| 摘要 |
PROBLEM TO BE SOLVED: To provide a system and method for detecting an event from multiple time-series data sequences.SOLUTION: The system receives multiple time-series data sequences generated by multiple sensors, where the multiple time-series data sequences comprise multiple sample points. The system pairs the sample points with one another for determining multiple pairs of the sample points. The system computes Euclidean distances and angles between the sample points for determining a distance matrix and an angle matrix corresponding to the multiple sample points. Further, the system determines global distribution of the multiple pairs of the sample points, where the global distribution of the multiple pairs of the sample points represent 2D shape histograms for the multiple time-series data sequences. The system concatenates the 2D shape histograms for respective time-series data sequences to generate a concatenated shape histogram. Finally the system matches the concatenated shape histogram to pre-stored multiple shape histograms for determining the event. |
