¿ëÇÑ È£¸ð±×·¡ÇÇ ÃßÁ¤ °á°ú¸¦ Çâ»ó½ÃŲ´Ù. ÀÌ·¯ÇÑ ¹æ¹ýÀ» ±âÁ¸ÀÇ È£¸ð±×·¡ÇÇ ÃßÁ¤ ¹æ¹ý¿¡ Àû¿ëÇÑ °á°ú ¼öÇà½Ã°£ÀÌ 71.7% ´ÜÃàµÇ¾úÀ¸¸ç, RANSAC ¾Ë°í¸®ÁòÀÇ ¹Ýº¹ Ƚ¼ö°¡ 69.4% ÁÙ¾îµé¾ú°í, ÂüÁ¤º¸ ºñÀ²ÀÌ 16.9% Áõ°¡ÇÏ¿´´Ù. |
¿µ¹®³»¿ë (English Abstract) |
Augmented reality is a technology that makes a virtual object appear as if it exists in reality by composing a virtual object in real time with the image captured by the camera. In order to augment the virtual object on the object existing in reality, the homography of images utilized to estimate the position and orientation of the object. The homography can be estimated by applying the RANSAC algorithm to the feature points of the images. But the homography estimation method using the RANSAC algorithm has a problem that accurate homography can not be es |
Å°¿öµå(Keyword) |
Áõ°Çö½Ç
È£¸ð±×·¡ÇÇ
RANSAC
±íÀÌ ¿µ»ó
Augmented Reality
Homography
RANSAC
Depth Image
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|