The alternative irregularity reduction algorithm built on 2-stage identification with AMF on FMIO

Abstract

Built on neighborhood correlation, a 2-stage identification technique was offered to incorporate on the irregularity reduction algorithm for random magnitude impulse outlier (RMIO). As a result, this algorithm has an ultimate efficacy thereby a 2-stage identification technique becomes to be one of the high efficacy identification technique. Accordingly, this paper attempts to propose an alternative irregularity reduction algorithm built on a 2-stage identification and adaptive median filter (AMF) with under fix magnitude impulsive outlier (FMIO) at little, mild and immense massiveness. First, by examining great number of depictions, the optimized window dimension for 2-stage scheme from computation and performance is disposed. Second, comprehensive examinations represent that the 2-stage identification technique is disposed to identify between regular and irregular pixels at all massiveness, especially little and mild massiveness. Third, the identification efficacy on great depictions at all massiveness is examined on regular, irregular, regular-irregular efficacy perspective to estimate the optimal window size and optimal 2-stage constant value. Finally, the overall outlier reduction efficacy of an outlier reduction built on 2-stage technique and AMF is examined on great depictions at all massiveness related with other up-to-the-minute outlier reductions. From these results, the outlier reduction has remarkable efficacy than other up-to-the-minute outlier reductions.