Abstract: In view of the drawbacks of the traditional histogram equalization that may cause graph details loss and noise amplification, this article improves the traditional histogram equalization by dealing with the high frequency weight and low frequency weight separately, and then dissolving the noise amalgamately. As a result, image contrast is enhanced and image details are presered simultaneously, and at the same time image noise can be restricted. Experiment results reveal that this technology can greatly impr...