https://github.com/mshr-h/image_dehazing

Haze removal using Dark channel prior
https://github.com/mshr-h/image_dehazing

computer-vision haze-removal

Last synced: 9 months ago
JSON representation

Haze removal using Dark channel prior

• Host: GitHub
• URL: https://github.com/mshr-h/image_dehazing
• Owner: mshr-h
• License: mit
• Archived: true
• Created: 2015-12-13T08:42:35.000Z (over 10 years ago)
• Default Branch: master
• Last Pushed: 2016-01-23T03:20:14.000Z (over 10 years ago)
• Last Synced: 2025-01-31T08:34:36.988Z (over 1 year ago)
• Topics: computer-vision, haze-removal
• Language: C
• Size: 822 KB
• Stars: 6
• Watchers: 1
• Forks: 5
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Haze removal using Dark channel prior [![travis-ci](https://travis-ci.org/mshr-h/Image_Dehazing.svg?branch=master)](https://travis-ci.org/mshr-h/Image_Dehazing)

## Description

This is an implementation of dehaze(haze removal) algorithm using C/C++ on Qt platform.

The algorithm is [written in C](lib_c/Dehaze.c).

## What's the Dehaze

Dehaze is a technique to improve the visibility of the outdoor images.

## Example

|         hazy image         |              output             |

| :------------------------: | :-----------------------------: |

| ![](inImg_db/Fattal_2.png) | ![](outImg_db/Fattal_2_out.png) |

- window size=5x5 pixels

- ω=0.95

- t_0=0.01

- p=0.1

## Requirement

- Qt 5.3 or newer (probably works on older than 5.3 but not tested)

## Directory tree

```

include/      header files

inImg_db/     example input images

lib_c/        C code

lib_qc/       C structure <-> C++ data conversion

outImg_db/    example output images

```

## Reference

[Kaiming He, Jian Sun, Xiaoou Tang, "Single image haze removal using dark channel prior", CVPR, 2009, 2013 IEEE Conference on Computer Vision and Pattern Recognition, 2013 IEEE Conference on Computer Vision and Pattern Recognition 2009, pp. 1956-1963, doi:10.1109/CVPRW.2009.5206515](http://research.microsoft.com/en-us/um/people/jiansun/papers/Dehaze_CVPR2009.pdf)