A novel dehazing model for remote sensing images

作者：Singh Dilbag^*; Kumar Vijay

来源：Computers & Electrical Engineering, 2018, 69: 14-27.

DOI：10.1016/j.compeleceng.2018.05.015

• Dehazing
• Gradient profile prior
• Gradient-based guided image filter
• Modified restoration model

The visibility of remote sensing images taken in poor environmental conditions is often degraded due to the occurrence of fog, haze, mist, and smog. To restore these images, many visibility restoration techniques have been developed. However, the existing techniques perform poorly to remove the effect of haze from remote sensing images with a complex background and large haze gradient. In this paper, a gradient profile prior is proposed to remove the haze from remote sensing images. The coarse estimated atmospheric veil has been refined by using gradient-based guided image filter. Moreover, the visibility restoration model has also been modified to overcome the over saturation and color distortion problems. Extensive experiments have shown that the proposed dehazing technique outperforms the existing image dehazing techniques. The proposed technique has significantly improved the spatial and radiometric information of remote sensing images.

• 出版日期2018-7