Morphological Transformation based Fractal Texture Image Analysis of Multi-Model Views in MR Brain Slices

Perumal K

Abstract


This paper presents the novel feature extraction and tumor detection technique for an automatic different view of MR (Magnetic Resonance) brain image classification. This approach consists of four main phases, (i) primarily, an input brain images are pre-processed using Hierarchical Transformation Technique (HTT) to extract the dark and white details in an image. Consequently, the additional image transformations of image complement and minima transform operations are applied. In next phase, the fractal texture based image segmentation and feature extraction technique has used to extract the brain abnormality and classify the appropriate brain image patterns of normal and abnormal slices of MR brain images with the use of K th Nearest Neighbour (KNN) and ensemble bagged tree classifier. Finally, some of the post-processing is handled for tumor region extraction and tumor cells computation.

 

 


Keywords


morphological operations; fractal texture; image segmentation; Kth nearest neighbour, ensemble bagged tree



DOI: http://doi.org/10.11591/ijeecs.v21.i1.pp%25p
Total views : 19 times

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

shopify stats IJEECS visitor statistics