No abstract

The purpose of this study was to analyse the explosive forensic data contained in digital image. The explosive forensic data used in this study was mainly targeted on the cross-section image of steel rod shrapnel. Digital image processing techniques were used to find the areas and types of the shrapnel. The process began with imaging segmentation to find the boundary box of the shrapnel area and then extracted the areas which contained key features and computed the statistical values of those areas, Finally, the statistical values were administered to the Artificial Neural Network to classify the types of shrapnel. The results showed 71% accuracy which was acceptable since each type of cross-section image had a very slightly different and much different to detect by human eyes.

No abstract