Generally, a cluster of seismic events which share similar source locations and focal mechanisms will show similar waveforms on the record. Based on this assumption, a method have been developed for microseismic event detection and arrival picking based on waveform cross-correlation. This method achieves moveout correction for the seismic records based on cross-correlation functions, then calculates a multi-channel semblance coefficient to identify the microseismic events. Meanwhile, the seismic records after moveout correction are superposed. The STA/LTA method is adopt to pick the arrivals for the stacked traces, the arrival times of the microseismic events are then automatically obtained. The performance of the method is evaluated using both synthetic and real datasets. Analysis of the results demonstrates that the proposed method can not only detect the microseismic events, but also obtain relatively accurate arrival picks at the same time.