Considering that traditional damage detection needs to install contact sensors, which is not conducive for protection of ancient timber structures, a non-contact damage identification method based on digital image correlation (DIC) by using painted patterns of ancient buildings itself was proposed. Firstly, the time series images of the painted wooden beam during vibration are captured by the digital camera. Then, use the integer and sub-pixel displacement search algorithm to calculate the displacement response time history of each point on the wooden beam as well as the Fourier transform to obtain the amplitude-frequency and phase-frequency curves. Finally, the operational mode of the wooden beam is extracted by the amplitude-frequency amplitude and phase-frequency phase angle.This study then adopts the rotation mode and the curvature mode as damage characteristic indicators for damage identification by the advantage of using the DIC method to obtain modal information of all pixels along the beam length. Results from simulation and experiment of painted simply supported beam show that the operational mode obtained by the DIC method can accurately identify the damage location and the relative damage degree of the wooden beam, which provides a new technique for preventive protection of ancient timber structures.