A partition method, with an efficient divide and conquer partition strategy, for the non-uniform sampling signal reconstruction based on Fourier reconstruction sparse inversion (FRSI) is developed. The novel partition FRSI(P-FRSI) is motivated by the observation that the partition processing of multi-dimensional signals can reduce the reconstruction difficulty and save the reconstruction time. Moreover, it is helpful to choose suitable reconstruction parameters. The P-FRSI employs divide and conquer strategy, and the signal is firstly partitioned into some blocks. Following that, traditional FRSI is applied to reconstruct signals in each block. We adopt linear or nonlinear superposition to determine the weight coefficients during integrating these blocks. Finally, P-FRSI is applied to two-dimensional seismic signal reconstruction. The superiority of the new method over conventional FRSI is demonstrated by numerical reconstruction experiments.