This paper explores the efficient filter to reduce the noises in the digital images corrupted highly with the fixed-value impulse noise using fuzzy α-cut sets and median measure. The efficiency of the proposed filter is analyzed and proved that it is a high-performing fixed-value impulse noise filter qualitatively in terms of peak signal-to-noise ratio (PSNR) and mean structural similarity matrix (MSSIM) values. The human visual perception (HVP) of the filtered images is too validated the merit of the proposed method. It is also proved additionally that the proposed filter has less time complexity and assures higher degree of edge and detail preservation. © 2018, Springer Nature Switzerland AG.