Title :
A fast variance-ordered domain block search algorithm for fractal encoding
Author_Institution :
Agency for Defense Dev., Daejeon
fDate :
5/1/1999 12:00:00 AM
Abstract :
A new fast fractal encoding algorithm based on the variances of image blocks is proposed. With the domain blocks sorted according to their variances, and the best matched domain block to a given range block is searched in the order that the variance-distance is closer. A great number of domain blocks could be safely rejected by the prior comparison of the current minimum distortion and variance difference between the candidate domain block and the range block during the search process. It was proved that our algorithm produces completely identical fractal codes with that of the conventional full search in reduced time. The simulation results confirmed the effectiveness of the proposed algorithm
Keywords :
fractals; image coding; image matching; search problems; fast fractal encoding algorithm; fractal codes; fractal image coding; full search; image blocks; matched domain block; minimum distortion; range block; simulation results; variance difference; variance-distance; variance-ordered domain block search algorithm; Approximation algorithms; Computational modeling; Decoding; Encoding; Euclidean distance; Fractals; Image coding; Image quality; Image reconstruction; Statistics;
Journal_Title :
Consumer Electronics, IEEE Transactions on
