Application of Fuzzy Points in MSIFT for Threat Objects Detection

The detection of threat objects using X-ray luggage scan images is a challenging task for security. Most Computer Aided Screening (CAS) is still based on the manual detection of potential threat objects by human expert´s where chances of human errors is quite high as thousands of commuter need to be scanned every day. Proposed sytem uses the very naïve concept of extended fuzzy logic based shape detection technique. In this paper Scale-invariant feature transform (SIFT) algorithm has been modified by using the concept of fuzzy points of fuzzy geometry for finger point matching. The criterion used in CAS system is the number of matched key-points obtained from the MSIFT algorithm with objects key-points data base. The proposed naive approach shows significant impact in the luggage screening.