Study on Power Instrument Symbols Identifying Based on Support Vector Machine

An instrument symbol new identifier of using support vector machine (SVM) is presented in this paper. SVM is a artificial intelligence methodology, which operates on the principle of structure risk minimization instead of traditional pattern recognition methods that only keep the empirical risk minimization, and a better performance is guaranteed. An image-based measuring instrument recognition based on support vector classifiers is constructed, via pretreatment and uniform process; the binary particle horizon and vertical edge projections are adopted for the SVMs input vectors. The multi-classifiers of the dial plate symbols have been designed based on SVM principle, and the experiments indicate that SVMs simple architecture and fast training process can satisfy the real-time perfectly.