Cooperative target localization technology of aircraft cluster has been attracting great research interest due to its potentially superior target positioning accuracy. The goal is to fuse the target localization information obtained from multi-view images to produce a refined localization information. However, most previous fusion methods ignored the influence of source confidence, scene recognition difficulty and cluster position error on target positioning accuracy. To solve this problem, we propose a novel image-based cooperative target localization framework. Specifically, we use target localization and image matching technology to convert the target positions from different view into a unified coordinate system, avoiding involving the cluster’s own position error into the target positioning. Then, a multi-source information fusion algorithm based on multi-factor causal reasoning is constructed to mine the information of source confidence and scene recognition difficulty hidden in the data. In the end, high-precision cooperative target localization is achieved. Comprehensive experiments on the multi-view ship dataset are conducted to demonstrate the effectiveness of the proposed framework for cooperative target localization.