Contrast Enhancement And Brightness Preservation Using Multi- Decomposition Histogram Equalization

TL;DR

This paper introduces Multi-Decomposition Histogram Equalization (MDHE), a novel image enhancement technique that improves contrast while preserving brightness better than existing histogram equalization methods.

Contribution

The paper proposes a new MDHE method that decomposes images into sixty-four parts, applies CHE to each, and interpolates them to enhance contrast and preserve brightness effectively.

Findings

MDHE outperforms CHE, ADHE, BHE, and RMSHE in contrast enhancement.

Quantitative metrics show improved PSNR, SNR, and reduced RMSE and MSE.

Results are supported by visual graphs and histograms.

Abstract

Histogram Equalization (HE) has been an essential addition to the Image Enhancement world. Enhancement techniques like Classical Histogram Equalization (CHE), Adaptive Histogram Equalization (ADHE), Bi-Histogram Equalization (BHE) and Recursive Mean Separate Histogram Equalization (RMSHE) methods enhance contrast, however, brightness is not well preserved with these methods, which gives an unpleasant look to the final image obtained. Thus, we introduce a novel technique Multi-Decomposition Histogram Equalization (MDHE) to eliminate the drawbacks of the earlier methods. In MDHE, we have decomposed the input sixty-four parts, applied CHE in each of the sub-images and then finally interpolated them in correct order. The final image after MDHE results in contrast enhanced and brightness preserved image compared to all other techniques mentioned above. We have calculated the various parameters like PSNR, SNR, RMSE, MSE, etc. for every technique. Our results are well supported by bar graphs, histograms and the parameter calculations at the end.