Integration of Multiple Sensors using Binary Features in a Bernoulli Mixture Model

This article reports on the use of a Bernoulli mixture model to integrate features extracted independently from two or more distinct sensors. Local image features (SIFT) and multiple types of features from a 2D laser range scan are all converted into Binary form and integrated into a single binary feature incidence matrix (FIM). The correlation between the different features is captured by modeling the resultant FIM in terms of a Bernoulli mixture model. The integration of binary features from different sensors allows for good place recognition. The use of binary features also promises a much simpler integration of features from dissimilar sensors