Motion Fall Detection System for the Elderly Based on Multi-feature Fusion Robot

Abstract

With the improvement of people's living standards, countries around the world are facing the serious problem of aging in their own countries. Unexpected events such as falls in the elderly have a serious and inestimable impact on the health of the elderly. The purpose of this paper is to study the elderly fall detection system based on multi-feature fusion robot. This article first introduces a fall detection algorithm. Second, it improves the way of feature fusion. It uses feature weighted kernel functions to achieve the weighted fusion of acceleration and angular velocity features to build a fall recognition model based on support vector machines. Then, it analyzes the key parts of the fall recognition model. To verify the generalization performance of the algorithm. Finally, a fall detection system based on the recognition model constructed in this paper is implemented on the Android phone platform, and the practical application effect of the fall detection algorithm is verified. The experiments show that the feature fusion fall monitoring algorithm constructed in this paper has a sensitivity of 90.67%, a specificity of 92.96%, and an accuracy of 92.14%, which verifies the detection performance of the detection algorithm constructed in this paper. The robustness and stability of the drop detection software based on the Android platform.