Explicit motion estimation is considered a major factor in the performance of classical motion-based super\r\nresolution (SR) algorithms. To reconstruct video frames sequentially, we applied a dynamic SR algorithm based on\r\nthe Kalman recursive estimator. Our approach includes a novel measurement validation process to attain robust\r\nimage reconstruction results under inexplicit motion estimation. In our method, the suitability for high-resolution\r\npixel estimation is determined by the accuracy of motion estimation. We measured the accuracy of the image\r\nregistration result using the Mahalanobis distance between the input low-resolution frame and the motion\r\ncompensated high-resolution estimation. We also incorporate an effective scene change detection method\r\ndedicated to the proposed SR approach for minimizing erroneous results when abrupt scene changes occur in the\r\nvideo frames. According to the ratio of well-aligned pixels (i.e., motion is compensated accurately) to the total\r\nnumber of pixels, we are able to detect sudden changes of scene and context in the input video. Representative\r\nexperiments on synthetic and real video data show robust performance of the proposed algorithm in terms of its\r\nreconstruction quality even with errors in the estimated motion.
Loading....