Automatic Body Fall Detection System for Elderly People using Accelerometer and Vision Based Technique

Body Falls in older adults are the significant cause of injury. Falls incorporate dropping from a standing position or from uncovered positions, for example, those on stepping stools or stepladders. The seriousness of damage is commonly identified with the height of fall often leading to disability or death. In this research generally we uses wearable sensor and vision based technique that is automatically detect body fall as early as possible. Accelerometer is used for measuring or maintaining orientation and angular velocity. In vision based procedure first we procure casings or video arrangements from the camera. The division module separates the body outline from the foundation. For Feature Extraction we used GLCM method. SVM method is used for classification. By using those methods we can surely detect the human body fall and can take the preventive measures.