Fall incidents have been reported as the second most common cause of death, especially for elderly people. Human fall detection is necessary in smart home healthcare systems. The aim of this study was to build health monitoring for elderly by computer vision technology.
Fist, moving object is segmented from each frame of the input video using GMM method. Then, an ellipse model is built from the segmented part and used for extracting five features. These features are applied for modeling HMM. From the trained HMMs, probability of each observed input data will be calculated to each model, then the best matched HMM model will be selected