Á¤º¸Ã³¸®ÇÐȸ ³í¹®Áö ¼ÒÇÁÆ®¿þ¾î ¹× µ¥ÀÌÅÍ °øÇÐ
Current Result Document :
ÇѱÛÁ¦¸ñ(Korean Title) |
Àΰø À§¼º »çÁø ³» ¼±¹Ú ŽÁö Á¤È®µµ Çâ»óÀ» À§ÇÑ Watershed ¾Ë°í¸®Áò ±â¹Ý RoI Ãà¼Ò ±â¹ý |
¿µ¹®Á¦¸ñ(English Title) |
Watershed Algorithm-Based RoI Reduction Techniques for Improving Ship Detection Accuracy in Satellite Imagery |
ÀúÀÚ(Author) |
À̽ÂÀç
À±Áö¿ø
Seung Jae Lee
Ji Won Yoon
|
¿ø¹®¼ö·Ïó(Citation) |
VOL 10 NO. 08 PP. 0311 ~ 0318 (2021. 08) |
Çѱ۳»¿ë (Korean Abstract) |
ÇØ»ó ¾Èº¸, ±¹Á¦ µ¿Çâ ÆÄ¾Ç µî ´Ù¾çÇÑ ÀÌÀ¯·Î ÇØ»ó »çÁø¿¡¼ ¼±¹ÚÀ» ŽÁöÇÏ°íÀÚÇÏ´Â ¿¬±¸´Â Áö¼ÓµÇ¾î ¿Ô´Ù. ÀΰøÁö´ÉÀÇ ¹ß´Þ·Î ÀÎÇØ »çÁø ¹× ¿µ»ó ³» °´Ã¼ ŽÁö¸¦ À§ÇÑ R-CNN ¸ðµ¨ÀÌ µîÀåÇÏ¿´°í °´Ã¼Å½ÁöÀÇ ¼º´ÉÀÌ ºñ¾àÀûÀ¸·Î »ó½ÂÇÏ¿´´Ù. R-CNN ¸ðµ¨À» ÀÌ¿ëÇÑ ÇØ»ó »çÁø¿¡¼ÀÇ ¼±¹Ú ŽÁö´Â ÀΰøÀ§¼º »çÁø¿¡µµ Àû¿ëµÇ±â ½ÃÀÛÇÏ¿´´Ù. ÇÏÁö¸¸ ÀΰøÀ§¼º »çÁøÀº ³ÐÀº Áö¿ªÀ» Åõ»çÇϱ⠶§¹®¿¡ ¼±¹Ú ¿Ü¿¡µµ Â÷·®, ÁöÇü, °Ç¹° µî ´Ù¾çÇÑ °´Ã¼µéÀÌ ¼±¹ÚÀ¸·Î ÀνĵǴ °æ¿ì°¡ ÀÖ´Ù. º» ³í¹®¿¡¼´Â R-CNN°è¿ ¸ðµ¨À» ÀÌ¿ëÇÑ ÀΰøÀ§¼º »çÁø¿¡¼ÀÇ ¼±¹Ú ŽÁöÀÇ ¼º´ÉÀ» °³¼±Çϱâ À§ÇÑ »õ·Î¿î ¹æ¹ý·ÐÀ» Á¦¾ÈÇÑ´Ù. Ç¥ÁöÀÚ ±â¹Ý watershed ¾Ë°í¸®ÁòÀ» ÅëÇØ À°Áö¿Í ¹Ù´Ù¸¦ ºÐ¸®ÇÏ°í morphology ¿¬»êÀ» ¼öÇàÇÏ¿© RoI¸¦ ÇÑ Â÷·Ê ´õ ƯÁ¤ÇÑ µÚ ƯÁ¤µÈ RoI¿¡ R-CNN °è¿ ¸ðµ¨À» »ç¿ëÇÏ¿© ¼±¹ÚÀ» ŽÁöÇÏ¿© ¿ÀŽÀ» ÁÙÀδÙ. ÇØ´ç ¹æ¹ýÀ» ÀÌ¿ëÇÏ¿© Faster R-CNNÀ» »ç¿ëÇÏ¿´À» °æ¿ì, Faster R-CNN¸¸À» »ç¿ëÇßÀ» ¶§¿¡ ºñÇØ ¿ÀŽ·üÀ» 80% ÁÙÀÏ ¼ö ÀÖ¾ú´Ù. |
¿µ¹®³»¿ë (English Abstract) |
Research has been ongoing to detect ships from offshore photographs for a variety of reasons, including maritime security, identifying international trends, and social scientific research. Due to the development of artificial intelligence, R-CNN models for object detection in photographs and images have emerged, and the performance of object detection has risen dramatically. Ship detection in offshore photographs using the R-CNN model has also begun to apply to satellite photography. However, satellite images project large areas, so various objects such as vehicles, landforms, and buildings are sometimes recognized as ships. In this paper, we propose a novel methodology to improve the performance of ship detection in satellite photographs using R-CNN series models. We separate land and sea via marker-based watershed algorithm and perform morphology operations to specify RoI one more time, then detect vessels using R-CNN family models on specific RoI to reduce typology. Using this method, we could reduce the misdetection rate by 80% compared to using only the Fast R-CNN. |
Å°¿öµå(Keyword) |
Çؾȼ± ÃßÃâ
ÀΰøÀ§¼º »çÁø
R-CNN
Watershed ¾Ë°í¸®Áò
Coastline Extraction
Satellite Image
R-CNN
Watershed Algorithm
|
ÆÄÀÏ÷ºÎ |
PDF ´Ù¿î·Îµå
|