Image segmentation of UV pattern for automatic paper-money inspection

We propose an image segmentation method for automatic paper-money inspection system. The UV (ultra violet) patterns embedded in the paper money should be segmented to determine whether the money is genuine or not. We acquired the image by directional lighting system using UV LEDs. In order to segment the pattern from the background, we apply the Gaussian mixture model to consider the multi-modal characteristics of the histogram. Experimental results are presented to verify the performance of the proposed method.