Multi-Scale Symmetry Transform with Application to Location of Feature Points on Human Face Image
PENG Jin-ye1,2, YU Bian-zhang1, WANG Da-kai2, LI Nan2
1. Department of Electronics,Northwestern Polytechnic University,Xi’an,Shaanxi 710068,China;2. Department of Electronics,Northwest University,Xi’an,Shaanxi 710069,China
Abstract：A method to implement multi-scale symmetry transform in the wavelet coefficients domain decomposed by using anti-symmetrical biorthogonal wavelets (ASBW) is presented with application to location of feature points on human face image.Compared to the original symmetry transform,which is performed on a single resolution level,our method can yield results with higher location accuracy without increment of computational burden.In addition,if an image is compressed based on ASBW decomposition,then our method provides a way to extract image features in compressed data domain without need of completely reconstructing the image.This is one of the most important technical consideration in the content-based indexing (CBI) problem.
彭进业;俞卞章;王大凯;李 楠. 多尺度对称变换及其应用于定位人脸特征点[J]. 电子学报, 2002, 30(3): 363-366.
PENG Jin-ye;YU Bian-zhang;WANG Da-kai;LI Nan. Multi-Scale Symmetry Transform with Application to Location of Feature Points on Human Face Image. Chinese Journal of Electronics, 2002, 30(3): 363-366.
2002, Vol. 30 
