| （英） |
In recent years, aging has progressed, and the number of elderly people who need nursing care has increased. There is also growing concern that the number of elderly workers is increasing and nursing care workers are in short supply. As a result, nursing staff cannot always monitor. Therefore, accidents such as falling when leaving the bed are increasing. In this study, CNN and pressure images were used to observe changes in behavior between sleeping posture and getting out of bed, and 6 types of predictions were made, including sleeping posture and behavior to get out of bed. As a result, the overall recognition rate was 97.6%, and a high recognition rate was obtained. As future work, it is necessary to make predictions using time-series data instead of using static pressure distribution images, and to proceed with verification for natural motion. |