Abstract: Compact ground wave radar is an important means of monitoring and early warning of maritime targets within the exclusive economic zone. However, due to its low transmission power and low signal-to-noise ratio of target echoes, the low detection probability can easily lead to missed detection of maritime vessel targets, making track initiation with sequential methods more challenging. This paper proposes a compact ground wave radar maritime target track initiation method based on multi-frame clustering by analyzing the characteristics of clutter and vessel targets position and velocity changes over time. This method leverages the discrepancy in motion features between the target and clutter across multiple consecutive frames. By incorporating motion feature constraints into the Density-Based Spatial Clustering of Applications with Noise (DBSCAN), the method clusters the targets of each vessel in multiple frames into distinct clusters, thereby distinguishing the vessel target trace points from the clutter points. The trace points within each cluster are connected in chronological order to obtain the starting trajectory. Track initiation experiments were conducted using simulation and actual measurement data. The results show that compared with logical methods, the proposed method advances the average starting time by 5.95 minutes on average while reducing the loss rate by 12.68%. The proposed method can solve problems of delayed weak target trajectory initiation process and is suitable for initiating tracks of vessels targets in cluttered areas or distant targets.