MRP

Fast Haze Removal for Nighttime Image Using Maximum Reflectance Prior, CVPR 2017.

The code has been tested on Win7/10 with Opencv 2.7.

Installation

Please install opencv 2.4.9 (or copy "opencv_core249.dll" "opencv_highgui249.dll " "opencv_imgproc249.dll" from "OPENCV_DIR/build/x64/vc10/bin/" to the same directory with "NighttimeDehaze.exe") before running this code.

Then, run the executable code as: "NighttimeDehazeMRP.exe name.bmp 0", where "name.bmp" is the input nighttime hazy image, the follwowing parameter indicates the algorithm being used, e.g., 0 for MRP and 1 for MRP-Faster, the output dehazed result is named as "name_J.bmp".

Folder Structure

MRP
    -data
        Some test images and the corresponding results generated by this code.
    -NighttimeDehazeMRP.exe
        The executable code
    -*.dll
        The dependencies
    -flickr14.bmp
        A test image

Citation

Please cite our paper in your publications if it helps your research:

@CONFERENCE{CVPR_2017_Jing,
    author = {Jing Zhang and Yang Cao and Shuai Fang and Yu Kang and Chang Wen Chen},
	title = {Fast Haze Removal for Nighttime Image Using Maximum Reflectance Prior},
	booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
	year = {2017}
}

Related Work

[1]. Nighttime haze removal based on a new imaging model, ICIP 2014. NighttimeDehaze: Project, NighttimeDehaze: github

[2]. Fully Point-wise Convolutional Neural Network for Modeling Statistical Regularities in Natural Images, ACM MM 2018. FPC-Net: Project, FPC-Net: github

[3]. FAMED-Net: A Fast and Accurate Multi-scale End-to-end Dehazing Network, T-IP, 2019. FAMED-Net: Project, FAMED-Net: github

Contact

Email