Nonlinear Multi-Scale Contrast Enhancement for X-Ray Chest Radiography

An algorithm for multi-scale nonlinear X-ray chest radiography contrast enhancement based on edge(MSCE) is presented to overcome the drawback of introducing visible artifacts in conventional methods.Images are decomposed within the multi-scale decomposition architecture,and then local contrast of images is obtained based on edge information.The enhancement coefficient is calculated from image's local contrast information,and the enhanced image is then obtained.The MSCE algorithm can provide a smooth enhancement of large structures to avoid these visible artifacts,and to maintain the same detail enhancement ability.Experimental results show that the average local variance in the edge region(ALVE) decreases about 36.78% and 29.64% with almost the same detail enhancement ability comparing with the Vuylsteke and Stahl's methods,and that the proposed method can achieve better perceptual quality.