Deep learning based methods have achieved remarkable success in image restoration and enhancement, but most such methods rely on RGB input images. These methods fail to take into account the rich spectral distribution of natural images. We propose a deep architecture, SpecNet, which computes spectral profile to estimate pixel-wise dynamic range adjustment of a given image. First, we employ an unpaired cycle-consistent framework to generate hyperspectral images (HSI) from low-light input images. HSI is further used to generate a normal light image of the same scene. We incorporate a self-supervision and a spectral profile regularization network to infer a plausible HSI from an RGB image. We evaluate the benefits of optimizing the spectral profile for real and fake images in low-light conditions on the LOL Dataset.

Haze removal in aerial images is a challenging problem due to considerable variation in spatial details and varying contrast. Changes in particulate matter density often lead to degradation in visibility. Therefore, several approaches utilize multi-spectral data as auxiliary information for haze removal. In this paper, we propose SkyGAN for haze removal in aerial images. SkyGAN comprises of 1) a domain-aware hazy-to-hyperspectral (H2H) module, and 2) a conditional GAN (cGAN) based multi-cue image-to-image translation module (I2I) for dehazing. The proposed H2H module reconstructs several visual bands from RGB images in an unsupervised manner, which overcomes the lack of hazy hyperspectral aerial image datasets. The module utilizes task supervision and domain adaptation in order to create a ‘hyperspectral catalyst’ for image dehazing. The I2I module uses the hyperspectral catalyst along with a 12-channel multi-cue input and performs effective image dehazing by utilizing the entire visual spectrum. In addition, this work introduces a new dataset, called Hazy Aerial-Image (HAI) dataset, that contains more than 65,000 pairs of hazy and ground truth aerial images with realistic, non-homogeneous haze of varying density. The performance of SkyGAN is evaluated on the recent SateHaze1k dataset as well as the HAI dataset. We also present a comprehensive evaluation of HAI dataset with a representative set of state-of-the-art techniques in terms of PSNR and SSIM.

Haze removal in images captured from a diverse set of scenarios is a very challenging problem. The existing dehazing methods either reconstruct the transmission map or directly estimate the dehazed image in RGB color space. In this paper, we make a first attempt to propose a Hyperspectral-guided Image Dehazing Generative Adversarial Network (HIDEGAN). The HIDEGAN architecture is formulated by designing a enhanced version of CYCLEGAN named R2HCYCLE and an enhanced conditional GAN named H2RGAN. The R2HCYCLE makes use of the hyperspectral-image (HSI) in combination with cycle-consistency and skeleton losses in order to improve the quality of information recovery by analyzing the entire spectrum. The H2RGAN estimates the clean RGB image from the hazy hyperspectral image generated by the R2HCYCLE. The models designed for spatial-spectralspatial mapping generate visually better haze-free images. To facilitate HSI generation, datasets from spectral reconstruction challenge at NTIRE 2018 and NTIRE 2020 are used. A comprehensive set of experiments were conducted on the D-Hazy,and the recent RESIDE-Standard (SOTS), RESIDE-β (OTS) and RESIDE-Standard (HSTS) datasets. The proposed HIDEGAN outperforms the existing state-ofthe-art in all these datasets.