To compensate different rates of image motion

a fast image algorithm was presented

and it was implemented on aerial camera system. The complicate and changeful flying attitude of aircraft will influence the working status of aerial camera which may create various kinds of image motion blur. When aircraft is steep flying

aerial camera works with a side-oblique angle

and then many image motions with different rates occur on the focal plane array simultaneously. Through analyzing how different rates of image motion blur generated in side-oblique situation

the entire blurred image was segmented into many slices by their different rates of image motion. For avoiding the ring effect introduced by 2D FFT

the slices continued to be divided into pixel lines

and then a parallel computing scheme was presented with the 1D Wiener filters being used to restore the all pixel lines. Combining all result pixel lines

the restoration image came into being. The experiment results show that the new algorithm can effectively restore the details in the side-oblique images with different rates of image motion

and the PSNR can reach 30.469; with the help of the parallel computing capability of GPU

our parallel computing algorithm restore a 2048x2048 gray blurred image in 17ms

realizing the real-time restoration.