Abstract: There are some shortages of traditional image resolution merge algorithm based on point matching, such as huge quantity of calculation, a little low accuracy, and too many restrictions of application. In order to solve these problems, an optimized remote sensing image matching algorithm was introduced in this paper. Moreover, a comparison experiment on traditional and modified image matching algorithm was performed with an ASTER image and a SPOT image. According to the accuracy comparison of classified image, which increased from 81% to 89%, it can be concluded that the modified algorithm is superior to the traditional algorithm, it has much higher accuracy and efficiency than the latter, and it should have more adaptability and applicable value.