Pregled bibliografske jedinice broj: 1236031
Unsupervised Image Dehazing Using Smooth Approximation of Dark Channel Prior
Unsupervised Image Dehazing Using Smooth Approximation of Dark Channel Prior // Proceedings of 2022 7th International Conference on Frontiers of Signal Processing (ICFSP 2022) / Blanc-Talon, Jacques ; Ghogho, Mounir ; Szczypiorski, Krzysztof (ur.).
Pariz: Institute of Electrical and Electronics Engineers (IEEE), 2022. str. 104-108 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
CROSBI ID: 1236031 Za ispravke kontaktirajte CROSBI podršku putem web obrasca
Naslov
Unsupervised Image Dehazing Using Smooth
Approximation of Dark Channel Prior
Autori
Stipetić, Vedran ; Lončrarić, Sven
Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni
Izvornik
Proceedings of 2022 7th International Conference on Frontiers of Signal Processing (ICFSP 2022)
/ Blanc-Talon, Jacques ; Ghogho, Mounir ; Szczypiorski, Krzysztof - Pariz : Institute of Electrical and Electronics Engineers (IEEE), 2022, 104-108
ISBN
978-1-6654-8157-1
Skup
7th International Conference on Frontiers of Signal Processing 2022 (ICFSP 2022)
Mjesto i datum
Pariz, Francuska, 07.09.2022. - 09.09.2022
Vrsta sudjelovanja
Predavanje
Vrsta recenzije
Međunarodna recenzija
Ključne riječi
image restoration ; image dehazing ; unsupervised learning
Sažetak
In this paper we propose a new unsupervised deep learning method for single image dehazing. The method is based on a new loss function that incorporates a smooth approximation of the famous dark channel prior. The method is used to train a neural network and results are compared to state of the art results of supervised neural networks. Evaluation is done by comparing increase in object detection on dehazed images as well as by visual inspection of results of dehazing on natural images.
Izvorni jezik
Engleski
Znanstvena područja
Računarstvo
POVEZANOST RADA
Ustanove:
Fakultet elektrotehnike i računarstva, Zagreb
