Research of isolator´s status image recognition based on Gabor wavelet and SVM

In the operation of substation, judgment of isolator´s status is very important. But there are potential safety problems by the current method, such as electric shock accidents and outage accidents. This paper proposes a isolator´s status image recognition technique based on Gabor wavelet and SVM(support vector machine). The isolator´s status video images are preprocessed before an analysis of the texture feature based on Gabor transformation with Gobar filter. The isolator´s status is recognized by the SVM classifiers with the RBF (radial basis function). Experiment indicates that the proposed isolator´s status recognition operates well.