Affiliations: School of Information and Computer Technology,
Sirindhorn International Institute of Technology, Thammasat University, 131 Moo
5, Tiwanont Road, Bangkadi Campus, Pathum Thani 12000, Thailand. E-mail:
{sittisak,makhanov}@siit.tu.ac.th
Note: [] Corresponding author
Abstract: We propose new rotation moment invariants based on multiresolution
filter bank techniques. The multiresolution pyramid motivates our simple but
efficient feature selection procedure based on fuzzy C-mean clustering,
combined with the Mahalanobis distance. The procedure verifies an impact of
random noise as well as an interesting and less known impact of noise due to
spatial transformations. The recognition accuracy of the proposed techniques
has been tested with the Zernike moments, the Fourier-Mellin moments as well as
with some wavelet based schemes. The numerical experiments, with more than
30,000 images, demonstrate a tangible accuracy increase of about 3% for low
noise, 8% for the average noise and 15% for high-level noise.
