A method of and a system for identifying objects using histogram segment
features from multi-energy CT images are provided. The multi-energy CT
images include a CT image, which approximates density measurements of
scanned objects, and a Z image, which approximates effective atomic
number measurements of scanned objects. The method comprises: computing a
density histogram for each potential threat object; smoothing the density
histogram using a low-pass filter; identifying peaks in the smoothed
density histogram; assigning a segment to each peak; computing histogram
segment features for each segment; classifying each potential threat
object into a threat or a non-threat using computed features.