A novel method to recognize meter parameters symbols, based on the combined features of Hu moment & compactness and support vector machine, is proposed. In the process of analyzing invariant moment features of discrete status, a new method that a feature is divided another feature is introduced to eliminate the influence of scale factor for these features. At last, support vector machine is used to recognize meter parameters symbols. Experiments results show the present method achieves recognition accuracy of 97.77% and has more obvious advantages.
Recognizing Meter Parameters Symbols Based on the Combined Features and Support Vector Machine
2008 Congress on Image and Signal Processing ; 4 ; 641-645
2008-05-01
244492 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Representing and Recognizing Visual Dynamic Events with Support Vector Machines
British Library Conference Proceedings | 1999
|Optimal Features and Classes for Estimating Mobile Robot Orientation Based on Support Vector Machine
British Library Conference Proceedings | 2011
|Optimal Features and Classes for Estimating Mobile Robot Orientation Based on Support Vector Machine
Springer Verlag | 2011
|