Author:
Wang Zheng,Zhang Long-fei,Feng Xiao-xing,Liu Meng-xi
Abstract
Abstract
An algorithm of activity pattern recognition based on fuzzy mathematics used the excellent characteristics and the classification ability of multiple classifier for human pasture identification is analyzed and compared. Time-domain analysis is used as the only method through large amounts of data, and corresponding BP neural network and Support vector machine (SVM) was established were compared with the algorithm of fuzzy mathematics, and some method were evaluated at last. The results showed that the algorithm based on fuzzy mathematics had a good performance and value.
Subject
General Physics and Astronomy
Reference6 articles.
1. A comparison of feature extraction methods for the classification of dynamic activities from accelerometer data[J];Preece;IEEE Transactions on Biomedical Engineering,2009
2. Radar emitter recognition based on POS-BP network[J];Zhifu;AASRI Procedia
3. Improved particle swarm optimization algorithm for fuzzy multi-class SVM[J];Ying;Journey of Systems Engineering and Electronics,2010
4. Application of BP neural network algorithm in sustainable development of highway construction projects[J];mingshun;Physics Procedia