217 Optimized and Automatic Multiscale Contrast Enhancement

In medical x-ray imaging, low-contrast images would interfere with regular diagnosis. Thus, multiscale contrast enhancement is critical to provide users with satisfying images. However, this method has too many parameters which may not be easily controlled by users and may not produce the best image quality. Hence, a new method is proposed. Firstly, we decompose the image in a multiscale framework. Then, we get the statistical histogram in different scales. Finally, the parameters could be automatically obtained through optimization from a contrast measurement combining histograms with the intensity of enhancement. The results show that our method could optimize the enhanced image to advancement and save much time for parameters' tuning.