| ³í¹®¸í |
ÄÄÇ»ÅÍ ºñÀü ±â¹Ý ¿Ü´Ü¿ °ø»çÀÇ Á¢ÂøÁ¦ µµÆ÷ǰÁú °¨¸® ÀÚµ¿È ¸ðµ¨ / Computer Vision-based Automated Adhesive Quality Inspection Model of Exterior Insulation and Finishing System |
| ÀúÀÚ¸í |
À±¼¼ºó(Yoon, Sebeen) ; °¹Î±Õ(Kang, Mingyun) ; ÀåÇö½Â(Jang, Hyounseung) ; ±èÅÂÈÆ(Kim, Taehoon) |
| ¼ö·Ï»çÇ× |
Çѱ¹°ÇÃà½Ã°øÇÐȸ ³í¹®Áý, Vol.23 No.2 (2023-04) |
| ÆäÀÌÁö |
½ÃÀÛÆäÀÌÁö(165) ÃÑÆäÀÌÁö(9) |
| ÁÖÁ¦¾î |
¿Ü´Ü¿ °ø¹ý; ÀνºÅϽº ¼¼±×¸àÅ×À̼Ç; ǰÁú°¨¸® ÀÚµ¿È ; exterior insulation and finish system; instance segmentation; automated quality supervision |
| ¿ä¾à1 |
º» ¿¬±¸¿¡¼´Â ¿Ü´Ü¿ °ø»çÀÇ ´Ü¿Àç Á¢ÂøÁ¦ µµÆ÷ ǰÁúÀ» ÀÚµ¿À¸·Î °¨¸®ÇÒ ¼ö ÀÖ´Â ¸ðµ¨À» Á¦¾ÈÇÏ¿´´Ù. »ç·Ê Àû¿ë °á°ú,¿µ¿ª ºÐÇÒ ¸ðµ¨Àº mAP 92.3%ÀÇ Á¤È®µµ¸¦ ³ªÅ¸³Â°í, Á¦¾È ¸ðµ¨ÀÇ Á¢ÂøÁ¦ ¸éÀû ºñÀ² »êÃâ Á¤È®µµ´Â 98.8%, Á¢ÂøÁ¦ µ¢¾î¸® Á᫐ °£ °Å¸® »êÃâ Á¤È®µµ´Â 96.7%·Î ³ªÅ¸³µ´Ù. º» ¿¬±¸ °á°ú´Â ¿Ü´Ü¿ °ø»çÀÇ °¨¸®¸¦ À§ÇÑ ÇöÀåÅõÀÔ ÀηÂÀ» ÃÖ¼ÒÈÇÏ¸é¼ ¿Ü´Ü¿ °ø»çÀÇ °¡Àå ºó¹øÇÑ ÇÏÀÚÀÎ ´Ü¿Àç Å»¶ô ÇÏÀÚ¸¦ ¿¹¹æÇÒ ¼ö ÀÖÀ¸¸ç ³ª¾Æ°¡ ¿Ü´Ü¿ ½Ã½ºÅÛÀÇ È°¼ºÈ¿¡ ±â¿©ÇÒ ¼ö ÀÖÀ» °ÍÀ¸·Î ÆÇ´ÜµÈ´Ù. ÇâÈÄ¿¡´Â ´Ù¾çÇÑ È¯°æ¿¡¼ ¿Ü´Ü¿ °ø¹ýÀÇ ½Ã°ø ¿µ»óÀ» ¼öÁýÇÏ¿© ¿µ»ó ºÐÇÒ ¸ðµ¨ÀÇ ¼º´ÉÀ» ³ôÀ̰í, ¿µ»ó ³»¿¡´Ù¼öÀÇ ´Ü¿Àç°¡ Æ÷ÇÔµÈ °æ¿ì¿¡µµ ÀÚµ¿ °¨¸®ÇÒ ¼ö ÀÖ´Â ¸ðµ¨À» °³¹ßÇϰíÀÚ ÇÑ´Ù. |
| ¿ä¾à2 |
This research proposed a model for automatically monitoring the quality of insulation adhesiveapplication in external insulation construction. Upon case implementation, the area segmentationmodel demonstrated a 92.3% accuracy, while the area and distance calculation accuracies of theproposed model were 98.8% and 96.7%, respectively. These findings suggest that the model caneffectively prevent the most common insulation defect, insulation failure, while simultaneously minimizingthe need for on-site supervisory personnel during external insulation construction. This, in turn,contributes to the enhancement of the external insulation system. Moving forward, we plan to gatherconstruction images of various external insulation methods to refine the image segmentation model¡¯sperformance and develop a model capable of automatically monitoring scenarios with a considerablenumber of insulation materials in the image. |