YOLOv5 기반 재실자-객체 검출 모델을 이용한 재실자 행동 분류 프로세스 개발

다국어 초록 (Multilingual Abstract)

Recently, as the need for efforts to improve building energy efficiency and secure the comfort of the occupants has emerged, research on the activity of the occupants has been actively conducted. Based on YOLOv5, an object detection algorithm, a detection model that detects objects used in behavior, and a behavior classification model that classifies the behavior of each object by receiving information from the detection model were developed in this study. As a result, a model was developed to classify four behaviors: cooking, eating, working, and exercise, and the developed occupant activity classification model’s performance was calculated as Accuracy 0.99, and F1-Score 0.98.

목차 (Table of Contents)

• Abstract
• 1. 서론
• 2. 연구방법
• 3. 결과
• 4. 결론
• 참고문헌