A united classification system of X-ray image based on fuzzy rule and neural networks

The detection of explosives in passengers¿ luggage is an important area in public traffic security. This paper presents a united classification system for detecting explosives based on fuzzy rule and neural networks. Due to imaging and influence of outer environment, preprocessing is firstly needed to improve the quality of X-ray images. Then, a test pattern may be considered as several possible objects with different degrees through the multi-level fuzzy classifier. Finally, the result of multi-level fuzzy classifier will be reconsidered through the parallel neural networks classifier. From the experience results, the united classification system performs well.