Abnormal Attitude Recognition of the Elderly Based on Mobile Devices
Download as PDF
DOI: 10.38007/Proceedings.0000144
Corresponding Author
Chengjun Xie
Abstract
With the aggravation of aging in China, people pay more and more attention to the safety of the elderly. The combination of the use of mobile portable devices and the study of abnormal attitude identification provides a new idea for people to monitor the life safety of the elderly. This paper designs an abnormal attitude recognition system that uses multi-feature fusion algorithm to extract features based on mobile devices, and compares it with other single-feature recognition algorithms. The advantages and disadvantages of this algorithm are comprehensively evaluated through the comparison of recognition accuracy and recognition efficiency. The result of research shows that the accuracy of the multi-feature fusion was 2.60% higher than the SurfFeatures, which is the best single feature; It is 0.35% higher than Surf&HuFeatures, which is the best dual features. The multi-feature is more accurate in the expression of information, has the highest recognition rate in the experiment, and is more reliable for the protection of the elderly. This paper studies the abnormal attitude recognition of the elderly based on mobile devices, which has important guiding significance for the establishment of human feature extraction and mobile device combination system.
Keywords
Mobile Devices; The Elderly; Abnormal Attitude; Fusion Feature