This paper proposes a parameter identification method for single-degree-of-freedom nonlinear systems based on backbone and envelope curves. The method uses the system's backbone and envelope curves as observational data, combining them with their analytical expressions to obtain the posterior joint distribution of physical parameters through Bayesian estimation. Subsequently, the Markov Chain Monte Carlo method is employed to derive the marginal probability distributions of each physical parameter. The proposed approach avoids the need for complex time-domain numerical integration, significantly improving computational efficiency. Furthermore, the effect of noise is considered, and the results of the Hilbert transform and the zero-crossing method for estimating the backbone and envelope curves are compared. To validate the accuracy of the proposed method, it is applied to identify the Duffing oscillator with a discussion of the impact of noise at different levels. The results demonstrate that the proposed method is accurate in identifying the physical parameters of single-degree-of-freedom nonlinear systems under noisy environments.