Mcnn : Conditional Focus Probability Learning to Multi-Focus Images Via Mutually Coupled Neural Network

In this paper, a novel conditional focus probability learning model, termed MCNN, is proposed for multi-focus image fusion (MFIF). Given a pair of source images, their conditional focus probabilities can be generated by using our MCNN trained well, which is further converted into the binary focus masks to directly produce an all-focus image with no post-processing. To this end, we design a fully convolutional encoder with two mutually coupled Siamese branches in MCNN, where a coupling block is designed as a bridge between the two branches to provide conditional information to each other, at different layers. Such that the encoder can more strongly extract conditional focus features and further encourage the decoder pixel-wisely to give more robust conditional focus probabilities. Moreover, we design a hybrid loss with a structural sparse fidelity loss and a structural similarity loss to force our network to learn more accurate conditional focus probabilities. Particularly, we propose a convolutional norm with good structural group sparse, to construct the structural sparse fidelity loss. Simulation results substantiate the superiority of our MCNN over other state-of-the-arts, in terms of both visual perception and quantitative evaluation