Color Image Multi-Dimension Histogram using the Improved Dimension Reduction Algorithm

Abstract

The multi-channel histogram modeling technology can improve the above defects, namely compute multi-component histogram joint probability density function and cumulative distribution function to get the pixel grey mapping function of the color image. But computing of the joint histogram also leads to too higher computing complexity. To reduce computational complexity, this paper proposes the mask dimension reduction method. This method transforms computing of multi-dimension histogram into computing of 1D histogram, so it can greatly reduce computing complexity, improve computing speed and enhance the comparison.