In the Window Fourier Transform (WFT) phase extraction technique, there is a contradiction between suppressing the linear phase error and background intensity interference on the window size, and it is impossible to suppress both simultaneously. Therefore, a linear phase error suppression technique based on changing the spectral components of the input signal was proposed. The Fourier space-frequency analysis method was used to filter the input signal and preserve the fundamental frequency components.This ensured that the measurement results were not affected by the background intensity. Considering that the linear phase error was also affected by the window size selection, the optimal window size was determined by simulation analysis.Finally, the parabolic shape was reconstructed by this method and compared with the surface shape obtained by the phase-shifting. The results show that compared to the traditional phase extraction technique, the proposed method achievesa comprehensive suppression effect of the linear phase error and background intensity interference.Moreover, it improves the traditional WFT phase extraction accuracy.