No abstract

No abstract

We propose a new image denoising algorithm, dubbed as Fully Convolutional Adaptive Image DEnoiser (FC-AIDE), that can learn from offline supervised training set with a fully convolutional neural network as well as adaptively fine-tune the supervised model for each given noisy image. We significantly extend the framework of the recently proposed Neural AIDE, which formulates the denoiser to be context-based pixelwise mappings and utilizes the unbiased estimator of MSE for such denoisers. The two main contributions we make are; 1) implementing a novel fully convolutional architecture that boosts the base supervised model, and 2) introducing regularization methods for the adaptive fine-tuning such that a stronger and more robust adaptivity can be attained. As a result, FC-AIDE is shown to possess many desirable features; it outperforms the recent CNNbased state-of-the-art denoisers on all of the benchmark datasets we tested, and gets particularly strong for various challenging scenarios, e.g., with mismatched image/noise characteristics or with scarce supervised training data.

In capacitive deionization, the salt-adsorption capacity of the electrode is critical for the efficient softening of brackish water. To improve the water-deionization capacity, the carbon electrode surface is modified with ion-exchange resins. Herein, we introduce the encapsulation of zwitterionic polymers over activated carbon to provide a resistant barrier that stabilizes the structure of electrode during electrochemical performance and enhances the capacitive deionization efficiency. Compared to conventional activated carbon, the surface-modified activated carbon exhibits significantly enhanced capacitive deionization, with a salt adsorption capacity of ∼2.0 × 10 mg/mL and a minimum conductivity of ∼43 μS/cm in the alkali-metal ions solution. Encapsulating the activated-carbon surface increased the number of ions adsorption sites and the surface area of the electrode, which improved the charge separation and deionization efficiency. In addition, the coating layer suppresses side reactions between the electrode and electrolyte, thus providing a stable cyclability. Our experimental findings suggest that the well-distributed coating layer leads to a synergistic effect on the enhanced electrochemical performance. In addition, density functional theory calculation reveals that a favorable binding affinity exists between the alkali-metal ion and zwitterionic polymer, which supports the preferable salt ions adsorption on the coating layer. The results provide useful information for designing more efficient capacitive-deionization electrodes that require high electrochemical stability.

No abstract