Khosro Rezaee - Biomedical Engineering Group, Department of Electrical and Computer Engineering
Javad Haddadnia - New Technologies Research Center, Medical University of Sabzevar, Sabzevar, Iran
Ahmad Delbari - Department of Neurobiology, Care Sciences and Society (NVS) Karolinska Institute, Stockholm, Sweden

Automatic detection of the abnormal walking inpeople, especially such accidents as the falls in the elderly, basedon image processing techniques and computer vision can helpdevelop an efficient system that its implementation in variouscontexts enables us to monitor people’s movements. This paperproposes a new algorithm, which drawing on fuzzy rules inclassification of movements as well as the implementation of themotion estimation, allows the rapid processing of the input data.At the testing stage, 57425 video frames received from MotherNursing Home in Farzanegan and the video sequences containingthe falls of the elderly were used. The results show that the valuesof average accuracy (AAC), detection rate (DR) and false alarmrate (FAR) were at an acceptable level, respectively with93%, 89% and 5%. Compared to the similar techniques, theimplementation of the proposed system in nursing homes andresidential areas allow the real time and intelligent monitoring ofthe people.

Video Processing, Gaussian mixture model, HSV Conversion, the elderly’s falls, Fuzzy Inference System, Motion Estimation