Abstract: ‍ ‍Passive radar system is a kind of passive detection system， which has the characteristics of high concealment， strong anti-interference ability and long working distance. Therefore， it has been widely used in many fields such as indoor positioning， signal reconnaissance， and electronic warfare， etc. Bearing-crossing localization is a kind of positioning technology， which is widely used in passive radar detection. However， passive radar will produce a large number of false crossing points when performing bearing-crossing localization on multiple interference sources. In order to accurately detect the position of the real interference source and ensure the tracking performance of the passive radar towards the interference source， it is necessary to perform false point removal processing. Traditional methods for removing false crossing points， such as minimum distance method， baseline clustering method， Hough transform method， etc， existed some problems， such as low positioning accuracy， high computational complexity， and deficiency in real-time processing. Aiming at above problems， In this paper， a false crossing points elimination method for passive radar cross location based on Hough transform was proposed. This method of eliminating false crossing points in passive radar cross location based on Hough transform combined the baseline clustering and Hough transform method to eliminate false crossing points. First of all， according to the angle of the interference source measured by the passive radar in a single observation period， all the cross-positioning points that needed to be processed was determined by the bearing-crossing localization technology. Then， through the baseline clustering method， all the obtained cross-positioning points was processed to eliminate some false crossing points and reduced the subsequent calculation. Finally， on the basis of baseline clustering， according to the number of interference source angles measured by passive radar， Hough transform method was further used to eliminate false crossing points to improve the positioning accuracy of the interference source. The proposed method was verified by simulation in this paper. The simulation results show that the proposed method can not only effectively modify the elimination effect of the baseline clustering method on the false crossing points of cross positioning， and improve the positioning accuracy， but also effectively reduce the computational complexity.