Improving the non-extensive medical image segmentation based on Tsallis entropy

Formal Pattern Analysis & Applications (Impact Factor: 0.65). 11/2011; 14(4):369-379. DOI: 10.1007/s10044-011-0225-y
Source: DBLP


Thresholding techniques for image segmentation is one of the most popular approaches in Computational Vision systems. Recently,
M. Albuquerque has proposed a thresholding method (Albuquerque et al. in Pattern Recognit Lett 25:1059–1065, 2004) based on the Tsallis entropy, which is a generalization of the traditional Shannon entropy through the introduction of an
entropic parameter q. However, the solution may be very dependent on the q value and the development of an automatic approach to compute a suitable value for q remains also an open problem. In this paper, we propose a generalization of the Tsallis theory in order to improve the non-extensive
segmentation method. Specifically, we work out over a suitable property of Tsallis theory, named the pseudo-additive property,
which states the formalism to compute the whole entropy from two probability distributions given an unique q value. Our idea is to use the original M. Albuquerque’s algorithm to compute an initial threshold and then update the q value using the ratio of the areas observed in the image histogram for the background and foreground. The proposed technique
is less sensitive to the q value and overcomes the M. Albuquerque and k-means algorithms, as we will demonstrate for both ultrasound breast cancer images and synthetic data.

12 Reads
  • Source
    • "There are several applications of the equation (2) such as fluid transport in porous media [3], MRI relaxometry in liquids [4] and many other applications in medicine.[5], [6], [7] The properties of q-Gaussian probability distribuition, which arise from the solution of equation (2), has an interesting behavior when we study the influence of each pixel at all filtering process. We can assume that the information of the neighborhood has more or less restricted influence that depends the value of q parameter. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Noise is inherent to Diffusion-Weighted Magnetic Resonance Imaging (DWI) and noise reduction methods are necessary. Although process based on classical diffusion is one of the most used approaches for digital image, anomalous diffusion has the potential for image enhancement and it has not been tested for DWI noise reduction. This study evaluates Anomalous Diffusion (AD) filter as DWI enhancement method. The proposed method was applied to magnetic resonance diffusion weighted images (DW-MRI) with different noise levels. Results show better performance for anomalous diffusion when compared to classical diffusion approach. The proposed method has shown potential in DWI enhancement and can be an important process to improve quality in DWI for neuroimage-based diagnosis.
    Full-text · Article · Jul 2013 · Conference proceedings: ... Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference
  • [Show abstract] [Hide abstract]
    ABSTRACT: Thresholding methods based on entropy have been proposed and developed over the years. In this paper, an improved Tsallis entropy based thresholding method is proposed for segmenting the images which presenting local long-range correlation rather than global long-range correlation. The advantage of the proposed method is to distinguish the pixels' local long-range correlation by the nonextensive parameter q. And the experimental results of various infrared images as well as nondestructive test ones show the effectiveness of the proposed method.
    No preview · Article · Dec 2012 · Signal Processing
  • [Show abstract] [Hide abstract]
    ABSTRACT: Thresholding-based image segmentation algorithms are usually developed for a specific set of images because the objective of these algorithms is strongly related to their applications. The binarization of the image is generally preferred over multisegmentation, mainly because it’s simple and easy to implement. However, in this paper we demonstrate that a scene separation with three threshold levels can be more effective and closer to a manually performed segmentation. Also, we show that similar results can be achieved through a firefly-based meta-heuristic. Finally, we suggest a similarity measure that can be used for the comparison between the distances of the automatic and manual segmentation.
    No preview · Chapter · Oct 2013
Show more