Remote sensing is widely used to assess the destruction from natural disasters and to plan relief
and recovery operations. How to automatically extract useful features and segment interesting objects from
digital images, including remote sensing imagery, becomes a critical task for image understanding.
Unfortunately, the data collection of aerial digital images is constrained with bad weather, muzzy
atmosphere, and unstable camera or camcorder. As a result, remote sensing imagery is shown as lowcontrast,
blurred, and dark from time to time. Here, we introduce a new method integrating image local
statistics and image natural characteristics to enhance remote sensing imagery. This method computes the
adaptive histogram equalization to each distinct region of the input image and then redistributes the lightness
values of the image. The natural characteristic of image is applied to adjust the restoration contrast. The
experiments on real data show the effectiveness of the algorithm.