The Indian Space Research Organisation (ISRO) has just developed a new algorithm that is able to dramatically improve the quality of satellite images, such that haze can be removed from the images to make surface features distinctly visible.
Developed by ISRO’s Space Applications Centre (SAC) at Ahmedabad, the algorithm has applied to multi-spectral remote sensing data acquired by ISRO’s Cartosat-2 satellite, which can produce submeter resolution imagery in the Visible and Near-Infrared (VNIR) wavelengths.
The need for such an algorithm arose because satellite data was imprecise due to atmospheric absorption and scattering effects, which are caused by haze and other environmental factors. To solve this, SAC combined haze-covered satellite data with aerosol models, which evaluate particle distribution in the atmosphere, as well as pre-computed lookup tables generated with vector radiative transfer calculations. ISRO’s press release stated that this algorithm significantly removes the haze from the images, making surface features distinctly visible, and hence more useable for qualitative as well as quantitative analysis and further applications.
Such an algorithm is especially useful in areas with high pollution or haze problems, such as parts of India and much of Southeast Asia. Although satellite imagery is used very frequently in handling haze-related crises, it is often tricky to detect the exact cause of the haze, to pinpoint affected areas accurately, and to completely understand the effects of haze on the landscape and environment.