nttcslab / rope

Code of the paper "Reflectance-oriented probabilistic equalization for image enhancement" published in ICASSP 2021.

ROPE

Code of the paper "Reflectance-oriented probabilistic equalization for image enhancement" published in ICASSP 2021.

• In this paper, we proposed a novel 2D histogram equalization approach known as reflectance-oriented probabilistic equalization (ROPE), which allows for adaptive regulation of global brightness.

• Please read LICENSE.md for license details.

• The original code was tested using MATLAB2020b on a machine running CentOS 7, but should also work on other OSs.

• Requirements

  • MATLAB
  • Image Processing Toolbox (Required by LIME)

• run demo.m for the demo

• LIME.p is provided by Guo et al. , which is used as an edge preserving filter for illumination estimation.

• Test images are provided by USC-SIPI and by Guo et al.

• Technical questions should be directed to the contact author, Xiaomeng Wu, using the following email address: