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

»çÀÌÆ®¸Ê

Loading..

Please wait....

±¹³» ³í¹®Áö

Ȩ Ȩ > ¿¬±¸¹®Çå > ±¹³» ³í¹®Áö > Çѱ¹Á¤º¸°úÇÐȸ ³í¹®Áö > Á¤º¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)

Á¤º¸°úÇÐȸ ÄÄÇ»ÆÃÀÇ ½ÇÁ¦ ³í¹®Áö (KIISE Transactions on Computing Practices)

Current Result Document :

ÇѱÛÁ¦¸ñ(Korean Title) YOLO ¾Ë°í¸®ÁòÀ» ÀÌ¿ëÇÑ ÀüÂ÷ ±¹Àû ½Äº° ¹× Æò°¡
¿µ¹®Á¦¸ñ(English Title) Identifications and Evaluation of Tank Nationality using YOLO Algorithm
ÀúÀÚ(Author) Àӽ±ՠ  °­µ¿¼ö   Seunggyun Lim   Dongsu Kang  
¿ø¹®¼ö·Ïó(Citation) VOL 27 NO. 12 PP. 0555 ~ 0562 (2021. 11)
Çѱ۳»¿ë
(Korean Abstract)
ÀΰøÁö´É µö·¯´× ±â¼úÀÌ Àû¿ëµÈ ¹«±âü°è°¡ Áö¼ÓÀûÀ¸·Î °³¹ßµÈ´Ù. ±âÁ¸ÀÇ ÀüÂ÷ ÇǾƽĺ° ½Ã½ºÅÛÀº »ç¶÷ÀÇ ´«À¸·Î Ç¥ÀûÀ» ȹµæÇÏ°í °ø°Ý ¿©ºÎ¸¦ ÆÇ´ÜÇÑ´Ù. µû¶ó¼­ ½Å¼Ó¼º°ú Á¤È®¼º Ãø¸é¿¡¼­ ÇÑ°è°¡ Á¸ÀçÇÑ´Ù. º» ¿¬±¸¿¡¼­´Â ÀÌ·¯ÇÑ Á¦ÇÑ»çÇ×À» °³¼±Çϱâ À§ÇØ YOLO(You Only Look Once) ¾Ë°í¸®Áò ±â¹ÝÀÇ ÀüÂ÷ ±¹Àû ½Äº° ¹æ¹ýÀ» Á¦¾ÈÇÑ´Ù. ¸ÕÀú, Çѱ¹, ¹Ì±¹, ÀϺ», ºÏÇÑÀÇ 4°³±¹ÀÇ ÁÖ·Â ÀüÂ÷ »çÁøÀ» ¼öÁýÇÑ´Ù. ƯÈ÷, ½ÇÁ¦ ±â°© ÀüÅõÀÇ »óȲ°ú À¯»çÇϵµ·Ï ³ëÀÌÁ Ãß°¡ÇÏ°í À̹ÌÁö Àüó¸® ÀÛ¾÷À» ÇÑ´Ù. ÀÌÈÄ µ¥ÀÌÅͼÂÀº ÇнÀ µ¥ÀÌÅÍÀÇ ÀûÀýÇÑ ±Ô¸ð¸¦ È®ÀÎÇϱâ À§ÇØ 8°³ÀÇ ±×·ìÀ¸·Î ±¸¼ºÇÑ´Ù. ¸¶Áö¸·À¸·Î Æò°¡ ôµµÀÎ mAP¿Í IoU¸¦ ±â¹ÝÀ¸·Î ÀûÀýÇÑ µ¥ÀÌÅÍ ±Ô¸ð¸¦ ºÐ¼®ÇÑ´Ù.
¿µ¹®³»¿ë
(English Abstract)
Advanced weapon systems with artificial intelligence deep learning technology will be continuously. The existing identification system of friend or foe on tank targets with human eyes determine whether to attack. Therefore, there are limitations in terms of speed and accuracy. In this paper, we propose a YOLO(You Only Look Once) Algorithm-based tank nationality identification method to improve these limitations. First, we collect photos of the main tank from four countries South Korea, the United States, Japan and North Korea. In particular, similar to the actual armored battle, we add noise and do image preprocessing. Afterward, the dataset organized into eight groups to check the appropriate size of the learning data. Finally, we analyze the appropriate data size based on the evaluation scales, mAP and IoU.
Å°¿öµå(Keyword) °´Ã¼Å½Áö   YOLO(You Only Look Once)   ¾Ë°í¸®Áò   µö·¯´×   µ¥ÀÌÅͼ   ÀüÂ÷   ÇǾƽĺ° ½Ã½ºÅÛ   mAP   IoU   object detection   YOLO(You Only Look Once) algorithm   deep learning   dataset   tank   Identification of Friend or Foe   mAP   IoU  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå