Abstract: Marine radar is a primary means for coastal vessel traffic management systems and ships underway to monitor and perceive the surrounding environment and its dynamic changes in real time. To address the problems of limited scene diversity, insufficient target types and quantities, and incomplete ground-truth annotations in existing open marine radar data, this paper proposes a marine radar ship-target imaging method based on multi-scattering-center echo convolution. Key factors affecting marine radar imaging, including ship structural characteristics, radar antenna directivity, and receiver signal processing, are comprehensively considered. First, the ship is modeled as multiple independent scattering centers according to its structural characteristics. Then, the radar image generated by the reflected echo from a single scattering center is calculated based on the radar equation and marine radar scanning principles. Finally, the radar video image is synthesized using the multi-scattering-center convolution method. In addition, the proposed echo synthesis model can incorporate the self-occlusion effect caused by ship structures by using the height information of each scattering center. Using measured data from a disc-antenna marine radar mounted on a small unmanned surface vehicle as a reference, typical anchored and underway ships are selected to validate the accuracy of the simulated data. The results show that the proposed multi-scattering-center convolution-based echo synthesis method can reasonably generate radar video image data for various complex ship targets under different scenarios.