Image Improvement by Using Noise Suppressing Unsharp Masking and Evaluation without Reference

The technique of Unsharp Masking(UM) are often applied for improving the image quality. However, for images with added noises, UM causes the amplification of noise, which often makes the method not usable in practice. In this paper, we propose a new UM method with suppressing the noise amplification. The proposed method reduces noise amplification and permits to obtain perceptually pleasant results. Furthermore we show a method for evaluating the improvement after unsharp masking processes. The new UM method has parameters to be set in advance. The evaluation method can be used to set the parameters for obtaining well improved results.