Random Sample Consensus (RANSAC) algorithm is widely used in the field of image matching, which helps the image matching algorithm to filter and eliminate mismatched pairs and improve the accuracy of image matching.The classical RANSAC algorithm has the problems such as large amount of computation, slow operation speed, and poor sample adaptability, which cannot meet the requirements on real-time performance and sample variability in image matching.Aiming at the above problems, an improved RANSAC algorithm is proposed based on adaptive pre-test and adaptive iterative threshold,and the working principles are given.The experimental results show that the proposed algorithm has a significant improvement in computation speed and sample adaptability compared with the classical RANSAC algorithm.