Quality-factor inspired deep neural network solver for solving inverse scattering problems
This is an incremental improvement for electromagnetic imaging applications, addressing specific bottlenecks in inverse scattering problems.
The paper tackles electromagnetic inverse scattering problems by developing a quality-factor inspired deep neural network (QuaDNN) solver that optimizes training data composition, integrates residual connections and channel attention, and designs a multi-component loss function to suppress artifacts and improve reconstruction accuracy, with numerical and experimental tests verifying its superiority.
Deep neural networks have been applied to address electromagnetic inverse scattering problems (ISPs) and shown superior imaging performances, which can be affected by the training dataset, the network architecture and the applied loss function. Here, the quality of data samples is cared and valued by the defined quality factor. Based on the quality factor, the composition of the training dataset is optimized. The network architecture is integrated with the residual connections and channel attention mechanism to improve feature extraction. A loss function that incorporates data-fitting error, physical-information constraints and the desired feature of the solution is designed and analyzed to suppress the background artifacts and improve the reconstruction accuracy. Various numerical analysis are performed to demonstrate the superiority of the proposed quality-factor inspired deep neural network (QuaDNN) solver and the imaging performance is finally verified by experimental imaging test.