In order to solve the low right track association probability of statistical association method under condition of dense target and cross track which fails to right associate to the target accurately, this paper proposes a track association algorithm on intutionistic fuzzy bi-threshold based on traditional track association using fuzzy bi-threshold through introduction of non-membership functions, and determines membership functions and non-membership functions in line with association peculiarity. Finally, the paper associates the crossing tracks observed by two radars by utilizing Monte-Carlo simulation, and compares the fuzzy bi-threshold algorithm with the traditional track association algorithm.The results show that the correct association probability is improved from 45.1% to 49.8% , and this indicstes that the algorithm is better in adapting to the association of intensive track.