Structure dynamic imaging using a scanning Laser Doppler Velocimeter provides a feasible and efficient means for high-resolution rotational velocity extraction. It shifts the experimentalists’ ground from physical measurement of rotations to mathematical data manipulation and the differentiation of the experimental translational velocity. Consequently, the differentiability of velocity data becomes a major issue since such data are sequences of discrete numbers and are usually noise contaminated. This paper presents an effective method for two-dimensional data set smoothing using one-dimensional Discrete Fourier Transform-Inverse Discrete Fourier Transform (DFT-IDFT) process as a lowpass spatial filter. The angular velocities can then be defined analytically by differentiation in the spatial frequency domain. The technique is prototyped with several real experimental data sets from a steel plate excited at its resonance modes. Special attention is paid to the efforts on making 2-D data set periodic in the data frame as well as to the related error analysis. A preliminary criterion is suggested for determination of “spatial frequency truncation” in IDFT.