• 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 > 2019³â Ãß°èÇмú´ëȸ

2019³â Ãß°èÇмú´ëȸ

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) µö·¯´×À» ÀÌ¿ëÇÑ Ç×°ø ¿µ»ó¿¡¼­ÀÇ µµ·Î ÀνÄ
¿µ¹®Á¦¸ñ(English Title) Road Recognition in Aerial Images Using Deep Learning
ÀúÀÚ(Author) Á¤À¯¼®   À±ÇüÁø   Á¶Á¤¿ø   À̹ÎÇý   ÀÌâ¿ì   Yu-seok jeong   Hyeong-jin Youn   Jeong-won Jo   Min-hye Lee   Chang-woo Lee  
¿ø¹®¼ö·Ïó(Citation) VOL 23 NO. 02 PP. 0622 ~ 0623 (2019. 10)
Çѱ۳»¿ë
(Korean Abstract)
µµ·Î ¿µ¿ª ÀνĿ¡ ´ëÇÑ ¹æ¹ýÀº µö·¯´×À» ÅëÇØ ¸¹Àº ¹æ¹ýÀ¸·Î ½ÃµµµÇ¾ú´Ù. º» ³í¹®Àº u-netÀ» ÀÌ¿ëÇÏ¿© °íÇØ»óµµ Ç×°ø À̹ÌÁö¿¡¼­ µµ·Î ¿µ¿ª¿¡ ´ëÇÑ ºÎºÐÀ» °ËÃâÇÑ´Ù. °ËÃâ ¹× »óȲ¿¡ µû¸¥ µµ·Î ÀνķüÀ» ¿Ã¸®±â À§ÇÏ¿© µµ·Î À§ÀÇ ¸ðµç °´Ã¼¸¦ ÃßÃâÇÑ µ¥ÀÌÅÍ¿Í µµ·Î Àüü¸¦ ÅëÇÕÇÑ µ¥ÀÌÅ͸¦ ÀÌ¿ëÇÏ¿© ºñ±³ ÁøÇàÇÏ¿´´Ù. U-net¿¡¼­ µµ·ÎÀÇ Á÷¼±ÀûÀÎ ÇüŸ¦ °ËÃâÇϱâ À§ÇØ ¹Ýº¹ÀûÀÎ ÆÐÅÏ ÇüÅ¿¡ ÁÁÀº ¼º´ÉÀ» ³»´Â ¹Ì·¯¸µ ÆеùÀÌ ¾Æ´Ñ Á¦·Î ÆеùÀ¸·Î º¯°æÇÏ¿´´Ù. µµ·Î À§ÀÇ ¸ðµç °´Ã¼¸¦ Á¦¿ÜÇÏ¿© ÁøÇàÇÏ´Â °ÍÀÌ ´õ ¼±¸íÇÏ°í Á¤È®ÇÑ °á°ú¸¦ º¸¿©ÁÖ¾ú´Ù.
¿µ¹®³»¿ë
(English Abstract)
The method for road area recognition has been tried in many ways through deep learning. This paper uses u-net to detect parts of road area in high resolution aerial image. In order to increase the road recognition rate according to detection and situation, all objects on the road were compared with the data extracted from the road as a whole. In order to detect the straight-line shape of the road on U-net, it was changed to zero padding rather than mirroring padding, which gives good performance to repetitive pattern forms. The process of excluding all objects on the road showed clearer and more accurate results.
Å°¿öµå(Keyword) segmentation   u-net   GeoAI   µµ·Î ÀνĠ 
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå