Human fall detection based on event pattern matching with ultrasonic array sensors

Abstract

We proposed an algorithm for fall detection based on event pattern matching with ultrasonic array sensors signal. In several references cited later using ultrasonic sensors to detect human fall. But shows miss detection of event pattern matching algorithm. Our contributions are the algorithm and improvement of the human fall detection results. We predict that the major bottleneck would be the implementation of ultrasonic array sensors. The type of falling events was determined through the implemented SVM model. The CEP and Event Pattern Matching are then applied to verify the falling event to ensure no false positive result taken place. The results demonstrate that a better performance can be achieved based on the designed algorithm. The quality of images also plays an important role of accuracy.