• 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) ÇÕ¼º°ö ½Å°æ¸ÁÀ» ÀÌ¿ëÇÑ ¾ß»ý ¸Å°³¸ð±âÁ¾ÀÇ ºÐ·ù
¿µ¹®Á¦¸ñ(English Title) Classification of Wild Vector Mosquito Species using Convolutional Neural Networks
ÀúÀÚ(Author) ¹ÚÁØ¿µ   ±èµ¿ÀΠ  ±ÇÇü¿í   °­¿ìö   Junyoung Park   Dong In Kim   Hyung Wook Kwon   Woochul Kang  
¿ø¹®¼ö·Ïó(Citation) VOL 27 NO. 11 PP. 0503 ~ 0509 (2021. 11)
Çѱ۳»¿ë
(Korean Abstract)
ÃÖ±Ù °¨¿°º´À» ¸Å°³ÇÏ´Â ¸ð±âÀÇ ¹ß»ý ºÐÆ÷°¡ È®´ëµÊ¿¡ µû¶ó ÀÌµé °³Ã¼ÀÇ ½Å¼ÓÇÑ ¹æÁ¦¸¦ À§ÇØ ÀÌµé °³Ã¼ÀÇ ºÐÆ÷¸¦ ºü¸£°Ô ÆľÇÇÏ´Â °ÍÀÌ ¿ä±¸µÇ°í ÀÖ´Ù. ±×·¯³ª ±âÁ¸ ½Ã½ºÅÛ¿¡ Àû¿ëµÈ ¸ð±â ½Äº° ¾Ë°í¸®ÁòÀº ¾ß»ý ¸ð±âÀÇ Á¾º° ºÐ·ù°¡ ºÒ°¡´ÉÇÏ´Ù´Â ÇÑ°è°¡ ÀÖ´Ù. ÀÌ·¯ÇÑ ¹®Á¦¸¦ ÇØ°áÇϱâ À§ÇØ, ÀÌ ¿¬±¸¿¡¼­´Â ¾ß»ý¿¡¼­ ³ªÅ¸³ª´Â ¸ð±âÀÇ Á¾ ºÐ·ù°¡ °¡´ÉÇÑ ÇÕ¼º°ö ½Å°æ¸Á ¸ðµ¨À» ÇнÀÇÏ°í Æò°¡ÇÑ´Ù. ÇнÀ¿¡ ÇÊ¿äÇÑ µ¥ÀÌÅ͸¦ ÃëµæÇϱâ À§ÇØ »ì¾ÆÀÖ´Â ¸ð±âÀÇ À̹ÌÁö¸¦ ¾ß»ý¿¡¼­ È¿À²ÀûÀ¸·Î ÃëµæÇÒ ¼ö ÀÖ´Â Æ÷Áý±â ÇüÅÂÀÇ ÃÔ¿µÀåÄ¡¸¦ Á¦ÀÛÇÏ¿´°í À̸¦ »ç¿ëÇÏ¿© ÁÖ¿ä °¨¿°º´ ¸Å°³ ¸ð±âÀÎ ÈòÁÙ½£¸ð±â, »¡°£Áý¸ð±â, ¾ó·è³¯°³¸ð±â¼ÓÀ» Æ÷ÇÔÇÑ 1¸¸ Àå ÀÌ»óÀÇ À̹ÌÁö¸¦ ÃëµæÇÏ¿© µ¥ÀÌÅÍ ¼¼Æ®¸¦ ±¸¼ºÇÏ¿´´Ù. ±× °á°ú, ÇнÀÇÑ ¸ðµ¨¿¡¼­ °ËÁõ µ¥ÀÌÅÍ ¼¼Æ®¿¡ ´ëÇÏ¿© ÃÖ´ë 96.87%, ¾ß»ý µ¥ÀÌÅÍ ¼¼Æ®¿¡ ´ëÇÏ¿© ÃÖ´ë 67.89%ÀÇ ºÐ·ù Á¤È®µµ¸¦ È®ÀÎÇÏ¿© ÁöÇâÇÏ´Â Æ÷Áý±â ½Ã½ºÅÛ¿¡¼­ÀÇ Àû¿ë °¡´É¼º°ú °³¼± ¹æÇâÀ» È®ÀÎÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
The distribution of infectious mosquitoes has been constantly expanding, thus identifying the species is required for rapid pest control. However, the current mosquito identification algorithm could not apply to wild mosquito species classification. To solve this problem, we propose a convolutional neural network model for classifying vector mosquito species in the wild. To acquire data for training and evaluation, we developed a trap-shaped imaging device to efficiently acquire live mosquito images in the wild and built datasets including more than 10,000 images of Aedes albopictus, Culex pipiens, and Anopheles Spp. As a result, our model achieved up to 96.87% of validation accuracy and 67.89% of wild mosquito classification accuracy, which shows great prospects for the future trap system and a way for further improvement.
Å°¿öµå(Keyword) ¸Å°³ ¸ð±â   °ïÃæ ºÐ·ù   ÇÕ¼º°ö ½Å°æ¸Á   µö·¯´×   vector mosquitoes   insects classification   convolutional neural networks   deep learning  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå