In care facilities, continuous, non-contact measurement of respiratory rate is crucial for early response to sudden health deterioration in residents. However, it is challenging with traditional non-contact respiratory rate measurement systems on beds to accurately measure respiratory rates when residents change position or posture significantly. To address this issue, this study proposes a method using Principal Component Analysis (PCA) to extract features of center-of-gravity fluctuations influenced by respiration. This approach enables the development of a system that can stably measure respiratory rates in a non-contact manner, even in situations where the resident’s posture changes significantly, such as during turning movements.
