Showing papers by "Dimitris Maroulis published in 2014"

Automated Adjustment of Region-Based Active Contour Parameters Using Local Image Geometry

Abstract: A principled method for active contour (AC) parameterization remains a challenging issue in segmentation research, with a potential impact on the quality, objectivity, and robustness of the segmentation results. This paper introduces a novel framework for automated adjustment of region-based AC regularization and data fidelity parameters. Motivated by an isomorphism between the weighting factors of AC energy terms and the eigenvalues of structure tensors, we encode local geometry information by mining the orientation coherence in edge regions. In this light, the AC is repelled from regions of randomly oriented edges and guided toward structured edge regions. Experiments are performed on four state-of-the-art AC models, which are automatically adjusted and applied on benchmark datasets of natural, textured and biomedical images and two image restoration models. The experimental results demonstrate that the obtained segmentation quality is comparable to the one obtained by empirical parameter adjustment, without the cumbersome and time-consuming process of trial and error.

An effective approach for detection and segmentation of protein spots on 2-d gel images

Abstract: Two-dimensional gel image analysis is widely recognized as a particularly challenging and arduous process in proteomics field. The detection and segmentation of protein spots are two significant stages of this process as they can considerably affect the final biological conclusions of a proteomic experiment. The available techniques and commercial software packages deal with the existing challenges of 2-D gel images in a different degree of success. Furthermore, they require extensive human intervention which not only limits the throughput but unavoidably questions the objectivity and reproducibility of results. This paper introduces a novel approach for the detection and segmentation of protein spots on 2-D gel images. The proposed approach is based on 2-D image histograms as well as on 3-D spots morphology. It is automatic and capable to deal with the most common deficiencies of existing software programs and techniques in an effective manner. Experimental evaluation includes tests on several real and synthetic 2-D gel images produced by different technology setups, containing a total of ~ 21 400 spots. Furthermore, the proposed approach has been compared with two commercial software packages as well as with two state-of-the-art techniques. Results have demonstrated the effectiveness of the proposed approach and its superiority against compared software packages and techniques.

FLBP: Fuzzy Local Binary Patterns

Abstract: In this chapter, a variant of the Local Binary Patterns method that extends the ability of the method to cope with noisy texture by utilizing fuzzy modelling techniques is presented. The presented generalised Fuzzy Binary Patterns model is applied to the classic Local Binary Patterns method as well as to the Local Binary Patterns with Contrast measure method, resulting to the respective fuzzy logic based methods. Supervised classification experiments were conducted on a wide range of natural and medical texture images, degraded by different types and intensities of additive noise, in order to evaluate the efficiency of the Fuzzy Local Binary Patterns method and its fusion with other proposed methods. Fuzzy Binary Patterns based methods outperform the respective methods based on the classic Binary Patterns model for all types of images and noise, indicating the efficiency of fuzzy modelling in coping with the uncertainty introduced to texture due to noise.

Self-parameterized active contours based on regional edge structure for medical image segmentation

Abstract: This work introduces a novel framework for unsupervised parameterization of region-based active contour regularization and data fidelity terms, which is applied for medical image segmentation. The work aims to relieve MDs from the laborious, time-consuming task of empirical parameterization and bolster the objectivity of the segmentation results. The proposed framework is inspired by an observed isomorphism between the eigenvalues of structure tensors and active contour parameters. Both may act as descriptors of the orientation coherence in regions containing edges. The experimental results demonstrate that the proposed framework maintains a high segmentation quality without the need of trial-and-error parameter adjustment.

Towards Self-Parameterized Active Contours for Medical Image Segmentation with Emphasis on Abdomen

Abstract: Medical doctors are typically required to segment medical images by means of computational tools, which suffer from parameters that are empirically selected through a cumbersome and time-consuming process. This chapter presents a framework for automated parameterization of region-based active contour regularization and data fidelity terms, which aims to relieve medical doctors from this process, as well as to enhance objectivity and reproducibility. Leaned on an observed isomorphism between the eigenvalues of structure tensors and active contour parameters, the presented framework automatically adjusts active contour parameters so as to reflect the orientation coherence in edge regions by means of the “orientation entropy.” To this end, the active contour is repelled from randomly oriented edge regions and is navigated towards structured ones, accelerating contour convergence. Experiments are conducted on abdominal imaging domains, which include colon and lung images. The experimental evaluation demonstrates that the presented framework is capable of speeding up contour convergence, whereas it achieves high-quality segmentation results, albeit in an unsupervised fashion.

Erratum to “An Effective Approach for Detection and Segmentation of Protein Spots on 2-D Gel Images” [Jan 14 67-76]

Abstract: In the above-named work [ibid., vol. 18, no. 1, pp. 67-76, Jan. 2014], the sub-figures of Fig. 11 appeared in a different/incorrect order. The correct figure is presented here.