Predicting the suitability for scene matching using SVM

Interrelation and constraint between multi-measure parameters of reference image and suitable matching area have not been considered in most of methods for selecting matching area. In order to overcome this drawback, we present a novel method of predicting the suitability using support vector machine (SVM). Firstly, gray-based and edge-based measure parameters are selected. Then the sample images data set are trained with radial basis kernel function after normalization of input vectors composed of measure parameters. Finally, we separate suitable matching area class and unsuitable matching area class using decision function. Thus we can predict the suitability and guide scene matching process. The experimental results show that this method holds the capability of flexibility and jamming resistance as well as proper guide to selection of suitable matching area from complex reference image.