Open Access Open Access  Restricted Access Subscription or Fee Access

Autoencoder based Visibility Enhancement Technique for Hazy Scenes

R. Ahila Priyadharshini, S. Arivazhagan, S. Aruna

Abstract



This paper proposes a single image dehazing model built with the autoencoder network. The Outdoor images captured during the inclement weather conditions such as haze, fog and sandstorms generally exhibit visibility degradation. Due to the unclear visibility, transportation system may suffer a lot. So, an automatic method has been proposed to enhance the visibility of images even in the bad weather condition too. There are so many traditional haze removal methods available, but there is a scope to improve the visibility of the image. To improve the visibility of the images still better, deep learning techniques are preferred in this work. The proposed method enhances the visual quality of the input hazy image. Experimentation is carried out with the FRIDA database images and natural hazy images. The result of this work is superior than the state of art in terms of the performance measures such as PSNR, E metric and metric.

Keywords


Dehaze, Image Restoration, Hidden Layers, Backpropagation.

Full Text:

PDF


Disclaimer/Regarding indexing issue:

We have provided the online access of all issues and papers to the indexing agencies (as given on journal web site). It’s depend on indexing agencies when, how and what manner they can index or not. Hence, we like to inform that on the basis of earlier indexing, we can’t predict the today or future indexing policy of third party (i.e. indexing agencies) as they have right to discontinue any journal at any time without prior information to the journal. So, please neither sends any question nor expects any answer from us on the behalf of third party i.e. indexing agencies.Hence, we will not issue any certificate or letter for indexing issue. Our role is just to provide the online access to them. So we do properly this and one can visit indexing agencies website to get the authentic information.