摘要
A new method based on the material removal and cross-section optical scanning is investigated.The advantage of this method is that the internal and external information of the specimen can be obtained at same precision. In order to eliminate the pulse and Gaussian noise, the multi-scale dyadic wavelet methods are presented and discussed. The experimental results show that the multi-scale dyadic wavelet methods can successfully extract the features from noise image.The accuracy of 2D edge detection is 5.4 μm with the resolution of 2.7 μm.
A new method based on the material removal and cross ― section opticalscanning is investigated. The advantage of this method is that the internal and external informationof the specimen can be obtained at same precision. In order to eliminate the pulse and Gaussiannoise, the multi ― scale dyadic wavelet methods are presented and discussed. The experimentalresults show that the multi ― scale dyadic wavelet methods can successfully extract the featuresfrom noise image. The accuracy of 2D edge detection is 5. 4 μm with the resolution of 2. 7 μm.
基金
KeyIndustrializationProjecfromNationalCommitteeofEconomicTradeofChina[1998(345)]
NaturalScienceFoundationofXianJiaotongUniversity(2001001)
