Abstract: To provide multi-target's real-time bearing time history data for target motion analysis and automatic military decision, extraction of multi-target's bearing time tracks on real time is investigated. Taking advantage of the time space accumulation effect of passive sonar's bearing time history display, a method is f roposed to extract multiple targets' bearing time tracks on real time. In order to extract multiple tracks from passive sonar's dynamic bearing time history display, the method takes four steps related with digital image processing: noise reduction, distinguished bearing point recognition, bearing time tracks extension and non-target tracks removal. Tests with data from sea trials and simulation experiments show that not only can the method improve the signal-to-noise ratio, but also solve the data association problem effectively when multiple bearing time tracks intersect. The correct extraction ratio of multi-target's bearing time tracks using the method is satisfactory. The method is also efficient, and can be applied in practice.