Improvement of Histogram Equalization for Minimum Mean Brightness Error

摘要

This Research presents the image enhancing using a mean-separate histogram equalization method. It can preserve the original brightness to a certain extends. To provide maximum brightness preservation, it separates the input image's histogram into 2{sup}n based on input mean before equalizing them independently. The image initially is separated from 2, 4 and 8 class by calculated threshold level and each class is histogram equalized to entire image, and gets lowest AMBE (AMBE: the absolute difference between input and output mean). The result found that AMBE gradually reduces when the separation is increased.