The Image Registration of TME Biomedical Images in variant Creutzfeldt-Jakob Diseases

摘要

This work was intended as an attempt to more global approach to the issue dealing with heuristic image matching (more complex images registration and landmarks arrangement). Proposed methodology is suitable for biomedical images congruencing of high-resolution images and large images size. The 2-D filter suitable to image features detection using reformulated stretched-Gabor wavelet has been presented. The additional over generated neighbouring feature points reduction algorithm has been also developed. The novel solution of generated landmarks coherences has been achieved by application of proposed feature points matching using similarity comparisons of triangles; they are build on the image feature points.