Hyperspectral image dehazing in the visible and infrared ranges for Enhanced Flight Vision Systems

Under haze or fog, the quality of the images is degraded due to the atmosphere, causing the details of the image to be difficult to identify by both an observer and computer vision systems. The techniques to eliminate this degradation, caused by the scattering and absorption of light, named as dehazing techniques, are crucial in various current applications with great scientific and technological interest: air and maritime transport, surveillance, driver assistance systems, remote sensing.

Dehazing is a multidisciplinary challenge, since it requires the knowledge of several fields: meteorology, physical models to predict how light is affected by the atmosphere, and computer vision & image processing to recover the parameters of the scene. The methods of dehazing that provide better results are those based on a physical model, whose objective is to reverse the atmospheric degradation. For most scientific and technological applications of dehazing it is essential that the methods work automatically, in real time, without the interaction of the user and without knowing atmospheric parameters.

We proposed in 2015 a method based on the dichromatic model of atmospheric scattering and on the constancy of the response ratio of a camera's RGB sensors. This method only requires a previous segmentation of the image to group the pixels located at the same distance.

This research project focuses on the use of multispectral and hyperspectral images in the visible and infrared range in a dehazing method based on a physical model.

https://euraxess.ec.europa.eu/jobs/hosting/msca-if-joint-application-university-granada-department-optics?fbclid=IwAR16a9nINZEWAlYWUd9P6Jy880inTuj_OLRBkRLfxh7WfV_EVmzeToVwPCg