Open Access Open Access  Restricted Access Subscription Access

Automatic Image Registration using 2D-Discrete Wavelet Transform


Affiliations
1 Bharath University, Chennai - 600073, Tamil Nadu, India
2 Department of ETC, Bharath University, Chennai - 600073, Tamil Nadu, India
 

The main objective of the paper is automatic image registration using discrete wavelet transform. In this paper to register the images I use the discrete wavelet transform. The characteristic of the approach is that the image is sub divided in to four components that as Less-Less, Less-High, High-Less and High-High from that we took Less-Less for the further processing. The DWT is applied for both the reference and sensed image to extract the LL components and the feature power points are extracted using scale invariant feature transform to match the two images. The simulation results done using MATLAB.

Keywords

Discrete Wavelet Transform, Image Registration, Rotation, SIFT, Scaling, Translation
User

Abstract Views: 234

PDF Views: 0




  • Automatic Image Registration using 2D-Discrete Wavelet Transform

Abstract Views: 234  |  PDF Views: 0

Authors

M. Deepa
Bharath University, Chennai - 600073, Tamil Nadu, India
T. Saravanan
Department of ETC, Bharath University, Chennai - 600073, Tamil Nadu, India

Abstract


The main objective of the paper is automatic image registration using discrete wavelet transform. In this paper to register the images I use the discrete wavelet transform. The characteristic of the approach is that the image is sub divided in to four components that as Less-Less, Less-High, High-Less and High-High from that we took Less-Less for the further processing. The DWT is applied for both the reference and sensed image to extract the LL components and the feature power points are extracted using scale invariant feature transform to match the two images. The simulation results done using MATLAB.

Keywords


Discrete Wavelet Transform, Image Registration, Rotation, SIFT, Scaling, Translation



DOI: https://doi.org/10.17485/ijst%2F2016%2Fv9i5%2F130622