Abstract: Aiming at the problem that there may be stitching seams, and occupy too much memory in unmanned aerial vehicles (UVA) image stitching, an aerial image stitching fusion algorithm based on brightness correction was proposed. Firstly, the aerial images were registered and preprocessed by brightness correction. Then, the best suture line was obtained by searching the images, and the image overlap area was divided into transition areas, which can reduce the Laplacian pyramid area. Finally, images were stitched in the different areas with weighted fusion algorithm and Laplacian pyramid fusion algorithm, respectively. Experiments show that the proposed algorithm can effectively improve the exposure difference, and at the same time the memory occupied during stitching can be reduced obviously.