Deep Learning for Efficient Microseismic Location using Source Migration-based Imaging

The uploaded .rar file contains data and network of mircoseismic location program. 'network.py' : network based on U-Net. The input is diffraction stacking images with the size of 1*64*64*64*32. The output has the same size as the input. 'Model_SeisLoca_Unet.hdf5' : The model of the network trained by 800 samples. 'ValidationSample_SNR1.mat' is corresponding to the validation sample showed in paper, and it includes 'input', 'label', and 'pred'. The 'pred' can be obtained by running 'prediction.py' with correct setups. 'TestSample_SNR0.5.mat' is corresponding to the test sample with the SNR equals 1/2 showed in paper, and it includes 'input', 'label', and 'pred'. The 'pred' can be obtained by running 'prediction.py' with correct setups. 'Draw_Synthetic_input_label_prediction.m' is a script to draw the input, label, and prediction.