Computationally simple Super-Resolution algorithm for video from endoscopic capsule

Super-Resolution algorithms are typically formulated as optimization problems with significant computational burden which disables real-time applications. We propose a computationally simple SR algorithm for increasing spatial resolution of images. The algorithm was designed for autonomous endoscopic capsule with objective of real time application. The significant computational speed-up was achieved by exchanging optimization stage present in typical SR algorithm by weighted averaging of upsampled and matched low resolution images. Conducted tests verified superior performance of our algorithm over standard interpolation methods.