<p>The development of DIC(Digital Image Correlation) technology has provided irreplaceable advantages for full-field vibration measurement, including the simplified setup, non-contact operation, and high spatial resolution. However, when used for modal analysis purposes, the substantial datasets of DIC expose inefficiencies for traditional modal analysis techniques that direct analyses of the complete data are almost impractical to perform. To address this issue, a novel two-step operational modal analysis method that simplifies the traditional frequency domain decomposition algorithm based on principal component analysis is proposed in this paper. Firstly, to efficiently extract frequency information, principal component analysis is applied to the raw DIC data to obtain virtual responses. Following this, the system’s order and natural frequency are determined through singular value decomposition of the power spectral density of the virtual responses at each frequency line. Subsequently, the power spectral density matrix of the full-field response is calculated near the natural frequency, and singular value decomposition is again utilized to identify the vibration mode shapes. With operational test of a cantilever beam, the feasibility and performance of the proposed method are evaluated and the first six orders of modal parameters of the beam are presented.</p>