This paper presents a feature extraction method to uncover the temperature effects on bridge responses, 5 which combines mode decomposition, data reduction and blind separation. For mode decomposition, 6 empirical mode decomposition (EMD) and ensemble empirical mode decomposition (EEMD) have 7 been selected, followed by principal component analysis (PCA) for data size compression. The 8 independent component analysis (ICA) is then employed for blind separation. The unique feature of the 9 proposed method is the blind separation, which means temperature-induced response can be extracted 10 from the mixed structural responses, without any prior information of the loading conditions and 11 structural physical models. This study further evaluates the effects of extracting temperature-induced 12 response on damage detectability when using Moving Principal Component Analysis (MPCA). The 13 numerical analysis of a truss bridge is first used to evaluate the proposed method for thermal feature 14 extraction, followed by a real truss bridge test in the structural laboratory in University of Warwick. 15 Results from the numerical case study show that the method enables the separation of temperature-16 induced response, and furthermore, the EEMD, in mode decomposition, has a positive influence on the 17 blind separation than EMD, when combined with PCA and ICA. Finally, the real truss bridge test 18 demonstrates that the feature extraction method can enhance the probability of MPCA to uncover the 19 damage, as the MPCA fails without proposed method. 20