In the cluster situational awareness, local sensors inevitably have the problem of random errors, systematic errors, false alarms and missed alarms. There, target association only based on the location information of the target may lead to erroneous results. To improve the accuracy of target association in complex scenes, this paper proposes a target association calibration method based on multi-view radial velocity information. This method first constructs a distance matrix for the target position, and obtains preliminary results for target association through the Hungarian algorithm based on the global minimum distance. Due to the perpendicular lines of the radial velocity of the same target intersecting at a point in multiple perspectives, the random sampling consensus (RANSAC) algorithm is used to identify the erroneous association terms in the preliminary results and re-associate them with the correct corresponding targets. This method has been verified in simulation, and the results show that it can effectively improve the stability and accuracy of target association in dense target scenarios.