• 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¸°úÇÐȸ³í¹®Áö (Journal of KIISE)

Á¤º¸°úÇÐȸ³í¹®Áö (Journal of KIISE)

Current Result Document : 8 / 38 ÀÌÀü°Ç ÀÌÀü°Ç   ´ÙÀ½°Ç ´ÙÀ½°Ç

ÇѱÛÁ¦¸ñ(Korean Title) C3D¿Í °´Ã¼ ±â¹ÝÀÇ ¿òÁ÷ÀÓ Á¤º¸ °áÇÕÀ» ÅëÇÑ °¨½Ã½Ã½ºÅÛ¿¡¼­ÀÇ ÀÌ»ó Çൿ ŽÁö
¿µ¹®Á¦¸ñ(English Title) Anomaly Detection by a Surveillance System through the Combination of C3D and Object-centric Motion Information
ÀúÀÚ(Author) ¹Ú½½±â   È«¸í´ö   Á¶±Ù½Ä   Seulgi Park   Myungduk Hong   Geunsik Jo  
¿ø¹®¼ö·Ïó(Citation) VOL 48 NO. 01 PP. 0091 ~ 0099 (2021. 01)
Çѱ۳»¿ë
(Korean Abstract)
±âÁ¸ CCTV ºñµð¿À¿¡¼­ µö·¯´× ±â¹ÝÀÇ ÀÌ»ó ŽÁö ¿¬±¸´Â °´Ã¼ÀÇ Çൿ °ª¸¸À» ÀÌ¿ëÇÏ¿© ÀÌ»ó À» ŽÁöÇϱ⠶§¹®¿¡, »óȲ¿¡ µû¸¥ °´Ã¼ Çൿ °ª ÃßÃâÀÌ ¾î·Æ°í, ½Ã°£ È帧¿¡ µû¸¥ Á¤º¸°¡ Ãà¼ÒµÇ´Â ¹®Á¦Á¡ÀÌ ÀÖ¾ú´Ù. ¶ÇÇÑ CCTV ºñµð¿À¿¡¼­ ÀÌ»óÀÇ ¿øÀÎÀº ÇÁ·¹ÀÓÀÇ º¹À⼺ µî ´Ù¾çÇÑ ¿ä¼Ò¿Í ½Ã°è¿­ ºÐ¼®¿¡ µû¸¥ Á¤º¸·Î ÀÌ·ç¾îÁ®, °´Ã¼ÀÇ Çൿ °ª¸¸À» ÀÌ¿ëÇÏ¿© ÀÌ»óÀ» ŽÁöÇϱ⿡´Â ÇÑ°è°¡ ÀÖ´Ù. ÀÌ·¯ÇÑ ¹®Á¦¸¦ ÇØ°á Çϱâ À§ÇØ º» ³í¹®¿¡¼­´Â °´Ã¼ Áß½ÉÀÇ ´Ù¾çÇÑ Æ¯Â¡°ªÀ» »ç¿ëÇÏ¿© C3D¿¡ ±¤ÇÐ È帧À» °áÇÕÇÑ ½Ã°ø°£Àû Á¤º¸¸¦ »ç¿ëÇÏ´Â »õ·Î¿î µö·¯´× ±â¹ÝÀÇ ÀÌ»ó ŽÁö ¸ðµ¨À» Á¦¾ÈÇÑ´Ù. Á¦¾ÈÇÏ´Â ÀÌ»ó ŽÁö ¸ðµ¨Àº UCF-Crime µ¥ÀÌÅÍ ¼¼Æ®¸¦ »ç¿ëÇÏ¿´À¸¸ç, ½ÇÇè °á°ú Á¤È®µµ¿¡ ÇØ´çÇÏ´Â AUC °ªÀÌ 76.44·Î, ±âÁ¸ ¿¬±¸¿Í ºñ±³ÇÏ¿© ºü¸¥ °´Ã¼°¡ ÀÖ´Â ºñµð¿À¿¡¼­ ´õ¿í È¿°úÀûÀ¸·Î µ¿ÀÛÇÏ´Â °ÍÀ» È®ÀÎÇÏ¿´´Ù. ÀÌ¿¡ °´Ã¼ÀÇ ´Ù¾çÇÑ Æ¯Â¡°ª°ú ½Ã°è¿­ ºÐ¼®¿¡ µû¸¥ Á¤º¸¸¦ »ç¿ëÇÏ´Â °ÍÀÌ ÀûÀýÇÏ´Ù´Â °á·ÐÀ» µµÃâÇÏ¿´´Ù.
¿µ¹®³»¿ë
(English Abstract)
In the existing closed-circuit television (CCTV) videos, the deep learning-based anomaly detection reported in the literature detected anomalies using only the object's action value. For this reason, it is difficult to extract the action value of an object depending upon the situation, and there is a problem that information is reduced over time. Since the cause of abnormalities in CCTV videos involves several factors such as frame complexity and information according to time series analysis, there is a limit to detecting an abnormality using only the action value of the object. To solve this problem, in this paper, we designed a new deep learning-based anomaly detection model that combined optical flow with C3D to use various feature values centered on the objects. The proposed anomaly detection model used the UCF-Crime dataset, and the experimental results achieved an area under the curve (AUC) of 76.44. Compared to previous studies, this study worked more effectively in fast-moving videos such as explosions. Finally, we concluded that it was appropriate to use the information according to different feature values and time series analysis considering various aspects of the behavior of an object when designing an anomaly detection model.
Å°¿öµå(Keyword) ÀÌ»ó ŽÁö   ±¤ÇÐ È帧   °´Ã¼ Á߽ɠ  µö·¯´×   ÀΰøÁö´É   anomaly detection   optical flow   object-centric   deep learning   artificial intelligence  
ÆÄÀÏ÷ºÎ PDF ´Ù¿î·Îµå