An algorithm for rapidly estimating the frequency hopping transition time based on compressive samplings is proposed, in order to estimate the frequency hopping transition time of the Frequency-Hopping signal accurately and rapidly under the condition of non-cooperation. Firstly, the frequency hopping signal is acquired by slide sampling, at a sampling rate far slow than that of the Nyquist sampling. Then the first two sparse coefficients with the maximum weight of Frequency-Hopping signal are reconstructed. Based on the relationship between the different slide time and the time of the hops before and after, the frequency hopping transition time is estimated. Simulation results show that this algorithm can estimate the hopping transition time of Frequency-Hopping signal effectively, and its speed is superior to traditional algorithm based on time-frequency analysis.