This letter attempts to design a surveillance scheme by adopting an active reconfigurable intelligent surface (RIS). Different from the conventional passive RIS, the active RIS could not only adjust the phase shift but also amplify the amplitude of the reflected signal. With such reflecting, the reflected signal of active RIS could jointly adjust the signal to interference plus noise ratio (SINR) of the suspicious receiver and the legitimate monitor, hence the proactive eavesdropping at the physical layer could be effectively realized. We formulate the optimization problem with the target of maximizing the eavesdropping rate to obtain the optimal reflecting coefficient matrix of the active RIS. The formulated optimization problem is nonconvex fractional programming and challenging to deal with. We then solve the problem by approximating it as a series of convex constraints. Simulation results validate the effectiveness of our designed surveillance scheme and show that the proposed active RIS aided surveillance scheme has good performance in terms of eavesdropping rate compared with the scheme with passive RIS.