Hydraulic fracturing is widely used in coal mines, and microseismic is usually used to detectthe fracturing effect. In order to accurately identify weak microseismic waveform and lay a foundationfor subsequent positioning, inversion and other waveform processing, the method of combining deeplearning CNN( convolutional neural network) with image is adopted to analyze the effect of the time do-main CNN model, the wavelet CNN model, AIC and STA/ LTA in identifying weak microseismic data of hydraulic fracturing in coal seam drilling. The results show that the training and testing accuracy of thetime-domain and wavelet convolution neural network models are above 99% , and the loss function isbelow 0. 02. In the detection of one hour continuous microseismic data, the identification accuracy oftime domain model and wavelet model for microseismic events reaches 100% and 84% respectively.The recall rate reaches 68% and 57% respectively, which is higher than AIC and STA/ LTA in identif-ying microseismic events with accuracy of 66% and 40% , and the recall rate is 42% and 25% . In therecognition of weak real events in coal mines, time domain CNN and wavelet CNN model are superior toAIC and STA/ LTA methods, and time domain models to wavelet models. The results indicate that thedeep learning CNN model has a stronger weak waveform recognition ability and generalization abilitythan AIC and STA/ LTA, a better microseismic detection method.