In recent years, there have been significant developments in the field of Unmanned Surface Vehicles (USVs), with a notable increase in the number of applications in both military and civilian contexts. Equipped with radar and optical systems, USVs are designed to markedly enhance detection capabilities. While radars are capable of determining target distance and bearing in all weather conditions, they are limited in their ability to classify targets. In contrast, optical systems have strong colour perception and classification abilities, with angular resolution comparable to that of lidar systems. However, their ranging capabilities are limited, and they are susceptible to instability in adverse weather conditions. The fusion of radar and optical systems, by leveraging their complementary advantages, effectively augments the detection capabilities of USVs.Radar-optical fusion methods can be broadly categorised into two primary approaches: the linkage method and the matrix transformation method. The linkage method entails the rotation of the optical servo in accordance with the azimuth and elevation angles of radar-detected targets, thus enabling the optical system to capture target images. Subsequently, the data is transmitted to the data centre, where it is analysed and decisions are made by the relevant personnel. Although this method is particularly focused on angular transformation and is effective in scenarios with fewer, unobstructed targets, it is unable to accurately detect multiple targets simultaneously. In contrast, the matrix transformation method unifies radar and optical information within a single coordinate system through the application of mathematical operations. By focusing on radar-detected points to generate Regions Of Interest (ROIs), this approach employs the Intersection over Union (IoU) algorithm for association. The matrix transformation method is predominantly applied in the autonomous driving sector. It facilitates the fusion of millimetre-wave, lidar, and onboard camera data, enabling multi-target detection and 360° capture via multiple fixed cameras for comprehensive radar association. Despite the high precision of automotive lidar systems, their detection range is typically limited to a few hundred metres. This limitation renders them unsuitable for maritime applications, where longer detection ranges are required, and errors are magnified due to the effects of sea waves and platform instability.In order to address these challenges, this article introduces a novel topological bidirectional fusion algorithm, which is aimed at improving the detection of radar and optical fusion for USVs. The YOLOv7-tiny algorithm for image detection is enhanced, and matrix transformation is employed to project radar points onto image coordinates. This study addresses the issue of vessel sway by applying the PROSAC algorithm to fit and correct the projected radar points. In order to reduce the computational demands of the system, a new coarse correlation gate has been designed which takes into account both sensor system error and position error. In order to advance the association process, topological association metrics are introduced, which serve to supplement the missing angular information that is a consequence of radar projection. This is achieved by integrating polygon angle similarity and the similarity of centreline connections, in addition to triangle similarity. Secondary detection and association are conducted on unassociated radar points in the vicinity of images and unassociated optical detection frames. This is achieved by selecting the nearest neighbouring radar points based on bearing lines.The empirical data analysis reveals a notable enhancement in the modified YOLOv7-tiny algorithm's mAP@0.5, which has increased from 0.883 to 0.93. The proposed topological bidirectional fusion algorithm achieved a remarkable accuracy rate of 92.76%, which surpassed the performance of the traditional IoU method. This enhancement provides a foundation for the potential of long-range radar and optical fusion detection for maritime Unmanned Surface Vehicles (USVs) and serves as a significant reference for research in the field of radar and optical association detection on USVs. The findings highlight the potential of advanced algorithmic approaches in addressing the challenges posed by environmental factors and platform instability, thereby paving the way for more reliable and effective situational awareness in marine environments.