In Direction-Finding (DF) crossing localization， the standard deviation of DF error is influenced by the nonuniform environment of the target area.To solve the problem， we proposed a Weighted Maximum Likelihood Estimation (WMLE) algorithm.In this algorithm， the effect of the target distance was introduced into Maximum Likelihood Estimation (MLE).A weighted vector was constructed to control and compensate for the standard deviation of the DF error increasing with the target distance.Theoretical analysis showed that the algorithm of WMLE can further improve the accuracy of the multi-station DF crossing localization.