Francesco Brandolin, Matteo Ravasi, and Tariq Alkhalifah

 GitHub: https://github.com/DeepWave-KAUST/PINNs_signal_separation-pub

Building on our previous slope assisted, physics-informed seismic interpolation framework, dubbed PINNslope, we introduce a series of enhancements that elevate the framework versatility. This ultimately enables its application to different signal separation problems, like deblending and beyond aliasing interpolation, with a specific focus on ground roll removal. With the new PINNslope framework we are able to constrain the slope estimation network in the ground roll removal problem, predicting only the clean reflections and avoiding the prediction of the ground roll.  We utilize a time derivative term in the loss function to emphasize the amplitude of the comparatively higher-frequency reflection arrivals. Ultimately, we modify the framework loss function and architecture to demonstrate the possibility of predicting two separate components of the seismic data according to the estimate of two local slopes.