Novel Biased Normalized Cuts Approach for the Automatic Segmentation of the Conjunctiva

Anemia is a common public health disease diffused worldwide. In many cases it affects the daily lives of patients needing medical assistance and continuous monitoring. Medical literature states empirical evidence of a correlation between conjunctival pallor on physical examinations and its association with anemia diagnosis. Although humans exhibit a natural expertise in pattern recognition and associative skills based on hue properties, the variance of estimates is high, requiring blood sampling even for monitoring. To design automatic systems for the objective evaluation of pallor utilizing digital images of the conjunctiva, it is necessary to obtain reliable automatic segmentation of the eyelid conjunctiva. In this study, we propose a graph partitioning segmentation approach. The semantic segmentation procedure of a diagnostically meaningful region of interest has been proposed for exploiting normalized cuts for perceptual grouping, thereby introducing a bias towards spectrophotometry features of hemoglobin. The reliability of the identification of the region of interest is demonstrated both with standard metrics and by measuring the correlation between the color of the ROI and the hemoglobin level based on 94 samples distributed in relation to age, sex and hemoglobin concentration. The region of interest automatically segmented is suitable for diagnostic procedures based on quantitative hemoglobin estimation of exposed tissues of the conjunctiva.

A Fast Color Image Segmentation Approach Using GDF with Improved Region-Level Ncut

Color image segmentation is fundamental in image processing and computer vision. A novel approach, GDF-Ncut, is proposed to segment color images by integrating generalized data field (GDF) and improved normalized cuts (Ncut). To start with, the hierarchy-grid structure is constructed in the color feature space of an image in an attempt to reduce the time complexity but preserve the quality of image segmentation. Then a fast hierarchy-grid clustering is performed under GDF potential estimation and therefore image pixels are merged into disjoint oversegmented but meaningful initial regions. Finally, these regions are presented as a weighted undirected graph, upon which Ncut algorithm merges homogenous initial regions to achieve final image segmentation. The use of the fast clustering improves the effectiveness of Ncut because regions-based graph is constructed instead of pixel-based graph. Meanwhile, during the processes of Ncut matrix computation, oversegmented regions are grouped into homogeneous parts for greatly ameliorating the intermediate problems from GDF and accordingly decreasing the sensitivity to noise. Experimental results on a variety of color images demonstrate that the proposed method significantly reduces the time complexity while partitioning image into meaningful and physically connected regions. The method is potentially beneficial to serve object extraction and pattern recognition.