In order to improve the reliability and matching precision of scene matching navigation system, a novel fusion algorithm is proposed based on Position Difference Filtering (PDF) and Least Squares Estimate (LSE) for dynamic scene matching. The essence of the matching results fusion problems for aircraft dynamic scene matching navigation system is analyzed. Then, according to the error characteristic of matching position point sequence, PDF method is designed to filtering the mismatching points, and LSE method is used for optimum parameter estimation, thus the mismatching points are estimated and the matching position are predicted. Furthermore, the principle and flowchart of the fusion algorithm are given. Simulation experimental results show the feasibility and availability of the proposed algorithm in eliminating the mismatching points and random matching disturbance for dynamic scene matching navigation system.