• 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¸Åë½ÅÇÐȸ ³í¹®Áö (Journal of the Korea Institute of Information and Communication Engineering)

Çѱ¹Á¤º¸Åë½ÅÇÐȸ ³í¹®Áö (Journal of the Korea Institute of Information and Communication Engineering)

Current Result Document : 6 / 21 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) CNN°ú Grad-CAM ±â¹ÝÀÇ ½Ç½Ã°£ È­Àç °¨Áö
¿µ¹®Á¦¸ñ(English Title) Real-Time Fire Detection based on CNN and Grad-CAM
ÀúÀÚ(Author) ±è¿µÁø   ±èÀº°æ   Young-Jin Kim   Eun-Gyung Kim  
¿ø¹®¼ö·Ïó(Citation) VOL 22 NO. 12 PP. 1596 ~ 1603 (2018. 12)
Çѱ۳»¿ë
(Korean Abstract)
È­Àç¿¡ ´ëÇÑ ½Å¼ÓÇÑ ¿¹Ãø°ú °æ°í´Â ÀÎ¸í ¹× Àç»êÇÇÇظ¦ ÃÖ¼ÒÈ­½Ãų ¼ö ÀÖ´Â ÇʼöÀûÀÎ ¿ä¼ÒÀÌ´Ù. ÀϹÝÀûÀ¸·Î È­Àç°¡ ¹ß»ýÇÏ¸é ¿¬±â¿Í È­¿°ÀÌ ÇÔ²² ¹ß»ýÇϱ⠶§¹®¿¡ È­Àç °¨Áö ½Ã½ºÅÛÀº ¿¬±â¿Í È­¿°À» ¸ðµÎ °¨ÁöÇÒ ÇÊ¿ä°¡ ÀÖ´Ù. ±×·¯³ª ´ëºÎºÐÀÇ È­Àç °¨Áö ½Ã½ºÅÛÀº È­¿° ȤÀº ¿¬±â¸¸ °¨ÁöÇϸç, È­Àç °¨Áö¸¦ À§ÇÑ Àüó¸® ÀÛ¾÷À» Ãß°¡ÇÔ¿¡ µû¶ó ó¸® ¼Óµµ°¡ ´À·ÁÁö´Â ´ÜÁ¡ÀÌ ÀÖ´Ù. º» ¿¬±¸¿¡¼­´Â ´ÙÁß ·¹ÀÌºí ºÐ·ù(Multi-labeled Classification)¸¦ Áö¿øÇÏ´Â CNN ¸ðµ¨À» ±¸¼ºÇؼ­ È­¿°°ú ¿¬±â¸¦ µ¿½Ã¿¡ ¿¹ÃøÇÏ°í, CNNÀÇ Æ¯Â¡À» ±â¹ÝÀ¸·Î Ŭ·¡½º¿¡ ´ëÇÑ À§Ä¡¸¦ ½Ã°¢È­ÇÏ´Â Grad-CAMÀ» ÀÌ¿ëÇؼ­ ½Ç½Ã°£À¸·Î È­Àç »óŸ¦ ¸ð´ÏÅ͸µ ÇÒ ¼ö ÀÖ´Â È­Àç °¨Áö ½Ã½ºÅÛÀ» ±¸ÇöÇÏ¿´´Ù. ¶ÇÇÑ, 13°³ÀÇ È­Àç µ¿¿µ»óÀ» »ç¿ëÇؼ­ Å×½ºÆ®ÇÑ °á°ú, È­¿°°ú ¿¬±â¿¡ ´ëÇØ °¢°¢ 98.73%¿Í 95.77%ÀÇ Á¤È®µµ¸¦ º¸¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
Rapidly detecting and warning of fires is necessary for minimizing human injury and property damage. Generally, when fires occur, both the smoke and the flames are generated, so fire detection systems need to detect both the smoke and the flames. However, most fire detection systems only detect flames or smoke and have the disadvantage of slower processing speed due to additional preprocessing task. In this paper, we implemented a fire detection system which predicts the flames and the smoke at the same time by constructing a CNN model that supports multi-labeled classification. Also, the system can monitor the fire status in real time by using Grad-CAM which visualizes the position of classes based on the characteristics of CNN. Also, we tested our proposed system with 13 fire videos and got an average accuracy of 98.73% and 95.77% respectively for the flames and the smoke.
Å°¿öµå(Keyword) ÄÁº¼·ç¼Ç ½Å°æ¸Á   ±×·¹µå-Ä·   ÀμÁ¼Ç ·¹½º³Ý ¹öÀü2   ¿µ»ó ±â¹Ý È­Àç °¨Áö   ´ÙÁß ·¹ÀÌºí ºÐ·ù   Convolutional Neural Network   Grad-CAM   nception ResNet V2   Image-based Fire Detection   Multi-labeled Classification  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå