• 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) U-Net ¾Ó»óºíÀ» ÀÌ¿ëÇÑ LeafNetÀÇ ¼º´ÉÇâ»ó
¿µ¹®Á¦¸ñ(English Title) Performance Improvement of LeafNet Using U-Net Ensemble
ÀúÀÚ(Author) Á¶Á¤¿ø   ÀÌâ¿ì   Jeong-won Jo   Chang-woo Lee  
¿ø¹®¼ö·Ïó(Citation) VOL 23 NO. 02 PP. 0377 ~ 0378 (2019. 10)
Çѱ۳»¿ë
(Korean Abstract)
À̹ÌÁö 󸮸¦ ¸ñÀûÀ¸·Î ÇÏ´Â ½Å°æ¸ÁÀº ´ëºÎºÐ RGB »ö°ø°£¿¡¼­ µ¥ÀÌÅ͸¦ ÇнÀÇÑ´Ù. ÇÏÁö¸¸ »ö°ø°£Àº RGB »Ó¸¸ ¾Æ´Ï¶ó LAB, YCrCbµî ¿©·¯ °³°¡ Á¸ÀçÇÑ´Ù. °¢°¢ÀÇ »ö°ø°£Àº µ¿ÀÏÇÑ È­¼Ò¿¡ ´ëÇØ ´Ù¸¥ °ªÀ¸·Î Ç¥ÇöÀ» Çϴ Ư¡ÀÌ ÀÖ´Ù. º» ³í¹®¿¡¼­´Â ¾Ó»óºí ±¸Á¶·Î ÀÌ·ç¾îÁø 3°³ÀÇ ¼ÒÇüÈ­ÇÑ U-Net¿¡ RGB, LAB, YCrCb »ö°ø°£À¸·Î º¯È¯ÇÑ ½Ä¹° À̹ÌÁö¸¦ ÇнÀÇÏ¿© ±âÁ¸ LeafNetÀÇ ¼º´ÉÇâ»óÀ» À§ÇÑ ¿¬±¸¸¦ ÁøÇàÇß´Ù.
¿µ¹®³»¿ë
(English Abstract)
Neural networks that are intended for image processing mostly learn data from RGB color space. However, not only RGB, but also LAB, YCrCb, etc Each color space is characterized by different values for the same pixel. In this paper, we studied the plant images converted into RGB, LAB and YCrCb color spaces in three miniaturized U-Net composed of ensemble structures, and conducted research for the performance enhancement of existing LeafNet.Neural networks, which are aimed at processing images, mostly learn data in the RGB color However, not only RGB, but also LAB, YCrCb, etc. Each color space is expressed differently for the same pixel. In this paper, we studied the performance improvement of LeafNet by learning RGB, LAB, and YCrCb color space using U-Net ensembles.
Å°¿öµå(Keyword) ÇÕ¼º°ö ½Å°æ¸Á   À̹ÌÁö ºÐÇÒ   U-Net   ½Ä¹°   ¾Ó»óºí   CNN   Image Segmentation   U-Net   Plant   Ensemble  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå