Feature selection is an important tool reducing necessary feature acquisition time in some applications. Standard methods, proposed in the literature, do not cope with the measurement cost issue. Including the measurement cost into the feature selection process is difficult when features are grouped together due to the implementation. If one feature from a group is requested, all others are available for zero additional measurement cost. In the paper, we investigate two approaches how to use the measurement cost and feature grouping in the selection process. We show, that employing grouping improves the performance significantly for low measurement costs. We discuss an application where limiting the computation time is a very important topic: the segmentation of backscatter images in product analysis.@en

In this paper we present an application of statistical pattern recognition for segmentation of backscatter images (BSE) in product analysis of laundry detergents. Currently, application experts segment BSE images interactively which is both time consuming and expert dependent. We present a new, automatic, procedure for supervised BSE segmentation which is trained using additional multi-spectral EDX images. Each time a new feature selection procedure is employed to find a convenient feature subset for a particular segmentation problem. The performance of the presented algorithm is evaluated using ground-truth segmentation results. It is compared with that of interactive segmentation performed by the analyst.@en