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

»çÀÌÆ®¸Ê

Loading..

Please wait....

Çмú´ëȸ ÇÁ·Î½Ãµù

Ȩ Ȩ > ¿¬±¸¹®Çå > Çмú´ëȸ ÇÁ·Î½Ãµù > Çѱ¹Á¤º¸Åë½ÅÇÐȸ Çмú´ëȸ > 2014³â Ãß°èÇмú´ëȸ

2014³â Ãß°èÇмú´ëȸ

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) SVR¿¡ ±â¹ÝÇÑ °³¼±µÈ ³×À̹ö ÀÓº£µù
¿µ¹®Á¦¸ñ(English Title) Advanced Neighbor Embedding based on Support Vector Regression
ÀúÀÚ(Author) ¾ö°æ¹è   Àüâ¿ì   ÃÖ¿µÈñ   ³²½ÂÅ   ÀÌÁ¾Âù   Kyoung-Bae Eum   Chang-Woo Jeon   Young-Hee Choi   Seung-Tae Nam   Jong-Chan Lee  
¿ø¹®¼ö·Ïó(Citation) VOL 18 NO. 02 PP. 0733 ~ 0735 (2014. 10)
Çѱ۳»¿ë
(Korean Abstract)
Ç¥º»±â¹Ý ÃÊÇØ»óµµ(Super Resolution ÀÌÇÏ SR) ±â¹ýÀº µ¥ÀÌÅͺ£À̽º¿¡ ÀúÀåµÈ °íÇØ»óµµ ¿µ»óÀÇ ÆÐÄ¡¿Í ÀúÇØ»óµµ ¿µ»óÀÇ ÆÐÄ¡ »çÀÌ¿¡ ´ëÀÀ°ü°è¸¦ ÀÌ¿ëÇÏ¿©, ÀúÇØ»óµµÀÇ ÀԷ¿µ»ó¿¡ °¡Àå À¯»çÇÑ °íÇØ»óµµ ÆÐÄ¡¸¦ µ¡ºÙ¿©¼­ °íÇػ󵵸¦ ±¸¼ºÇÏ´Â ¹æ½ÄÀÌ´Ù. ÀÌ·¯ÇÑ ¹æ½ÄÀº ÇÑ ÀåÀÇ ¿µ»ó¸¸À¸·Î °íÇØ»óµµ ¿µ»óÀ» ¾òÀ» ¼ö ÀÖ°í, À§ÀÇ °úÁ¤À» ¹Ýº¹ÇÏ¿© 2¹è ÀÌ»óÀÇ È®´ëµÈ ¿µ»óÀ» ¾òÀ» ¼ö À־ ±âÁ¸ÀÇ °íÀüÀû SRÀÇ ¹®Á¦Á¡À» ÇØ°áÇÒ ¼ö ÀÖ´Ù. Ç¥º»±â¹Ý SRÀÇ ¹æ¹ýµé Áß ³×À̹ö ÀÓº£µù(Neighbor Embedding ÀÌÇÏ NE) ±â¹ýÀÇ ±âº» ¿ø¸®´Â Áö¿ªÀû ¼±Çü ÀÓº£µùÀ̶ó´Â ¸Å´ÏÆúµå ÇнÀ¹æ¹ýÀÇ °³³ä°ú °°´Ù. ±×·¯³ª ³×À̹ö ÀÓº£µùÀÇ ºó¾àÇÑ ÀϹÝÈ­ ´É·ÂÀ¸·Î ÀÎÇÏ¿© ¾Ë°í¸®ÁòÀÇ ¼º´ÉÀ» Å©°Ô ÀúÇϽÃŲ´Ù. ÀÌÀ¯´Â ±¹ºÎÇнÀ µ¥ÀÌÅÍ ÁýÇÕÀÇ Å©±â°¡ ³Ê¹« À۾Ƽ­ NE ¾Ë°í¸®ÁòÀÇ ¼º´ÉÀ» ÇöÀúÈ÷ ÀúÇϽÃŲ´Ù. º» ³í¹®¿¡¼­´Â ÀÌ¿Í °°Àº ¹®Á¦Á¡À» ÇØ°áÇϱâ À§Çؼ­ ÀϹÝÈ­ ´É·ÂÀÌ ¶Ù¾î³­ Support Vector Regression(ÀÌÇÏ SVR)±â¹Ý °³¼±µÈ NE¸¦ Á¦¾ÈÇÏ¿´´Ù. ÀúÇØ»óµµ ÀÔ·Â ÆÐÄ¡°¡ ÁÖ¾îÁö¸é SVR ±â¹Ý °³¼±µÈ NE¸¦ ÀÌ¿ëÇÏ¿© °íÇØ»óµµÀÇ ÇØ´ç È­¼Ò °ªÀ» ¿¹ÃøÇÏ¿´´Ù. ½ÇÇèÀ» ÅëÇÏ¿© Á¦¾ÈµÈ ±â¹ýÀÌ ±âÁ¸ÀÇ º¸°£¹ý ¹× NE ±â¹ý µî¿¡ ºñÇØ Á¤·®ÀûÀΠôµµ ¹× ½Ã°¢ÀûÀ¸·Î Çâ»óµÈ °á°ú¸¦ º¸¿© ÁÖ¾ú´Ù.
¿µ¹®³»¿ë
(English Abstract)
Example based Super Resolution(SR) is using the correspondence between the low and high resolution image from a database. This method uses only one image to estimate a high resolution image and can get the larger image than 2 times. Example based SR is proposed to solve the problem of classical SR. Neighbor embedding(NE) has been inspired by manifold learning method, particularly locally linear embedding. However, the poor generalization of NE decreases the performance of such algorithm. The sizes of local training sets are always too small to improve the performance of NE. We propose the advanced NE baesd on SVR having an excellent generalization ability to solve this problem. Given a low resolution image, we estimate a pixel in its high resolution version by using SVR based NE. Through experimental results, we quantitatively and qualitatively confirm the improved results of the proposed algorithm when comparing with conventional interpolation methods and NE.
Å°¿öµå(Keyword) Super Resolution   Neighbor Embedding   Support Vector Regression  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå