Remote Sensing Images Classification/Data Fusion Using Distance Weighted Multiple Classifiers Systems

For a multiple classifiers system, a weighting policy is applied to fuse knowledge acquired by classifiers to arrive at an overall decision that is supposedly superior to that attainable by any one of them acting alone. The distance measured between the classifier output and its desired output can be used as a classifier performance indicator. By adopting this performance indicator, the rms and average distance weighted multiple classifiers systems are proposed in this paper. The classification performances of utilizing various multiple classifiers systems to the application of remote sensing image classification are demonstrated and compared. Experimental results show that the classification accuracy is considerably improved by making use of the multiple classifiers system. In addition, the multiple classifiers systems of using distance weighted algorithms are superior to those of using the conventional bagging and boosting algorithms. Moreover, average distance weighted multiple classifiers system outperform rms distance weighted multiple classifiers system slightly