Miguel Venâncio - Academia.edu (original) (raw)

Related Authors

Toru Tamaki

Jean-Francois E Hochedez

Michael D. Breitenstein

Jules Raymond Tapamo

ashraf kassim

ashraf kassim

Singapore University of Technology and Design (SUTD)

Ka-veng Yuen

Tonatiuh Tecuan

Uploads

Papers by Miguel Venâncio

Research paper thumbnail of Registration of Medical Images in 2D and 3D

Two image registration methods of different classes are evaluated and compared in this paper. Fro... more Two image registration methods of different classes are evaluated and compared in this paper. From the feature based methods, the chosen algorithm was the ICP (Iterative Closest Point) [1], which is one of the most commonly used image registration algorithms in 2D and 3D applications. From the area based methods, the studied algorithm permits the registration of images with different intensities since it simultaneously estimates the transformation parameters and the image normalization. In this paper, this method will be referred to as SRIN (Simultaneous Registration and Intensity Normalization) [2]. Since these two methods are from different nature, the SRIN method will be applied in intensity images created from the same features used by the ICP method. In this way, the SRIN method avoids the step of feature correspondence. Experimental results demonstrate that the SRIN algorithm has an excellent convergence capacity and registration precision using a variety of different images a...

Research paper thumbnail of Registration of Medical Images in 2D and 3D

Two image registration methods of different classes are evaluated and compared in this paper. Fro... more Two image registration methods of different classes are evaluated and compared in this paper. From the feature based methods, the chosen algorithm was the ICP (Iterative Closest Point) [1], which is one of the most commonly used image registration algorithms in 2D and 3D applications. From the area based methods, the studied algorithm permits the registration of images with different intensities since it simultaneously estimates the transformation parameters and the image normalization. In this paper, this method will be referred to as SRIN (Simultaneous Registration and Intensity Normalization) [2]. Since these two methods are from different nature, the SRIN method will be applied in intensity images created from the same features used by the ICP method. In this way, the SRIN method avoids the step of feature correspondence. Experimental results demonstrate that the SRIN algorithm has an excellent convergence capacity and registration precision using a variety of different images a...

Log In