Abstract: To solve the problem of projectile characteristic parameters identification in weapon test and evaluation with radar, an approach for characteristic parameters identification of projectile was proposed based on reconstructing and intelligent matching of radar echo time-frequency image. First, a radar echo signal model was constructed based on the projectile flight mechanism, and the inverted radar time-frequency image was obtained by performing time-frequency transformation on the echo signal generated by the model. Then, the feature envelops were extracted from the measured time-frequency image and inverted image through image digital processing technology and the root mean square of the feature envelope matching error was taken as the objective function. The global optimal solution was found efficiently through Bayesian optimization algorithm to obtain the projectile characteristic parameters. Finally, taking a certain type of terminal guided projectile as an example, the projectile characteristic parameters were identified by the proposed method. The simulation and verification results showed that multi-dimensional characteristic parameters of projectile could be identified effectively by the proposed method, and the results were in good agreement with the actual projectile characteristic parameters. The proposed method provides a new approach for identifying the projectile motion state, and it has practical application reference value.