Author(s)

Jeffrey C. Miecznikowski, Kimberly F. Sellers, William F. Eddy

Department of Biostatistics, Roswell Park Cancer Institute, SUNY University at Buffalo, Buffalo, USA.
Department of Mathematics and Statistics, Georgetown University, Washington DC, USA.
Department of Statistics, Carnegie Mellon University, Pittsburgh, USA.

Feature detection in chemical sensors images falls under the general topic of mathematical morphology, where the goal is to detect “image objects” e.g. peaks or spots in an image. Here, we propose a novel method for object detection that can be generalized for a k-dimensional object obtained from an analogous higher-dimensional technology source. Our method is based on the smoothing decomposition, Data = Smooth + Rough, where the “rough” (i.e. residual) object from a k-dimensional cross-shaped smoother provides information for object detection. We demonstrate properties of this procedure with chemical sensor applications from various biological fields, including genetic and proteomic data analysis.

Bump Hunting; Image Analysis; Spatial Smoothing; Feature Detection; Mathematical Morphology

J. C. Miecznikowski, K. F. Sellers and W. F. Eddy, "Multidimensional Median Filters for Finding Bumps in Chemical Sensor Datasets," Journal of Sensor Technology, Vol. 2 No. 1, 2012, pp. 23-37. doi: 10.4236/jst.2012.21005.