Assumptions about image continuity lead to oversmoothed edges in common ima
ge interpolation algorithms. A wavelet-based interpolation method that impo
ses no continuity constraints is introduced. The algorithm estimates the re
gularity of edges by measuring the decay of wavelet transform coefficients
across scales and preserves the underlying regularity by extrapolating a ne
w subband to be used in image resynthesis. The algorithm produces visibly s
harper edges than traditional techniques and exhibits an average peak signa
l-to-noise ratio (PSNR) improvement of 2.5 dB over bilinear and bicubic tec
hniques.