ANALYSIS OF SPECTROSCOPIC IMAGING DATA BY FUZZY C-MEANS CLUSTERING

A novel method of analyzing spectroscopic imaging data is presented. A fuzzy C-means clustering algorithm has been applied to the analysis o f near-infrared spectroscopic imaging data acquired with the combinati on of a CCD camera and a liquid crystal tunable filter. The use of fuz zy C-means clustering dramatically increased the information obtained from near-IR spectroscopic images and allowed for the detection of sma ll subregions of the image that contained novel and unanticipated spec tral features, without the need for a priori knowledge of the chemical composition of the sample. Two illustrative samples were analyzed, on e comprised of four different inks printed on label paper and the othe r containing indocyanine green and human blood patches. The regions co ntaining the different constituents were clearly demarcated and their mean spectra determined, The mean spectra of the second sample were sh own to match those obtained using a scanning near-IR spectrometer. In addition to probing the spatial and spectral characteristics of the sa mples, the fuzzy C-means clustering analysis also helped improve the s ignal-to-noise ratio of the spectra.