An Adaptive Local Fuzzy Enhancement Approach Based on Information Entropy for X-Ray Ammunition Image

For the problems of the low contrast and fuzzy flaw edge in X-ray ammunition image, an adaptive local fuzzy enhancement approach based information entropy was presented. It first dynamically normalized the ammunition image, obtained the regions of interest by gradient operator, and automatically selected the best crossover point gc with OTSU algorithm in order to get the transition μc of fuzzy enhancement operator. It then used the improved membership function to nonlinearly fuzzy process the gray value on the two sides μc to obtain the enhanced ammunition image. Experimental results demonstrated that the proposed approach could efficiently improve the contrast of the ammunition image and enhanced the flaw information.