作者: Félix Zapata , María López-López , José Manuel Amigo , Carmen García-Ruiz
DOI: 10.1016/J.FORSCIINT.2017.11.025
关键词:
摘要: Abstract Multispectral images of clothing targets shot at seven different distances (from 10 to 220 cm) were recorded 18 specific wavelengths in the 400–1000 nm range visualize gunshot residue (GSR) pattern. Principal component analysis (PCA) showed that use violet-blue (430, 450 and 470 nm) provided largest contrast between GSR particles white cotton fabric. Then, correlation amount on shooting distance was studied. By selecting blue frame multispectral (i.e. red-green-blue (RGB) system which falls 470 nm), pixels containing accounted based intensity frame. Results demonstrated number exponentially decreases with from 30 220 cm following a particular exponential equation. However, shortest (10 cm) did not satisfy above equation, probably due noticeable differences GSR-pattern these (e.g. high presence soot). equation applied validation samples estimate distances, obtaining results an error below 10%.