Adaptive Neuro-fuzzy Classifier for Weapon Detection in X-Ray Images of Luggage Using Zernike Moments and Shape Context Descriptor

Weapon detection in luggages using X-ray machines at various security tight areas especially airports are inevitable. The process tends to be time consuming and requires the skills of a human operator to identify the weapon from the contents of the luggage. An automated weapon (gun) detection method for luggage screening systems utilizing connected component analysis, zernike moments and shape context descriptor based feature extraction methods for adaptive neuro-fuzzy classifier is proposed here. The proposed method showed an efficiency of 98% in detecting weapon (gun) with least false alarm rate.