Chudanat SudthongkhongSiwat SuksriChanate RatanaubolSookyuen TepthongJira JitsupaPutawan Suksai2025-03-102025-03-102023Proceedings - 17th International Conference on Signal-Image Technology and Internet-Based Systems, SITIS 2023979-835037091-110.1109/SITIS61268.2023.000832-s2.0-85190146908https://repository.dusit.ac.th//handle/123456789/4555Every year, around one-third of elderly individuals experience falls at home, especially in high-risk areas like bathrooms and stairs. Uneven floor surfaces exacerbate these dangers, impeding elderly mobility and significantly increasing fall risks, with recurrent falls being common. Recognizing this pressing concern, our project introduces a 'Human Fall Detection and Estimation System' to mitigate harm. This system deploys a specialized camera with gesture recognition software to monitor for falls by detecting posture deviations. When a fall occurs, the system records the location and uses advanced Image Processing for precise Pose Estimation. Deep Learning analyzes Pose Estimation data to gauge fall severity and simultaneously alerts caregivers via the network for swift assistance. Incidents are logged in a database for root cause analysis, facilitating more effective elderly care systems. our system plays a crucial role in preventing and addressing elderly falls, swiftly detecting and assessing incidents, and alerting caregivers [1], enhancing elderly safety and well-being. © 2023 IEEE.Caregiver alert systemDeep LearningElderly fallsFall detection systemHuman posture analysisConference paperScopus