Dataset and dehazing methods for non-homogeneous and dense hazy scenes

Project description

An improved method of removing haze from images

Under bad weather conditions such as haze, the quality of images degrades severely due to the presence of floating particles in the air that scatter light. Objects are therefore difficult to be identified either by human vision or computer vision systems. Existing haze removal (dehazing) techniques perform poorly when it comes to dense homogeneous hazy scenes. The EU-funded NH-DEHAZE project will develop dehazing techniques that do not assume homogeneous distribution of light and haze. It will build up the first image dataset that includes pairs of real scenes, hazy as well as clear. Deep neural networks will then be trained to derive clear images out of the hazy ones.

Fields of science

• natural sciences › computer and information sciences › artificial intelligence › computer vision
• natural sciences › computer and information sciences › artificial intelligence › computational intelligence

Programme(s)

• H2020-EU.1.3. - EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions Main Programme
• H2020-EU.1.3.2. - Nurturing excellence by means of cross-border and cross-sector mobility

Topic(s)

• MSCA-IF-2019 - Individual Fellowships

Call for proposal

H2020-MSCA-IF-2019

See other projects for this call

Funding Scheme

MSCA-IF-EF-ST - Standard EF

Coordinator

UNIVERSITE CATHOLIQUE DE LOUVAIN

Net EU contribution

€ 178 320,00

Address

Place De L Universite 1
1348 Louvain La Neuve
Belgium

Region

Région wallonne › Prov. Brabant Wallon › Arr. Nivelles

Activity type

Higher or Secondary Education Establishments

Links

Other funding

€ 178 320,00