Image blurring artifact is the main challenge to any spatial, denoisingfilters. This artifact is contributed by the heterogeneous intensities withinthe given neighborhood or window of fixed size. Selection of most similarintensities (G-Neighbors) helps to adapt the window shape which is ofedge-aware nature and subsequently reduce this blurring artifact. The paperpresents a memristive circuit design to implement this variable pixelG-Neighbor filter. The memristive circuits exhibits parallel processingcapabilities (near real-time) and neuromorphic architectures. The proposeddesign is demonstrated as simulations of both algorithm (MATLAB) and circuit(SPICE). Circuit design is evaluated for various parameters such as processingtime, fabrication area used, and power consumption. Denoising performance isdemonstrated using image quality metrics such as peak signal-to-noise ratio(PSNR), mean square error (MSE), and structural similarity index measure(SSIM). Combining adaptive filtering method with mean filter resulted inaverage improvement of MSE to about 65\% reduction, increase of PSNR and SSIMto nearly 18\% and 12\% correspondingly.
展开▼
