Diffusion based multiresolution filtering algorithms for accurate abnormility detection in medical images
Shilpa Joshi, Ramesh K. Kulkarni · 2017
In present scenario as medical science is progressing, its dependency on medical imaging technology is increasing. As medical images are very complex and noisy in nature and for the treatment planning medical science is very much depend on the medical imaging technology. This gives rise to research in medical image analysis and improves quality of output. Starting late various diffusion based filtering techniques have been made, for instance, anisotropic diffusion (AD) or nonlinear diffusion (ND), which can decrease the speckle noise in Medical images while ensuring and enhancing the edges in ultrasound picture. In any case, in perspective of the granular speckle noise, it is difficult to reduce the same accurately through a specific diffusion based systems. via computerizing or encouraging the depiction of anatomical structures and different locales of intrigue. In this paper two methods of image segmentations are implemented. Based on merits and demerits of each method of the clustering and Region Growing segmentation, seed selection was done and Region of Interest (ROI) was selected. This paper proposes application of super-Resolution(SR) on these filtered and segmented medical images of different imaging modalities as some give anatomical and uncover data about the structure of the human body, and others give practical data, areas of specific activity at particular area. The results produced by proposed hybrid technique gives excellent result in terms of visuals well as statistical analysis.