A 2-D Inverse synthetic aperture radar (ISAR) imaging method is proposed, which only requires a short period of time for target observation. Typically, a long period of observing time is necessary to achieve high cross-range resolution in ISAR imaging since a wide observing angle is critical. However, long-term observation accompanies disadvantage in discriminating the target especially when the target has complicated motion. In this paper, we propose a 2-D ISAR imaging method with short-term observation. The down-range of scatterers is obtained by inverse fast Fourier transform (IFFT) and the cross-range of scatterers is obtained by estimation of signal parameters via rotational invariance technique (ESPRIT). To improve computational efficiency, we sort out the signal region and noise region with a certain threshold level and then the ESPRIT method is only applied to the signal region. A set of simulated and measured data of Boeing 727 as an aircraft target is used to validate the proposed imaging method in this paper. By adopting the proposed method, the required observation time of target can be reduced to 12.5%.