A new strategy was demonstrated to establish typical spectra database. For the study, bone fragments of three different species MBM were analyzed on near-infrared microscopic imaging system, and both visible and near-infrared images were acquired at the same size. To extract and mark the position of every single bone fragment in visible image, a graphical user-friendly interface, based on marker-controlled watershed segmentation method, written in Matlab for extracting and marking bone fragments has been developed. The recognition ratio was 96.4%. By the position information in visible image, near-infrared microscopic imaging data was decompounded to spectrum of each bone fragment. Next, typical spectra were selected by the multivariate ‘distances’—Global Hood (GH) & Neighbor Hood (NH) statistics.