Evaluating KNN, LDA and QDA classification for embedded online feature fusion

In this paper we evaluate k-nearest neighbor (KNN), linear and quadratic discriminant analysis (LDA and QDA, respectively) for embedded, online feature fusion which poses strong limitations on computing resources and timing. These algorithms are implemented on our multisensor data fusion (MSDF) architecture and are applied to traffic monitoring, i.e., classifying vehicles using distributed image, acoustic and laser sensors. We performed several tests of the algorithms on our embedded platform and evaluated CPU performance and memory consumption for training as well as classification. The results obtained are very promising for further use, especially of LDA and QDA for embedded online fusion at feature-level.